deepbox 0.1.0

package/dist/chunk-AD436M45.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../src/preprocess/index.ts","../src/preprocess/_internal.ts","../src/preprocess/encoders.ts","../src/preprocess/scalers.ts","../src/preprocess/split.ts"],"names":["value","out","row","nClasses","classes","result","fittedClasses","getNumericData","q","base"],"mappings":";;;;AAAA,IAAA,kBAAA,GAAA;AAAA,QAAA,CAAA,kBAAA,EAAA;AAAA,EAAA,UAAA,EAAA,MAAA,UAAA;AAAA,EAAA,KAAA,EAAA,MAAA,KAAA;AAAA,EAAA,cAAA,EAAA,MAAA,cAAA;AAAA,EAAA,YAAA,EAAA,MAAA,YAAA;AAAA,EAAA,WAAA,EAAA,MAAA,WAAA;AAAA,EAAA,SAAA,EAAA,MAAA,SAAA;AAAA,EAAA,YAAA,EAAA,MAAA,YAAA;AAAA,EAAA,YAAA,EAAA,MAAA,YAAA;AAAA,EAAA,mBAAA,EAAA,MAAA,mBAAA;AAAA,EAAA,UAAA,EAAA,MAAA,UAAA;AAAA,EAAA,aAAA,EAAA,MAAA,aAAA;AAAA,EAAA,cAAA,EAAA,MAAA,cAAA;AAAA,EAAA,gBAAA,EAAA,MAAA,gBAAA;AAAA,EAAA,mBAAA,EAAA,MAAA,mBAAA;AAAA,EAAA,YAAA,EAAA,MAAA,YAAA;AAAA,EAAA,cAAA,EAAA,MAAA,cAAA;AAAA,EAAA,eAAA,EAAA,MAAA,eAAA;AAAA,EAAA,cAAA,EAAA,MAAA;AAAA,CAAA,CAAA;;;ACiBO,SAAS,mBAAA,CAAoB,GAAW,IAAA,EAAoB;AACjE,EAAA,IAAI,CAAA,CAAE,UAAU,QAAA,EAAU;AACxB,IAAA,MAAM,IAAI,UAAA,CAAW,CAAA,EAAG,IAAI,CAAA,gBAAA,CAAkB,CAAA;AAAA,EAChD;AACF;AAOO,SAAS,QAAA,CAAS,GAAW,IAAA,EAAoB;AACtD,EAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,IAAA,MAAM,IAAI,UAAA,CAAW,CAAA,EAAG,IAAI,CAAA,0BAAA,EAA6B,CAAA,CAAE,IAAI,CAAA,CAAA,CAAG,CAAA;AAAA,EACpE;AACF;AAOO,SAAS,WAAW,CAAA,EAA6B;AACtD,EAAA,IAAI,CAAA,CAAE,IAAA,KAAS,CAAA,IAAK,CAAA,CAAE,KAAA,CAAM,CAAC,CAAA,KAAM,MAAA,IAAa,CAAA,CAAE,KAAA,CAAM,CAAC,CAAA,KAAM,MAAA,EAAW;AACxE,IAAA,MAAM,IAAI,WAAW,CAAA,gDAAA,EAAmD,CAAA,CAAE,MAAM,IAAA,CAAK,IAAI,CAAC,CAAA,CAAA,CAAG,CAAA;AAAA,EAC/F;AACA,EAAA,OAAO,CAAC,EAAE,KAAA,CAAM,CAAC,GAAG,CAAA,CAAE,KAAA,CAAM,CAAC,CAAC,CAAA;AAChC;AAOO,SAAS,YAAY,CAAA,EAAmB;AAC7C,EAAA,MAAM,MAAA,GAAS,CAAA,CAAE,OAAA,CAAQ,CAAC,CAAA;AAC1B,EAAA,IAAI,WAAW,MAAA,EAAW;AACxB,IAAA,MAAM,IAAI,aAAa,8CAA8C,CAAA;AAAA,EACvE;AACA,EAAA,OAAO,MAAA;AACT;AAOO,SAAS,aAAa,CAAA,EAA6B;AACxD,EAAA,MAAM,OAAA,GAAU,CAAA,CAAE,OAAA,CAAQ,CAAC,CAAA;AAC3B,EAAA,MAAM,OAAA,GAAU,CAAA,CAAE,OAAA,CAAQ,CAAC,CAAA;AAC3B,EAAA,IAAI,OAAA,KAAY,MAAA,IAAa,OAAA,KAAY,MAAA,EAAW;AAClD,IAAA,MAAM,IAAI,aAAa,+CAA+C,CAAA;AAAA,EACxE;AACA,EAAA,OAAO,CAAC,SAAS,OAAO,CAAA;AAC1B;AAWO,SAAS,mBAAmB,IAAA,EAA4B;AAC7D,EAAA,MAAM,CAAA,GAAI,UAAA;AACV,EAAA,MAAM,CAAA,GAAI,KAAA;AACV,EAAA,MAAM,IAAI,CAAA,IAAK,EAAA;AAEf,EAAA,IACE,CAAC,MAAA,CAAO,QAAA,CAAS,IAAI,CAAA,IACrB,CAAC,MAAA,CAAO,SAAA,CAAU,IAAI,CAAA,IACtB,CAAC,MAAA,CAAO,aAAA,CAAc,IAAI,CAAA,IAC1B,OAAO,CAAA,EACP;AACA,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,iDAAA;AAAA,MACA,aAAA;AAAA,MACA;AAAA,KACF;AAAA,EACF;AAEA,EAAA,IAAI,QAAQ,IAAA,GAAO,CAAA;AAEnB,EAAA,OAAO,MAAM;AACX,IAAA,KAAA,GAAA,CAAS,CAAA,GAAI,QAAQ,CAAA,IAAK,CAAA;AAC1B,IAAA,OAAO,KAAA,GAAQ,CAAA;AAAA,EACjB,CAAA;AACF;AAUO,SAAS,qBAAA,CAAsB,SAAmB,MAAA,EAA4B;AACnF,EAAA,KAAA,IAAS,IAAI,OAAA,CAAQ,MAAA,GAAS,CAAA,EAAG,CAAA,GAAI,GAAG,CAAA,EAAA,EAAK;AAC3C,IAAA,MAAM,IAAI,IAAA,CAAK,KAAA,CAAM,MAAA,EAAO,IAAK,IAAI,CAAA,CAAE,CAAA;AACvC,IAAA,MAAM,IAAA,GAAO,QAAQ,CAAC,CAAA;AACtB,IAAA,IAAI,SAAS,MAAA,EAAW;AACtB,MAAA,MAAM,IAAI,aAAa,8CAA8C,CAAA;AAAA,IACvE;AACA,IAAA,MAAM,IAAA,GAAO,QAAQ,CAAC,CAAA;AACtB,IAAA,IAAI,SAAS,MAAA,EAAW;AACtB,MAAA,MAAM,IAAI,aAAa,8CAA8C,CAAA;AAAA,IACvE;AACA,IAAA,OAAA,CAAQ,CAAC,CAAA,GAAI,IAAA;AACb,IAAA,OAAA,CAAQ,CAAC,CAAA,GAAI,IAAA;AAAA,EACf;AACF;;;AC3GA,SAAS,cAAc,CAAA,EAAqB;AAC1C,EAAA,IAAI,CAAA,CAAE,UAAU,QAAA,EAAU;AACxB,IAAA,MAAM,IAAI,WAAW,wBAAwB,CAAA;AAAA,EAC/C;AACA,EAAA,IAAI,CAAC,KAAA,CAAM,OAAA,CAAQ,CAAA,CAAE,IAAI,CAAA,EAAG;AAC1B,IAAA,MAAM,IAAI,aAAa,+CAA+C,CAAA;AAAA,EACxE;AACA,EAAA,OAAO,CAAA,CAAE,IAAA;AACX;AAEA,SAAS,eAAe,CAAA,EAAuC;AAC7D,EAAA,IAAI,CAAA,CAAE,UAAU,QAAA,EAAU;AACxB,IAAA,MAAM,IAAI,WAAW,yBAAyB,CAAA;AAAA,EAChD;AACA,EAAA,IAAI,KAAA,CAAM,OAAA,CAAQ,CAAA,CAAE,IAAI,CAAA,EAAG;AACzB,IAAA,MAAM,IAAI,aAAa,gDAAgD,CAAA;AAAA,EACzE;AACA,EAAA,OAAO,CAAA,CAAE,IAAA;AACX;AAEA,SAAS,iBAAA,CAAkB,QAAoB,SAAA,EAAiC;AAC9E,EAAA,IAAI,SAAA,GAAY,KAAA;AAChB,EAAA,IAAI,SAAA,GAAY,KAAA;AAChB,EAAA,IAAI,SAAA,GAAY,KAAA;AAEhB,EAAA,KAAA,MAAW,SAAS,MAAA,EAAQ;AAC1B,IAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,MAAA,SAAA,GAAY,IAAA;AAAA,IACd,CAAA,MAAA,IAAW,OAAO,KAAA,KAAU,QAAA,EAAU;AACpC,MAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,KAAK,CAAA,EAAG;AAC3B,QAAA,MAAM,IAAI,qBAAA,CAAsB,wCAAA,EAA0C,SAAA,EAAW,KAAK,CAAA;AAAA,MAC5F;AACA,MAAA,SAAA,GAAY,IAAA;AAAA,IACd,CAAA,MAAA,IAAW,OAAO,KAAA,KAAU,QAAA,EAAU;AACpC,MAAA,SAAA,GAAY,IAAA;AAAA,IACd;AAAA,EACF;AAEA,EAAA,MAAM,SAAA,GAAA,CAAa,YAAY,CAAA,GAAI,CAAA,KAAM,YAAY,CAAA,GAAI,CAAA,CAAA,IAAM,YAAY,CAAA,GAAI,CAAA,CAAA;AAE/E,EAAA,IAAI,cAAc,CAAA,EAAG;AACnB,IAAA,OAAO,QAAA;AAAA,EACT;AACA,EAAA,IAAI,YAAY,CAAA,EAAG;AACjB,IAAA,MAAM,IAAI,qBAAA,CAAsB,wCAAA,EAA0C,SAAS,CAAA;AAAA,EACrF;AACA,EAAA,IAAI,WAAW,OAAO,QAAA;AACtB,EAAA,IAAI,WAAW,OAAO,QAAA;AACtB,EAAA,OAAO,QAAA;AACT;AAEA,SAAS,cAAA,CAAe,QAA4B,SAAA,EAA+B;AACjF,EAAA,MAAM,GAAA,GAAM,KAAA,CAAM,IAAA,CAAK,MAAM,CAAA;AAC7B,EAAA,IAAI,GAAA,CAAI,MAAA,KAAW,CAAA,EAAG,OAAO,GAAA;AAE7B,EAAA,MAAM,YAAA,GAAe,iBAAA,CAAkB,GAAA,EAAK,SAAS,CAAA;AAErD,EAAA,IAAI,iBAAiB,QAAA,EAAU;AAC7B,IAAA,GAAA,CAAI,IAAA,CAAK,CAAC,CAAA,EAAG,CAAA,KAAM;AACjB,MAAA,IAAI,OAAO,CAAA,KAAM,QAAA,IAAY,OAAO,MAAM,QAAA,EAAU;AAClD,QAAA,MAAM,IAAI,aAAa,6CAA6C,CAAA;AAAA,MACtE;AACA,MAAA,OAAO,CAAA,CAAE,cAAc,CAAC,CAAA;AAAA,IAC1B,CAAC,CAAA;AACD,IAAA,OAAO,GAAA;AAAA,EACT;AAEA,EAAA,IAAI,iBAAiB,QAAA,EAAU;AAC7B,IAAA,GAAA,CAAI,IAAA,CAAK,CAAC,CAAA,EAAG,CAAA,KAAM;AACjB,MAAA,IAAI,OAAO,CAAA,KAAM,QAAA,IAAY,OAAO,MAAM,QAAA,EAAU;AAClD,QAAA,MAAM,IAAI,aAAa,6CAA6C,CAAA;AAAA,MACtE;AACA,MAAA,IAAI,CAAA,GAAI,GAAG,OAAO,EAAA;AAClB,MAAA,IAAI,CAAA,GAAI,GAAG,OAAO,CAAA;AAClB,MAAA,OAAO,CAAA;AAAA,IACT,CAAC,CAAA;AACD,IAAA,OAAO,GAAA;AAAA,EACT;AAEA,EAAA,GAAA,CAAI,IAAA,CAAK,CAAC,CAAA,EAAG,CAAA,KAAM;AACjB,IAAA,IAAI,OAAO,CAAA,KAAM,QAAA,IAAY,OAAO,MAAM,QAAA,EAAU;AAClD,MAAA,MAAM,IAAI,aAAa,6CAA6C,CAAA;AAAA,IACtE;AACA,IAAA,OAAO,CAAA,GAAI,CAAA;AAAA,EACb,CAAC,CAAA;AACD,EAAA,OAAO,GAAA;AACT;AAEA,SAAS,sBAAA,CAAuB,QAAiC,SAAA,EAA+B;AAC9F,EAAA,IAAI,MAAA,CAAO,WAAW,CAAA,EAAG;AACvB,IAAA,MAAM,IAAI,qBAAA,CAAsB,4CAAA,EAA8C,SAAS,CAAA;AAAA,EACzF;AACA,EAAA,MAAM,GAAA,GAAM,KAAA,CAAM,IAAA,CAAK,MAAM,CAAA;AAC7B,EAAA,iBAAA,CAAkB,KAAK,SAAS,CAAA;AAEhC,EAAA,MAAM,IAAA,uBAAW,GAAA,EAAc;AAC/B,EAAA,KAAA,MAAW,SAAS,GAAA,EAAK;AACvB,IAAA,IAAI,IAAA,CAAK,GAAA,CAAI,KAAK,CAAA,EAAG;AACnB,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,CAAA,2CAAA,EAA8C,MAAA,CAAO,KAAK,CAAC,CAAA,CAAA;AAAA,QAC3D,SAAA;AAAA,QACA;AAAA,OACF;AAAA,IACF;AACA,IAAA,IAAA,CAAK,IAAI,KAAK,CAAA;AAAA,EAChB;AACA,EAAA,OAAO,GAAA;AACT;AAEA,SAAS,uBAAA,CACP,gBAAA,EACA,SAAA,EACA,SAAA,EAC+C;AAC/C,EAAA,IAAI,qBAAqB,MAAA,EAAQ;AAC/B,IAAA,OAAO,IAAA;AAAA,EACT;AACA,EAAA,IAAI,CAAC,KAAA,CAAM,OAAA,CAAQ,gBAAgB,CAAA,EAAG;AACpC,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,0DAAA;AAAA,MACA,SAAA;AAAA,MACA;AAAA,KACF;AAAA,EACF;AACA,EAAA,IAAI,gBAAA,CAAiB,WAAW,SAAA,EAAW;AACzC,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,iDAAA;AAAA,MACA,SAAA;AAAA,MACA,gBAAA,CAAiB;AAAA,KACnB;AAAA,EACF;AACA,EAAA,OAAO,gBAAA;AACT;AAUA,SAAS,WAAA,CAAY,GAAW,CAAA,EAAqB;AACnD,EAAA,MAAM,MAAA,GAAS,YAAY,CAAC,CAAA;AAC5B,EAAA,MAAM,GAAA,GAAM,CAAA,CAAE,MAAA,GAAS,CAAA,GAAI,MAAA;AAC3B,EAAA,IAAI,CAAA,CAAE,UAAU,QAAA,EAAU;AACxB,IAAA,MAAMA,MAAAA,GAAQ,aAAA,CAAc,CAAC,CAAA,CAAE,GAAG,CAAA;AAClC,IAAA,IAAIA,WAAU,MAAA,EAAW;AACvB,MAAA,MAAM,IAAI,aAAa,oDAAoD,CAAA;AAAA,IAC7E;AACA,IAAA,OAAOA,MAAAA;AAAA,EACT;AACA,EAAA,MAAM,KAAA,GAAQ,cAAA,CAAe,CAAC,CAAA,CAAE,GAAG,CAAA;AACnC,EAAA,IAAI,UAAU,MAAA,EAAW;AACvB,IAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,EAC9E;AACA,EAAA,OAAO,OAAO,KAAA,KAAU,QAAA,GAAW,KAAA,GAAQ,OAAO,KAAK,CAAA;AACzD;AAWA,SAAS,WAAA,CAAY,CAAA,EAAW,GAAA,EAAa,GAAA,EAAuB;AAClE,EAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AACzC,EAAA,MAAM,GAAA,GAAM,CAAA,CAAE,MAAA,GAAS,GAAA,GAAM,UAAU,GAAA,GAAM,OAAA;AAC7C,EAAA,IAAI,CAAA,CAAE,UAAU,QAAA,EAAU;AACxB,IAAA,MAAMA,MAAAA,GAAQ,aAAA,CAAc,CAAC,CAAA,CAAE,GAAG,CAAA;AAClC,IAAA,IAAIA,WAAU,MAAA,EAAW;AACvB,MAAA,MAAM,IAAI,aAAa,oDAAoD,CAAA;AAAA,IAC7E;AACA,IAAA,OAAOA,MAAAA;AAAA,EACT;AACA,EAAA,MAAM,KAAA,GAAQ,cAAA,CAAe,CAAC,CAAA,CAAE,GAAG,CAAA;AACnC,EAAA,IAAI,UAAU,MAAA,EAAW;AACvB,IAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,EAC9E;AACA,EAAA,OAAO,OAAO,KAAA,KAAU,QAAA,GAAW,KAAA,GAAQ,OAAO,KAAK,CAAA;AACzD;AAEA,SAAS,QAAA,CAAS,GAAW,IAAA,EAAoB;AAC/C,EAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,IAAA,MAAM,IAAI,UAAA,CAAW,CAAA,EAAG,IAAI,CAAA,oBAAA,CAAsB,CAAA;AAAA,EACpD;AACF;AAEA,SAAS,eAAA,CAAgB,MAAA,EAAoB,KAAA,EAAe,OAAA,EAA2B;AACrF,EAAA,MAAM,KAAA,GAAQ,OAAO,KAAK,CAAA;AAC1B,EAAA,IAAI,UAAU,MAAA,EAAW;AACvB,IAAA,MAAM,IAAI,YAAA,CAAa,CAAA,0CAAA,EAA6C,KAAK,CAAA,EAAA,EAAK,OAAO,CAAA,CAAA,CAAG,CAAA;AAAA,EAC1F;AACA,EAAA,OAAO,KAAA;AACT;AAEA,SAAS,yBAAA,CAA0B,MAAoB,SAAA,EAAiC;AACtF,EAAA,MAAM,SAAqB,EAAC;AAC5B,EAAA,KAAA,MAAW,OAAO,IAAA,EAAM;AACtB,IAAA,KAAA,MAAW,SAAS,GAAA,EAAK;AACvB,MAAA,MAAA,CAAO,KAAK,KAAK,CAAA;AAAA,IACnB;AAAA,EACF;AACA,EAAA,OAAO,iBAAA,CAAkB,QAAQ,SAAS,CAAA;AAC5C;AAEA,SAAS,8BAAA,CAA+B,SAAqB,SAAA,EAA2B;AACtF,EAAA,MAAM,YAAA,GAAe,iBAAA,CAAkB,OAAA,EAAS,SAAS,CAAA;AACzD,EAAA,IAAI,iBAAiB,QAAA,EAAU;AAC7B,IAAA,OAAO,MAAM,CAAC,CAAC,GAAG,EAAE,KAAA,EAAO,UAAU,CAAA;AAAA,EACvC;AACA,EAAA,IAAI,iBAAiB,QAAA,EAAU;AAC7B,IAAA,OAAO,MAAM,CAAC,CAAC,GAAG,EAAE,KAAA,EAAO,SAAS,CAAA;AAAA,EACtC;AACA,EAAA,OAAO,MAAM,CAAC,CAAC,GAAG,EAAE,KAAA,EAAO,WAAW,CAAA;AACxC;AAEA,SAAS,iCAAA,CACP,UAAA,EACA,SAAA,EACA,SAAA,EACQ;AACR,EAAA,MAAM,YAAA,GAAe,yBAAA,CAA0B,UAAA,EAAY,SAAS,CAAA;AACpE,EAAA,IAAI,iBAAiB,QAAA,EAAU;AAC7B,IAAA,OAAO,KAAA,CAAM,CAAC,CAAA,EAAG,SAAS,GAAG,EAAE,KAAA,EAAO,UAAU,CAAA;AAAA,EAClD;AACA,EAAA,IAAI,iBAAiB,QAAA,EAAU;AAC7B,IAAA,OAAO,KAAA,CAAM,CAAC,CAAA,EAAG,SAAS,GAAG,EAAE,KAAA,EAAO,SAAS,CAAA;AAAA,EACjD;AACA,EAAA,OAAO,KAAA,CAAM,CAAC,CAAA,EAAG,SAAS,GAAG,EAAE,KAAA,EAAO,WAAW,CAAA;AACnD;AAEA,SAAS,sBAAA,CAAuB,MAAA,EAAoB,SAAA,GAAY,GAAA,EAAa;AAC3E,EAAA,MAAM,YAAA,GAAe,iBAAA,CAAkB,MAAA,EAAQ,SAAS,CAAA;AAExD,EAAA,IAAI,iBAAiB,QAAA,EAAU;AAC7B,IAAA,MAAMC,IAAAA,GAAM,IAAI,KAAA,CAAc,MAAA,CAAO,MAAM,CAAA;AAC3C,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,MAAA,CAAO,QAAQ,CAAA,EAAA,EAAK;AACtC,MAAA,MAAM,KAAA,GAAQ,OAAO,CAAC,CAAA;AACtB,MAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,QAAA,MAAM,IAAI,aAAa,gDAAgD,CAAA;AAAA,MACzE;AACA,MAAAA,IAAAA,CAAI,CAAC,CAAA,GAAI,KAAA;AAAA,IACX;AACA,IAAA,OAAO,OAAOA,IAAG,CAAA;AAAA,EACnB;AAEA,EAAA,IAAI,iBAAiB,QAAA,EAAU;AAC7B,IAAA,MAAMA,IAAAA,GAAM,IAAI,aAAA,CAAc,MAAA,CAAO,MAAM,CAAA;AAC3C,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,MAAA,CAAO,QAAQ,CAAA,EAAA,EAAK;AACtC,MAAA,MAAM,KAAA,GAAQ,OAAO,CAAC,CAAA;AACtB,MAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,QAAA,MAAM,IAAI,aAAa,gDAAgD,CAAA;AAAA,MACzE;AACA,MAAAA,IAAAA,CAAI,CAAC,CAAA,GAAI,KAAA;AAAA,IACX;AACA,IAAA,OAAO,OAAOA,IAAG,CAAA;AAAA,EACnB;AAEA,EAAA,MAAM,GAAA,GAAM,IAAI,YAAA,CAAa,MAAA,CAAO,MAAM,CAAA;AAC1C,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,MAAA,CAAO,QAAQ,CAAA,EAAA,EAAK;AACtC,IAAA,MAAM,KAAA,GAAQ,OAAO,CAAC,CAAA;AACtB,IAAA,IAAI,KAAA,KAAU,MAAA,IAAa,OAAO,KAAA,KAAU,QAAA,EAAU;AACpD,MAAA,MAAM,IAAI,aAAa,iDAAiD,CAAA;AAAA,IAC1E;AACA,IAAA,GAAA,CAAI,CAAC,CAAA,GAAI,KAAA;AAAA,EACX;AACA,EAAA,OAAO,OAAO,GAAG,CAAA;AACnB;AAEA,SAAS,sBAAA,CAAuB,MAAA,EAAsB,SAAA,GAAY,GAAA,EAAa;AAC7E,EAAA,MAAM,OAAO,MAAA,CAAO,MAAA;AACpB,EAAA,MAAM,OAAO,IAAA,GAAO,CAAA,GAAK,OAAO,CAAC,CAAA,EAAG,UAAU,CAAA,GAAK,CAAA;AAEnD,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,IAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,IAAA,IAAI,CAAC,GAAA,EAAK;AACR,MAAA,MAAM,IAAI,aAAa,gDAAgD,CAAA;AAAA,IACzE;AACA,IAAA,IAAI,GAAA,CAAI,WAAW,IAAA,EAAM;AACvB,MAAA,MAAM,IAAI,WAAW,sDAAsD,CAAA;AAAA,IAC7E;AAAA,EACF;AAEA,EAAA,MAAM,OAAmB,EAAC;AAC1B,EAAA,KAAA,MAAW,OAAO,MAAA,EAAQ;AACxB,IAAA,KAAA,MAAW,SAAS,GAAA,EAAK;AACvB,MAAA,IAAA,CAAK,KAAK,KAAK,CAAA;AAAA,IACjB;AAAA,EACF;AAEA,EAAA,MAAM,YAAA,GAAe,iBAAA,CAAkB,IAAA,EAAM,SAAS,CAAA;AACtD,EAAA,IAAI,iBAAiB,QAAA,EAAU;AAC7B,IAAA,MAAM,GAAA,GAAM,IAAI,KAAA,CAAgB,IAAI,CAAA;AACpC,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,MAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,MAAA,IAAI,CAAC,GAAA,EAAK;AACR,QAAA,MAAM,IAAI,aAAa,gDAAgD,CAAA;AAAA,MACzE;AACA,MAAA,MAAM,MAAA,GAAS,IAAI,KAAA,CAAc,IAAI,CAAA;AACrC,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,QAAA,MAAM,KAAA,GAAQ,IAAI,CAAC,CAAA;AACnB,QAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,UAAA,MAAM,IAAI,aAAa,gDAAgD,CAAA;AAAA,QACzE;AACA,QAAA,MAAA,CAAO,CAAC,CAAA,GAAI,KAAA;AAAA,MACd;AACA,MAAA,GAAA,CAAI,CAAC,CAAA,GAAI,MAAA;AAAA,IACX;AACA,IAAA,OAAO,OAAO,GAAG,CAAA;AAAA,EACnB;AACA,EAAA,IAAI,iBAAiB,QAAA,EAAU;AAC7B,IAAA,MAAM,GAAA,GAAM,IAAI,KAAA,CAAgB,IAAI,CAAA;AACpC,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,MAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,MAAA,IAAI,CAAC,GAAA,EAAK;AACR,QAAA,MAAM,IAAI,aAAa,gDAAgD,CAAA;AAAA,MACzE;AACA,MAAA,MAAM,MAAA,GAAS,IAAI,KAAA,CAAc,IAAI,CAAA;AACrC,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,QAAA,MAAM,KAAA,GAAQ,IAAI,CAAC,CAAA;AACnB,QAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,UAAA,MAAM,IAAI,aAAa,iDAAiD,CAAA;AAAA,QAC1E;AACA,QAAA,MAAA,CAAO,CAAC,CAAA,GAAI,KAAA;AAAA,MACd;AACA,MAAA,GAAA,CAAI,CAAC,CAAA,GAAI,MAAA;AAAA,IACX;AACA,IAAA,OAAO,MAAA,CAAO,GAAA,EAAK,EAAE,KAAA,EAAO,WAAW,CAAA;AAAA,EACzC;AAEA,EAAA,MAAM,IAAA,GAAO,IAAI,aAAA,CAAc,IAAA,GAAO,IAAI,CAAA;AAC1C,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,CAAK,QAAQ,CAAA,EAAA,EAAK;AACpC,IAAA,MAAM,KAAA,GAAQ,KAAK,CAAC,CAAA;AACpB,IAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,MAAA,MAAM,IAAI,aAAa,gDAAgD,CAAA;AAAA,IACzE;AACA,IAAA,IAAA,CAAK,CAAC,CAAA,GAAI,KAAA;AAAA,EACZ;AAEA,EAAA,MAAM,EAAE,aAAA,EAAc,GAAI,SAAA,EAAU;AACpC,EAAA,OAAO,OAAW,cAAA,CAAe;AAAA,IAC/B,IAAA;AAAA,IACA,KAAA,EAAO,CAAC,IAAA,EAAM,IAAI,CAAA;AAAA,IAClB,KAAA,EAAO,OAAA;AAAA,IACP,MAAA,EAAQ;AAAA,GACT,CAAA;AACH;AA8BO,IAAM,eAAN,MAAmB;AAAA;AAAA,EAEhB,MAAA,GAAS,KAAA;AAAA;AAAA,EAET,QAAA;AAAA;AAAA,EAEA,aAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAUR,IAAI,CAAA,EAAiB;AACnB,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,MAAA,MAAM,IAAI,qBAAA,CAAsB,wCAAA,EAA0C,GAAG,CAAA;AAAA,IAC/E;AAGA,IAAA,MAAM,SAAA,uBAAgB,GAAA,EAAc;AACpC,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,MAAM,CAAA,EAAA,EAAK;AAC/B,MAAA,SAAA,CAAU,GAAA,CAAI,WAAA,CAAY,CAAA,EAAG,CAAC,CAAC,CAAA;AAAA,IACjC;AAGA,IAAA,IAAA,CAAK,QAAA,GAAW,cAAA,CAAe,SAAA,EAAW,GAAG,CAAA;AAG7C,IAAA,IAAA,CAAK,aAAA,uBAAoB,GAAA,EAAI;AAC7B,IAAA,KAAA,IAAS,IAAI,CAAA,EAAG,CAAA,GAAI,IAAA,CAAK,QAAA,CAAS,QAAQ,CAAA,EAAA,EAAK;AAC7C,MAAA,IAAA,CAAK,aAAA,CAAc,IAAI,eAAA,CAAgB,IAAA,CAAK,UAAU,CAAA,EAAG,kBAAkB,GAAG,CAAC,CAAA;AAAA,IACjF;AAEA,IAAA,IAAA,CAAK,MAAA,GAAS,IAAA;AACd,IAAA,OAAO,IAAA;AAAA,EACT;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAWA,UAAU,CAAA,EAAmB;AAC3B,IAAA,IAAI,CAAC,KAAK,MAAA,EAAQ;AAChB,MAAA,MAAM,IAAI,eAAe,8CAA8C,CAAA;AAAA,IACzE;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,MAAA,OAAO,MAAA,CAAO,EAAE,CAAA;AAAA,IAClB;AAEA,IAAA,MAAM,SAAS,IAAA,CAAK,aAAA;AACpB,IAAA,IAAI,CAAC,IAAA,CAAK,QAAA,IAAY,CAAC,MAAA,EAAQ;AAC7B,MAAA,MAAM,IAAI,aAAa,mDAAmD,CAAA;AAAA,IAC5E;AAGA,IAAA,MAAM,MAAA,GAAS,IAAI,KAAA,CAAc,CAAA,CAAE,IAAI,CAAA;AAGvC,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,MAAM,CAAA,EAAA,EAAK;AAC/B,MAAA,MAAM,GAAA,GAAM,WAAA,CAAY,CAAA,EAAG,CAAC,CAAA;AAC5B,MAAA,MAAM,GAAA,GAAM,MAAA,CAAO,GAAA,CAAI,GAAG,CAAA;AAC1B,MAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,QAAA,MAAM,IAAI,qBAAA;AAAA,UACR,CAAA,eAAA,EAAkB,MAAA,CAAO,GAAG,CAAC,CAAA,mCAAA,CAAA;AAAA,UAC7B,GAAA;AAAA,UACA;AAAA,SACF;AAAA,MACF;AACA,MAAA,MAAA,CAAO,CAAC,CAAA,GAAI,GAAA;AAAA,IACd;AAEA,IAAA,OAAO,MAAA,CAAO,MAAA,EAAQ,EAAE,KAAA,EAAO,WAAW,CAAA;AAAA,EAC5C;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EASA,aAAa,CAAA,EAAmB;AAC9B,IAAA,OAAO,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,CAAE,UAAU,CAAC,CAAA;AAAA,EAChC;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAWA,iBAAiB,CAAA,EAAmB;AAClC,IAAA,IAAI,CAAC,KAAK,MAAA,EAAQ;AAChB,MAAA,MAAM,IAAI,eAAe,sDAAsD,CAAA;AAAA,IACjF;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,MAAM,UAAU,IAAA,CAAK,QAAA;AACrB,IAAA,IAAI,CAAC,OAAA,EAAS;AACZ,MAAA,MAAM,IAAI,aAAa,mDAAmD,CAAA;AAAA,IAC5E;AACA,IAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,MAAA,OAAO,8BAAA,CAA+B,SAAS,GAAG,CAAA;AAAA,IACpD;AAEA,IAAA,MAAM,aAAa,OAAA,CAAQ,MAAA;AAE3B,IAAA,MAAM,MAAA,GAAS,IAAI,KAAA,CAAgB,CAAA,CAAE,IAAI,CAAA;AACzC,IAAA,MAAM,MAAA,GAAS,YAAY,CAAC,CAAA;AAC5B,IAAA,MAAM,IAAA,GAAO,eAAe,CAAC,CAAA;AAG7B,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,MAAM,CAAA,EAAA,EAAK;AAC/B,MAAA,MAAM,GAAA,GAAM,IAAA,CAAK,CAAA,CAAE,MAAA,GAAS,IAAI,MAAM,CAAA;AACtC,MAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,QAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,MAC9E;AACA,MAAA,MAAM,GAAA,GAAM,OAAO,GAAG,CAAA;AAGtB,MAAA,IAAI,GAAA,GAAM,KAAK,GAAA,IAAO,UAAA,IAAc,CAAC,MAAA,CAAO,SAAA,CAAU,GAAG,CAAA,EAAG;AAC1D,QAAA,MAAM,IAAI,qBAAA;AAAA,UACR,CAAA,qBAAA,EAAwB,GAAG,CAAA,yBAAA,EAA4B,UAAA,GAAa,CAAC,CAAA,CAAA,CAAA;AAAA,UACrE,GAAA;AAAA,UACA;AAAA,SACF;AAAA,MACF;AAEA,MAAA,MAAA,CAAO,CAAC,CAAA,GAAI,eAAA,CAAgB,OAAA,EAAS,KAAK,+BAA+B,CAAA;AAAA,IAC3E;AAGA,IAAA,OAAO,sBAAA,CAAuB,QAAQ,GAAG,CAAA;AAAA,EAC3C;AACF;AA6BO,IAAM,gBAAN,MAAoB;AAAA;AAAA,EAEjB,MAAA,GAAS,KAAA;AAAA;AAAA,EAET,WAAA;AAAA;AAAA,EAEA,gBAAA;AAAA;AAAA,EAEA,MAAA;AAAA;AAAA,EAEA,aAAA;AAAA;AAAA,EAEA,IAAA;AAAA;AAAA,EAEA,YAAA;AAAA;AAAA,EAEA,gBAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAYR,WAAA,CACE,OAAA,GAMI,EAAC,EACL;AACA,IAAA,MAAM,YAAA,GAAe,OAAA,CAAQ,MAAA,IAAU,OAAA,CAAQ,YAAA,IAAgB,KAAA;AAC/D,IAAA,IAAI,OAAA,CAAQ,MAAA,KAAW,MAAA,IAAa,OAAA,CAAQ,iBAAiB,MAAA,EAAW;AACtE,MAAA,IAAI,OAAA,CAAQ,MAAA,KAAW,OAAA,CAAQ,YAAA,EAAc;AAC3C,QAAA,MAAM,IAAI,qBAAA;AAAA,UACR,2DAAA;AAAA,UACA,QAAA;AAAA,UACA,OAAA,CAAQ;AAAA,SACV;AAAA,MACF;AAAA,IACF;AACA,IAAA,IAAA,CAAK,MAAA,GAAS,YAAA;AACd,IAAA,IAAA,CAAK,aAAA,GAAgB,QAAQ,aAAA,IAAiB,OAAA;AAC9C,IAAA,IAAA,CAAK,IAAA,GAAO,QAAQ,IAAA,IAAQ,IAAA;AAC5B,IAAA,IAAA,CAAK,gBAAA,GAAmB,QAAQ,UAAA,IAAc,MAAA;AAE9C,IAAA,IAAI,OAAO,IAAA,CAAK,MAAA,KAAW,SAAA,EAAW;AACpC,MAAA,MAAM,IAAI,qBAAA,CAAsB,0BAAA,EAA4B,QAAA,EAAU,KAAK,MAAM,CAAA;AAAA,IACnF;AACA,IAAA,IAAI,IAAA,CAAK,aAAA,KAAkB,OAAA,IAAW,IAAA,CAAK,kBAAkB,QAAA,EAAU;AACrE,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,2CAAA;AAAA,QACA,eAAA;AAAA,QACA,IAAA,CAAK;AAAA,OACP;AAAA,IACF;AACA,IAAA,IAAI,IAAA,CAAK,SAAS,IAAA,IAAQ,IAAA,CAAK,SAAS,OAAA,IAAW,IAAA,CAAK,SAAS,WAAA,EAAa;AAC5E,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,4CAAA;AAAA,QACA,MAAA;AAAA,QACA,IAAA,CAAK;AAAA,OACP;AAAA,IACF;AAAA,EACF;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAWA,IAAI,CAAA,EAAiB;AACnB,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAE1C,IAAA,IAAI,QAAA,KAAa,CAAA,IAAK,SAAA,KAAc,CAAA,EAAG;AACrC,MAAA,MAAM,IAAI,qBAAA,CAAsB,yCAAA,EAA2C,GAAG,CAAA;AAAA,IAChF;AAGA,IAAA,IAAA,CAAK,cAAc,EAAC;AACpB,IAAA,IAAA,CAAK,mBAAmB,EAAC;AAEzB,IAAA,MAAM,kBAAA,GAAqB,uBAAA;AAAA,MACzB,IAAA,CAAK,gBAAA;AAAA,MACL,SAAA;AAAA,MACA;AAAA,KACF;AAGA,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,MAAA,IAAI,IAAA;AAEJ,MAAA,IAAI,kBAAA,EAAoB;AACtB,QAAA,MAAM,WAAA,GAAc,mBAAmB,CAAC,CAAA;AACxC,QAAA,IAAI,CAAC,WAAA,EAAa;AAChB,UAAA,MAAM,IAAI,qBAAA,CAAsB,gCAAA,EAAkC,YAAA,EAAc,CAAC,CAAA;AAAA,QACnF;AACA,QAAA,IAAI,CAAC,KAAA,CAAM,OAAA,CAAQ,WAAW,CAAA,EAAG;AAC/B,UAAA,MAAM,IAAI,qBAAA;AAAA,YACR,gDAAA;AAAA,YACA,YAAA;AAAA,YACA;AAAA,WACF;AAAA,QACF;AACA,QAAA,IAAA,GAAO,sBAAA,CAAuB,aAAa,YAAY,CAAA;AAAA,MACzD,CAAA,MAAO;AACL,QAAA,MAAM,SAAA,uBAAgB,GAAA,EAAc;AAGpC,QAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,UAAA,SAAA,CAAU,GAAA,CAAI,WAAA,CAAY,CAAA,EAAG,CAAA,EAAG,CAAC,CAAC,CAAA;AAAA,QACpC;AAGA,QAAA,IAAA,GAAO,cAAA,CAAe,WAAW,GAAG,CAAA;AAAA,MACtC;AAEA,MAAA,IAAI,IAAA,CAAK,WAAW,CAAA,EAAG;AACrB,QAAA,MAAM,IAAI,qBAAA,CAAsB,8CAAA,EAAgD,GAAA,EAAK,CAAC,CAAA;AAAA,MACxF;AAEA,MAAA,IAAA,CAAK,WAAA,CAAY,KAAK,IAAI,CAAA;AAG1B,MAAA,MAAM,GAAA,uBAAU,GAAA,EAAsB;AACtC,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,CAAK,QAAQ,CAAA,EAAA,EAAK;AACpC,QAAA,GAAA,CAAI,IAAI,eAAA,CAAgB,IAAA,EAAM,CAAA,EAAG,mBAAmB,GAAG,CAAC,CAAA;AAAA,MAC1D;AACA,MAAA,IAAA,CAAK,gBAAA,CAAiB,KAAK,GAAG,CAAA;AAG9B,MAAA,IAAI,kBAAA,EAAoB;AACtB,QAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,UAAA,MAAM,GAAA,GAAM,WAAA,CAAY,CAAA,EAAG,CAAA,EAAG,CAAC,CAAA;AAC/B,UAAA,IAAI,CAAC,GAAA,CAAI,GAAA,CAAI,GAAG,CAAA,EAAG;AACjB,YAAA,MAAM,IAAI,qBAAA;AAAA,cACR,CAAA,kBAAA,EAAqB,MAAA,CAAO,GAAG,CAAC,eAAe,CAAC,CAAA,CAAA;AAAA,cAChD,GAAA;AAAA,cACA;AAAA,aACF;AAAA,UACF;AAAA,QACF;AAAA,MACF;AAAA,IACF;AAEA,IAAA,IAAA,CAAK,YAAA,GAAe,IAAA,CAAK,WAAA,CAAY,GAAA,CAAI,CAAC,IAAA,KAAS;AACjD,MAAA,IAAI,IAAA,CAAK,IAAA,KAAS,IAAA,EAAM,OAAO,IAAA;AAC/B,MAAA,IAAI,KAAK,IAAA,KAAS,OAAA,SAAgB,IAAA,CAAK,MAAA,GAAS,IAAI,CAAA,GAAI,IAAA;AACxD,MAAA,IAAI,KAAK,IAAA,KAAS,WAAA,SAAoB,IAAA,CAAK,MAAA,KAAW,IAAI,CAAA,GAAI,IAAA;AAC9D,MAAA,OAAO,IAAA;AAAA,IACT,CAAC,CAAA;AAED,IAAA,IAAA,CAAK,MAAA,GAAS,IAAA;AACd,IAAA,OAAO,IAAA;AAAA,EACT;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAWA,UAAU,CAAA,EAA+B;AACvC,IAAA,IAAI,CAAC,KAAK,MAAA,EAAQ;AAChB,MAAA,MAAM,IAAI,eAAe,+CAA+C,CAAA;AAAA,IAC1E;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAE1C,IAAA,MAAM,aAAa,IAAA,CAAK,WAAA;AACxB,IAAA,MAAM,eAAe,IAAA,CAAK,gBAAA;AAC1B,IAAA,IAAI,CAAC,UAAA,IAAc,CAAC,YAAA,EAAc;AAChC,MAAA,MAAM,IAAI,aAAa,oDAAoD,CAAA;AAAA,IAC7E;AACA,IAAA,MAAM,iBAAiB,UAAA,CAAW,MAAA;AAClC,IAAA,IAAI,cAAc,cAAA,EAAgB;AAChC,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,iDAAA;AAAA,QACA,GAAA;AAAA,QACA;AAAA,OACF;AAAA,IACF;AAEA,IAAA,MAAM,cAAc,IAAA,CAAK,YAAA,IAAgB,UAAA,CAAW,GAAA,CAAI,MAAM,IAAI,CAAA;AAGlE,IAAA,IAAI,SAAA,GAAY,CAAA;AAChB,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,UAAA,CAAW,QAAQ,CAAA,EAAA,EAAK;AAC1C,MAAA,MAAM,IAAA,GAAO,WAAW,CAAC,CAAA;AACzB,MAAA,IAAI,CAAC,IAAA,EAAM;AACX,MAAA,MAAM,SAAA,GAAY,WAAA,CAAY,CAAC,CAAA,IAAK,IAAA;AACpC,MAAA,SAAA,IAAa,IAAA,CAAK,MAAA,IAAU,SAAA,KAAc,IAAA,GAAO,CAAA,GAAI,CAAA,CAAA;AAAA,IACvD;AAEA,IAAA,IAAI,QAAA,KAAa,CAAA,IAAK,SAAA,KAAc,CAAA,EAAG;AACrC,MAAA,OAAO,IAAA,CAAK,MAAA,GACR,SAAA,CAAU,OAAA,CAAQ;AAAA,QAChB,IAAA,EAAM,CAAA;AAAA,QACN,IAAA,EAAM,SAAA;AAAA,QACN,UAAA,EAAY,IAAI,UAAA,CAAW,CAAC,CAAA;AAAA,QAC5B,UAAA,EAAY,IAAI,UAAA,CAAW,CAAC,CAAA;AAAA,QAC5B,MAAA,EAAQ,IAAI,YAAA,CAAa,CAAC;AAAA,OAC3B,CAAA,GACD,KAAA,CAAM,CAAC,CAAA,EAAG,SAAS,CAAA,EAAG,EAAE,KAAA,EAAO,SAAA,EAAW,CAAA;AAAA,IAChD;AAEA,IAAA,IAAI,KAAK,MAAA,EAAQ;AACf,MAAA,MAAM,SAAmB,EAAC;AAC1B,MAAA,MAAM,SAAmB,EAAC;AAC1B,MAAA,MAAM,OAAiB,EAAC;AAExB,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,QAAA,IAAI,SAAA,GAAY,CAAA;AAChB,QAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,UAAA,MAAM,IAAA,GAAO,WAAW,CAAC,CAAA;AACzB,UAAA,MAAM,GAAA,GAAM,aAAa,CAAC,CAAA;AAC1B,UAAA,MAAM,SAAA,GAAY,WAAA,CAAY,CAAC,CAAA,IAAK,IAAA;AACpC,UAAA,IAAI,CAAC,IAAA,IAAQ,CAAC,GAAA,EAAK;AACjB,YAAA,MAAM,IAAI,aAAa,yDAAyD,CAAA;AAAA,UAClF;AACA,UAAA,MAAM,OAAA,GAAU,IAAA,CAAK,MAAA,IAAU,SAAA,KAAc,OAAO,CAAA,GAAI,CAAA,CAAA;AACxD,UAAA,MAAM,GAAA,GAAM,WAAA,CAAY,CAAA,EAAG,CAAA,EAAG,CAAC,CAAA;AAC/B,UAAA,MAAM,GAAA,GAAM,GAAA,CAAI,GAAA,CAAI,GAAG,CAAA;AACvB,UAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,YAAA,IAAI,IAAA,CAAK,kBAAkB,QAAA,EAAU;AACnC,cAAA,SAAA,IAAa,OAAA;AACb,cAAA;AAAA,YACF;AACA,YAAA,MAAM,IAAI,sBAAsB,CAAA,kBAAA,EAAqB,MAAA,CAAO,GAAG,CAAC,CAAA,CAAA,EAAI,KAAK,GAAG,CAAA;AAAA,UAC9E;AAEA,UAAA,IAAI,SAAA,KAAc,IAAA,IAAQ,GAAA,KAAQ,SAAA,EAAW;AAC3C,YAAA,SAAA,IAAa,OAAA;AACb,YAAA;AAAA,UACF;AAEA,UAAA,MAAM,WAAW,SAAA,KAAc,IAAA,IAAQ,GAAA,GAAM,SAAA,GAAY,MAAM,CAAA,GAAI,GAAA;AACnE,UAAA,MAAA,CAAO,KAAK,CAAC,CAAA;AACb,UAAA,MAAA,CAAO,IAAA,CAAK,YAAY,QAAQ,CAAA;AAChC,UAAA,IAAA,CAAK,KAAK,CAAC,CAAA;AACX,UAAA,SAAA,IAAa,OAAA;AAAA,QACf;AAAA,MACF;AAEA,MAAA,OAAO,UAAU,OAAA,CAAQ;AAAA,QACvB,IAAA,EAAM,QAAA;AAAA,QACN,IAAA,EAAM,SAAA;AAAA,QACN,UAAA,EAAY,UAAA,CAAW,IAAA,CAAK,MAAM,CAAA;AAAA,QAClC,UAAA,EAAY,UAAA,CAAW,IAAA,CAAK,MAAM,CAAA;AAAA,QAClC,MAAA,EAAQ,YAAA,CAAa,IAAA,CAAK,IAAI;AAAA,OAC/B,CAAA;AAAA,IACH;AAEA,IAAA,MAAM,MAAA,GAAS,KAAA,CAAM,IAAA,CAAK,EAAE,QAAQ,QAAA,EAAS,EAAG,MAAM,IAAI,KAAA,CAAc,SAAS,CAAA,CAAE,IAAA,CAAK,CAAC,CAAC,CAAA;AAE1F,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,IAAI,SAAA,GAAY,CAAA;AAChB,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,QAAA,MAAM,IAAA,GAAO,WAAW,CAAC,CAAA;AACzB,QAAA,MAAM,GAAA,GAAM,aAAa,CAAC,CAAA;AAC1B,QAAA,MAAM,SAAA,GAAY,WAAA,CAAY,CAAC,CAAA,IAAK,IAAA;AACpC,QAAA,IAAI,CAAC,IAAA,IAAQ,CAAC,GAAA,EAAK;AACjB,UAAA,MAAM,IAAI,aAAa,yDAAyD,CAAA;AAAA,QAClF;AACA,QAAA,MAAM,OAAA,GAAU,IAAA,CAAK,MAAA,IAAU,SAAA,KAAc,OAAO,CAAA,GAAI,CAAA,CAAA;AACxD,QAAA,MAAM,GAAA,GAAM,WAAA,CAAY,CAAA,EAAG,CAAA,EAAG,CAAC,CAAA;AAC/B,QAAA,MAAM,GAAA,GAAM,GAAA,CAAI,GAAA,CAAI,GAAG,CAAA;AACvB,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,IAAI,IAAA,CAAK,kBAAkB,QAAA,EAAU;AACnC,YAAA,SAAA,IAAa,OAAA;AACb,YAAA;AAAA,UACF;AACA,UAAA,MAAM,IAAI,sBAAsB,CAAA,kBAAA,EAAqB,MAAA,CAAO,GAAG,CAAC,CAAA,CAAA,EAAI,KAAK,GAAG,CAAA;AAAA,QAC9E;AACA,QAAA,IAAI,SAAA,KAAc,IAAA,IAAQ,GAAA,KAAQ,SAAA,EAAW;AAC3C,UAAA,SAAA,IAAa,OAAA;AACb,UAAA;AAAA,QACF;AACA,QAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,QACnE;AACA,QAAA,MAAM,WAAW,SAAA,KAAc,IAAA,IAAQ,GAAA,GAAM,SAAA,GAAY,MAAM,CAAA,GAAI,GAAA;AACnE,QAAA,GAAA,CAAI,SAAA,GAAY,QAAQ,CAAA,GAAI,CAAA;AAC5B,QAAA,SAAA,IAAa,OAAA;AAAA,MACf;AAAA,IACF;AAEA,IAAA,OAAO,MAAA,CAAO,QAAQ,EAAE,KAAA,EAAO,WAAW,MAAA,EAAQ,CAAA,CAAE,QAAQ,CAAA;AAAA,EAC9D;AAAA,EAEA,aAAa,CAAA,EAA+B;AAC1C,IAAA,OAAO,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,CAAE,UAAU,CAAC,CAAA;AAAA,EAChC;AAAA,EAEA,iBAAiB,CAAA,EAA+B;AAC9C,IAAA,IAAI,CAAC,KAAK,MAAA,EAAQ;AAChB,MAAA,MAAM,IAAI,eAAe,uDAAuD,CAAA;AAAA,IAClF;AACA,IAAA,MAAM,KAAA,GAAQ,CAAA,YAAa,SAAA,GAAY,CAAA,CAAE,SAAQ,GAAI,CAAA;AACrD,IAAA,QAAA,CAAS,OAAO,GAAG,CAAA;AACnB,IAAA,mBAAA,CAAoB,OAAO,GAAG,CAAA;AAC9B,IAAA,MAAM,CAAC,QAAA,EAAU,KAAK,CAAA,GAAI,WAAW,KAAK,CAAA;AAC1C,IAAA,MAAM,aAAa,IAAA,CAAK,WAAA;AACxB,IAAA,IAAI,CAAC,UAAA,EAAY;AACf,MAAA,MAAM,IAAI,aAAa,yDAAyD,CAAA;AAAA,IAClF;AACA,IAAA,MAAM,YAAY,UAAA,CAAW,MAAA;AAC7B,IAAA,MAAM,cAAc,IAAA,CAAK,YAAA,IAAgB,UAAA,CAAW,GAAA,CAAI,MAAM,IAAI,CAAA;AAClE,IAAA,MAAM,YAAY,UAAA,CAAW,MAAA,CAAO,CAAC,GAAA,EAAK,MAAM,GAAA,KAAQ;AACtD,MAAA,MAAM,SAAA,GAAY,WAAA,CAAY,GAAG,CAAA,IAAK,IAAA;AACtC,MAAA,OAAO,GAAA,GAAM,IAAA,CAAK,MAAA,IAAU,SAAA,KAAc,OAAO,CAAA,GAAI,CAAA,CAAA;AAAA,IACvD,GAAG,CAAC,CAAA;AACJ,IAAA,IAAI,UAAU,SAAA,EAAW;AACvB,MAAA,MAAM,IAAI,qBAAA,CAAsB,+CAAA,EAAiD,GAAA,EAAK,KAAK,CAAA;AAAA,IAC7F;AACA,IAAA,IAAI,aAAa,CAAA,EAAG;AAClB,MAAA,OAAO,iCAAA,CAAkC,UAAA,EAAY,SAAA,EAAW,GAAG,CAAA;AAAA,IACrE;AAEA,IAAA,MAAM,MAAA,GAAS,IAAI,KAAA,CAAkB,QAAQ,CAAA;AAC7C,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAA,CAAO,CAAC,CAAA,GAAI,IAAI,KAAA,CAAgB,SAAS,CAAA;AAAA,IAC3C;AACA,IAAA,MAAM,SAAA,GAAY,eAAe,KAAK,CAAA;AACtC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,KAAK,CAAA;AAE7C,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,IAAI,SAAA,GAAY,CAAA;AAChB,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,QAAA,MAAM,IAAA,GAAO,WAAW,CAAC,CAAA;AACzB,QAAA,MAAM,SAAA,GAAY,WAAA,CAAY,CAAC,CAAA,IAAK,IAAA;AACpC,QAAA,IAAI,CAAC,IAAA,EAAM;AACT,UAAA,MAAM,IAAI,aAAa,yDAAyD,CAAA;AAAA,QAClF;AACA,QAAA,MAAM,OAAA,GAAU,IAAA,CAAK,MAAA,IAAU,SAAA,KAAc,OAAO,CAAA,GAAI,CAAA,CAAA;AACxD,QAAA,IAAI,YAAY,CAAA,EAAG;AACjB,UAAA,MAAMC,IAAAA,GAAM,OAAO,CAAC,CAAA;AACpB,UAAA,IAAI,CAACA,IAAAA,EAAK;AACR,YAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,UACnE;AACA,UAAAA,KAAI,CAAC,CAAA,GAAI,gBAAgB,IAAA,EAAM,SAAA,IAAa,GAAG,gCAAgC,CAAA;AAC/E,UAAA;AAAA,QACF;AAEA,QAAA,IAAI,MAAA,GAAS,CAAA;AACb,QAAA,MAAM,OAAA,GAAU,KAAA,CAAM,MAAA,GAAS,CAAA,GAAI,UAAU,SAAA,GAAY,OAAA;AACzD,QAAA,MAAM,KAAA,GAAQ,UAAU,OAAO,CAAA;AAC/B,QAAA,IAAI,UAAU,MAAA,EAAW;AACvB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,IAAI,MAAA,GAAS,OAAO,KAAK,CAAA;AACzB,QAAA,IAAI,cAAc,MAAA,GAAS,CAAA;AAE3B,QAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,OAAA,EAAS,CAAA,EAAA,EAAK;AAChC,UAAA,MAAM,GAAA,GAAM,SAAA,CAAU,OAAA,GAAU,CAAA,GAAI,OAAO,CAAA;AAC3C,UAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,YAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,UAC9E;AACA,UAAA,MAAM,GAAA,GAAM,OAAO,GAAG,CAAA;AACtB,UAAA,IAAI,MAAM,MAAA,EAAQ;AAChB,YAAA,MAAA,GAAS,GAAA;AACT,YAAA,MAAA,GAAS,CAAA;AAAA,UACX;AACA,UAAA,IAAI,MAAM,CAAA,EAAG;AACX,YAAA,WAAA,GAAc,IAAA;AAAA,UAChB;AAAA,QACF;AAEA,QAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,QACnE;AACA,QAAA,IAAI,CAAC,WAAA,EAAa;AAChB,UAAA,IAAI,cAAc,IAAA,EAAM;AACtB,YAAA,GAAA,CAAI,CAAC,CAAA,GAAI,eAAA,CAAgB,IAAA,EAAM,WAAW,gCAAgC,CAAA;AAAA,UAC5E,CAAA,MAAA,IAAW,IAAA,CAAK,aAAA,KAAkB,QAAA,EAAU;AAC1C,YAAA,MAAM,IAAI,qBAAA;AAAA,cACR,gJAAA;AAAA,cACA;AAAA,aACF;AAAA,UACF,CAAA,MAAO;AACL,YAAA,MAAM,IAAI,qBAAA,CAAsB,qCAAA,EAAuC,GAAG,CAAA;AAAA,UAC5E;AAAA,QACF,CAAA,MAAO;AACL,UAAA,MAAM,YAAY,SAAA,KAAc,IAAA,IAAQ,MAAA,IAAU,SAAA,GAAY,SAAS,CAAA,GAAI,MAAA;AAC3E,UAAA,GAAA,CAAI,CAAC,CAAA,GAAI,eAAA,CAAgB,IAAA,EAAM,WAAW,gCAAgC,CAAA;AAAA,QAC5E;AAEA,QAAA,SAAA,IAAa,OAAA;AAAA,MACf;AAAA,IACF;AAEA,IAAA,OAAO,sBAAA,CAAuB,QAAQ,GAAG,CAAA;AAAA,EAC3C;AACF;AA2BO,IAAM,iBAAN,MAAqB;AAAA;AAAA,EAElB,MAAA,GAAS,KAAA;AAAA;AAAA,EAET,WAAA;AAAA;AAAA,EAEA,gBAAA;AAAA;AAAA,EAEA,aAAA;AAAA;AAAA,EAEA,YAAA;AAAA;AAAA,EAEA,gBAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAUR,WAAA,CACE,OAAA,GAII,EAAC,EACL;AACA,IAAA,IAAA,CAAK,aAAA,GAAgB,QAAQ,aAAA,IAAiB,OAAA;AAC9C,IAAA,IAAA,CAAK,YAAA,GAAe,QAAQ,YAAA,IAAgB,EAAA;AAC5C,IAAA,IAAA,CAAK,gBAAA,GAAmB,QAAQ,UAAA,IAAc,MAAA;AAE9C,IAAA,IAAI,IAAA,CAAK,aAAA,KAAkB,OAAA,IAAW,IAAA,CAAK,kBAAkB,iBAAA,EAAmB;AAC9E,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,oDAAA;AAAA,QACA,eAAA;AAAA,QACA,IAAA,CAAK;AAAA,OACP;AAAA,IACF;AACA,IAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,IAAA,CAAK,YAAY,CAAA,IAAK,CAAC,MAAA,CAAO,KAAA,CAAM,IAAA,CAAK,YAAY,CAAA,EAAG;AAC3E,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,6CAAA;AAAA,QACA,cAAA;AAAA,QACA,IAAA,CAAK;AAAA,OACP;AAAA,IACF;AACA,IAAA,IAAI,MAAA,CAAO,QAAA,CAAS,IAAA,CAAK,YAAY,CAAA,IAAK,CAAC,MAAA,CAAO,SAAA,CAAU,IAAA,CAAK,YAAY,CAAA,EAAG;AAC9E,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,6CAAA;AAAA,QACA,cAAA;AAAA,QACA,IAAA,CAAK;AAAA,OACP;AAAA,IACF;AAAA,EACF;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAUA,IAAI,CAAA,EAAiB;AACnB,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAE1C,IAAA,IAAI,aAAa,CAAA,EAAG;AAClB,MAAA,MAAM,IAAI,qBAAA,CAAsB,0CAAA,EAA4C,GAAG,CAAA;AAAA,IACjF;AAEA,IAAA,IAAA,CAAK,cAAc,EAAC;AACpB,IAAA,IAAA,CAAK,mBAAmB,EAAC;AAEzB,IAAA,MAAM,kBAAA,GAAqB,uBAAA;AAAA,MACzB,IAAA,CAAK,gBAAA;AAAA,MACL,SAAA;AAAA,MACA;AAAA,KACF;AAGA,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,MAAA,IAAI,MAAA;AAEJ,MAAA,IAAI,kBAAA,EAAoB;AACtB,QAAA,MAAM,WAAA,GAAc,mBAAmB,CAAC,CAAA;AACxC,QAAA,IAAI,CAAC,WAAA,EAAa;AAChB,UAAA,MAAM,IAAI,qBAAA,CAAsB,gCAAA,EAAkC,YAAA,EAAc,CAAC,CAAA;AAAA,QACnF;AACA,QAAA,IAAI,CAAC,KAAA,CAAM,OAAA,CAAQ,WAAW,CAAA,EAAG;AAC/B,UAAA,MAAM,IAAI,qBAAA;AAAA,YACR,gDAAA;AAAA,YACA,YAAA;AAAA,YACA;AAAA,WACF;AAAA,QACF;AACA,QAAA,MAAA,GAAS,sBAAA,CAAuB,aAAa,YAAY,CAAA;AAAA,MAC3D,CAAA,MAAO;AACL,QAAA,MAAM,SAAA,uBAAgB,GAAA,EAAc;AAGpC,QAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,UAAA,SAAA,CAAU,GAAA,CAAI,WAAA,CAAY,CAAA,EAAG,CAAA,EAAG,CAAC,CAAC,CAAA;AAAA,QACpC;AAGA,QAAA,MAAA,GAAS,cAAA,CAAe,WAAW,GAAG,CAAA;AAAA,MACxC;AAEA,MAAA,IAAI,MAAA,CAAO,WAAW,CAAA,EAAG;AACvB,QAAA,MAAM,IAAI,qBAAA,CAAsB,8CAAA,EAAgD,GAAA,EAAK,CAAC,CAAA;AAAA,MACxF;AAEA,MAAA,IAAA,CAAK,WAAA,CAAY,KAAK,MAAM,CAAA;AAG5B,MAAA,MAAM,GAAA,uBAAU,GAAA,EAAsB;AACtC,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,MAAA,CAAO,QAAQ,CAAA,EAAA,EAAK;AACtC,QAAA,GAAA,CAAI,IAAI,eAAA,CAAgB,MAAA,EAAQ,CAAA,EAAG,oBAAoB,GAAG,CAAC,CAAA;AAAA,MAC7D;AACA,MAAA,IAAA,CAAK,gBAAA,CAAiB,KAAK,GAAG,CAAA;AAE9B,MAAA,IAAI,kBAAA,EAAoB;AACtB,QAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,UAAA,MAAM,GAAA,GAAM,WAAA,CAAY,CAAA,EAAG,CAAA,EAAG,CAAC,CAAA;AAC/B,UAAA,IAAI,CAAC,GAAA,CAAI,GAAA,CAAI,GAAG,CAAA,EAAG;AACjB,YAAA,MAAM,IAAI,qBAAA;AAAA,cACR,CAAA,kBAAA,EAAqB,MAAA,CAAO,GAAG,CAAC,eAAe,CAAC,CAAA,CAAA;AAAA,cAChD,GAAA;AAAA,cACA;AAAA,aACF;AAAA,UACF;AAAA,QACF;AAAA,MACF;AAEA,MAAA,IAAI,IAAA,CAAK,kBAAkB,iBAAA,EAAmB;AAC5C,QAAA,IACE,MAAA,CAAO,QAAA,CAAS,IAAA,CAAK,YAAY,CAAA,IACjC,IAAA,CAAK,YAAA,IAAgB,CAAA,IACrB,IAAA,CAAK,YAAA,GAAe,MAAA,CAAO,MAAA,EAC3B;AACA,UAAA,MAAM,IAAI,qBAAA;AAAA,YACR,8DAAA;AAAA,YACA,cAAA;AAAA,YACA,IAAA,CAAK;AAAA,WACP;AAAA,QACF;AAAA,MACF;AAAA,IACF;AAEA,IAAA,IAAA,CAAK,MAAA,GAAS,IAAA;AACd,IAAA,OAAO,IAAA;AAAA,EACT;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAWA,UAAU,CAAA,EAAmB;AAC3B,IAAA,IAAI,CAAC,KAAK,MAAA,EAAQ;AAChB,MAAA,MAAM,IAAI,eAAe,gDAAgD,CAAA;AAAA,IAC3E;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,cAAA,GAAiB,IAAA,CAAK,WAAA,EAAa,MAAA,IAAU,CAAA;AACnD,IAAA,IAAI,cAAc,cAAA,EAAgB;AAChC,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,iDAAA;AAAA,QACA,GAAA;AAAA,QACA;AAAA,OACF;AAAA,IACF;AAEA,IAAA,IAAI,aAAa,CAAA,EAAG;AAClB,MAAA,OAAO,KAAA,CAAM,CAAC,CAAA,EAAG,SAAS,GAAG,EAAE,KAAA,EAAO,WAAW,CAAA;AAAA,IACnD;AAGA,IAAA,MAAM,MAAA,GAAS,IAAI,KAAA,CAAgB,QAAQ,CAAA;AAC3C,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAA,CAAO,CAAC,CAAA,GAAI,IAAI,KAAA,CAAc,SAAS,CAAA;AAAA,IACzC;AAGA,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,QAAA,MAAM,GAAA,GAAM,WAAA,CAAY,CAAA,EAAG,CAAA,EAAG,CAAC,CAAA;AAC/B,QAAA,MAAM,GAAA,GAAM,IAAA,CAAK,gBAAA,GAAmB,CAAC,CAAA;AACrC,QAAA,IAAI,CAAC,GAAA,EAAK;AACR,UAAA,MAAM,IAAI,aAAa,0DAA0D,CAAA;AAAA,QACnF;AAGA,QAAA,MAAM,GAAA,GAAM,GAAA,CAAI,GAAA,CAAI,GAAG,CAAA;AACvB,QAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,QAAA,IAAI,CAAC,GAAA,EAAK;AACR,UAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,QACnE;AACA,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,IAAI,IAAA,CAAK,kBAAkB,iBAAA,EAAmB;AAC5C,YAAA,GAAA,CAAI,CAAC,IAAI,IAAA,CAAK,YAAA;AACd,YAAA;AAAA,UACF;AACA,UAAA,MAAM,IAAI,qBAAA;AAAA,YACR,CAAA,kBAAA,EAAqB,MAAA,CAAO,GAAG,CAAC,eAAe,CAAC,CAAA,CAAA;AAAA,YAChD,GAAA;AAAA,YACA;AAAA,WACF;AAAA,QACF;AAEA,QAAA,GAAA,CAAI,CAAC,CAAA,GAAI,GAAA;AAAA,MACX;AAAA,IACF;AAEA,IAAA,OAAO,MAAA,CAAO,MAAA,EAAQ,EAAE,KAAA,EAAO,WAAW,CAAA;AAAA,EAC5C;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EASA,aAAa,CAAA,EAAmB;AAC9B,IAAA,OAAO,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,CAAE,UAAU,CAAC,CAAA;AAAA,EAChC;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAWA,iBAAiB,CAAA,EAAmB;AAClC,IAAA,IAAI,CAAC,KAAK,MAAA,EAAQ;AAChB,MAAA,MAAM,IAAI,eAAe,wDAAwD,CAAA;AAAA,IACnF;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,cAAA,GAAiB,IAAA,CAAK,WAAA,EAAa,MAAA,IAAU,CAAA;AACnD,IAAA,IAAI,cAAc,cAAA,EAAgB;AAChC,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,iDAAA;AAAA,QACA,GAAA;AAAA,QACA;AAAA,OACF;AAAA,IACF;AAEA,IAAA,IAAI,QAAA,KAAa,CAAA,IAAK,SAAA,KAAc,CAAA,EAAG;AACrC,MAAA,MAAM,YAAA,GAAe,IAAA,CAAK,WAAA,IAAe,EAAC;AAC1C,MAAA,MAAM,YAAA,GAAe,yBAAA,CAA0B,YAAA,EAAc,GAAG,CAAA;AAChE,MAAA,IAAI,iBAAiB,QAAA,EAAU;AAC7B,QAAA,OAAO,KAAA,CAAM,CAAC,CAAA,EAAG,SAAS,GAAG,EAAE,KAAA,EAAO,UAAU,CAAA;AAAA,MAClD;AACA,MAAA,IAAI,iBAAiB,QAAA,EAAU;AAC7B,QAAA,OAAO,KAAA,CAAM,CAAC,CAAA,EAAG,SAAS,GAAG,EAAE,KAAA,EAAO,SAAS,CAAA;AAAA,MACjD;AACA,MAAA,OAAO,KAAA,CAAM,CAAC,CAAA,EAAG,SAAS,GAAG,EAAE,KAAA,EAAO,WAAW,CAAA;AAAA,IACnD;AAGA,IAAA,MAAM,MAAA,GAAS,IAAI,KAAA,CAAkB,QAAQ,CAAA;AAC7C,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAA,CAAO,CAAC,CAAA,GAAI,IAAI,KAAA,CAAgB,SAAS,CAAA;AAAA,IAC3C;AAGA,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AACzC,IAAA,MAAM,IAAA,GAAO,eAAe,CAAC,CAAA;AAC7B,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,QAAA,MAAM,MAAM,IAAA,CAAK,CAAA,CAAE,SAAS,CAAA,GAAI,OAAA,GAAU,IAAI,OAAO,CAAA;AACrD,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,MAAM,GAAA,GAAM,OAAO,GAAG,CAAA;AACtB,QAAA,MAAM,cAAA,GACJ,IAAA,CAAK,aAAA,KAAkB,iBAAA,KACtB,OAAO,KAAA,CAAM,GAAG,CAAA,GAAI,MAAA,CAAO,KAAA,CAAM,IAAA,CAAK,YAAY,CAAA,GAAI,QAAQ,IAAA,CAAK,YAAA,CAAA;AACtE,QAAA,IAAI,cAAA,EAAgB;AAClB,UAAA,MAAM,IAAI,qBAAA;AAAA,YACR,gDAAA;AAAA,YACA,GAAA;AAAA,YACA;AAAA,WACF;AAAA,QACF;AACA,QAAA,MAAM,IAAA,GAAO,IAAA,CAAK,WAAA,GAAc,CAAC,CAAA;AAGjC,QAAA,IAAI,CAAC,IAAA,IAAQ,GAAA,GAAM,CAAA,IAAK,GAAA,IAAO,IAAA,CAAK,MAAA,IAAU,CAAC,MAAA,CAAO,SAAA,CAAU,GAAG,CAAA,EAAG;AACpE,UAAA,MAAM,IAAI,qBAAA;AAAA,YACR,CAAA,uBAAA,EAA0B,GAAG,CAAA,aAAA,EAAgB,CAAC,6BAA6B,IAAA,EAAM,MAAA,IAAU,KAAK,CAAC,CAAA,CAAA,CAAA;AAAA,YACjG,GAAA;AAAA,YACA;AAAA,WACF;AAAA,QACF;AAEA,QAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,QAAA,IAAI,CAAC,GAAA,EAAK;AACR,UAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,QACnE;AACA,QAAA,MAAM,MAAA,GAAS,KAAK,GAAG,CAAA;AACvB,QAAA,IAAI,WAAW,MAAA,EAAW;AACxB,UAAA,MAAM,IAAI,aAAa,wCAAwC,CAAA;AAAA,QACjE;AACA,QAAA,GAAA,CAAI,CAAC,CAAA,GAAI,MAAA;AAAA,MACX;AAAA,IACF;AAEA,IAAA,OAAO,sBAAA,CAAuB,QAAQ,GAAG,CAAA;AAAA,EAC3C;AACF;AAyBO,IAAM,iBAAN,MAAqB;AAAA;AAAA,EAElB,MAAA,GAAS,KAAA;AAAA;AAAA,EAET,QAAA;AAAA;AAAA,EAEA,aAAA;AAAA;AAAA,EAEA,QAAA;AAAA;AAAA,EAEA,QAAA;AAAA;AAAA,EAEA,MAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAWR,WAAA,CACE,OAAA,GAKI,EAAC,EACL;AACA,IAAA,IAAA,CAAK,QAAA,GAAW,QAAQ,QAAA,IAAY,CAAA;AACpC,IAAA,IAAA,CAAK,QAAA,GAAW,QAAQ,QAAA,IAAY,CAAA;AACpC,IAAA,MAAM,YAAA,GAAe,OAAA,CAAQ,MAAA,IAAU,OAAA,CAAQ,YAAA,IAAgB,KAAA;AAC/D,IAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,IAAA,CAAK,QAAQ,CAAA,IAAK,CAAC,MAAA,CAAO,QAAA,CAAS,IAAA,CAAK,QAAQ,CAAA,EAAG;AACtE,MAAA,MAAM,IAAI,qBAAA,CAAsB,8CAAA,EAAgD,UAAU,CAAA;AAAA,IAC5F;AACA,IAAA,IAAI,IAAA,CAAK,QAAA,IAAY,IAAA,CAAK,QAAA,EAAU;AAClC,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,wCAAA;AAAA,QACA,UAAA;AAAA,QACA,IAAA,CAAK;AAAA,OACP;AAAA,IACF;AACA,IAAA,IAAI,OAAA,CAAQ,MAAA,KAAW,MAAA,IAAa,OAAA,CAAQ,iBAAiB,MAAA,EAAW;AACtE,MAAA,IAAI,OAAA,CAAQ,MAAA,KAAW,OAAA,CAAQ,YAAA,EAAc;AAC3C,QAAA,MAAM,IAAI,qBAAA;AAAA,UACR,2DAAA;AAAA,UACA,QAAA;AAAA,UACA,OAAA,CAAQ;AAAA,SACV;AAAA,MACF;AAAA,IACF;AACA,IAAA,IAAI,OAAO,iBAAiB,SAAA,EAAW;AACrC,MAAA,MAAM,IAAI,qBAAA,CAAsB,0BAAA,EAA4B,QAAA,EAAU,YAAY,CAAA;AAAA,IACpF;AACA,IAAA,IAAI,YAAA,IAAgB,IAAA,CAAK,QAAA,KAAa,CAAA,EAAG;AACvC,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,yCAAA;AAAA,QACA,UAAA;AAAA,QACA,IAAA,CAAK;AAAA,OACP;AAAA,IACF;AACA,IAAA,IAAA,CAAK,MAAA,GAAS,YAAA;AAAA,EAChB;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAUA,IAAI,CAAA,EAAiB;AACnB,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,MAAA,MAAM,IAAI,qBAAA,CAAsB,0CAAA,EAA4C,GAAG,CAAA;AAAA,IACjF;AAGA,IAAA,MAAM,SAAA,uBAAgB,GAAA,EAAc;AACpC,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,MAAM,CAAA,EAAA,EAAK;AAC/B,MAAA,SAAA,CAAU,GAAA,CAAI,WAAA,CAAY,CAAA,EAAG,CAAC,CAAC,CAAA;AAAA,IACjC;AAGA,IAAA,IAAA,CAAK,QAAA,GAAW,cAAA,CAAe,SAAA,EAAW,GAAG,CAAA;AAC7C,IAAA,IAAA,CAAK,aAAA,uBAAoB,GAAA,EAAI;AAC7B,IAAA,KAAA,IAAS,IAAI,CAAA,EAAG,CAAA,GAAI,IAAA,CAAK,QAAA,CAAS,QAAQ,CAAA,EAAA,EAAK;AAC7C,MAAA,IAAA,CAAK,aAAA,CAAc,IAAI,eAAA,CAAgB,IAAA,CAAK,UAAU,CAAA,EAAG,oBAAoB,GAAG,CAAC,CAAA;AAAA,IACnF;AAEA,IAAA,IAAA,CAAK,MAAA,GAAS,IAAA;AACd,IAAA,OAAO,IAAA;AAAA,EACT;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAWA,UAAU,CAAA,EAA+B;AACvC,IAAA,IAAI,CAAC,KAAK,MAAA,EAAQ;AAChB,MAAA,MAAM,IAAI,eAAe,gDAAgD,CAAA;AAAA,IAC3E;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,MAAA,MAAMC,SAAAA,GAAW,IAAA,CAAK,QAAA,EAAU,MAAA,IAAU,CAAA;AAC1C,MAAA,OAAO,IAAA,CAAK,MAAA,GACR,SAAA,CAAU,OAAA,CAAQ;AAAA,QAChB,IAAA,EAAM,CAAA;AAAA,QACN,IAAA,EAAMA,SAAAA;AAAA,QACN,UAAA,EAAY,IAAI,UAAA,CAAW,CAAC,CAAA;AAAA,QAC5B,UAAA,EAAY,IAAI,UAAA,CAAW,CAAC,CAAA;AAAA,QAC5B,MAAA,EAAQ,IAAI,YAAA,CAAa,CAAC;AAAA,OAC3B,CAAA,GACD,KAAA,CAAM,CAAC,CAAA,EAAGA,SAAQ,CAAA,EAAG,EAAE,KAAA,EAAO,SAAA,EAAW,CAAA;AAAA,IAC/C;AAEA,IAAA,MAAM,WAAW,CAAA,CAAE,IAAA;AACnB,IAAA,MAAM,QAAA,GAAW,IAAA,CAAK,QAAA,EAAU,MAAA,IAAU,CAAA;AAC1C,IAAA,MAAM,SAAS,IAAA,CAAK,aAAA;AACpB,IAAA,IAAI,CAAC,MAAA,EAAQ;AACX,MAAA,MAAM,IAAI,aAAa,sDAAsD,CAAA;AAAA,IAC/E;AAEA,IAAA,IAAI,KAAK,MAAA,EAAQ;AACf,MAAA,MAAM,SAAmB,EAAC;AAC1B,MAAA,MAAM,SAAmB,EAAC;AAC1B,MAAA,MAAM,OAAiB,EAAC;AAExB,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,QAAA,MAAM,GAAA,GAAM,WAAA,CAAY,CAAA,EAAG,CAAC,CAAA;AAC5B,QAAA,MAAM,GAAA,GAAM,MAAA,CAAO,GAAA,CAAI,GAAG,CAAA;AAE1B,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,qBAAA;AAAA,YACR,CAAA,eAAA,EAAkB,MAAA,CAAO,GAAG,CAAC,CAAA,mCAAA,CAAA;AAAA,YAC7B,GAAA;AAAA,YACA;AAAA,WACF;AAAA,QACF;AAEA,QAAA,MAAA,CAAO,KAAK,CAAC,CAAA;AACb,QAAA,MAAA,CAAO,KAAK,GAAG,CAAA;AACf,QAAA,IAAA,CAAK,IAAA,CAAK,KAAK,QAAQ,CAAA;AAAA,MACzB;AAEA,MAAA,OAAO,UAAU,OAAA,CAAQ;AAAA,QACvB,IAAA,EAAM,QAAA;AAAA,QACN,IAAA,EAAM,QAAA;AAAA,QACN,UAAA,EAAY,UAAA,CAAW,IAAA,CAAK,MAAM,CAAA;AAAA,QAClC,UAAA,EAAY,UAAA,CAAW,IAAA,CAAK,MAAM,CAAA;AAAA,QAClC,MAAA,EAAQ,YAAA,CAAa,IAAA,CAAK,IAAI;AAAA,OAC/B,CAAA;AAAA,IACH;AAGA,IAAA,MAAM,MAAA,GAAS,IAAI,KAAA,CAAgB,QAAQ,CAAA;AAC3C,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAA,CAAO,CAAC,IAAI,IAAI,KAAA,CAAc,QAAQ,CAAA,CAAE,IAAA,CAAK,KAAK,QAAQ,CAAA;AAAA,IAC5D;AAGA,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAM,GAAA,GAAM,WAAA,CAAY,CAAA,EAAG,CAAC,CAAA;AAC5B,MAAA,MAAM,GAAA,GAAM,MAAA,CAAO,GAAA,CAAI,GAAG,CAAA;AAE1B,MAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,QAAA,MAAM,IAAI,qBAAA;AAAA,UACR,CAAA,eAAA,EAAkB,MAAA,CAAO,GAAG,CAAC,CAAA,mCAAA,CAAA;AAAA,UAC7B,GAAA;AAAA,UACA;AAAA,SACF;AAAA,MACF;AAEA,MAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,MAAA,IAAI,CAAC,GAAA,EAAK;AACR,QAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,MACnE;AACA,MAAA,GAAA,CAAI,GAAG,IAAI,IAAA,CAAK,QAAA;AAAA,IAClB;AAEA,IAAA,OAAO,MAAA,CAAO,MAAA,EAAQ,EAAE,KAAA,EAAO,WAAW,CAAA;AAAA,EAC5C;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EASA,aAAa,CAAA,EAA+B;AAC1C,IAAA,OAAO,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,CAAE,UAAU,CAAC,CAAA;AAAA,EAChC;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAWA,iBAAiB,CAAA,EAA+B;AAC9C,IAAA,IAAI,CAAC,KAAK,MAAA,EAAQ;AAChB,MAAA,MAAM,IAAI,eAAe,wDAAwD,CAAA;AAAA,IACnF;AACA,IAAA,IAAI,aAAa,SAAA,EAAW;AAC1B,MAAA,IAAI,IAAA,CAAK,aAAa,CAAA,EAAG;AACvB,QAAA,MAAM,IAAI,qBAAA;AAAA,UACR,oDAAA;AAAA,UACA,UAAA;AAAA,UACA,IAAA,CAAK;AAAA,SACP;AAAA,MACF;AACA,MAAA,MAAM,CAAC,IAAA,EAAM,IAAI,CAAA,GAAI,CAAA,CAAE,KAAA;AACvB,MAAA,IAAI,IAAA,KAAS,MAAA,IAAa,IAAA,KAAS,MAAA,EAAW;AAC5C,QAAA,MAAM,IAAI,WAAW,yBAAyB,CAAA;AAAA,MAChD;AACA,MAAA,MAAMA,SAAAA,GAAW,IAAA,CAAK,QAAA,EAAU,MAAA,IAAU,CAAA;AAC1C,MAAA,IAAI,SAASA,SAAAA,EAAU;AACrB,QAAA,MAAM,IAAI,qBAAA,CAAsB,+CAAA,EAAiD,GAAA,EAAK,IAAI,CAAA;AAAA,MAC5F;AACA,MAAA,MAAMC,WAAU,IAAA,CAAK,QAAA;AACrB,MAAA,IAAI,CAACA,QAAAA,EAAS;AACZ,QAAA,MAAM,IAAI,aAAa,uDAAuD,CAAA;AAAA,MAChF;AACA,MAAA,IAAI,SAAS,CAAA,EAAG;AACd,QAAA,OAAO,8BAAA,CAA+BA,UAAS,GAAG,CAAA;AAAA,MACpD;AAEA,MAAA,MAAMC,OAAAA,GAAS,IAAI,KAAA,CAAgB,IAAI,CAAA;AACvC,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,QAAA,IAAI,MAAA,GAAS,CAAA;AACb,QAAA,IAAI,SAAS,IAAA,CAAK,QAAA;AAClB,QAAA,MAAM,KAAA,GAAQ,CAAA,CAAE,MAAA,CAAO,CAAC,CAAA,IAAK,CAAA;AAC7B,QAAA,MAAM,GAAA,GAAM,CAAA,CAAE,MAAA,CAAO,CAAA,GAAI,CAAC,CAAA,IAAK,KAAA;AAC/B,QAAA,KAAA,IAAS,CAAA,GAAI,KAAA,EAAO,CAAA,GAAI,GAAA,EAAK,CAAA,EAAA,EAAK;AAChC,UAAA,MAAM,GAAA,GAAM,CAAA,CAAE,OAAA,CAAQ,CAAC,CAAA;AACvB,UAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,YAAA,MAAM,IAAI,aAAa,6CAA6C,CAAA;AAAA,UACtE;AACA,UAAA,IAAI,GAAA,GAAM,CAAA,IAAK,GAAA,IAAOF,SAAAA,EAAU;AAC9B,YAAA,MAAM,IAAI,qBAAA;AAAA,cACR,+CAAA;AAAA,cACA,GAAA;AAAA,cACA;AAAA,aACF;AAAA,UACF;AACA,UAAA,MAAM,GAAA,GAAM,CAAA,CAAE,IAAA,CAAK,CAAC,CAAA;AACpB,UAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,YAAA,MAAM,IAAI,aAAa,sCAAsC,CAAA;AAAA,UAC/D;AACA,UAAA,MAAM,GAAA,GAAM,OAAO,GAAG,CAAA;AACtB,UAAA,IAAI,MAAM,MAAA,EAAQ;AAChB,YAAA,MAAA,GAAS,GAAA;AACT,YAAA,MAAA,GAAS,GAAA;AAAA,UACX;AAAA,QACF;AAEA,QAAA,IAAI,MAAA,IAAU,KAAK,QAAA,EAAU;AAC3B,UAAA,MAAM,IAAI,qBAAA;AAAA,YACR,oCAAoC,CAAC,CAAA,kDAAA,CAAA;AAAA,YACrC;AAAA,WACF;AAAA,QACF;AACA,QAAAE,QAAO,CAAC,CAAA,GAAI,eAAA,CAAgBD,QAAAA,EAAS,QAAQ,iCAAiC,CAAA;AAAA,MAChF;AACA,MAAA,OAAO,sBAAA,CAAuBC,SAAQ,GAAG,CAAA;AAAA,IAC3C;AAEA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,MAAM,CAAC,QAAA,EAAU,KAAK,CAAA,GAAI,WAAW,CAAC,CAAA;AAEtC,IAAA,MAAM,QAAA,GAAW,IAAA,CAAK,QAAA,EAAU,MAAA,IAAU,CAAA;AAC1C,IAAA,IAAI,UAAU,QAAA,EAAU;AACtB,MAAA,MAAM,IAAI,qBAAA,CAAsB,+CAAA,EAAiD,GAAA,EAAK,KAAK,CAAA;AAAA,IAC7F;AACA,IAAA,MAAM,UAAU,IAAA,CAAK,QAAA;AACrB,IAAA,IAAI,CAAC,OAAA,EAAS;AACZ,MAAA,MAAM,IAAI,aAAa,uDAAuD,CAAA;AAAA,IAChF;AACA,IAAA,IAAI,aAAa,CAAA,EAAG;AAClB,MAAA,OAAO,8BAAA,CAA+B,SAAS,GAAG,CAAA;AAAA,IACpD;AACA,IAAA,MAAM,MAAA,GAAS,IAAI,KAAA,CAAgB,QAAQ,CAAA;AAC3C,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AACzC,IAAA,MAAM,IAAA,GAAO,eAAe,CAAC,CAAA;AAG7B,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,IAAI,MAAA,GAAS,CAAA;AACb,MAAA,MAAM,OAAA,GAAU,CAAA,CAAE,MAAA,GAAS,CAAA,GAAI,OAAA;AAC/B,MAAA,MAAM,KAAA,GAAQ,KAAK,OAAO,CAAA;AAC1B,MAAA,IAAI,UAAU,MAAA,EAAW;AACvB,QAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,MAC9E;AACA,MAAA,IAAI,MAAA,GAAS,OAAO,KAAK,CAAA;AAGzB,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,KAAA,EAAO,CAAA,EAAA,EAAK;AAC9B,QAAA,MAAM,GAAA,GAAM,IAAA,CAAK,OAAA,GAAU,CAAA,GAAI,OAAO,CAAA;AACtC,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,MAAM,GAAA,GAAM,OAAO,GAAG,CAAA;AACtB,QAAA,IAAI,MAAM,MAAA,EAAQ;AAChB,UAAA,MAAA,GAAS,GAAA;AACT,UAAA,MAAA,GAAS,CAAA;AAAA,QACX;AAAA,MACF;AAEA,MAAA,IAAI,MAAA,IAAU,KAAK,QAAA,EAAU;AAC3B,QAAA,MAAM,IAAI,qBAAA;AAAA,UACR,oCAAoC,CAAC,CAAA,kDAAA,CAAA;AAAA,UACrC;AAAA,SACF;AAAA,MACF;AAEA,MAAA,MAAA,CAAO,CAAC,CAAA,GAAI,eAAA,CAAgB,OAAA,EAAS,QAAQ,iCAAiC,CAAA;AAAA,IAChF;AAGA,IAAA,OAAO,sBAAA,CAAuB,QAAQ,GAAG,CAAA;AAAA,EAC3C;AACF;AA0BO,IAAM,sBAAN,MAA0B;AAAA;AAAA,EAEvB,MAAA,GAAS,KAAA;AAAA;AAAA,EAET,QAAA;AAAA;AAAA,EAEA,aAAA;AAAA;AAAA,EAEA,MAAA;AAAA;AAAA,EAEA,aAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAUR,WAAA,CACE,OAAA,GAII,EAAC,EACL;AACA,IAAA,MAAM,YAAA,GAAe,OAAA,CAAQ,MAAA,IAAU,OAAA,CAAQ,YAAA,IAAgB,KAAA;AAC/D,IAAA,IAAI,OAAA,CAAQ,MAAA,KAAW,MAAA,IAAa,OAAA,CAAQ,iBAAiB,MAAA,EAAW;AACtE,MAAA,IAAI,OAAA,CAAQ,MAAA,KAAW,OAAA,CAAQ,YAAA,EAAc;AAC3C,QAAA,MAAM,IAAI,qBAAA;AAAA,UACR,2DAAA;AAAA,UACA,QAAA;AAAA,UACA,OAAA,CAAQ;AAAA,SACV;AAAA,MACF;AAAA,IACF;AACA,IAAA,IAAA,CAAK,MAAA,GAAS,YAAA;AACd,IAAA,IAAI,OAAO,IAAA,CAAK,MAAA,KAAW,SAAA,EAAW;AACpC,MAAA,MAAM,IAAI,qBAAA,CAAsB,0BAAA,EAA4B,QAAA,EAAU,KAAK,MAAM,CAAA;AAAA,IACnF;AACA,IAAA,IAAI,OAAA,CAAQ,YAAY,MAAA,EAAW;AACjC,MAAA,IAAA,CAAK,aAAA,GAAgB,sBAAA,CAAuB,OAAA,CAAQ,OAAA,EAAS,SAAS,CAAA;AAAA,IACxE;AAAA,EACF;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAUA,IAAI,CAAA,EAAiD;AACnD,IAAA,IAAI,CAAA,CAAE,WAAW,CAAA,EAAG;AAClB,MAAA,MAAM,IAAI,qBAAA,CAAsB,+CAAA,EAAiD,GAAG,CAAA;AAAA,IACtF;AACA,IAAA,KAAA,MAAW,UAAU,CAAA,EAAG;AACtB,MAAA,IAAI,CAAC,KAAA,CAAM,OAAA,CAAQ,MAAM,CAAA,EAAG;AAC1B,QAAA,MAAM,IAAI,qBAAA,CAAsB,0CAAA,EAA4C,GAAA,EAAK,MAAM,CAAA;AAAA,MACzF;AACA,MAAA,KAAA,MAAW,SAAS,MAAA,EAAQ;AAC1B,QAAA,IAAI,OAAO,UAAU,QAAA,IAAY,OAAO,UAAU,QAAA,IAAY,OAAO,UAAU,QAAA,EAAU;AACvF,UAAA,MAAM,IAAI,qBAAA;AAAA,YACR,iEAAA;AAAA,YACA,GAAA;AAAA,YACA;AAAA,WACF;AAAA,QACF;AAAA,MACF;AAAA,IACF;AAEA,IAAA,IAAI,IAAA,CAAK,aAAA,IAAiB,IAAA,CAAK,aAAA,CAAc,WAAW,CAAA,EAAG;AACzD,MAAA,MAAM,IAAI,qBAAA,CAAsB,yCAAA,EAA2C,SAAS,CAAA;AAAA,IACtF;AAEA,IAAA,IAAI,KAAK,aAAA,EAAe;AACtB,MAAA,IAAA,CAAK,QAAA,GAAW,KAAA,CAAM,IAAA,CAAK,IAAA,CAAK,aAAa,CAAA;AAAA,IAC/C,CAAA,MAAO;AAEL,MAAA,MAAM,SAAA,uBAAgB,GAAA,EAAc;AACpC,MAAA,KAAA,MAAW,UAAU,CAAA,EAAG;AACtB,QAAA,KAAA,MAAW,SAAS,MAAA,EAAQ;AAC1B,UAAA,SAAA,CAAU,IAAI,KAAK,CAAA;AAAA,QACrB;AAAA,MACF;AAGA,MAAA,IAAA,CAAK,QAAA,GAAW,cAAA,CAAe,SAAA,EAAW,GAAG,CAAA;AAAA,IAC/C;AACA,IAAA,IAAA,CAAK,aAAA,uBAAoB,GAAA,EAAI;AAC7B,IAAA,KAAA,IAAS,IAAI,CAAA,EAAG,CAAA,GAAI,IAAA,CAAK,QAAA,CAAS,QAAQ,CAAA,EAAA,EAAK;AAC7C,MAAA,IAAA,CAAK,aAAA,CAAc,IAAI,eAAA,CAAgB,IAAA,CAAK,UAAU,CAAA,EAAG,yBAAyB,GAAG,CAAC,CAAA;AAAA,IACxF;AACA,IAAA,IAAI,KAAK,aAAA,EAAe;AACtB,MAAA,KAAA,MAAW,UAAU,CAAA,EAAG;AACtB,QAAA,KAAA,MAAW,SAAS,MAAA,EAAQ;AAC1B,UAAA,IAAI,CAAC,IAAA,CAAK,aAAA,CAAc,GAAA,CAAI,KAAK,CAAA,EAAG;AAClC,YAAA,MAAM,IAAI,qBAAA;AAAA,cACR,CAAA,eAAA,EAAkB,MAAA,CAAO,KAAK,CAAC,CAAA,mCAAA,CAAA;AAAA,cAC/B,GAAA;AAAA,cACA;AAAA,aACF;AAAA,UACF;AAAA,QACF;AAAA,MACF;AAAA,IACF;AAEA,IAAA,IAAA,CAAK,MAAA,GAAS,IAAA;AACd,IAAA,OAAO,IAAA;AAAA,EACT;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAWA,UAAU,CAAA,EAA+D;AACvE,IAAA,IAAI,CAAC,KAAK,MAAA,EAAQ;AAChB,MAAA,MAAM,IAAI,eAAe,qDAAqD,CAAA;AAAA,IAChF;AACA,IAAA,KAAA,MAAW,UAAU,CAAA,EAAG;AACtB,MAAA,IAAI,CAAC,KAAA,CAAM,OAAA,CAAQ,MAAM,CAAA,EAAG;AAC1B,QAAA,MAAM,IAAI,qBAAA,CAAsB,0CAAA,EAA4C,GAAA,EAAK,MAAM,CAAA;AAAA,MACzF;AACA,MAAA,KAAA,MAAW,SAAS,MAAA,EAAQ;AAC1B,QAAA,IAAI,OAAO,UAAU,QAAA,IAAY,OAAO,UAAU,QAAA,IAAY,OAAO,UAAU,QAAA,EAAU;AACvF,UAAA,MAAM,IAAI,qBAAA;AAAA,YACR,iEAAA;AAAA,YACA,GAAA;AAAA,YACA;AAAA,WACF;AAAA,QACF;AAAA,MACF;AAAA,IACF;AACA,IAAA,IAAI,CAAA,CAAE,WAAW,CAAA,EAAG;AAClB,MAAA,MAAMF,SAAAA,GAAW,IAAA,CAAK,QAAA,EAAU,MAAA,IAAU,CAAA;AAC1C,MAAA,OAAO,IAAA,CAAK,MAAA,GACR,SAAA,CAAU,OAAA,CAAQ;AAAA,QAChB,IAAA,EAAM,CAAA;AAAA,QACN,IAAA,EAAMA,SAAAA;AAAA,QACN,UAAA,EAAY,IAAI,UAAA,CAAW,CAAC,CAAA;AAAA,QAC5B,UAAA,EAAY,IAAI,UAAA,CAAW,CAAC,CAAA;AAAA,QAC5B,MAAA,EAAQ,IAAI,YAAA,CAAa,CAAC;AAAA,OAC3B,CAAA,GACD,KAAA,CAAM,CAAC,CAAA,EAAGA,SAAQ,CAAA,EAAG,EAAE,KAAA,EAAO,SAAA,EAAW,CAAA;AAAA,IAC/C;AAEA,IAAA,MAAM,WAAW,CAAA,CAAE,MAAA;AACnB,IAAA,MAAM,QAAA,GAAW,IAAA,CAAK,QAAA,EAAU,MAAA,IAAU,CAAA;AAC1C,IAAA,MAAM,SAAS,IAAA,CAAK,aAAA;AACpB,IAAA,IAAI,CAAC,MAAA,EAAQ;AACX,MAAA,MAAM,IAAI,aAAa,2DAA2D,CAAA;AAAA,IACpF;AAEA,IAAA,IAAI,KAAK,MAAA,EAAQ;AACf,MAAA,MAAM,SAAmB,EAAC;AAC1B,MAAA,MAAM,SAAmB,EAAC;AAC1B,MAAA,MAAM,OAAiB,EAAC;AAExB,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,QAAA,MAAM,IAAA,GAAO,EAAE,CAAC,CAAA;AAChB,QAAA,IAAI,CAAC,IAAA,EAAM;AACX,QAAA,MAAM,IAAA,uBAAW,GAAA,EAAY;AAC7B,QAAA,KAAA,MAAW,SAAS,IAAA,EAAM;AACxB,UAAA,MAAM,GAAA,GAAM,MAAA,CAAO,GAAA,CAAI,KAAK,CAAA;AAC5B,UAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,YAAA,MAAM,IAAI,qBAAA;AAAA,cACR,CAAA,eAAA,EAAkB,MAAA,CAAO,KAAK,CAAC,CAAA,mCAAA,CAAA;AAAA,cAC/B,GAAA;AAAA,cACA;AAAA,aACF;AAAA,UACF;AACA,UAAA,IAAI,IAAA,CAAK,GAAA,CAAI,GAAG,CAAA,EAAG;AACnB,UAAA,IAAA,CAAK,IAAI,GAAG,CAAA;AACZ,UAAA,MAAA,CAAO,KAAK,CAAC,CAAA;AACb,UAAA,MAAA,CAAO,KAAK,GAAG,CAAA;AACf,UAAA,IAAA,CAAK,KAAK,CAAC,CAAA;AAAA,QACb;AAAA,MACF;AAEA,MAAA,OAAO,UAAU,OAAA,CAAQ;AAAA,QACvB,IAAA,EAAM,QAAA;AAAA,QACN,IAAA,EAAM,QAAA;AAAA,QACN,UAAA,EAAY,UAAA,CAAW,IAAA,CAAK,MAAM,CAAA;AAAA,QAClC,UAAA,EAAY,UAAA,CAAW,IAAA,CAAK,MAAM,CAAA;AAAA,QAClC,MAAA,EAAQ,YAAA,CAAa,IAAA,CAAK,IAAI;AAAA,OAC/B,CAAA;AAAA,IACH;AAGA,IAAA,MAAM,MAAA,GAAS,IAAI,KAAA,CAAgB,QAAQ,CAAA;AAC3C,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAA,CAAO,CAAC,CAAA,GAAI,IAAI,MAAc,QAAQ,CAAA,CAAE,KAAK,CAAC,CAAA;AAAA,IAChD;AAGA,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAM,IAAA,GAAO,EAAE,CAAC,CAAA;AAChB,MAAA,IAAI,CAAC,IAAA,EAAM;AAEX,MAAA,KAAA,MAAW,SAAS,IAAA,EAAM;AACxB,QAAA,MAAM,GAAA,GAAM,MAAA,CAAO,GAAA,CAAI,KAAK,CAAA;AAC5B,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,qBAAA;AAAA,YACR,CAAA,eAAA,EAAkB,MAAA,CAAO,KAAK,CAAC,CAAA,mCAAA,CAAA;AAAA,YAC/B,GAAA;AAAA,YACA;AAAA,WACF;AAAA,QACF;AAEA,QAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,QAAA,IAAI,CAAC,GAAA,EAAK;AACR,UAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,QACnE;AACA,QAAA,GAAA,CAAI,GAAG,CAAA,GAAI,CAAA;AAAA,MACb;AAAA,IACF;AAEA,IAAA,OAAO,MAAA,CAAO,MAAA,EAAQ,EAAE,KAAA,EAAO,WAAW,CAAA;AAAA,EAC5C;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EASA,aAAa,CAAA,EAA+D;AAC1E,IAAA,OAAO,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,CAAE,UAAU,CAAC,CAAA;AAAA,EAChC;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAWA,iBAAiB,CAAA,EAAqC;AACpD,IAAA,IAAI,CAAC,KAAK,MAAA,EAAQ;AAChB,MAAA,MAAM,IAAI,eAAe,6DAA6D,CAAA;AAAA,IACxF;AACA,IAAA,IAAI,aAAa,SAAA,EAAW;AAC1B,MAAA,MAAM,CAAC,IAAA,EAAM,IAAI,CAAA,GAAI,CAAA,CAAE,KAAA;AACvB,MAAA,IAAI,IAAA,KAAS,MAAA,IAAa,IAAA,KAAS,MAAA,EAAW;AAC5C,QAAA,MAAM,IAAI,WAAW,yBAAyB,CAAA;AAAA,MAChD;AACA,MAAA,MAAMG,cAAAA,GAAgB,IAAA,CAAK,QAAA,EAAU,MAAA,IAAU,CAAA;AAC/C,MAAA,IAAI,SAASA,cAAAA,EAAe;AAC1B,QAAA,MAAM,IAAI,qBAAA,CAAsB,+CAAA,EAAiD,GAAA,EAAK,IAAI,CAAA;AAAA,MAC5F;AACA,MAAA,IAAI,SAAS,CAAA,EAAG;AACd,QAAA,OAAO,EAAC;AAAA,MACV;AAEA,MAAA,MAAMF,WAAU,IAAA,CAAK,QAAA;AACrB,MAAA,IAAI,CAACA,QAAAA,EAAS;AACZ,QAAA,MAAM,IAAI,aAAa,4DAA4D,CAAA;AAAA,MACrF;AAEA,MAAA,MAAMC,UAAuB,EAAC;AAC9B,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,QAAA,MAAM,SAAqB,EAAC;AAC5B,QAAA,MAAM,KAAA,GAAQ,CAAA,CAAE,MAAA,CAAO,CAAC,CAAA,IAAK,CAAA;AAC7B,QAAA,MAAM,GAAA,GAAM,CAAA,CAAE,MAAA,CAAO,CAAA,GAAI,CAAC,CAAA,IAAK,KAAA;AAC/B,QAAA,KAAA,IAAS,CAAA,GAAI,KAAA,EAAO,CAAA,GAAI,GAAA,EAAK,CAAA,EAAA,EAAK;AAChC,UAAA,MAAM,GAAA,GAAM,CAAA,CAAE,OAAA,CAAQ,CAAC,CAAA;AACvB,UAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,YAAA,MAAM,IAAI,aAAa,6CAA6C,CAAA;AAAA,UACtE;AACA,UAAA,IAAI,GAAA,GAAM,CAAA,IAAK,GAAA,IAAOC,cAAAA,EAAe;AACnC,YAAA,MAAM,IAAI,qBAAA;AAAA,cACR,+CAAA;AAAA,cACA,GAAA;AAAA,cACA;AAAA,aACF;AAAA,UACF;AACA,UAAA,MAAM,GAAA,GAAM,CAAA,CAAE,IAAA,CAAK,CAAC,CAAA;AACpB,UAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,YAAA,MAAM,IAAI,aAAa,sCAAsC,CAAA;AAAA,UAC/D;AACA,UAAA,MAAM,KAAA,GAAQ,OAAO,GAAG,CAAA;AACxB,UAAA,IAAI,QAAQ,CAAA,EAAG;AACb,YAAA,MAAA,CAAO,IAAA,CAAK,eAAA,CAAgBF,QAAAA,EAAS,GAAA,EAAK,sCAAsC,CAAC,CAAA;AAAA,UACnF;AAAA,QACF;AACA,QAAAC,OAAAA,CAAO,KAAK,MAAM,CAAA;AAAA,MACpB;AACA,MAAA,OAAOA,OAAAA;AAAA,IACT;AAEA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,MAAM,QAAA,GAAW,CAAA,CAAE,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA;AAC/B,IAAA,MAAM,QAAA,GAAW,CAAA,CAAE,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA;AAC/B,IAAA,MAAM,aAAA,GAAgB,IAAA,CAAK,QAAA,EAAU,MAAA,IAAU,CAAA;AAC/C,IAAA,IAAI,aAAa,aAAA,EAAe;AAC9B,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,+CAAA;AAAA,QACA,GAAA;AAAA,QACA;AAAA,OACF;AAAA,IACF;AAEA,IAAA,IAAI,aAAa,CAAA,EAAG;AAClB,MAAA,OAAO,EAAC;AAAA,IACV;AAEA,IAAA,MAAM,UAAU,IAAA,CAAK,QAAA;AACrB,IAAA,IAAI,CAAC,OAAA,EAAS;AACZ,MAAA,MAAM,IAAI,aAAa,4DAA4D,CAAA;AAAA,IACrF;AAEA,IAAA,MAAM,SAAuB,EAAC;AAC9B,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AACzC,IAAA,MAAM,IAAA,GAAO,eAAe,CAAC,CAAA;AAG7B,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAM,SAAqB,EAAC;AAC5B,MAAA,MAAM,OAAA,GAAU,CAAA,CAAE,MAAA,GAAS,CAAA,GAAI,OAAA;AAG/B,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,QAAA,MAAM,GAAA,GAAM,IAAA,CAAK,OAAA,GAAU,CAAA,GAAI,OAAO,CAAA;AACtC,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,MAAM,GAAA,GAAM,OAAO,GAAG,CAAA;AAEtB,QAAA,IAAI,MAAM,CAAA,EAAG;AACX,UAAA,MAAA,CAAO,IAAA,CAAK,eAAA,CAAgB,OAAA,EAAS,CAAA,EAAG,sCAAsC,CAAC,CAAA;AAAA,QACjF;AAAA,MACF;AAEA,MAAA,MAAA,CAAO,KAAK,MAAM,CAAA;AAAA,IACpB;AAEA,IAAA,OAAO,MAAA;AAAA,EACT;AACF;;;ACpgEA,SAASE,eAAAA,CAAe,GAAW,IAAA,EAA0C;AAC3E,EAAA,IAAI,CAAA,CAAE,UAAU,QAAA,EAAU;AACxB,IAAA,MAAM,IAAI,UAAA,CAAW,CAAA,EAAG,IAAI,CAAA,gBAAA,CAAkB,CAAA;AAAA,EAChD;AACA,EAAA,IAAI,KAAA,CAAM,OAAA,CAAQ,CAAA,CAAE,IAAI,CAAA,EAAG;AACzB,IAAA,MAAM,IAAI,aAAa,gDAAgD,CAAA;AAAA,EACzE;AACA,EAAA,OAAO,CAAA,CAAE,IAAA;AACX;AAEA,SAAS,kBAAA,CAAmB,KAAA,EAAgB,IAAA,EAAc,YAAA,EAAgC;AACxF,EAAA,IAAI,UAAU,MAAA,EAAW;AACvB,IAAA,OAAO,YAAA;AAAA,EACT;AACA,EAAA,IAAI,OAAO,UAAU,SAAA,EAAW;AAC9B,IAAA,MAAM,IAAI,qBAAA,CAAsB,CAAA,EAAG,IAAI,CAAA,kBAAA,CAAA,EAAsB,MAAM,KAAK,CAAA;AAAA,EAC1E;AACA,EAAA,OAAO,KAAA;AACT;AAEA,SAAS,kBAAA,CAAmB,GAAW,IAAA,EAAoB;AACzD,EAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,EAAA,MAAM,IAAA,GAAOA,eAAAA,CAAe,CAAA,EAAG,IAAI,CAAA;AACnC,EAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AACzC,EAAA,IAAI,SAAA,GAAY,CAAA;AAEhB,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,IAAA,MAAM,OAAA,GAAU,CAAA,CAAE,MAAA,GAAS,CAAA,GAAI,OAAA;AAC/B,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,MAAA,MAAM,GAAA,GAAM,IAAA,CAAK,OAAA,GAAU,CAAA,GAAI,OAAO,CAAA;AACtC,MAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,QAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,MAC9E;AACA,MAAA,MAAM,GAAA,GAAM,OAAO,GAAG,CAAA;AACtB,MAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,GAAG,CAAA,EAAG;AACzB,QAAA,MAAM,IAAI,mBAAA,CAAoB,CAAA,EAAG,IAAI,CAAA,mCAAA,EAAsC,SAAS,CAAA,CAAE,CAAA;AAAA,MACxF;AACA,MAAA,SAAA,IAAa,CAAA;AAAA,IACf;AAAA,EACF;AACF;AAEA,SAAS,iBAAiB,KAAA,EAAuB;AAC/C,EAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,KAAK,GAAG,OAAO,KAAA;AACpC,EAAA,MAAM,OAAA,GAAU,IAAA,CAAK,KAAA,CAAM,KAAK,CAAA;AAChC,EAAA,IAAI,KAAK,GAAA,CAAI,KAAA,GAAQ,OAAO,CAAA,GAAI,OAAO,OAAO,OAAA;AAC9C,EAAA,MAAM,MAAA,GAAS,IAAA,CAAK,KAAA,CAAM,KAAA,GAAQ,IAAI,CAAA,GAAI,IAAA;AAC1C,EAAA,IAAI,KAAK,GAAA,CAAI,KAAA,GAAQ,MAAM,CAAA,GAAI,OAAO,OAAO,MAAA;AAC7C,EAAA,OAAO,KAAA;AACT;AAEA,SAAS,eAAe,CAAA,EAAmB;AACzC,EAAA,IAAI,CAAC,OAAO,QAAA,CAAS,CAAC,KAAK,CAAA,IAAK,CAAA,IAAK,KAAK,CAAA,EAAG;AAC3C,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,uDAAA;AAAA,MACA,GAAA;AAAA,MACA;AAAA,KACF;AAAA,EACF;AAEA,EAAA,MAAM,EAAA,GAAK,kBAAA;AACX,EAAA,MAAM,EAAA,GAAK,iBAAA;AACX,EAAA,MAAM,EAAA,GAAK,kBAAA;AACX,EAAA,MAAM,EAAA,GAAK,gBAAA;AACX,EAAA,MAAM,EAAA,GAAK,kBAAA;AACX,EAAA,MAAM,EAAA,GAAK,iBAAA;AAEX,EAAA,MAAM,EAAA,GAAK,kBAAA;AACX,EAAA,MAAM,EAAA,GAAK,iBAAA;AACX,EAAA,MAAM,EAAA,GAAK,kBAAA;AACX,EAAA,MAAM,EAAA,GAAK,iBAAA;AACX,EAAA,MAAM,EAAA,GAAK,kBAAA;AAEX,EAAA,MAAM,EAAA,GAAK,qBAAA;AACX,EAAA,MAAM,EAAA,GAAK,mBAAA;AACX,EAAA,MAAM,EAAA,GAAK,kBAAA;AACX,EAAA,MAAM,EAAA,GAAK,kBAAA;AACX,EAAA,MAAM,EAAA,GAAK,iBAAA;AACX,EAAA,MAAM,EAAA,GAAK,iBAAA;AAEX,EAAA,MAAM,EAAA,GAAK,oBAAA;AACX,EAAA,MAAM,EAAA,GAAK,kBAAA;AACX,EAAA,MAAM,EAAA,GAAK,iBAAA;AACX,EAAA,MAAM,EAAA,GAAK,iBAAA;AAEX,EAAA,MAAM,IAAA,GAAO,OAAA;AACb,EAAA,MAAM,QAAQ,CAAA,GAAI,IAAA;AAElB,EAAA,IAAI,IAAI,IAAA,EAAM;AACZ,IAAA,MAAMC,KAAI,IAAA,CAAK,IAAA,CAAK,KAAK,IAAA,CAAK,GAAA,CAAI,CAAC,CAAC,CAAA;AACpC,IAAA,OAAA,CAAA,CAAA,CAAA,CAAA,CACO,KAAKA,EAAAA,GAAI,EAAA,IAAMA,KAAI,EAAA,IAAMA,EAAAA,GAAI,MAAMA,EAAAA,GAAI,EAAA,IAAMA,EAAAA,GAAI,EAAA,KAAA,CAAA,CAAA,CAClD,KAAKA,EAAAA,GAAI,EAAA,IAAMA,KAAI,EAAA,IAAMA,EAAAA,GAAI,MAAMA,EAAAA,GAAI,CAAA,CAAA;AAAA,EAE/C;AACA,EAAA,IAAI,IAAI,KAAA,EAAO;AACb,IAAA,MAAMA,EAAAA,GAAI,KAAK,IAAA,CAAK,EAAA,GAAK,KAAK,GAAA,CAAI,CAAA,GAAI,CAAC,CAAC,CAAA;AACxC,IAAA,OAAO,OACA,EAAA,GAAKA,EAAAA,GAAI,MAAMA,EAAAA,GAAI,EAAA,IAAMA,KAAI,EAAA,IAAMA,EAAAA,GAAI,MAAMA,EAAAA,GAAI,EAAA,KAAA,CAAA,CAAA,CAClD,KAAKA,EAAAA,GAAI,EAAA,IAAMA,KAAI,EAAA,IAAMA,EAAAA,GAAI,MAAMA,EAAAA,GAAI,CAAA,CAAA,CAAA;AAAA,EAE/C;AAEA,EAAA,MAAM,IAAI,CAAA,GAAI,GAAA;AACd,EAAA,MAAM,IAAI,CAAA,GAAI,CAAA;AACd,EAAA,OAAA,CAAA,CAAA,CAAA,CAAA,CACQ,EAAA,GAAK,IAAI,EAAA,IAAM,CAAA,GAAI,MAAM,CAAA,GAAI,EAAA,IAAM,IAAI,EAAA,IAAM,CAAA,GAAI,MAAM,CAAA,IAAA,CAAA,CAAA,CAAA,CACxD,EAAA,GAAK,IAAI,EAAA,IAAM,CAAA,GAAI,MAAM,CAAA,GAAI,EAAA,IAAM,CAAA,GAAI,EAAA,IAAM,CAAA,GAAI,CAAA,CAAA;AAE1D;AAwBO,IAAM,iBAAN,MAAqB;AAAA,EAClB,MAAA,GAAS,KAAA;AAAA,EACT,KAAA;AAAA,EACA,MAAA;AAAA,EACA,QAAA;AAAA,EACA,OAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAUR,WAAA,CAAY,OAAA,GAAqE,EAAC,EAAG;AACnF,IAAA,IAAA,CAAK,QAAA,GAAW,kBAAA,CAAmB,OAAA,CAAQ,QAAA,EAAU,YAAY,IAAI,CAAA;AACrE,IAAA,IAAA,CAAK,OAAA,GAAU,kBAAA,CAAmB,OAAA,CAAQ,OAAA,EAAS,WAAW,IAAI,CAAA;AAClE,IAAA,kBAAA,CAAmB,OAAA,CAAQ,IAAA,EAAM,MAAA,EAAQ,IAAI,CAAA;AAAA,EAC/C;AAAA,EAEA,IAAI,CAAA,EAAiB;AACnB,IAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,MAAA,MAAM,IAAI,qBAAA,CAAsB,oCAAA,EAAsC,GAAG,CAAA;AAAA,IAC3E;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,kBAAA,CAAmB,GAAG,GAAG,CAAA;AACzB,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,IAAA,GAAOD,eAAAA,CAAe,CAAA,EAAG,GAAG,CAAA;AAClC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AAEzC,IAAA,IAAI,KAAA;AACJ,IAAA,IAAI,IAAA,CAAK,QAAA,IAAY,IAAA,CAAK,OAAA,EAAS;AACjC,MAAA,KAAA,GAAQ,IAAI,KAAA,CAAc,SAAS,CAAA,CAAE,KAAK,CAAC,CAAA;AAC3C,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,QAAA,IAAI,GAAA,GAAM,CAAA;AACV,QAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,UAAA,MAAM,MAAM,IAAA,CAAK,CAAA,CAAE,SAAS,CAAA,GAAI,OAAA,GAAU,IAAI,OAAO,CAAA;AACrD,UAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,YAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,UAC9E;AACA,UAAA,GAAA,IAAO,OAAO,GAAG,CAAA;AAAA,QACnB;AACA,QAAA,IAAI,KAAA,EAAO;AACT,UAAA,KAAA,CAAM,CAAC,IAAI,GAAA,GAAM,QAAA;AAAA,QACnB;AAAA,MACF;AAAA,IACF;AAEA,IAAA,IAAI,KAAK,OAAA,EAAS;AAChB,MAAA,MAAM,OAAO,IAAI,KAAA,CAAc,SAAS,CAAA,CAAE,KAAK,CAAC,CAAA;AAChD,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,QAAA,MAAM,IAAA,GAAO,KAAA,GAAS,KAAA,CAAM,CAAC,KAAK,CAAA,GAAK,CAAA;AACvC,QAAA,IAAI,KAAA,GAAQ,CAAA;AACZ,QAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,UAAA,MAAM,MAAM,IAAA,CAAK,CAAA,CAAE,SAAS,CAAA,GAAI,OAAA,GAAU,IAAI,OAAO,CAAA;AACrD,UAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,YAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,UAC9E;AACA,UAAA,MAAM,GAAA,GAAM,MAAA,CAAO,GAAG,CAAA,GAAI,IAAA;AAC1B,UAAA,KAAA,IAAS,GAAA,GAAM,GAAA;AAAA,QACjB;AACA,QAAA,IAAA,CAAK,CAAC,CAAA,GAAI,IAAA,CAAK,IAAA,CAAK,QAAQ,QAAQ,CAAA;AAAA,MACtC;AACA,MAAA,IAAA,CAAK,SAAS,MAAA,CAAO,IAAA,EAAM,EAAE,KAAA,EAAO,WAAW,CAAA;AAAA,IACjD,CAAA,MAAO;AACL,MAAA,IAAA,CAAK,MAAA,GAAS,MAAA;AAAA,IAChB;AAEA,IAAA,IAAA,CAAK,KAAA,GAAQ,IAAA,CAAK,QAAA,IAAY,KAAA,GAAQ,MAAA,CAAO,OAAO,EAAE,KAAA,EAAO,SAAA,EAAW,CAAA,GAAI,MAAA;AAE5E,IAAA,IAAA,CAAK,MAAA,GAAS,IAAA;AACd,IAAA,OAAO,IAAA;AAAA,EACT;AAAA,EAEA,UAAU,CAAA,EAAmB;AAC3B,IAAA,IAAI,CAAC,KAAK,MAAA,EAAQ;AAChB,MAAA,MAAM,IAAI,eAAe,gDAAgD,CAAA;AAAA,IAC3E;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,kBAAA,CAAmB,GAAG,GAAG,CAAA;AACzB,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,IAAA,GAAOA,eAAAA,CAAe,CAAA,EAAG,GAAG,CAAA;AAClC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AACzC,IAAA,MAAM,OAAO,IAAA,CAAK,KAAA;AAClB,IAAA,MAAM,QAAQ,IAAA,CAAK,MAAA;AACnB,IAAA,MAAM,QAAA,GAAW,IAAA,GAAOA,eAAAA,CAAe,IAAA,EAAM,OAAO,CAAA,GAAI,MAAA;AACxD,IAAA,MAAM,SAAA,GAAY,KAAA,GAAQA,eAAAA,CAAe,KAAA,EAAO,QAAQ,CAAA,GAAI,MAAA;AAC5D,IAAA,MAAM,UAAA,GAAa,IAAA,GAAO,WAAA,CAAY,IAAI,CAAA,GAAI,CAAA;AAC9C,IAAA,MAAM,WAAA,GAAc,KAAA,GAAQ,WAAA,CAAY,KAAK,CAAA,GAAI,CAAA;AAEjD,IAAA,IAAI,IAAA,CAAK,QAAA,IAAY,CAAC,IAAA,EAAM;AAC1B,MAAA,MAAM,IAAI,aAAa,8CAA8C,CAAA;AAAA,IACvE;AACA,IAAA,IAAI,IAAA,CAAK,OAAA,IAAW,CAAC,KAAA,EAAO;AAC1B,MAAA,MAAM,IAAI,aAAa,+CAA+C,CAAA;AAAA,IACxE;AAEA,IAAA,MAAM,MAAA,GAAS,KAAA,CAAM,IAAA,CAAK,EAAE,QAAQ,QAAA,EAAS,EAAG,MAAM,IAAI,KAAA,CAAc,SAAS,CAAA,CAAE,IAAA,CAAK,CAAC,CAAC,CAAA;AAE1F,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAM,OAAA,GAAU,CAAA,CAAE,MAAA,GAAS,CAAA,GAAI,OAAA;AAC/B,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,QAAA,MAAM,GAAA,GAAM,IAAA,CAAK,OAAA,GAAU,CAAA,GAAI,OAAO,CAAA;AACtC,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,IAAI,GAAA,GAAM,OAAO,GAAG,CAAA;AAEpB,QAAA,IAAI,IAAA,CAAK,QAAA,IAAY,IAAA,IAAQ,QAAA,EAAU;AACrC,UAAA,MAAM,SAAA,GAAY,QAAA,CAAS,IAAA,CAAK,MAAA,GAAS,IAAI,UAAU,CAAA;AACvD,UAAA,IAAI,cAAc,MAAA,EAAW;AAC3B,YAAA,MAAM,IAAI,aAAa,kDAAkD,CAAA;AAAA,UAC3E;AACA,UAAA,GAAA,IAAO,OAAO,SAAS,CAAA;AAAA,QACzB;AAEA,QAAA,IAAI,IAAA,CAAK,OAAA,IAAW,KAAA,IAAS,SAAA,EAAW;AACtC,UAAA,MAAM,QAAA,GAAW,SAAA,CAAU,KAAA,CAAM,MAAA,GAAS,IAAI,WAAW,CAAA;AACzD,UAAA,IAAI,aAAa,MAAA,EAAW;AAC1B,YAAA,MAAM,IAAI,aAAa,mDAAmD,CAAA;AAAA,UAC5E;AACA,UAAA,MAAM,GAAA,GAAM,OAAO,QAAQ,CAAA;AAC3B,UAAA,MAAM,OAAA,GAAU,GAAA,KAAQ,CAAA,GAAI,CAAA,GAAI,GAAA;AAChC,UAAA,GAAA,IAAO,OAAA;AAAA,QACT;AAEA,QAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,QACnE;AACA,QAAA,GAAA,CAAI,CAAC,CAAA,GAAI,GAAA;AAAA,MACX;AAAA,IACF;AAEA,IAAA,OAAO,MAAA,CAAO,QAAQ,EAAE,KAAA,EAAO,WAAW,MAAA,EAAQ,CAAA,CAAE,QAAQ,CAAA;AAAA,EAC9D;AAAA,EAEA,aAAa,CAAA,EAAmB;AAC9B,IAAA,OAAO,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,CAAE,UAAU,CAAC,CAAA;AAAA,EAChC;AAAA,EAEA,iBAAiB,CAAA,EAAmB;AAClC,IAAA,IAAI,CAAC,KAAK,MAAA,EAAQ;AAChB,MAAA,MAAM,IAAI,eAAe,wDAAwD,CAAA;AAAA,IACnF;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,IAAA,GAAOA,eAAAA,CAAe,CAAA,EAAG,GAAG,CAAA;AAClC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AACzC,IAAA,MAAM,OAAO,IAAA,CAAK,KAAA;AAClB,IAAA,MAAM,QAAQ,IAAA,CAAK,MAAA;AACnB,IAAA,MAAM,QAAA,GAAW,IAAA,GAAOA,eAAAA,CAAe,IAAA,EAAM,OAAO,CAAA,GAAI,MAAA;AACxD,IAAA,MAAM,SAAA,GAAY,KAAA,GAAQA,eAAAA,CAAe,KAAA,EAAO,QAAQ,CAAA,GAAI,MAAA;AAC5D,IAAA,MAAM,UAAA,GAAa,IAAA,GAAO,WAAA,CAAY,IAAI,CAAA,GAAI,CAAA;AAC9C,IAAA,MAAM,WAAA,GAAc,KAAA,GAAQ,WAAA,CAAY,KAAK,CAAA,GAAI,CAAA;AAEjD,IAAA,IAAI,IAAA,CAAK,QAAA,IAAY,CAAC,IAAA,EAAM;AAC1B,MAAA,MAAM,IAAI,aAAa,8CAA8C,CAAA;AAAA,IACvE;AACA,IAAA,IAAI,IAAA,CAAK,OAAA,IAAW,CAAC,KAAA,EAAO;AAC1B,MAAA,MAAM,IAAI,aAAa,+CAA+C,CAAA;AAAA,IACxE;AAEA,IAAA,MAAM,MAAA,GAAS,KAAA,CAAM,IAAA,CAAK,EAAE,QAAQ,QAAA,EAAS,EAAG,MAAM,IAAI,KAAA,CAAc,SAAS,CAAA,CAAE,IAAA,CAAK,CAAC,CAAC,CAAA;AAE1F,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAM,OAAA,GAAU,CAAA,CAAE,MAAA,GAAS,CAAA,GAAI,OAAA;AAC/B,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,QAAA,MAAM,GAAA,GAAM,IAAA,CAAK,OAAA,GAAU,CAAA,GAAI,OAAO,CAAA;AACtC,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,IAAI,GAAA,GAAM,OAAO,GAAG,CAAA;AAEpB,QAAA,IAAI,IAAA,CAAK,OAAA,IAAW,KAAA,IAAS,SAAA,EAAW;AACtC,UAAA,MAAM,QAAA,GAAW,SAAA,CAAU,KAAA,CAAM,MAAA,GAAS,IAAI,WAAW,CAAA;AACzD,UAAA,IAAI,aAAa,MAAA,EAAW;AAC1B,YAAA,MAAM,IAAI,aAAa,mDAAmD,CAAA;AAAA,UAC5E;AACA,UAAA,MAAM,GAAA,GAAM,OAAO,QAAQ,CAAA;AAC3B,UAAA,MAAM,OAAA,GAAU,GAAA,KAAQ,CAAA,GAAI,CAAA,GAAI,GAAA;AAChC,UAAA,GAAA,IAAO,OAAA;AAAA,QACT;AAEA,QAAA,IAAI,IAAA,CAAK,QAAA,IAAY,IAAA,IAAQ,QAAA,EAAU;AACrC,UAAA,MAAM,SAAA,GAAY,QAAA,CAAS,IAAA,CAAK,MAAA,GAAS,IAAI,UAAU,CAAA;AACvD,UAAA,IAAI,cAAc,MAAA,EAAW;AAC3B,YAAA,MAAM,IAAI,aAAa,kDAAkD,CAAA;AAAA,UAC3E;AACA,UAAA,GAAA,IAAO,OAAO,SAAS,CAAA;AAAA,QACzB;AAEA,QAAA,MAAM,SAAA,GAAY,OAAO,CAAC,CAAA;AAC1B,QAAA,IAAI,cAAc,MAAA,EAAW;AAC3B,UAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,QACnE;AACA,QAAA,SAAA,CAAU,CAAC,CAAA,GAAI,gBAAA,CAAiB,GAAG,CAAA;AAAA,MACrC;AAAA,IACF;AAEA,IAAA,OAAO,MAAA,CAAO,QAAQ,EAAE,KAAA,EAAO,WAAW,MAAA,EAAQ,CAAA,CAAE,QAAQ,CAAA;AAAA,EAC9D;AACF;AASO,IAAM,eAAN,MAAmB;AAAA,EAChB,MAAA,GAAS,KAAA;AAAA,EACT,QAAA;AAAA,EACA,QAAA;AAAA,EACA,YAAA;AAAA,EACA,IAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAUR,WAAA,CACE,OAAA,GAII,EAAC,EACL;AACA,IAAA,IAAA,CAAK,YAAA,GAAe,OAAA,CAAQ,YAAA,IAAgB,CAAC,GAAG,CAAC,CAAA;AACjD,IAAA,IAAA,CAAK,IAAA,GAAO,kBAAA,CAAmB,OAAA,CAAQ,IAAA,EAAM,QAAQ,KAAK,CAAA;AAC1D,IAAA,kBAAA,CAAmB,OAAA,CAAQ,IAAA,EAAM,MAAA,EAAQ,IAAI,CAAA;AAC7C,IAAA,MAAM,CAAC,QAAA,EAAU,QAAQ,CAAA,GAAI,IAAA,CAAK,YAAA;AAClC,IAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,QAAQ,CAAA,IAAK,CAAC,MAAA,CAAO,QAAA,CAAS,QAAQ,CAAA,IAAK,QAAA,IAAY,QAAA,EAAU;AACpF,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,gDAAA;AAAA,QACA,cAAA;AAAA,QACA,IAAA,CAAK;AAAA,OACP;AAAA,IACF;AAAA,EACF;AAAA,EAEA,IAAI,CAAA,EAAiB;AACnB,IAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,MAAA,MAAM,IAAI,qBAAA,CAAsB,oCAAA,EAAsC,GAAG,CAAA;AAAA,IAC3E;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,kBAAA,CAAmB,GAAG,GAAG,CAAA;AACzB,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,IAAA,GAAOA,eAAAA,CAAe,CAAA,EAAG,GAAG,CAAA;AAClC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AAEzC,IAAA,MAAM,OAAO,IAAI,KAAA,CAAc,SAAS,CAAA,CAAE,IAAA,CAAK,OAAO,iBAAiB,CAAA;AACvE,IAAA,MAAM,OAAO,IAAI,KAAA,CAAc,SAAS,CAAA,CAAE,IAAA,CAAK,OAAO,iBAAiB,CAAA;AAEvE,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,QAAA,MAAM,MAAM,IAAA,CAAK,CAAA,CAAE,SAAS,CAAA,GAAI,OAAA,GAAU,IAAI,OAAO,CAAA;AACrD,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,MAAM,GAAA,GAAM,OAAO,GAAG,CAAA;AACtB,QAAA,MAAM,UAAA,GAAa,KAAK,CAAC,CAAA;AACzB,QAAA,MAAM,UAAA,GAAa,KAAK,CAAC,CAAA;AACzB,QAAA,IAAI,UAAA,KAAe,MAAA,IAAa,UAAA,KAAe,MAAA,EAAW;AACxD,UAAA,MAAM,IAAI,aAAa,6CAA6C,CAAA;AAAA,QACtE;AACA,QAAA,IAAA,CAAK,CAAC,CAAA,GAAI,IAAA,CAAK,GAAA,CAAI,YAAY,GAAG,CAAA;AAClC,QAAA,IAAA,CAAK,CAAC,CAAA,GAAI,IAAA,CAAK,GAAA,CAAI,YAAY,GAAG,CAAA;AAAA,MACpC;AAAA,IACF;AAEA,IAAA,IAAA,CAAK,WAAW,MAAA,CAAO,IAAA,EAAM,EAAE,KAAA,EAAO,WAAW,CAAA;AACjD,IAAA,IAAA,CAAK,WAAW,MAAA,CAAO,IAAA,EAAM,EAAE,KAAA,EAAO,WAAW,CAAA;AACjD,IAAA,IAAA,CAAK,MAAA,GAAS,IAAA;AACd,IAAA,OAAO,IAAA;AAAA,EACT;AAAA,EAEA,UAAU,CAAA,EAAmB;AAC3B,IAAA,IAAI,CAAC,KAAK,MAAA,EAAQ;AAChB,MAAA,MAAM,IAAI,eAAe,8CAA8C,CAAA;AAAA,IACzE;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,kBAAA,CAAmB,GAAG,GAAG,CAAA;AACzB,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,CAAC,QAAA,EAAU,QAAQ,CAAA,GAAI,IAAA,CAAK,YAAA;AAClC,IAAA,MAAM,IAAA,GAAOA,eAAAA,CAAe,CAAA,EAAG,GAAG,CAAA;AAClC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AACzC,IAAA,MAAM,UAAU,IAAA,CAAK,QAAA;AACrB,IAAA,MAAM,UAAU,IAAA,CAAK,QAAA;AAErB,IAAA,IAAI,CAAC,OAAA,IAAW,CAAC,OAAA,EAAS;AACxB,MAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,IAC9E;AACA,IAAA,MAAM,OAAA,GAAUA,eAAAA,CAAe,OAAA,EAAS,UAAU,CAAA;AAClD,IAAA,MAAM,OAAA,GAAUA,eAAAA,CAAe,OAAA,EAAS,UAAU,CAAA;AAClD,IAAA,MAAM,SAAA,GAAY,YAAY,OAAO,CAAA;AACrC,IAAA,MAAM,SAAA,GAAY,YAAY,OAAO,CAAA;AAErC,IAAA,MAAM,MAAA,GAAS,KAAA,CAAM,IAAA,CAAK,EAAE,QAAQ,QAAA,EAAS,EAAG,MAAM,IAAI,KAAA,CAAc,SAAS,CAAA,CAAE,IAAA,CAAK,CAAC,CAAC,CAAA;AAE1F,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAM,OAAA,GAAU,CAAA,CAAE,MAAA,GAAS,CAAA,GAAI,OAAA;AAC/B,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,QAAA,MAAM,GAAA,GAAM,IAAA,CAAK,OAAA,GAAU,CAAA,GAAI,OAAO,CAAA;AACtC,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,MAAM,GAAA,GAAM,OAAO,GAAG,CAAA;AACtB,QAAA,MAAM,MAAA,GAAS,OAAA,CAAQ,OAAA,CAAQ,MAAA,GAAS,IAAI,SAAS,CAAA;AACrD,QAAA,MAAM,MAAA,GAAS,OAAA,CAAQ,OAAA,CAAQ,MAAA,GAAS,IAAI,SAAS,CAAA;AACrD,QAAA,IAAI,MAAA,KAAW,MAAA,IAAa,MAAA,KAAW,MAAA,EAAW;AAChD,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,MAAM,GAAA,GAAM,OAAO,MAAM,CAAA;AACzB,QAAA,MAAM,GAAA,GAAM,OAAO,MAAM,CAAA;AACzB,QAAA,MAAM,QAAQ,GAAA,GAAM,GAAA;AAEpB,QAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,QACnE;AACA,QAAA,IAAI,MAAA,GACF,UAAU,CAAA,GAAA,CAAM,GAAA,GAAM,OAAO,KAAA,IAAU,QAAA,GAAW,YAAY,QAAA,GAAW,QAAA;AAC3E,QAAA,IAAI,KAAK,IAAA,EAAM;AACb,UAAA,MAAA,GAAS,KAAK,GAAA,CAAI,QAAA,EAAU,KAAK,GAAA,CAAI,QAAA,EAAU,MAAM,CAAC,CAAA;AAAA,QACxD;AACA,QAAA,GAAA,CAAI,CAAC,CAAA,GAAI,MAAA;AAAA,MACX;AAAA,IACF;AAEA,IAAA,OAAO,MAAA,CAAO,QAAQ,EAAE,KAAA,EAAO,WAAW,MAAA,EAAQ,CAAA,CAAE,QAAQ,CAAA;AAAA,EAC9D;AAAA,EAEA,aAAa,CAAA,EAAmB;AAC9B,IAAA,OAAO,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,CAAE,UAAU,CAAC,CAAA;AAAA,EAChC;AAAA,EAEA,iBAAiB,CAAA,EAAmB;AAClC,IAAA,IAAI,CAAC,KAAK,MAAA,EAAQ;AAChB,MAAA,MAAM,IAAI,eAAe,sDAAsD,CAAA;AAAA,IACjF;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,CAAC,QAAA,EAAU,QAAQ,CAAA,GAAI,IAAA,CAAK,YAAA;AAClC,IAAA,MAAM,IAAA,GAAOA,eAAAA,CAAe,CAAA,EAAG,GAAG,CAAA;AAClC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AACzC,IAAA,MAAM,UAAU,IAAA,CAAK,QAAA;AACrB,IAAA,MAAM,UAAU,IAAA,CAAK,QAAA;AAErB,IAAA,IAAI,CAAC,OAAA,IAAW,CAAC,OAAA,EAAS;AACxB,MAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,IAC9E;AACA,IAAA,MAAM,OAAA,GAAUA,eAAAA,CAAe,OAAA,EAAS,UAAU,CAAA;AAClD,IAAA,MAAM,OAAA,GAAUA,eAAAA,CAAe,OAAA,EAAS,UAAU,CAAA;AAClD,IAAA,MAAM,SAAA,GAAY,YAAY,OAAO,CAAA;AACrC,IAAA,MAAM,SAAA,GAAY,YAAY,OAAO,CAAA;AAErC,IAAA,MAAM,MAAA,GAAS,KAAA,CAAM,IAAA,CAAK,EAAE,QAAQ,QAAA,EAAS,EAAG,MAAM,IAAI,KAAA,CAAc,SAAS,CAAA,CAAE,IAAA,CAAK,CAAC,CAAC,CAAA;AAE1F,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAM,OAAA,GAAU,CAAA,CAAE,MAAA,GAAS,CAAA,GAAI,OAAA;AAC/B,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,QAAA,MAAM,GAAA,GAAM,IAAA,CAAK,OAAA,GAAU,CAAA,GAAI,OAAO,CAAA;AACtC,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,MAAM,GAAA,GAAM,OAAO,GAAG,CAAA;AACtB,QAAA,MAAM,MAAA,GAAS,OAAA,CAAQ,OAAA,CAAQ,MAAA,GAAS,IAAI,SAAS,CAAA;AACrD,QAAA,MAAM,MAAA,GAAS,OAAA,CAAQ,OAAA,CAAQ,MAAA,GAAS,IAAI,SAAS,CAAA;AACrD,QAAA,IAAI,MAAA,KAAW,MAAA,IAAa,MAAA,KAAW,MAAA,EAAW;AAChD,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,MAAM,GAAA,GAAM,OAAO,MAAM,CAAA;AACzB,QAAA,MAAM,GAAA,GAAM,OAAO,MAAM,CAAA;AACzB,QAAA,MAAM,QAAQ,GAAA,GAAM,GAAA;AAEpB,QAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,QACnE;AACA,QAAA,GAAA,CAAI,CAAC,CAAA,GAAA,CAAM,GAAA,GAAM,QAAA,KAAa,QAAA,GAAW,YAAa,KAAA,GAAQ,GAAA;AAAA,MAChE;AAAA,IACF;AAEA,IAAA,OAAO,MAAA,CAAO,QAAQ,EAAE,KAAA,EAAO,WAAW,MAAA,EAAQ,CAAA,CAAE,QAAQ,CAAA;AAAA,EAC9D;AACF;AASO,IAAM,eAAN,MAAmB;AAAA,EAChB,MAAA,GAAS,KAAA;AAAA,EACT,OAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAQR,WAAA,CAAY,OAAA,GAA8B,EAAC,EAAG;AAC5C,IAAA,kBAAA,CAAmB,OAAA,CAAQ,IAAA,EAAM,MAAA,EAAQ,IAAI,CAAA;AAAA,EAC/C;AAAA,EAEA,IAAI,CAAA,EAAiB;AACnB,IAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,MAAA,MAAM,IAAI,qBAAA,CAAsB,oCAAA,EAAsC,GAAG,CAAA;AAAA,IAC3E;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,kBAAA,CAAmB,GAAG,GAAG,CAAA;AACzB,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,IAAA,GAAOA,eAAAA,CAAe,CAAA,EAAG,GAAG,CAAA;AAClC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AAEzC,IAAA,MAAM,SAAS,IAAI,KAAA,CAAc,SAAS,CAAA,CAAE,KAAK,CAAC,CAAA;AAElD,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,QAAA,MAAM,MAAM,IAAA,CAAK,CAAA,CAAE,SAAS,CAAA,GAAI,OAAA,GAAU,IAAI,OAAO,CAAA;AACrD,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,MAAM,UAAA,GAAa,OAAO,CAAC,CAAA;AAC3B,QAAA,IAAI,eAAe,MAAA,EAAW;AAC5B,UAAA,MAAM,IAAI,aAAa,4CAA4C,CAAA;AAAA,QACrE;AACA,QAAA,MAAA,CAAO,CAAC,CAAA,GAAI,IAAA,CAAK,GAAA,CAAI,UAAA,EAAY,KAAK,GAAA,CAAI,MAAA,CAAO,GAAG,CAAC,CAAC,CAAA;AAAA,MACxD;AAAA,IACF;AAEA,IAAA,IAAA,CAAK,UAAU,MAAA,CAAO,MAAA,EAAQ,EAAE,KAAA,EAAO,WAAW,CAAA;AAClD,IAAA,IAAA,CAAK,MAAA,GAAS,IAAA;AACd,IAAA,OAAO,IAAA;AAAA,EACT;AAAA,EAEA,UAAU,CAAA,EAAmB;AAC3B,IAAA,IAAI,CAAC,KAAK,MAAA,EAAQ;AAChB,MAAA,MAAM,IAAI,eAAe,8CAA8C,CAAA;AAAA,IACzE;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,kBAAA,CAAmB,GAAG,GAAG,CAAA;AACzB,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,IAAA,GAAOA,eAAAA,CAAe,CAAA,EAAG,GAAG,CAAA;AAClC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AACzC,IAAA,MAAM,SAAS,IAAA,CAAK,OAAA;AACpB,IAAA,IAAI,CAAC,MAAA,EAAQ;AACX,MAAA,MAAM,IAAI,aAAa,oDAAoD,CAAA;AAAA,IAC7E;AACA,IAAA,MAAM,OAAA,GAAUA,eAAAA,CAAe,MAAA,EAAQ,SAAS,CAAA;AAChD,IAAA,MAAM,SAAA,GAAY,YAAY,MAAM,CAAA;AAEpC,IAAA,MAAM,MAAA,GAAS,KAAA,CAAM,IAAA,CAAK,EAAE,QAAQ,QAAA,EAAS,EAAG,MAAM,IAAI,KAAA,CAAc,SAAS,CAAA,CAAE,IAAA,CAAK,CAAC,CAAC,CAAA;AAE1F,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAM,OAAA,GAAU,CAAA,CAAE,MAAA,GAAS,CAAA,GAAI,OAAA;AAC/B,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,QAAA,MAAM,GAAA,GAAM,IAAA,CAAK,OAAA,GAAU,CAAA,GAAI,OAAO,CAAA;AACtC,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,MAAM,GAAA,GAAM,OAAO,GAAG,CAAA;AACtB,QAAA,MAAM,QAAA,GAAW,OAAA,CAAQ,MAAA,CAAO,MAAA,GAAS,IAAI,SAAS,CAAA;AACtD,QAAA,IAAI,aAAa,MAAA,EAAW;AAC1B,UAAA,MAAM,IAAI,aAAa,oDAAoD,CAAA;AAAA,QAC7E;AACA,QAAA,MAAM,KAAA,GAAQ,OAAO,QAAQ,CAAA;AAC7B,QAAA,MAAM,SAAA,GAAY,KAAA,KAAU,CAAA,GAAI,CAAA,GAAI,KAAA;AACpC,QAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,QACnE;AACA,QAAA,GAAA,CAAI,CAAC,IAAI,GAAA,GAAM,SAAA;AAAA,MACjB;AAAA,IACF;AAEA,IAAA,OAAO,MAAA,CAAO,QAAQ,EAAE,KAAA,EAAO,WAAW,MAAA,EAAQ,CAAA,CAAE,QAAQ,CAAA;AAAA,EAC9D;AAAA,EAEA,aAAa,CAAA,EAAmB;AAC9B,IAAA,OAAO,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,CAAE,UAAU,CAAC,CAAA;AAAA,EAChC;AAAA,EAEA,iBAAiB,CAAA,EAAmB;AAClC,IAAA,IAAI,CAAC,KAAK,MAAA,EAAQ;AAChB,MAAA,MAAM,IAAI,eAAe,sDAAsD,CAAA;AAAA,IACjF;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,IAAA,GAAOA,eAAAA,CAAe,CAAA,EAAG,GAAG,CAAA;AAClC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AACzC,IAAA,MAAM,SAAS,IAAA,CAAK,OAAA;AACpB,IAAA,IAAI,CAAC,MAAA,EAAQ;AACX,MAAA,MAAM,IAAI,aAAa,oDAAoD,CAAA;AAAA,IAC7E;AACA,IAAA,MAAM,OAAA,GAAUA,eAAAA,CAAe,MAAA,EAAQ,SAAS,CAAA;AAChD,IAAA,MAAM,SAAA,GAAY,YAAY,MAAM,CAAA;AAEpC,IAAA,MAAM,MAAA,GAAS,KAAA,CAAM,IAAA,CAAK,EAAE,QAAQ,QAAA,EAAS,EAAG,MAAM,IAAI,KAAA,CAAc,SAAS,CAAA,CAAE,IAAA,CAAK,CAAC,CAAC,CAAA;AAE1F,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAM,OAAA,GAAU,CAAA,CAAE,MAAA,GAAS,CAAA,GAAI,OAAA;AAC/B,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,QAAA,MAAM,GAAA,GAAM,IAAA,CAAK,OAAA,GAAU,CAAA,GAAI,OAAO,CAAA;AACtC,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,MAAM,GAAA,GAAM,OAAO,GAAG,CAAA;AACtB,QAAA,MAAM,QAAA,GAAW,OAAA,CAAQ,MAAA,CAAO,MAAA,GAAS,IAAI,SAAS,CAAA;AACtD,QAAA,IAAI,aAAa,MAAA,EAAW;AAC1B,UAAA,MAAM,IAAI,aAAa,oDAAoD,CAAA;AAAA,QAC7E;AACA,QAAA,MAAM,KAAA,GAAQ,OAAO,QAAQ,CAAA;AAC7B,QAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,QACnE;AACA,QAAA,GAAA,CAAI,CAAC,IAAI,GAAA,GAAM,KAAA;AAAA,MACjB;AAAA,IACF;AAEA,IAAA,OAAO,MAAA,CAAO,QAAQ,EAAE,KAAA,EAAO,WAAW,MAAA,EAAQ,CAAA,CAAE,QAAQ,CAAA;AAAA,EAC9D;AACF;AASO,IAAM,eAAN,MAAmB;AAAA,EAChB,MAAA,GAAS,KAAA;AAAA,EACT,OAAA;AAAA,EACA,MAAA;AAAA,EACA,aAAA;AAAA,EACA,WAAA;AAAA,EACA,aAAA;AAAA,EACA,YAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAYR,WAAA,CACE,OAAA,GAMI,EAAC,EACL;AACA,IAAA,IAAA,CAAK,aAAA,GAAgB,kBAAA,CAAmB,OAAA,CAAQ,aAAA,EAAe,iBAAiB,IAAI,CAAA;AACpF,IAAA,IAAA,CAAK,WAAA,GAAc,kBAAA,CAAmB,OAAA,CAAQ,WAAA,EAAa,eAAe,IAAI,CAAA;AAC9E,IAAA,IAAA,CAAK,aAAA,GAAgB,OAAA,CAAQ,aAAA,IAAiB,CAAC,IAAI,EAAE,CAAA;AACrD,IAAA,IAAA,CAAK,YAAA,GAAe,kBAAA,CAAmB,OAAA,CAAQ,YAAA,EAAc,gBAAgB,KAAK,CAAA;AAClF,IAAA,kBAAA,CAAmB,OAAA,CAAQ,IAAA,EAAM,MAAA,EAAQ,IAAI,CAAA;AAC7C,IAAA,MAAM,CAAC,KAAA,EAAO,KAAK,CAAA,GAAI,IAAA,CAAK,aAAA;AAC5B,IAAA,IACE,CAAC,MAAA,CAAO,QAAA,CAAS,KAAK,KACtB,CAAC,MAAA,CAAO,QAAA,CAAS,KAAK,KACtB,KAAA,GAAQ,CAAA,IACR,KAAA,GAAQ,GAAA,IACR,SAAS,KAAA,EACT;AACA,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,0DAAA;AAAA,QACA,eAAA;AAAA,QACA,IAAA,CAAK;AAAA,OACP;AAAA,IACF;AAAA,EACF;AAAA,EAEA,IAAI,CAAA,EAAiB;AACnB,IAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,MAAA,MAAM,IAAI,qBAAA,CAAsB,oCAAA,EAAsC,GAAG,CAAA;AAAA,IAC3E;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,kBAAA,CAAmB,GAAG,GAAG,CAAA;AACzB,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,IAAA,GAAOA,eAAAA,CAAe,CAAA,EAAG,GAAG,CAAA;AAClC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AAEzC,IAAA,MAAM,UAAU,IAAI,KAAA,CAAc,SAAS,CAAA,CAAE,KAAK,CAAC,CAAA;AACnD,IAAA,MAAM,SAAS,IAAI,KAAA,CAAc,SAAS,CAAA,CAAE,KAAK,CAAC,CAAA;AAGlD,IAAA,MAAM,CAAC,eAAA,EAAiB,eAAe,CAAA,GAAI,IAAA,CAAK,aAAA;AAChD,IAAA,MAAM,gBAAgB,eAAA,GAAkB,GAAA;AACxC,IAAA,MAAM,gBAAgB,eAAA,GAAkB,GAAA;AACxC,IAAA,MAAM,UAAA,GAAa,KAAK,YAAA,GACpB,cAAA,CAAe,aAAa,CAAA,GAAI,cAAA,CAAe,aAAa,CAAA,GAC5D,CAAA;AACJ,IAAA,IAAI,IAAA,CAAK,iBAAiB,CAAC,MAAA,CAAO,SAAS,UAAU,CAAA,IAAK,cAAc,CAAA,CAAA,EAAI;AAC1E,MAAA,MAAM,IAAI,aAAa,+DAA+D,CAAA;AAAA,IACxF;AAEA,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,MAAA,MAAM,SAAmB,EAAC;AAC1B,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,QAAA,MAAM,MAAM,IAAA,CAAK,CAAA,CAAE,SAAS,CAAA,GAAI,OAAA,GAAU,IAAI,OAAO,CAAA;AACrD,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,MAAA,CAAO,IAAA,CAAK,MAAA,CAAO,GAAG,CAAC,CAAA;AAAA,MACzB;AACA,MAAA,MAAA,CAAO,IAAA,CAAK,CAAC,CAAA,EAAG,CAAA,KAAM,IAAI,CAAC,CAAA;AAE3B,MAAA,MAAM,WAAA,GAAc,CAAC,CAAA,KAAsB;AACzC,QAAA,IAAI,MAAA,CAAO,WAAW,CAAA,EAAG;AACvB,UAAA,MAAM,IAAI,aAAa,iDAAiD,CAAA;AAAA,QAC1E;AACA,QAAA,IAAI,MAAA,CAAO,WAAW,CAAA,EAAG;AACvB,UAAA,MAAM,IAAA,GAAO,OAAO,CAAC,CAAA;AACrB,UAAA,IAAI,SAAS,MAAA,EAAW;AACtB,YAAA,MAAM,IAAI,aAAa,sCAAsC,CAAA;AAAA,UAC/D;AACA,UAAA,OAAO,IAAA;AAAA,QACT;AAEA,QAAA,MAAM,QAAA,GAAW,CAAA,IAAK,MAAA,CAAO,MAAA,GAAS,CAAA,CAAA;AACtC,QAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,KAAA,CAAM,QAAQ,CAAA;AACjC,QAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,IAAA,CAAK,QAAQ,CAAA;AAChC,QAAA,MAAM,UAAA,GAAa,OAAO,KAAK,CAAA;AAC/B,QAAA,MAAM,UAAA,GAAa,OAAO,KAAK,CAAA;AAC/B,QAAA,IAAI,UAAA,KAAe,MAAA,IAAa,UAAA,KAAe,MAAA,EAAW;AACxD,UAAA,MAAM,IAAI,aAAa,4DAA4D,CAAA;AAAA,QACrF;AACA,QAAA,IAAI,UAAU,KAAA,EAAO;AACnB,UAAA,OAAO,UAAA;AAAA,QACT;AACA,QAAA,MAAM,SAAS,QAAA,GAAW,KAAA;AAC1B,QAAA,OAAO,UAAA,IAAc,CAAA,GAAI,MAAA,CAAA,GAAU,UAAA,GAAa,MAAA;AAAA,MAClD,CAAA;AAGA,MAAA,OAAA,CAAQ,CAAC,CAAA,GAAI,WAAA,CAAY,GAAG,CAAA;AAC5B,MAAA,MAAM,MAAA,GAAS,YAAY,aAAa,CAAA;AACxC,MAAA,MAAM,MAAA,GAAS,YAAY,aAAa,CAAA;AACxC,MAAA,MAAM,MAAM,MAAA,GAAS,MAAA;AACrB,MAAA,MAAA,CAAO,CAAC,CAAA,GAAI,IAAA,CAAK,YAAA,GAAe,MAAM,UAAA,GAAa,GAAA;AAAA,IACrD;AAEA,IAAA,IAAA,CAAK,OAAA,GAAU,KAAK,aAAA,GAAgB,MAAA,CAAO,SAAS,EAAE,KAAA,EAAO,SAAA,EAAW,CAAA,GAAI,MAAA;AAC5E,IAAA,IAAA,CAAK,MAAA,GAAS,KAAK,WAAA,GAAc,MAAA,CAAO,QAAQ,EAAE,KAAA,EAAO,SAAA,EAAW,CAAA,GAAI,MAAA;AACxE,IAAA,IAAA,CAAK,MAAA,GAAS,IAAA;AACd,IAAA,OAAO,IAAA;AAAA,EACT;AAAA,EAEA,UAAU,CAAA,EAAmB;AAC3B,IAAA,IAAI,CAAC,KAAK,MAAA,EAAQ;AAChB,MAAA,MAAM,IAAI,eAAe,8CAA8C,CAAA;AAAA,IACzE;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,kBAAA,CAAmB,GAAG,GAAG,CAAA;AACzB,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,IAAA,GAAOA,eAAAA,CAAe,CAAA,EAAG,GAAG,CAAA;AAClC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AACzC,IAAA,MAAM,SAAS,IAAA,CAAK,OAAA;AACpB,IAAA,MAAM,QAAQ,IAAA,CAAK,MAAA;AACnB,IAAA,MAAM,UAAA,GAAa,MAAA,GAASA,eAAAA,CAAe,MAAA,EAAQ,SAAS,CAAA,GAAI,MAAA;AAChE,IAAA,MAAM,SAAA,GAAY,KAAA,GAAQA,eAAAA,CAAe,KAAA,EAAO,QAAQ,CAAA,GAAI,MAAA;AAC5D,IAAA,MAAM,YAAA,GAAe,MAAA,GAAS,WAAA,CAAY,MAAM,CAAA,GAAI,CAAA;AACpD,IAAA,MAAM,WAAA,GAAc,KAAA,GAAQ,WAAA,CAAY,KAAK,CAAA,GAAI,CAAA;AAEjD,IAAA,IAAI,IAAA,CAAK,aAAA,IAAiB,CAAC,MAAA,EAAQ;AACjC,MAAA,MAAM,IAAI,aAAa,8CAA8C,CAAA;AAAA,IACvE;AACA,IAAA,IAAI,IAAA,CAAK,WAAA,IAAe,CAAC,KAAA,EAAO;AAC9B,MAAA,MAAM,IAAI,aAAa,6CAA6C,CAAA;AAAA,IACtE;AAEA,IAAA,MAAM,MAAA,GAAS,KAAA,CAAM,IAAA,CAAK,EAAE,QAAQ,QAAA,EAAS,EAAG,MAAM,IAAI,KAAA,CAAc,SAAS,CAAA,CAAE,IAAA,CAAK,CAAC,CAAC,CAAA;AAE1F,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAM,OAAA,GAAU,CAAA,CAAE,MAAA,GAAS,CAAA,GAAI,OAAA;AAC/B,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,QAAA,MAAM,GAAA,GAAM,IAAA,CAAK,OAAA,GAAU,CAAA,GAAI,OAAO,CAAA;AACtC,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,IAAI,GAAA,GAAM,OAAO,GAAG,CAAA;AAEpB,QAAA,IAAI,IAAA,CAAK,aAAA,IAAiB,MAAA,IAAU,UAAA,EAAY;AAC9C,UAAA,MAAM,SAAA,GAAY,UAAA,CAAW,MAAA,CAAO,MAAA,GAAS,IAAI,YAAY,CAAA;AAC7D,UAAA,IAAI,cAAc,MAAA,EAAW;AAC3B,YAAA,MAAM,IAAI,aAAa,oDAAoD,CAAA;AAAA,UAC7E;AACA,UAAA,GAAA,IAAO,OAAO,SAAS,CAAA;AAAA,QACzB;AAEA,QAAA,IAAI,IAAA,CAAK,WAAA,IAAe,KAAA,IAAS,SAAA,EAAW;AAC1C,UAAA,MAAM,QAAA,GAAW,SAAA,CAAU,KAAA,CAAM,MAAA,GAAS,IAAI,WAAW,CAAA;AACzD,UAAA,IAAI,aAAa,MAAA,EAAW;AAC1B,YAAA,MAAM,IAAI,aAAa,mDAAmD,CAAA;AAAA,UAC5E;AACA,UAAA,MAAM,UAAA,GAAa,OAAO,QAAQ,CAAA;AAClC,UAAA,MAAM,SAAA,GAAY,UAAA,KAAe,CAAA,GAAI,CAAA,GAAI,UAAA;AACzC,UAAA,GAAA,IAAO,SAAA;AAAA,QACT;AAEA,QAAA,MAAM,SAAA,GAAY,OAAO,CAAC,CAAA;AAC1B,QAAA,IAAI,cAAc,MAAA,EAAW;AAC3B,UAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,QACnE;AACA,QAAA,SAAA,CAAU,CAAC,CAAA,GAAI,GAAA;AAAA,MACjB;AAAA,IACF;AAEA,IAAA,OAAO,MAAA,CAAO,QAAQ,EAAE,KAAA,EAAO,WAAW,MAAA,EAAQ,CAAA,CAAE,QAAQ,CAAA;AAAA,EAC9D;AAAA,EAEA,aAAa,CAAA,EAAmB;AAC9B,IAAA,OAAO,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,CAAE,UAAU,CAAC,CAAA;AAAA,EAChC;AAAA,EAEA,iBAAiB,CAAA,EAAmB;AAClC,IAAA,IAAI,CAAC,KAAK,MAAA,EAAQ;AAChB,MAAA,MAAM,IAAI,eAAe,sDAAsD,CAAA;AAAA,IACjF;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,IAAA,GAAOA,eAAAA,CAAe,CAAA,EAAG,GAAG,CAAA;AAClC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AACzC,IAAA,MAAM,SAAS,IAAA,CAAK,OAAA;AACpB,IAAA,MAAM,QAAQ,IAAA,CAAK,MAAA;AACnB,IAAA,MAAM,UAAA,GAAa,MAAA,GAASA,eAAAA,CAAe,MAAA,EAAQ,SAAS,CAAA,GAAI,MAAA;AAChE,IAAA,MAAM,SAAA,GAAY,KAAA,GAAQA,eAAAA,CAAe,KAAA,EAAO,QAAQ,CAAA,GAAI,MAAA;AAC5D,IAAA,MAAM,YAAA,GAAe,MAAA,GAAS,WAAA,CAAY,MAAM,CAAA,GAAI,CAAA;AACpD,IAAA,MAAM,WAAA,GAAc,KAAA,GAAQ,WAAA,CAAY,KAAK,CAAA,GAAI,CAAA;AAEjD,IAAA,IAAI,IAAA,CAAK,aAAA,IAAiB,CAAC,MAAA,EAAQ;AACjC,MAAA,MAAM,IAAI,aAAa,8CAA8C,CAAA;AAAA,IACvE;AACA,IAAA,IAAI,IAAA,CAAK,WAAA,IAAe,CAAC,KAAA,EAAO;AAC9B,MAAA,MAAM,IAAI,aAAa,6CAA6C,CAAA;AAAA,IACtE;AAEA,IAAA,MAAM,MAAA,GAAS,KAAA,CAAM,IAAA,CAAK,EAAE,QAAQ,QAAA,EAAS,EAAG,MAAM,IAAI,KAAA,CAAc,SAAS,CAAA,CAAE,IAAA,CAAK,CAAC,CAAC,CAAA;AAE1F,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAM,OAAA,GAAU,CAAA,CAAE,MAAA,GAAS,CAAA,GAAI,OAAA;AAC/B,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,QAAA,MAAM,GAAA,GAAM,IAAA,CAAK,OAAA,GAAU,CAAA,GAAI,OAAO,CAAA;AACtC,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,IAAI,GAAA,GAAM,OAAO,GAAG,CAAA;AAEpB,QAAA,IAAI,IAAA,CAAK,WAAA,IAAe,KAAA,IAAS,SAAA,EAAW;AAC1C,UAAA,MAAM,QAAA,GAAW,SAAA,CAAU,KAAA,CAAM,MAAA,GAAS,IAAI,WAAW,CAAA;AACzD,UAAA,IAAI,aAAa,MAAA,EAAW;AAC1B,YAAA,MAAM,IAAI,aAAa,mDAAmD,CAAA;AAAA,UAC5E;AACA,UAAA,MAAM,UAAA,GAAa,OAAO,QAAQ,CAAA;AAClC,UAAA,MAAM,SAAA,GAAY,UAAA,KAAe,CAAA,GAAI,CAAA,GAAI,UAAA;AACzC,UAAA,GAAA,IAAO,SAAA;AAAA,QACT;AAEA,QAAA,IAAI,IAAA,CAAK,aAAA,IAAiB,MAAA,IAAU,UAAA,EAAY;AAC9C,UAAA,MAAM,SAAA,GAAY,UAAA,CAAW,MAAA,CAAO,MAAA,GAAS,IAAI,YAAY,CAAA;AAC7D,UAAA,IAAI,cAAc,MAAA,EAAW;AAC3B,YAAA,MAAM,IAAI,aAAa,oDAAoD,CAAA;AAAA,UAC7E;AACA,UAAA,GAAA,IAAO,OAAO,SAAS,CAAA;AAAA,QACzB;AAEA,QAAA,MAAM,SAAA,GAAY,OAAO,CAAC,CAAA;AAC1B,QAAA,IAAI,cAAc,MAAA,EAAW;AAC3B,UAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,QACnE;AACA,QAAA,SAAA,CAAU,CAAC,CAAA,GAAI,GAAA;AAAA,MACjB;AAAA,IACF;AAEA,IAAA,OAAO,MAAA,CAAO,QAAQ,EAAE,KAAA,EAAO,WAAW,MAAA,EAAQ,CAAA,CAAE,QAAQ,CAAA;AAAA,EAC9D;AACF;AASO,IAAM,aAAN,MAAiB;AAAA,EACd,IAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EASR,WAAA,CAAY,OAAA,GAA0D,EAAC,EAAG;AACxE,IAAA,IAAA,CAAK,IAAA,GAAO,QAAQ,IAAA,IAAQ,IAAA;AAC5B,IAAA,IAAI,IAAA,CAAK,SAAS,IAAA,IAAQ,IAAA,CAAK,SAAS,IAAA,IAAQ,IAAA,CAAK,SAAS,KAAA,EAAO;AACnE,MAAA,MAAM,IAAI,qBAAA,CAAsB,kCAAA,EAAoC,MAAA,EAAQ,KAAK,IAAI,CAAA;AAAA,IACvF;AACA,IAAA,kBAAA,CAAmB,OAAA,CAAQ,IAAA,EAAM,MAAA,EAAQ,IAAI,CAAA;AAAA,EAC/C;AAAA,EAEA,IAAI,EAAA,EAAkB;AACpB,IAAA,OAAO,IAAA;AAAA,EACT;AAAA,EAEA,UAAU,CAAA,EAAmB;AAC3B,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,kBAAA,CAAmB,GAAG,GAAG,CAAA;AACzB,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,IAAA,GAAOA,eAAAA,CAAe,CAAA,EAAG,GAAG,CAAA;AAClC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AAEzC,IAAA,MAAM,MAAA,GAAS,KAAA,CAAM,IAAA,CAAK,EAAE,QAAQ,QAAA,EAAS,EAAG,MAAM,IAAI,KAAA,CAAc,SAAS,CAAA,CAAE,IAAA,CAAK,CAAC,CAAC,CAAA;AAE1F,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,IAAI,IAAA,GAAO,CAAA;AACX,MAAA,MAAM,OAAA,GAAU,CAAA,CAAE,MAAA,GAAS,CAAA,GAAI,OAAA;AAE/B,MAAA,IAAI,IAAA,CAAK,SAAS,IAAA,EAAM;AACtB,QAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,UAAA,MAAM,GAAA,GAAM,IAAA,CAAK,OAAA,GAAU,CAAA,GAAI,OAAO,CAAA;AACtC,UAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,YAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,UAC9E;AACA,UAAA,MAAM,GAAA,GAAM,OAAO,GAAG,CAAA;AACtB,UAAA,IAAA,IAAQ,GAAA,GAAM,GAAA;AAAA,QAChB;AACA,QAAA,IAAA,GAAO,IAAA,CAAK,KAAK,IAAI,CAAA;AAAA,MACvB,CAAA,MAAA,IAAW,IAAA,CAAK,IAAA,KAAS,IAAA,EAAM;AAC7B,QAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,UAAA,MAAM,GAAA,GAAM,IAAA,CAAK,OAAA,GAAU,CAAA,GAAI,OAAO,CAAA;AACtC,UAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,YAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,UAC9E;AACA,UAAA,IAAA,IAAQ,IAAA,CAAK,GAAA,CAAI,MAAA,CAAO,GAAG,CAAC,CAAA;AAAA,QAC9B;AAAA,MACF,CAAA,MAAA,IAAW,IAAA,CAAK,IAAA,KAAS,KAAA,EAAO;AAC9B,QAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,UAAA,MAAM,GAAA,GAAM,IAAA,CAAK,OAAA,GAAU,CAAA,GAAI,OAAO,CAAA;AACtC,UAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,YAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,UAC9E;AACA,UAAA,IAAA,GAAO,IAAA,CAAK,IAAI,IAAA,EAAM,IAAA,CAAK,IAAI,MAAA,CAAO,GAAG,CAAC,CAAC,CAAA;AAAA,QAC7C;AAAA,MACF;AAEA,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,QAAA,MAAM,GAAA,GAAM,IAAA,CAAK,OAAA,GAAU,CAAA,GAAI,OAAO,CAAA;AACtC,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,MAAM,GAAA,GAAM,OAAO,GAAG,CAAA;AACtB,QAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,QACnE;AACA,QAAA,GAAA,CAAI,CAAC,CAAA,GAAI,IAAA,KAAS,CAAA,GAAI,MAAM,GAAA,GAAM,IAAA;AAAA,MACpC;AAAA,IACF;AAEA,IAAA,OAAO,MAAA,CAAO,QAAQ,EAAE,KAAA,EAAO,WAAW,MAAA,EAAQ,CAAA,CAAE,QAAQ,CAAA;AAAA,EAC9D;AAAA,EAEA,aAAa,CAAA,EAAmB;AAC9B,IAAA,OAAO,IAAA,CAAK,UAAU,CAAC,CAAA;AAAA,EACzB;AACF;AASO,IAAM,sBAAN,MAA0B;AAAA,EACvB,MAAA,GAAS,KAAA;AAAA,EACT,UAAA;AAAA,EACA,kBAAA;AAAA,EACA,UAAA;AAAA,EACA,SAAA;AAAA,EACA,WAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAYR,WAAA,CACE,OAAA,GAMI,EAAC,EACL;AACA,IAAA,IAAA,CAAK,UAAA,GAAa,QAAQ,UAAA,IAAc,GAAA;AACxC,IAAA,IAAA,CAAK,kBAAA,GAAqB,QAAQ,kBAAA,IAAsB,SAAA;AACxD,IAAA,IAAA,CAAK,YAAY,OAAA,CAAQ,SAAA;AACzB,IAAA,IAAA,CAAK,cAAc,OAAA,CAAQ,WAAA;AAC3B,IAAA,kBAAA,CAAmB,OAAA,CAAQ,IAAA,EAAM,MAAA,EAAQ,IAAI,CAAA;AAC7C,IAAA,IACE,CAAC,MAAA,CAAO,QAAA,CAAS,IAAA,CAAK,UAAU,CAAA,IAChC,CAAC,MAAA,CAAO,SAAA,CAAU,IAAA,CAAK,UAAU,CAAA,IACjC,IAAA,CAAK,aAAa,CAAA,EAClB;AACA,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,+BAAA;AAAA,QACA,YAAA;AAAA,QACA,IAAA,CAAK;AAAA,OACP;AAAA,IACF;AACA,IAAA,IAAI,IAAA,CAAK,kBAAA,KAAuB,SAAA,IAAa,IAAA,CAAK,uBAAuB,QAAA,EAAU;AACjF,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,kDAAA;AAAA,QACA,oBAAA;AAAA,QACA,IAAA,CAAK;AAAA,OACP;AAAA,IACF;AACA,IAAA,IAAI,IAAA,CAAK,cAAc,MAAA,EAAW;AAChC,MAAA,IACE,CAAC,MAAA,CAAO,QAAA,CAAS,IAAA,CAAK,SAAS,CAAA,IAC/B,CAAC,MAAA,CAAO,SAAA,CAAU,IAAA,CAAK,SAAS,CAAA,IAChC,IAAA,CAAK,YAAY,CAAA,EACjB;AACA,QAAA,MAAM,IAAI,qBAAA;AAAA,UACR,mCAAA;AAAA,UACA,WAAA;AAAA,UACA,IAAA,CAAK;AAAA,SACP;AAAA,MACF;AAAA,IACF;AAAA,EACF;AAAA,EAEA,IAAI,CAAA,EAAiB;AACnB,IAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,MAAA,MAAM,IAAI,qBAAA,CAAsB,oCAAA,EAAsC,GAAG,CAAA;AAAA,IAC3E;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,kBAAA,CAAmB,GAAG,GAAG,CAAA;AACzB,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,IAAA,GAAOA,eAAAA,CAAe,CAAA,EAAG,GAAG,CAAA;AAClC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AAEzC,IAAA,IAAA,CAAK,UAAA,uBAAiB,GAAA,EAAI;AAC1B,IAAA,MAAM,WAAA,GACJ,KAAK,SAAA,KAAc,MAAA,GAAY,KAAK,GAAA,CAAI,IAAA,CAAK,SAAA,EAAW,QAAQ,CAAA,GAAI,QAAA;AACtE,IAAA,MAAM,mBAAA,GAAsB,IAAA,CAAK,GAAA,CAAI,IAAA,CAAK,YAAY,WAAW,CAAA;AACjE,IAAA,MAAM,aACJ,mBAAA,IAAuB,CAAA,GACnB,CAAC,GAAG,IACJ,KAAA,CAAM,IAAA,CAAK,EAAE,MAAA,EAAQ,qBAAoB,EAAG,CAAC,GAAG,CAAA,KAAM,CAAA,IAAK,sBAAsB,CAAA,CAAE,CAAA;AAEzF,IAAA,IAAI,aAAA;AACJ,IAAA,IAAI,cAAc,QAAA,EAAU;AAC1B,MAAA,aAAA,GAAgB,KAAA,CAAM,KAAK,EAAE,MAAA,EAAQ,UAAS,EAAG,CAAC,CAAA,EAAG,CAAA,KAAM,CAAC,CAAA;AAC5D,MAAA,MAAM,MAAA,GACJ,KAAK,WAAA,KAAgB,MAAA,GAAY,mBAAmB,IAAA,CAAK,WAAW,IAAI,IAAA,CAAK,MAAA;AAC/E,MAAA,qBAAA,CAAsB,eAAe,MAAM,CAAA;AAC3C,MAAA,aAAA,GAAgB,aAAA,CAAc,KAAA,CAAM,CAAA,EAAG,WAAW,CAAA;AAAA,IACpD;AAEA,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,MAAA,MAAM,SAAmB,EAAC;AAC1B,MAAA,IAAI,aAAA,EAAe;AACjB,QAAA,KAAA,MAAW,OAAO,aAAA,EAAe;AAC/B,UAAA,MAAM,MAAM,IAAA,CAAK,CAAA,CAAE,SAAS,GAAA,GAAM,OAAA,GAAU,IAAI,OAAO,CAAA;AACvD,UAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,YAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,UAC9E;AACA,UAAA,MAAA,CAAO,IAAA,CAAK,MAAA,CAAO,GAAG,CAAC,CAAA;AAAA,QACzB;AAAA,MACF,CAAA,MAAO;AACL,QAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,UAAA,MAAM,MAAM,IAAA,CAAK,CAAA,CAAE,SAAS,CAAA,GAAI,OAAA,GAAU,IAAI,OAAO,CAAA;AACrD,UAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,YAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,UAC9E;AACA,UAAA,MAAA,CAAO,IAAA,CAAK,MAAA,CAAO,GAAG,CAAC,CAAA;AAAA,QACzB;AAAA,MACF;AACA,MAAA,MAAM,MAAA,GAAS,CAAC,GAAG,MAAM,CAAA,CAAE,KAAK,CAAC,CAAA,EAAG,CAAA,KAAM,CAAA,GAAI,CAAC,CAAA;AAC/C,MAAA,MAAM,SAAA,GAAY,WAAW,GAAA,CAAI,CAAC,MAAM,IAAA,CAAK,qBAAA,CAAsB,MAAA,EAAQ,CAAC,CAAC,CAAA;AAC7E,MAAA,IAAA,CAAK,WAAW,GAAA,CAAI,CAAA,EAAG,EAAE,SAAA,EAAW,YAAY,CAAA;AAAA,IAClD;AAEA,IAAA,IAAA,CAAK,MAAA,GAAS,IAAA;AACd,IAAA,OAAO,IAAA;AAAA,EACT;AAAA,EAEA,UAAU,CAAA,EAAmB;AAC3B,IAAA,IAAI,CAAC,IAAA,CAAK,MAAA,IAAU,CAAC,KAAK,UAAA,EAAY;AACpC,MAAA,MAAM,IAAI,eAAe,qDAAqD,CAAA;AAAA,IAChF;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,kBAAA,CAAmB,GAAG,GAAG,CAAA;AACzB,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,IAAA,GAAOA,eAAAA,CAAe,CAAA,EAAG,GAAG,CAAA;AAClC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AAEzC,IAAA,IAAI,aAAa,CAAA,EAAG;AAClB,MAAA,OAAO,KAAA,CAAM,CAAC,CAAA,EAAG,SAAS,GAAG,EAAE,KAAA,EAAO,WAAW,CAAA;AAAA,IACnD;AAEA,IAAA,MAAM,MAAA,GAAS,IAAI,KAAA,CAAgB,QAAQ,CAAA;AAC3C,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAA,CAAO,CAAC,CAAA,GAAI,IAAI,KAAA,CAAc,SAAS,CAAA;AAAA,IACzC;AAEA,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,MAAA,MAAM,OAAA,GAAU,IAAA,CAAK,UAAA,CAAW,GAAA,CAAI,CAAC,CAAA;AACrC,MAAA,IAAI,CAAC,OAAA,EAAS;AACZ,QAAA,MAAM,IAAI,YAAA,CAAa,CAAA,qDAAA,EAAwD,CAAC,CAAA,CAAE,CAAA;AAAA,MACpF;AAEA,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,QAAA,MAAM,MAAM,IAAA,CAAK,CAAA,CAAE,SAAS,CAAA,GAAI,OAAA,GAAU,IAAI,OAAO,CAAA;AACrD,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,MAAM,GAAA,GAAM,OAAO,GAAG,CAAA;AACtB,QAAA,MAAM,WAAW,IAAA,CAAK,kBAAA,CAAmB,KAAK,OAAA,CAAQ,SAAA,EAAW,QAAQ,UAAU,CAAA;AAEnF,QAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,QAAA,IAAI,CAAC,GAAA,EAAK;AACR,UAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,QACnE;AAEA,QAAA,IAAI,IAAA,CAAK,uBAAuB,SAAA,EAAW;AACzC,UAAA,GAAA,CAAI,CAAC,CAAA,GAAI,QAAA;AAAA,QACX,CAAA,MAAO;AAGL,UAAA,MAAM,eAAA,GAAkB,KAAK,GAAA,CAAI,IAAA,EAAM,KAAK,GAAA,CAAI,CAAA,GAAI,IAAA,EAAM,QAAQ,CAAC,CAAA;AACnE,UAAA,MAAM,CAAA,GAAI,KAAK,IAAA,CAAK,CAAC,IAAI,IAAA,CAAK,MAAA,CAAO,CAAA,GAAI,eAAA,GAAkB,CAAC,CAAA;AAC5D,UAAA,GAAA,CAAI,CAAC,CAAA,GAAI,CAAA;AAAA,QACX;AAAA,MACF;AAAA,IACF;AAEA,IAAA,OAAO,MAAA,CAAO,QAAQ,EAAE,KAAA,EAAO,WAAW,MAAA,EAAQ,CAAA,CAAE,QAAQ,CAAA;AAAA,EAC9D;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAYA,iBAAiB,CAAA,EAAmB;AAClC,IAAA,IAAI,CAAC,IAAA,CAAK,MAAA,IAAU,CAAC,KAAK,UAAA,EAAY;AACpC,MAAA,MAAM,IAAI,eAAe,6DAA6D,CAAA;AAAA,IACxF;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,kBAAA,CAAmB,GAAG,GAAG,CAAA;AACzB,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,IAAA,GAAOA,eAAAA,CAAe,CAAA,EAAG,GAAG,CAAA;AAClC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AAEzC,IAAA,IAAI,aAAa,CAAA,EAAG;AAClB,MAAA,OAAO,KAAA,CAAM,CAAC,CAAA,EAAG,SAAS,GAAG,EAAE,KAAA,EAAO,WAAW,CAAA;AAAA,IACnD;AAEA,IAAA,MAAM,MAAA,GAAS,IAAI,KAAA,CAAgB,QAAQ,CAAA;AAC3C,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAA,CAAO,CAAC,CAAA,GAAI,IAAI,KAAA,CAAc,SAAS,CAAA;AAAA,IACzC;AAEA,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,MAAA,MAAM,OAAA,GAAU,IAAA,CAAK,UAAA,CAAW,GAAA,CAAI,CAAC,CAAA;AACrC,MAAA,IAAI,CAAC,OAAA,EAAS;AACZ,QAAA,MAAM,IAAI,YAAA,CAAa,CAAA,qDAAA,EAAwD,CAAC,CAAA,CAAE,CAAA;AAAA,MACpF;AAEA,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,QAAA,MAAM,MAAM,IAAA,CAAK,CAAA,CAAE,SAAS,CAAA,GAAI,OAAA,GAAU,IAAI,OAAO,CAAA;AACrD,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,MAAM,KAAA,GAAQ,OAAO,GAAG,CAAA;AACxB,QAAA,IAAI,WAAW,IAAA,CAAK,kBAAA,KAAuB,WAAW,IAAA,CAAK,SAAA,CAAU,KAAK,CAAA,GAAI,KAAA;AAE9E,QAAA,QAAA,GAAW,KAAK,GAAA,CAAI,CAAA,EAAG,KAAK,GAAA,CAAI,CAAA,EAAG,QAAQ,CAAC,CAAA;AAC5C,QAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,QAAA,IAAI,CAAC,GAAA,EAAK;AACR,UAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,QACnE;AACA,QAAA,GAAA,CAAI,CAAC,IAAI,IAAA,CAAK,kBAAA,CAAmB,UAAU,OAAA,CAAQ,SAAA,EAAW,QAAQ,UAAU,CAAA;AAAA,MAClF;AAAA,IACF;AAEA,IAAA,OAAO,MAAA,CAAO,QAAQ,EAAE,KAAA,EAAO,WAAW,MAAA,EAAQ,CAAA,CAAE,QAAQ,CAAA;AAAA,EAC9D;AAAA,EAEQ,IAAI,CAAA,EAAmB;AAE7B,IAAA,MAAM,IAAA,GAAO,CAAA,GAAI,CAAA,GAAI,EAAA,GAAK,CAAA;AAC1B,IAAA,MAAM,IAAA,GAAO,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA;AACvB,IAAA,MAAM,CAAA,GAAI,CAAA,IAAK,CAAA,GAAI,SAAA,GAAY,IAAA,CAAA;AAC/B,IAAA,MAAM,EAAA,GAAK,WAAA;AACX,IAAA,MAAM,EAAA,GAAK,YAAA;AACX,IAAA,MAAM,EAAA,GAAK,WAAA;AACX,IAAA,MAAM,EAAA,GAAK,YAAA;AACX,IAAA,MAAM,EAAA,GAAK,WAAA;AACX,IAAA,MAAM,IAAA,GAAA,CAAA,CAAA,CAAA,CAAW,KAAK,CAAA,GAAI,EAAA,IAAM,IAAI,EAAA,IAAM,CAAA,GAAI,EAAA,IAAM,CAAA,GAAI,EAAA,IAAM,CAAA;AAC9D,IAAA,OAAO,QAAQ,CAAA,GAAI,IAAA,GAAO,KAAK,GAAA,CAAI,CAAC,OAAO,IAAI,CAAA,CAAA;AAAA,EACjD;AAAA,EAEQ,UAAU,CAAA,EAAmB;AACnC,IAAA,OAAO,GAAA,IAAO,IAAI,IAAA,CAAK,GAAA,CAAI,IAAI,IAAA,CAAK,IAAA,CAAK,CAAC,CAAC,CAAA,CAAA;AAAA,EAC7C;AAAA,EAEQ,OAAO,CAAA,EAAmB;AAChC,IAAA,MAAM,CAAA,GAAI,KAAA;AACV,IAAA,MAAM,CAAA,GAAI,CAAA,IAAK,IAAA,CAAK,EAAA,GAAK,CAAA,CAAA,GAAK,KAAK,GAAA,CAAI,CAAA,GAAI,CAAA,GAAI,CAAC,CAAA,GAAI,CAAA;AACpD,IAAA,MAAM,IAAA,GAAO,CAAA,GAAI,CAAA,GAAI,EAAA,GAAK,CAAA;AAC1B,IAAA,OAAO,IAAA,GAAO,IAAA,CAAK,IAAA,CAAK,IAAA,CAAK,KAAK,CAAA,GAAI,CAAA,GAAI,IAAA,CAAK,GAAA,CAAI,IAAI,CAAA,GAAI,CAAC,CAAA,GAAI,CAAC,IAAI,CAAC,CAAA;AAAA,EACxE;AAAA,EAEQ,qBAAA,CAAsB,QAAkB,CAAA,EAAmB;AACjE,IAAA,IAAI,MAAA,CAAO,WAAW,CAAA,EAAG;AACvB,MAAA,MAAM,IAAI,aAAa,wDAAwD,CAAA;AAAA,IACjF;AACA,IAAA,IAAI,MAAA,CAAO,WAAW,CAAA,EAAG;AACvB,MAAA,MAAM,IAAA,GAAO,OAAO,CAAC,CAAA;AACrB,MAAA,IAAI,SAAS,MAAA,EAAW;AACtB,QAAA,MAAM,IAAI,aAAa,sCAAsC,CAAA;AAAA,MAC/D;AACA,MAAA,OAAO,IAAA;AAAA,IACT;AAEA,IAAA,MAAM,QAAA,GAAW,CAAA,IAAK,MAAA,CAAO,MAAA,GAAS,CAAA,CAAA;AACtC,IAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,KAAA,CAAM,QAAQ,CAAA;AACjC,IAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,IAAA,CAAK,QAAQ,CAAA;AAChC,IAAA,MAAM,UAAA,GAAa,OAAO,KAAK,CAAA;AAC/B,IAAA,MAAM,UAAA,GAAa,OAAO,KAAK,CAAA;AAC/B,IAAA,IAAI,UAAA,KAAe,MAAA,IAAa,UAAA,KAAe,MAAA,EAAW;AACxD,MAAA,MAAM,IAAI,aAAa,4DAA4D,CAAA;AAAA,IACrF;AAEA,IAAA,IAAI,UAAU,KAAA,EAAO;AACnB,MAAA,OAAO,UAAA;AAAA,IACT;AAEA,IAAA,MAAM,SAAS,QAAA,GAAW,KAAA;AAC1B,IAAA,OAAO,UAAA,IAAc,CAAA,GAAI,MAAA,CAAA,GAAU,UAAA,GAAa,MAAA;AAAA,EAClD;AAAA,EAEQ,kBAAA,CAAmB,KAAA,EAAe,SAAA,EAAqB,UAAA,EAA8B;AAC3F,IAAA,MAAM,IAAI,SAAA,CAAU,MAAA;AACpB,IAAA,IAAI,MAAM,CAAA,EAAG;AACX,MAAA,OAAO,CAAA;AAAA,IACT;AACA,IAAA,IAAI,MAAM,CAAA,EAAG;AACX,MAAA,MAAM,aAAA,GAAgB,WAAW,CAAC,CAAA;AAClC,MAAA,IAAI,kBAAkB,MAAA,EAAW;AAC/B,QAAA,MAAM,IAAI,aAAa,4CAA4C,CAAA;AAAA,MACrE;AACA,MAAA,OAAO,aAAA;AAAA,IACT;AAEA,IAAA,MAAM,aAAA,GAAgB,UAAU,CAAC,CAAA;AACjC,IAAA,MAAM,YAAA,GAAe,SAAA,CAAU,CAAA,GAAI,CAAC,CAAA;AACpC,IAAA,IAAI,aAAA,KAAkB,MAAA,IAAa,YAAA,KAAiB,MAAA,EAAW;AAC7D,MAAA,MAAM,IAAI,aAAa,4CAA4C,CAAA;AAAA,IACrE;AACA,IAAA,IAAI,SAAS,aAAA,EAAe;AAC1B,MAAA,OAAO,CAAA;AAAA,IACT;AACA,IAAA,IAAI,SAAS,YAAA,EAAc;AACzB,MAAA,OAAO,CAAA;AAAA,IACT;AAEA,IAAA,IAAI,IAAA,GAAO,CAAA;AACX,IAAA,IAAI,QAAQ,CAAA,GAAI,CAAA;AAChB,IAAA,OAAO,IAAA,GAAO,IAAI,KAAA,EAAO;AACvB,MAAA,MAAM,GAAA,GAAM,IAAA,CAAK,KAAA,CAAA,CAAO,IAAA,GAAO,SAAS,CAAC,CAAA;AACzC,MAAA,MAAM,QAAA,GAAW,UAAU,GAAG,CAAA;AAC9B,MAAA,IAAI,aAAa,MAAA,EAAW;AAC1B,QAAA,MAAM,IAAI,aAAa,2CAA2C,CAAA;AAAA,MACpE;AACA,MAAA,IAAI,YAAY,KAAA,EAAO;AACrB,QAAA,IAAA,GAAO,GAAA;AAAA,MACT,CAAA,MAAO;AACL,QAAA,KAAA,GAAQ,GAAA;AAAA,MACV;AAAA,IACF;AAEA,IAAA,MAAM,KAAA,GAAQ,UAAU,IAAI,CAAA;AAC5B,IAAA,MAAM,MAAA,GAAS,UAAU,KAAK,CAAA;AAC9B,IAAA,MAAM,KAAA,GAAQ,WAAW,IAAI,CAAA;AAC7B,IAAA,MAAM,MAAA,GAAS,WAAW,KAAK,CAAA;AAC/B,IAAA,IACE,UAAU,MAAA,IACV,MAAA,KAAW,UACX,KAAA,KAAU,MAAA,IACV,WAAW,MAAA,EACX;AACA,MAAA,MAAM,IAAI,aAAa,uDAAuD,CAAA;AAAA,IAChF;AACA,IAAA,IAAI,UAAU,KAAA,EAAO;AACnB,MAAA,OAAA,CAAQ,QAAQ,MAAA,IAAU,CAAA;AAAA,IAC5B;AAEA,IAAA,MAAM,KAAA,GAAA,CAAS,KAAA,GAAQ,KAAA,KAAU,MAAA,GAAS,KAAA,CAAA;AAC1C,IAAA,OAAO,KAAA,GAAQ,SAAS,MAAA,GAAS,KAAA,CAAA;AAAA,EACnC;AAAA,EAEQ,kBAAA,CAAmB,QAAA,EAAkB,SAAA,EAAqB,UAAA,EAA8B;AAC9F,IAAA,MAAM,IAAI,UAAA,CAAW,MAAA;AACrB,IAAA,IAAI,MAAM,CAAA,EAAG;AACX,MAAA,OAAO,CAAA;AAAA,IACT;AACA,IAAA,IAAI,MAAM,CAAA,EAAG;AACX,MAAA,MAAM,YAAA,GAAe,UAAU,CAAC,CAAA;AAChC,MAAA,IAAI,iBAAiB,MAAA,EAAW;AAC9B,QAAA,MAAM,IAAI,aAAa,wCAAwC,CAAA;AAAA,MACjE;AACA,MAAA,OAAO,YAAA;AAAA,IACT;AAEA,IAAA,MAAM,QAAA,GAAW,WAAW,CAAC,CAAA;AAC7B,IAAA,MAAM,OAAA,GAAU,UAAA,CAAW,CAAA,GAAI,CAAC,CAAA;AAChC,IAAA,IAAI,QAAA,KAAa,MAAA,IAAa,OAAA,KAAY,MAAA,EAAW;AACnD,MAAA,MAAM,IAAI,aAAa,6CAA6C,CAAA;AAAA,IACtE;AACA,IAAA,IAAI,YAAY,QAAA,EAAU;AACxB,MAAA,MAAM,aAAA,GAAgB,UAAU,CAAC,CAAA;AACjC,MAAA,IAAI,kBAAkB,MAAA,EAAW;AAC/B,QAAA,MAAM,IAAI,aAAa,4CAA4C,CAAA;AAAA,MACrE;AACA,MAAA,OAAO,aAAA;AAAA,IACT;AACA,IAAA,IAAI,YAAY,OAAA,EAAS;AACvB,MAAA,MAAM,YAAA,GAAe,SAAA,CAAU,CAAA,GAAI,CAAC,CAAA;AACpC,MAAA,IAAI,iBAAiB,MAAA,EAAW;AAC9B,QAAA,MAAM,IAAI,aAAa,4CAA4C,CAAA;AAAA,MACrE;AACA,MAAA,OAAO,YAAA;AAAA,IACT;AAEA,IAAA,IAAI,IAAA,GAAO,CAAA;AACX,IAAA,IAAI,QAAQ,CAAA,GAAI,CAAA;AAChB,IAAA,OAAO,IAAA,GAAO,IAAI,KAAA,EAAO;AACvB,MAAA,MAAM,GAAA,GAAM,IAAA,CAAK,KAAA,CAAA,CAAO,IAAA,GAAO,SAAS,CAAC,CAAA;AACzC,MAAA,MAAM,MAAA,GAAS,WAAW,GAAG,CAAA;AAC7B,MAAA,IAAI,WAAW,MAAA,EAAW;AACxB,QAAA,MAAM,IAAI,aAAa,4CAA4C,CAAA;AAAA,MACrE;AACA,MAAA,IAAI,UAAU,QAAA,EAAU;AACtB,QAAA,IAAA,GAAO,GAAA;AAAA,MACT,CAAA,MAAO;AACL,QAAA,KAAA,GAAQ,GAAA;AAAA,MACV;AAAA,IACF;AAEA,IAAA,MAAM,KAAA,GAAQ,WAAW,IAAI,CAAA;AAC7B,IAAA,MAAM,MAAA,GAAS,WAAW,KAAK,CAAA;AAC/B,IAAA,MAAM,KAAA,GAAQ,UAAU,IAAI,CAAA;AAC5B,IAAA,MAAM,MAAA,GAAS,UAAU,KAAK,CAAA;AAC9B,IAAA,IACE,UAAU,MAAA,IACV,MAAA,KAAW,UACX,KAAA,KAAU,MAAA,IACV,WAAW,MAAA,EACX;AACA,MAAA,MAAM,IAAI,aAAa,uDAAuD,CAAA;AAAA,IAChF;AACA,IAAA,IAAI,UAAU,KAAA,EAAO;AACnB,MAAA,OAAA,CAAQ,QAAQ,MAAA,IAAU,CAAA;AAAA,IAC5B;AACA,IAAA,MAAM,KAAA,GAAA,CAAS,QAAA,GAAW,KAAA,KAAU,MAAA,GAAS,KAAA,CAAA;AAC7C,IAAA,OAAO,KAAA,GAAQ,SAAS,MAAA,GAAS,KAAA,CAAA;AAAA,EACnC;AAAA,EAEA,aAAa,CAAA,EAAmB;AAC9B,IAAA,OAAO,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,CAAE,UAAU,CAAC,CAAA;AAAA,EAChC;AACF;AASO,IAAM,mBAAN,MAAuB;AAAA,EACpB,MAAA,GAAS,KAAA;AAAA,EACT,MAAA;AAAA,EACA,QAAA;AAAA,EACA,WAAA;AAAA,EACA,KAAA;AAAA,EACA,MAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAUR,WAAA,CACE,OAAA,GAII,EAAC,EACL;AACA,IAAA,IAAA,CAAK,MAAA,GAAS,QAAQ,MAAA,IAAU,aAAA;AAChC,IAAA,IAAI,IAAA,CAAK,MAAA,KAAW,SAAA,IAAa,IAAA,CAAK,WAAW,aAAA,EAAe;AAC9D,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,2CAAA;AAAA,QACA,QAAA;AAAA,QACA,IAAA,CAAK;AAAA,OACP;AAAA,IACF;AACA,IAAA,IAAA,CAAK,WAAA,GAAc,kBAAA,CAAmB,OAAA,CAAQ,WAAA,EAAa,eAAe,KAAK,CAAA;AAC/E,IAAA,kBAAA,CAAmB,OAAA,CAAQ,IAAA,EAAM,MAAA,EAAQ,IAAI,CAAA;AAAA,EAC/C;AAAA,EAEA,IAAI,CAAA,EAAiB;AACnB,IAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,MAAA,MAAM,IAAI,qBAAA,CAAsB,oCAAA,EAAsC,GAAG,CAAA;AAAA,IAC3E;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,kBAAA,CAAmB,GAAG,GAAG,CAAA;AACzB,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,IAAA,GAAOA,eAAAA,CAAe,CAAA,EAAG,GAAG,CAAA;AAClC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AAEzC,IAAA,MAAM,OAAA,GAAU,IAAI,KAAA,CAAc,SAAS,CAAA;AAC3C,IAAA,MAAM,KAAA,GAAQ,KAAK,WAAA,GAAc,IAAI,MAAc,SAAS,CAAA,CAAE,IAAA,CAAK,CAAC,CAAA,GAAI,MAAA;AACxE,IAAA,MAAM,MAAA,GAAS,KAAK,WAAA,GAAc,IAAI,MAAc,SAAS,CAAA,CAAE,IAAA,CAAK,CAAC,CAAA,GAAI,MAAA;AAEzE,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,MAAA,MAAM,aAAA,GAAgB,IAAI,KAAA,CAAc,QAAQ,CAAA;AAChD,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,QAAA,MAAM,MAAM,IAAA,CAAK,CAAA,CAAE,SAAS,CAAA,GAAI,OAAA,GAAU,IAAI,OAAO,CAAA;AACrD,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,MAAM,KAAA,GAAQ,OAAO,GAAG,CAAA;AACxB,QAAA,IAAI,IAAA,CAAK,MAAA,KAAW,SAAA,IAAa,KAAA,IAAS,CAAA,EAAG;AAC3C,UAAA,MAAM,IAAI,qBAAA;AAAA,YACR,kEAAkE,CAAC,CAAA,CAAA,CAAA;AAAA,YACnE,GAAA;AAAA,YACA;AAAA,WACF;AAAA,QACF;AACA,QAAA,aAAA,CAAc,CAAC,CAAA,GAAI,KAAA;AAAA,MACrB;AACA,MAAA,MAAM,MAAA,GAAS,IAAA,CAAK,cAAA,CAAe,aAAa,CAAA;AAChD,MAAA,OAAA,CAAQ,CAAC,CAAA,GAAI,MAAA;AAEb,MAAA,IAAI,IAAA,CAAK,WAAA,IAAe,KAAA,IAAS,MAAA,EAAQ;AACvC,QAAA,IAAI,GAAA,GAAM,CAAA;AAEV,QAAA,KAAA,MAAW,SAAS,aAAA,EAAe;AACjC,UAAA,MAAM,WAAA,GACJ,IAAA,CAAK,MAAA,KAAW,SAAA,GACZ,IAAA,CAAK,oBAAA,CAAqB,KAAA,EAAO,MAAM,CAAA,GACvC,IAAA,CAAK,wBAAA,CAAyB,KAAA,EAAO,MAAM,CAAA;AACjD,UAAA,GAAA,IAAO,WAAA;AAAA,QACT;AACA,QAAA,MAAM,OAAO,GAAA,GAAM,QAAA;AACnB,QAAA,KAAA,CAAM,CAAC,CAAA,GAAI,IAAA;AAGX,QAAA,IAAI,SAAA,GAAY,CAAA;AAChB,QAAA,KAAA,MAAW,SAAS,aAAA,EAAe;AACjC,UAAA,MAAM,WAAA,GACJ,IAAA,CAAK,MAAA,KAAW,SAAA,GACZ,IAAA,CAAK,oBAAA,CAAqB,KAAA,EAAO,MAAM,CAAA,GACvC,IAAA,CAAK,wBAAA,CAAyB,KAAA,EAAO,MAAM,CAAA;AACjD,UAAA,MAAM,OAAO,WAAA,GAAc,IAAA;AAC3B,UAAA,SAAA,IAAa,IAAA,GAAO,IAAA;AAAA,QACtB;AACA,QAAA,MAAM,WAAW,SAAA,GAAY,QAAA;AAC7B,QAAA,MAAM,MAAM,IAAA,CAAK,IAAA,CAAK,KAAK,GAAA,CAAI,QAAA,EAAU,CAAC,CAAC,CAAA;AAC3C,QAAA,MAAA,CAAO,CAAC,CAAA,GAAI,GAAA,KAAQ,CAAA,GAAI,CAAA,GAAI,GAAA;AAAA,MAC9B;AAAA,IACF;AAEA,IAAA,IAAA,CAAK,QAAA,GAAW,OAAA;AAChB,IAAA,IAAA,CAAK,KAAA,GAAQ,IAAA,CAAK,WAAA,GAAc,KAAA,GAAQ,MAAA;AACxC,IAAA,IAAA,CAAK,MAAA,GAAS,IAAA,CAAK,WAAA,GAAc,MAAA,GAAS,MAAA;AAC1C,IAAA,IAAA,CAAK,MAAA,GAAS,IAAA;AACd,IAAA,OAAO,IAAA;AAAA,EACT;AAAA,EAEA,UAAU,CAAA,EAAmB;AAC3B,IAAA,IAAI,CAAC,IAAA,CAAK,MAAA,IAAU,CAAC,KAAK,QAAA,EAAU;AAClC,MAAA,MAAM,IAAI,eAAe,kDAAkD,CAAA;AAAA,IAC7E;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,kBAAA,CAAmB,GAAG,GAAG,CAAA;AACzB,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,IAAA,GAAOA,eAAAA,CAAe,CAAA,EAAG,GAAG,CAAA;AAClC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AAEzC,IAAA,IAAI,KAAK,WAAA,KAAgB,CAAC,KAAK,KAAA,IAAS,CAAC,KAAK,MAAA,CAAA,EAAS;AACrD,MAAA,MAAM,IAAI,aAAa,gEAAgE,CAAA;AAAA,IACzF;AAEA,IAAA,MAAM,MAAA,GAAS,KAAA,CAAM,IAAA,CAAK,EAAE,QAAQ,QAAA,EAAS,EAAG,MAAM,IAAI,KAAA,CAAc,SAAS,CAAA,CAAE,IAAA,CAAK,CAAC,CAAC,CAAA;AAE1F,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAM,OAAA,GAAU,CAAA,CAAE,MAAA,GAAS,CAAA,GAAI,OAAA;AAC/B,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,QAAA,MAAM,GAAA,GAAM,IAAA,CAAK,OAAA,GAAU,CAAA,GAAI,OAAO,CAAA;AACtC,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,MAAM,GAAA,GAAM,OAAO,GAAG,CAAA;AACtB,QAAA,MAAM,MAAA,GAAS,IAAA,CAAK,QAAA,CAAS,CAAC,CAAA;AAC9B,QAAA,IAAI,WAAW,MAAA,EAAW;AACxB,UAAA,MAAM,IAAI,YAAA,CAAa,CAAA,kDAAA,EAAqD,CAAC,CAAA,CAAE,CAAA;AAAA,QACjF;AAEA,QAAA,IAAI,WAAA;AACJ,QAAA,IAAI,IAAA,CAAK,WAAW,SAAA,EAAW;AAC7B,UAAA,IAAI,OAAO,CAAA,EAAG;AACZ,YAAA,MAAM,IAAI,qBAAA,CAAsB,2CAAA,EAA6C,GAAA,EAAK,GAAG,CAAA;AAAA,UACvF;AACA,UAAA,WAAA,GAAc,IAAA,CAAK,oBAAA,CAAqB,GAAA,EAAK,MAAM,CAAA;AAAA,QACrD,CAAA,MAAO;AACL,UAAA,WAAA,GAAc,IAAA,CAAK,wBAAA,CAAyB,GAAA,EAAK,MAAM,CAAA;AAAA,QACzD;AAEA,QAAA,IAAI,IAAA,CAAK,WAAA,IAAe,IAAA,CAAK,KAAA,IAAS,KAAK,MAAA,EAAQ;AACjD,UAAA,MAAM,IAAA,GAAO,IAAA,CAAK,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA;AAC9B,UAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,MAAA,CAAO,CAAC,CAAA,IAAK,CAAA;AAChC,UAAA,WAAA,GAAA,CAAe,cAAc,IAAA,IAAQ,KAAA;AAAA,QACvC;AAEA,QAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,QACnE;AACA,QAAA,GAAA,CAAI,CAAC,CAAA,GAAI,WAAA;AAAA,MACX;AAAA,IACF;AAEA,IAAA,OAAO,MAAA,CAAO,QAAQ,EAAE,KAAA,EAAO,WAAW,MAAA,EAAQ,CAAA,CAAE,QAAQ,CAAA;AAAA,EAC9D;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAUA,iBAAiB,CAAA,EAAmB;AAClC,IAAA,IAAI,CAAC,IAAA,CAAK,MAAA,IAAU,CAAC,KAAK,QAAA,EAAU;AAClC,MAAA,MAAM,IAAI,eAAe,0DAA0D,CAAA;AAAA,IACrF;AACA,IAAA,QAAA,CAAS,GAAG,GAAG,CAAA;AACf,IAAA,mBAAA,CAAoB,GAAG,GAAG,CAAA;AAC1B,IAAA,kBAAA,CAAmB,GAAG,GAAG,CAAA;AACzB,IAAA,MAAM,CAAC,QAAA,EAAU,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAC1C,IAAA,MAAM,IAAA,GAAOA,eAAAA,CAAe,CAAA,EAAG,GAAG,CAAA;AAClC,IAAA,MAAM,CAAC,OAAA,EAAS,OAAO,CAAA,GAAI,aAAa,CAAC,CAAA;AAEzC,IAAA,IAAI,KAAK,WAAA,KAAgB,CAAC,KAAK,KAAA,IAAS,CAAC,KAAK,MAAA,CAAA,EAAS;AACrD,MAAA,MAAM,IAAI,aAAa,gEAAgE,CAAA;AAAA,IACzF;AAEA,IAAA,MAAM,MAAA,GAAS,KAAA,CAAM,IAAA,CAAK,EAAE,QAAQ,QAAA,EAAS,EAAG,MAAM,IAAI,KAAA,CAAc,SAAS,CAAA,CAAE,IAAA,CAAK,CAAC,CAAC,CAAA;AAE1F,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAM,OAAA,GAAU,CAAA,CAAE,MAAA,GAAS,CAAA,GAAI,OAAA;AAC/B,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,QAAA,MAAM,GAAA,GAAM,IAAA,CAAK,OAAA,GAAU,CAAA,GAAI,OAAO,CAAA;AACtC,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,qDAAqD,CAAA;AAAA,QAC9E;AACA,QAAA,IAAI,GAAA,GAAM,OAAO,GAAG,CAAA;AAEpB,QAAA,IAAI,IAAA,CAAK,WAAA,IAAe,IAAA,CAAK,KAAA,IAAS,KAAK,MAAA,EAAQ;AACjD,UAAA,MAAM,IAAA,GAAO,IAAA,CAAK,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA;AAC9B,UAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,MAAA,CAAO,CAAC,CAAA,IAAK,CAAA;AAChC,UAAA,GAAA,GAAM,MAAM,KAAA,GAAQ,IAAA;AAAA,QACtB;AAEA,QAAA,MAAM,MAAA,GAAS,IAAA,CAAK,QAAA,CAAS,CAAC,CAAA;AAC9B,QAAA,IAAI,WAAW,MAAA,EAAW;AACxB,UAAA,MAAM,IAAI,YAAA,CAAa,CAAA,kDAAA,EAAqD,CAAC,CAAA,CAAE,CAAA;AAAA,QACjF;AAEA,QAAA,IAAI,QAAA;AACJ,QAAA,IAAI,IAAA,CAAK,WAAW,SAAA,EAAW;AAC7B,UAAA,QAAA,GAAW,IAAA,CAAK,kBAAA,CAAmB,GAAA,EAAK,MAAM,CAAA;AAAA,QAChD,CAAA,MAAO;AACL,UAAA,QAAA,GAAW,IAAA,CAAK,sBAAA,CAAuB,GAAA,EAAK,MAAM,CAAA;AAAA,QACpD;AAEA,QAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,QAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,UAAA,MAAM,IAAI,aAAa,0CAA0C,CAAA;AAAA,QACnE;AACA,QAAA,GAAA,CAAI,CAAC,CAAA,GAAI,QAAA;AAAA,MACX;AAAA,IACF;AAEA,IAAA,OAAO,MAAA,CAAO,QAAQ,EAAE,KAAA,EAAO,WAAW,MAAA,EAAQ,CAAA,CAAE,QAAQ,CAAA;AAAA,EAC9D;AAAA,EAEQ,oBAAA,CAAqB,OAAe,MAAA,EAAwB;AAClE,IAAA,OAAO,IAAA,CAAK,GAAA,CAAI,MAAM,CAAA,GAAI,KAAA,GAAQ,IAAA,CAAK,GAAA,CAAI,KAAK,CAAA,GAAA,CAAK,KAAA,IAAS,MAAA,GAAS,CAAA,IAAK,MAAA;AAAA,EAC9E;AAAA,EAEQ,wBAAA,CAAyB,OAAe,MAAA,EAAwB;AACtE,IAAA,IAAI,SAAS,CAAA,EAAG;AACd,MAAA,OAAO,IAAA,CAAK,GAAA,CAAI,MAAM,CAAA,GAAI,KAAA,GAAQ,IAAA,CAAK,GAAA,CAAI,KAAA,GAAQ,CAAC,CAAA,GAAA,CAAA,CAAM,KAAA,GAAQ,CAAA,KAAM,SAAS,CAAA,IAAK,MAAA;AAAA,IACxF;AACA,IAAA,MAAM,iBAAiB,CAAA,GAAI,MAAA;AAC3B,IAAA,OAAO,IAAA,CAAK,GAAA,CAAI,cAAc,CAAA,GAAI,QAC9B,CAAC,IAAA,CAAK,GAAA,CAAI,CAAA,GAAI,KAAK,CAAA,GACnB,EAAA,CAAG,CAAA,GAAI,KAAA,KAAU,iBAAiB,CAAA,CAAA,GAAK,cAAA;AAAA,EAC7C;AAAA,EAEQ,kBAAA,CAAmB,OAAe,MAAA,EAAwB;AAChE,IAAA,IAAI,IAAA,CAAK,GAAA,CAAI,MAAM,CAAA,GAAI,KAAA,EAAO;AAC5B,MAAA,OAAO,IAAA,CAAK,IAAI,KAAK,CAAA;AAAA,IACvB;AACA,IAAA,MAAM,IAAA,GAAO,QAAQ,MAAA,GAAS,CAAA;AAC9B,IAAA,IAAI,QAAQ,CAAA,EAAG;AACb,MAAA,MAAM,IAAI,qBAAA,CAAsB,2CAAA,EAA6C,GAAA,EAAK,KAAK,CAAA;AAAA,IACzF;AACA,IAAA,OAAO,SAAS,CAAA,GAAI,MAAA,CAAA;AAAA,EACtB;AAAA,EAEQ,sBAAA,CAAuB,OAAe,MAAA,EAAwB;AACpE,IAAA,IAAI,SAAS,CAAA,EAAG;AACd,MAAA,IAAI,IAAA,CAAK,GAAA,CAAI,MAAM,CAAA,GAAI,KAAA,EAAO;AAC5B,QAAA,OAAO,IAAA,CAAK,GAAA,CAAI,KAAK,CAAA,GAAI,CAAA;AAAA,MAC3B;AACA,MAAA,MAAME,KAAAA,GAAO,QAAQ,MAAA,GAAS,CAAA;AAC9B,MAAA,IAAIA,SAAQ,CAAA,EAAG;AACb,QAAA,MAAM,IAAI,qBAAA;AAAA,UACR,+CAAA;AAAA,UACA,GAAA;AAAA,UACA;AAAA,SACF;AAAA,MACF;AACA,MAAA,OAAOA,KAAAA,KAAS,IAAI,MAAA,CAAA,GAAU,CAAA;AAAA,IAChC;AAEA,IAAA,MAAM,iBAAiB,CAAA,GAAI,MAAA;AAC3B,IAAA,IAAI,IAAA,CAAK,GAAA,CAAI,cAAc,CAAA,GAAI,KAAA,EAAO;AACpC,MAAA,OAAO,CAAA,GAAI,IAAA,CAAK,GAAA,CAAI,CAAC,KAAK,CAAA;AAAA,IAC5B;AACA,IAAA,MAAM,IAAA,GAAO,IAAI,KAAA,GAAQ,cAAA;AACzB,IAAA,IAAI,QAAQ,CAAA,EAAG;AACb,MAAA,MAAM,IAAI,qBAAA,CAAsB,+CAAA,EAAiD,GAAA,EAAK,KAAK,CAAA;AAAA,IAC7F;AACA,IAAA,OAAO,CAAA,GAAI,SAAS,CAAA,GAAI,cAAA,CAAA;AAAA,EAC1B;AAAA,EAEQ,aAAA,CAAc,QAA2B,MAAA,EAAwB;AACvE,IAAA,MAAM,WAAA,GAAc,IAAI,KAAA,CAAc,MAAA,CAAO,MAAM,CAAA;AACnD,IAAA,IAAI,QAAA,GAAW,CAAA;AAEf,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,MAAA,CAAO,QAAQ,CAAA,EAAA,EAAK;AACtC,MAAA,MAAM,KAAA,GAAQ,OAAO,CAAC,CAAA;AACtB,MAAA,IAAI,UAAU,MAAA,EAAW;AACvB,QAAA,MAAM,IAAI,aAAa,2DAA2D,CAAA;AAAA,MACpF;AAEA,MAAA,IAAI,gBAAA;AACJ,MAAA,IAAI,IAAA,CAAK,WAAW,SAAA,EAAW;AAC7B,QAAA,IAAI,SAAS,CAAA,EAAG;AACd,UAAA,OAAO,MAAA,CAAO,iBAAA;AAAA,QAChB;AACA,QAAA,gBAAA,GAAmB,IAAA,CAAK,oBAAA,CAAqB,KAAA,EAAO,MAAM,CAAA;AAC1D,QAAA,QAAA,IAAA,CAAa,MAAA,GAAS,CAAA,IAAK,IAAA,CAAK,GAAA,CAAI,KAAK,CAAA;AAAA,MAC3C,CAAA,MAAO;AACL,QAAA,gBAAA,GAAmB,IAAA,CAAK,wBAAA,CAAyB,KAAA,EAAO,MAAM,CAAA;AAC9D,QAAA,QAAA,IACE,KAAA,IAAS,CAAA,GAAA,CAAK,MAAA,GAAS,CAAA,IAAK,KAAK,GAAA,CAAI,KAAA,GAAQ,CAAC,CAAA,GAAA,CAAK,CAAA,GAAI,MAAA,IAAU,IAAA,CAAK,GAAA,CAAI,IAAI,KAAK,CAAA;AAAA,MACvF;AAEA,MAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,gBAAgB,CAAA,EAAG;AACtC,QAAA,OAAO,MAAA,CAAO,iBAAA;AAAA,MAChB;AACA,MAAA,WAAA,CAAY,CAAC,CAAA,GAAI,gBAAA;AAAA,IACnB;AAEA,IAAA,IAAI,GAAA,GAAM,CAAA;AACV,IAAA,KAAA,MAAW,SAAS,WAAA,EAAa;AAC/B,MAAA,GAAA,IAAO,KAAA;AAAA,IACT;AACA,IAAA,MAAM,IAAA,GAAO,MAAM,WAAA,CAAY,MAAA;AAE/B,IAAA,IAAI,WAAA,GAAc,CAAA;AAClB,IAAA,KAAA,MAAW,SAAS,WAAA,EAAa;AAC/B,MAAA,MAAM,QAAQ,KAAA,GAAQ,IAAA;AACtB,MAAA,WAAA,IAAe,KAAA,GAAQ,KAAA;AAAA,IACzB;AACA,IAAA,MAAM,QAAA,GAAW,cAAc,WAAA,CAAY,MAAA;AAC3C,IAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,QAAQ,CAAA,IAAK,YAAY,KAAA,EAAO;AACnD,MAAA,OAAO,MAAA,CAAO,iBAAA;AAAA,IAChB;AAEA,IAAA,OAAO,OAAO,WAAA,CAAY,MAAA,GAAS,IAAA,CAAK,GAAA,CAAI,QAAQ,CAAA,GAAI,QAAA;AAAA,EAC1D;AAAA,EAEQ,eAAe,MAAA,EAAmC;AACxD,IAAA,IAAI,MAAA,CAAO,SAAS,CAAA,EAAG;AACrB,MAAA,OAAO,CAAA;AAAA,IACT;AAEA,IAAA,IAAI,WAAW,MAAA,CAAO,iBAAA;AACtB,IAAA,IAAI,WAAW,MAAA,CAAO,iBAAA;AACtB,IAAA,KAAA,MAAW,SAAS,MAAA,EAAQ;AAC1B,MAAA,IAAI,KAAA,GAAQ,UAAU,QAAA,GAAW,KAAA;AACjC,MAAA,IAAI,KAAA,GAAQ,UAAU,QAAA,GAAW,KAAA;AAAA,IACnC;AAEA,IAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,QAAQ,CAAA,IAAK,CAAC,MAAA,CAAO,QAAA,CAAS,QAAQ,CAAA,IAAK,QAAA,GAAW,QAAA,IAAY,KAAA,EAAO;AAC5F,MAAA,OAAO,CAAA;AAAA,IACT;AAEA,IAAA,IAAI,IAAA,GAAO,EAAA;AACX,IAAA,IAAI,KAAA,GAAQ,CAAA;AACZ,IAAA,MAAM,GAAA,GAAA,CAAO,IAAA,CAAK,IAAA,CAAK,CAAC,IAAI,CAAA,IAAK,CAAA;AACjC,IAAA,IAAI,CAAA,GAAI,KAAA,GAAQ,GAAA,IAAO,KAAA,GAAQ,IAAA,CAAA;AAC/B,IAAA,IAAI,CAAA,GAAI,IAAA,GAAO,GAAA,IAAO,KAAA,GAAQ,IAAA,CAAA;AAC9B,IAAA,IAAI,EAAA,GAAK,IAAA,CAAK,aAAA,CAAc,MAAA,EAAQ,CAAC,CAAA;AACrC,IAAA,IAAI,EAAA,GAAK,IAAA,CAAK,aAAA,CAAc,MAAA,EAAQ,CAAC,CAAA;AAErC,IAAA,KAAA,IAAS,IAAA,GAAO,CAAA,EAAG,IAAA,GAAO,EAAA,EAAI,IAAA,EAAA,EAAQ;AACpC,MAAA,IAAI,IAAA,CAAK,GAAA,CAAI,KAAA,GAAQ,IAAI,IAAI,IAAA,EAAM;AACnC,MAAA,IAAI,KAAK,EAAA,EAAI;AACX,QAAA,KAAA,GAAQ,CAAA;AACR,QAAA,CAAA,GAAI,CAAA;AACJ,QAAA,EAAA,GAAK,EAAA;AACL,QAAA,CAAA,GAAI,KAAA,GAAQ,OAAO,KAAA,GAAQ,IAAA,CAAA;AAC3B,QAAA,EAAA,GAAK,IAAA,CAAK,aAAA,CAAc,MAAA,EAAQ,CAAC,CAAA;AAAA,MACnC,CAAA,MAAO;AACL,QAAA,IAAA,GAAO,CAAA;AACP,QAAA,CAAA,GAAI,CAAA;AACJ,QAAA,EAAA,GAAK,EAAA;AACL,QAAA,CAAA,GAAI,IAAA,GAAO,OAAO,KAAA,GAAQ,IAAA,CAAA;AAC1B,QAAA,EAAA,GAAK,IAAA,CAAK,aAAA,CAAc,MAAA,EAAQ,CAAC,CAAA;AAAA,MACnC;AAAA,IACF;AAEA,IAAA,MAAM,UAAA,GAAa,CAAC,IAAA,EAAM,KAAA,EAAA,CAAQ,IAAA,GAAO,SAAS,CAAA,EAAG,CAAA,EAAG,CAAA,EAAG,CAAA,EAAG,EAAE,CAAA;AAChE,IAAA,IAAI,UAAA,GAAa,CAAA;AACjB,IAAA,IAAI,YAAY,MAAA,CAAO,iBAAA;AACvB,IAAA,KAAA,MAAW,UAAU,UAAA,EAAY;AAC/B,MAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,aAAA,CAAc,MAAA,EAAQ,MAAM,CAAA;AAC/C,MAAA,IAAI,QAAQ,SAAA,EAAW;AACrB,QAAA,SAAA,GAAY,KAAA;AACZ,QAAA,UAAA,GAAa,MAAA;AAAA,MACf;AAAA,IACF;AAEA,IAAA,OAAO,MAAA,CAAO,QAAA,CAAS,UAAU,CAAA,GAAI,UAAA,GAAa,CAAA;AAAA,EACpD;AAAA,EAEA,aAAa,CAAA,EAAmB;AAC9B,IAAA,OAAO,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,CAAE,UAAU,CAAC,CAAA;AAAA,EAChC;AACF;;;ACz0DA,SAAS,gBAAgB,OAAA,EAAuB;AAC9C,EAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,OAAO,CAAA,IAAK,CAAC,MAAA,CAAO,SAAA,CAAU,OAAO,CAAA,IAAK,OAAA,GAAU,CAAA,EAAG;AAC1E,IAAA,MAAM,IAAI,qBAAA,CAAsB,uCAAA,EAAyC,SAAA,EAAW,OAAO,CAAA;AAAA,EAC7F;AACF;AAOA,SAAS,cAAA,CAAe,OAA2B,IAAA,EAAqC;AACtF,EAAA,IAAI,UAAU,MAAA,EAAW;AACvB,IAAA,OAAO,MAAA;AAAA,EACT;AACA,EAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,KAAK,CAAA,IAAK,SAAS,CAAA,EAAG;AACzC,IAAA,MAAM,IAAI,qBAAA,CAAsB,CAAA,EAAG,IAAI,CAAA,0BAAA,CAAA,EAA8B,MAAM,KAAK,CAAA;AAAA,EAClF;AACA,EAAA,IAAI,QAAQ,CAAA,EAAG;AACb,IAAA,OAAO,EAAE,IAAA,EAAM,UAAA,EAAY,KAAA,EAAM;AAAA,EACnC;AACA,EAAA,IAAI,CAAC,MAAA,CAAO,SAAA,CAAU,KAAK,CAAA,EAAG;AAC5B,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,GAAG,IAAI,CAAA,qDAAA,CAAA;AAAA,MACP,IAAA;AAAA,MACA;AAAA,KACF;AAAA,EACF;AACA,EAAA,OAAO,EAAE,IAAA,EAAM,OAAA,EAAS,KAAA,EAAM;AAChC;AAEA,SAAS,iBAAA,CAAkB,IAAA,EAAiB,QAAA,EAAkB,OAAA,EAA0B;AACtF,EAAA,IAAI,IAAA,CAAK,SAAS,OAAA,EAAS;AACzB,IAAA,OAAO,IAAA,CAAK,KAAA;AAAA,EACd;AACA,EAAA,MAAM,KAAA,GAAQ,WAAW,IAAA,CAAK,KAAA;AAC9B,EAAA,OAAO,UAAU,IAAA,CAAK,KAAA,CAAM,KAAK,CAAA,GAAI,IAAA,CAAK,KAAK,KAAK,CAAA;AACtD;AAEA,SAAS,sBAAA,CACP,QAAA,EACA,SAAA,EACA,QAAA,EACkB;AAClB,EAAA,MAAM,eAAA,GAAkB,SAAA,KAAc,MAAA,IAAa,QAAA,KAAa,SAAY,IAAA,GAAO,QAAA;AACnF,EAAA,MAAM,SAAA,GAAY,cAAA,CAAe,SAAA,EAAW,WAAW,CAAA;AACvD,EAAA,MAAM,QAAA,GAAW,cAAA,CAAe,eAAA,EAAiB,UAAU,CAAA;AAE3D,EAAA,IAAI,SAAA,EAAW,IAAA,KAAS,OAAA,IAAW,SAAA,CAAU,QAAQ,QAAA,EAAU;AAC7D,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,6CAAA;AAAA,MACA,WAAA;AAAA,MACA,SAAA,CAAU;AAAA,KACZ;AAAA,EACF;AACA,EAAA,IAAI,QAAA,EAAU,IAAA,KAAS,OAAA,IAAW,QAAA,CAAS,QAAQ,QAAA,EAAU;AAC3D,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,4CAAA;AAAA,MACA,UAAA;AAAA,MACA,QAAA,CAAS;AAAA,KACX;AAAA,EACF;AAEA,EAAA,IACE,SAAA,EAAW,IAAA,KAAS,UAAA,IACpB,QAAA,EAAU,IAAA,KAAS,cACnB,SAAA,CAAU,KAAA,GAAQ,QAAA,CAAS,KAAA,GAAQ,CAAA,EACnC;AACA,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,wDAAA;AAAA,MACA,WAAA;AAAA,MACA,SAAA,CAAU;AAAA,KACZ;AAAA,EACF;AAEA,EAAA,IAAI,SAAS,SAAA,KAAc,MAAA,GAAY,SAAY,iBAAA,CAAkB,SAAA,EAAW,UAAU,IAAI,CAAA;AAC9F,EAAA,IAAI,QAAQ,QAAA,KAAa,MAAA,GAAY,SAAY,iBAAA,CAAkB,QAAA,EAAU,UAAU,KAAK,CAAA;AAE5F,EAAA,IAAI,MAAA,KAAW,MAAA,IAAa,KAAA,KAAU,MAAA,EAAW;AAC/C,IAAA,MAAM,IAAI,aAAa,+CAA+C,CAAA;AAAA,EACxE;AAEA,EAAA,IAAI,WAAW,MAAA,EAAW;AACxB,IAAA,MAAA,GAAS,YAAY,KAAA,IAAS,CAAA,CAAA;AAAA,EAChC;AACA,EAAA,IAAI,UAAU,MAAA,EAAW;AACvB,IAAA,KAAA,GAAQ,QAAA,GAAW,MAAA;AAAA,EACrB;AAEA,EAAA,IAAI,MAAA,GAAS,QAAQ,QAAA,EAAU;AAC7B,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,iDAAA;AAAA,MACA,WAAA;AAAA,MACA;AAAA,KACF;AAAA,EACF;AAEA,EAAA,IAAI,SAAS,CAAA,EAAG;AACd,IAAA,MAAM,IAAI,qBAAA,CAAsB,qCAAA,EAAuC,WAAA,EAAa,SAAS,CAAA;AAAA,EAC/F;AACA,EAAA,IAAI,QAAQ,CAAA,EAAG;AACb,IAAA,MAAM,IAAI,qBAAA,CAAsB,oCAAA,EAAsC,UAAA,EAAY,QAAQ,CAAA;AAAA,EAC5F;AAEA,EAAA,OAAO,CAAC,QAAQ,KAAK,CAAA;AACvB;AAEA,SAAS,aAAA,CAAc,GAAY,CAAA,EAAoB;AACrD,EAAA,IAAI,OAAO,CAAA,KAAM,QAAA,IAAY,OAAO,CAAA,KAAM,QAAA,SAAiB,CAAA,GAAI,CAAA;AAC/D,EAAA,IAAI,OAAO,CAAA,KAAM,QAAA,IAAY,OAAO,MAAM,QAAA,EAAU;AAClD,IAAA,IAAI,CAAA,GAAI,GAAG,OAAO,EAAA;AAClB,IAAA,IAAI,CAAA,GAAI,GAAG,OAAO,CAAA;AAClB,IAAA,OAAO,CAAA;AAAA,EACT;AACA,EAAA,OAAO,OAAO,CAAC,CAAA,CAAE,aAAA,CAAc,MAAA,CAAO,CAAC,CAAC,CAAA;AAC1C;AAEA,SAAS,aAAA,CAAc,OAAe,OAAA,EAA2B;AAC/D,EAAA,MAAM,IAAA,GAAO,IAAA,CAAK,KAAA,CAAM,KAAA,GAAQ,OAAO,CAAA;AACvC,EAAA,MAAM,YAAY,KAAA,GAAQ,OAAA;AAC1B,EAAA,OAAO,KAAA,CAAM,IAAA,CAAK,EAAE,MAAA,EAAQ,OAAA,EAAQ,EAAG,CAAC,CAAA,EAAG,CAAA,KAAM,IAAA,IAAQ,CAAA,GAAI,SAAA,GAAY,IAAI,CAAA,CAAE,CAAA;AACjF;AAEA,SAAS,eAAA,CAAgB,GAAW,OAAA,EAA6C;AAC/E,EAAA,MAAM,KAAA,GAAQ,CAAA,CAAE,EAAA,CAAG,GAAG,OAAO,CAAA;AAC7B,EAAA,IAAI,OAAO,UAAU,QAAA,IAAY,OAAO,UAAU,QAAA,IAAY,OAAO,UAAU,QAAA,EAAU;AACvF,IAAA,OAAO,KAAA;AAAA,EACT;AACA,EAAA,MAAM,IAAI,aAAa,+CAA+C,CAAA;AACxE;AAEA,SAAS,gBAAA,CAAiB,CAAA,EAAW,SAAA,EAAmB,KAAA,EAAuC;AAC7F,EAAA,IAAI,CAAA,CAAE,UAAU,QAAA,EAAU;AACxB,IAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,MAAA,MAAM,IAAI,aAAa,yDAAyD,CAAA;AAAA,IAClF;AACA,IAAA,CAAA,CAAE,IAAA,CAAK,SAAS,CAAA,GAAI,KAAA;AACpB,IAAA;AAAA,EACF;AAEA,EAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,IAAA,MAAM,IAAI,aAAa,4DAA4D,CAAA;AAAA,EACrF;AAEA,EAAA,IAAI,CAAA,CAAE,gBAAgB,aAAA,EAAe;AACnC,IAAA,CAAA,CAAE,IAAA,CAAK,SAAS,CAAA,GAAI,OAAO,UAAU,QAAA,GAAW,KAAA,GAAQ,OAAO,KAAK,CAAA;AACpE,IAAA;AAAA,EACF;AAEA,EAAA,CAAA,CAAE,IAAA,CAAK,SAAS,CAAA,GAAI,MAAA,CAAO,KAAK,CAAA;AAClC;AAEA,SAAS,UAAA,CAAW,GAAW,aAAA,EAAiC;AAC9D,EAAA,MAAM,GAAG,SAAS,CAAA,GAAI,WAAW,CAAC,CAAA;AAClC,EAAA,MAAM,GAAA,GAAM,KAAA,CAAM,CAAC,aAAA,CAAc,MAAA,EAAQ,SAAS,CAAA,EAAG,EAAE,KAAA,EAAO,CAAA,CAAE,KAAA,EAAO,CAAA;AAEvE,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,aAAA,CAAc,QAAQ,CAAA,EAAA,EAAK;AAC7C,IAAA,MAAM,WAAA,GAAc,cAAc,CAAC,CAAA;AACnC,IAAA,IAAI,gBAAgB,MAAA,EAAW;AAC7B,MAAA,MAAM,IAAI,aAAa,4CAA4C,CAAA;AAAA,IACrE;AACA,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,SAAA,EAAW,CAAA,EAAA,EAAK;AAClC,MAAA,MAAM,QAAQ,eAAA,CAAgB,CAAA,EAAG,CAAC,WAAA,EAAa,CAAC,CAAC,CAAA;AACjD,MAAA,gBAAA,CAAiB,KAAK,GAAA,CAAI,MAAA,GAAS,CAAA,GAAI,SAAA,GAAY,GAAG,KAAK,CAAA;AAAA,IAC7D;AAAA,EACF;AAEA,EAAA,OAAO,GAAA;AACT;AAEA,SAAS,UAAA,CAAW,GAAW,aAAA,EAAiC;AAC9D,EAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,IAAA,MAAM,IAAI,UAAA,CAAW,CAAA,2BAAA,EAA8B,CAAA,CAAE,IAAI,CAAA,CAAA,CAAG,CAAA;AAAA,EAC9D;AACA,EAAA,MAAM,GAAA,GAAM,KAAA,CAAM,CAAC,aAAA,CAAc,MAAM,GAAG,EAAE,KAAA,EAAO,CAAA,CAAE,KAAA,EAAO,CAAA;AAE5D,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,aAAA,CAAc,QAAQ,CAAA,EAAA,EAAK;AAC7C,IAAA,MAAM,WAAA,GAAc,cAAc,CAAC,CAAA;AACnC,IAAA,IAAI,gBAAgB,MAAA,EAAW;AAC7B,MAAA,MAAM,IAAI,aAAa,4CAA4C,CAAA;AAAA,IACrE;AACA,IAAA,MAAM,KAAA,GAAQ,eAAA,CAAgB,CAAA,EAAG,CAAC,WAAW,CAAC,CAAA;AAC9C,IAAA,gBAAA,CAAiB,GAAA,EAAK,GAAA,CAAI,MAAA,GAAS,CAAA,EAAG,KAAK,CAAA;AAAA,EAC7C;AAEA,EAAA,OAAO,GAAA;AACT;AA0BO,SAAS,cAAA,CACd,CAAA,EACA,CAAA,EACA,OAAA,EAOU;AACV,EAAA,MAAM,IAAA,GAAO,WAAW,EAAC;AACzB,EAAA,MAAM,OAAA,GAAU,KAAK,OAAA,IAAW,IAAA;AAChC,EAAA,MAAM,cAAc,IAAA,CAAK,WAAA;AAEzB,EAAA,MAAM,CAAC,QAAQ,CAAA,GAAI,UAAA,CAAW,CAAC,CAAA;AAE/B,EAAA,IAAI,aAAa,CAAA,EAAG;AAClB,IAAA,MAAM,IAAI,qBAAA,CAAsB,0BAAA,EAA4B,GAAG,CAAA;AAAA,EACjE;AAEA,EAAA,IAAI,CAAA,EAAG;AACL,IAAA,MAAM,OAAA,GAAU,CAAA,CAAE,KAAA,CAAM,CAAC,CAAA;AACzB,IAAA,IAAI,OAAA,KAAY,MAAA,IAAa,OAAA,KAAY,QAAA,EAAU;AACjD,MAAA,MAAM,IAAI,qBAAA,CAAsB,0CAAA,EAA4C,GAAA,EAAK,OAAO,CAAA;AAAA,IAC1F;AAAA,EACF;AAEA,EAAA,IAAI,KAAK,QAAA,EAAU;AACjB,IAAA,IAAI,IAAA,CAAK,QAAA,CAAS,IAAA,KAAS,CAAA,EAAG;AAC5B,MAAA,MAAM,IAAI,UAAA,CAAW,CAAA,kCAAA,EAAqC,IAAA,CAAK,QAAA,CAAS,IAAI,CAAA,CAAA,CAAG,CAAA;AAAA,IACjF;AACA,IAAA,MAAM,cAAA,GAAiB,IAAA,CAAK,QAAA,CAAS,KAAA,CAAM,CAAC,CAAA;AAC5C,IAAA,IAAI,cAAA,KAAmB,MAAA,IAAa,cAAA,KAAmB,QAAA,EAAU;AAC/D,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,gDAAA;AAAA,QACA,UAAA;AAAA,QACA;AAAA,OACF;AAAA,IACF;AAAA,EACF;AAEA,EAAA,MAAM,CAAC,QAAQ,KAAK,CAAA,GAAI,uBAAuB,QAAA,EAAU,IAAA,CAAK,SAAA,EAAW,IAAA,CAAK,QAAQ,CAAA;AAEtF,EAAA,MAAM,OAAA,GAAU,KAAA,CAAM,IAAA,CAAK,EAAE,MAAA,EAAQ,UAAS,EAAG,CAAC,CAAA,EAAG,CAAA,KAAM,CAAC,CAAA;AAC5D,EAAA,MAAM,SAAS,WAAA,KAAgB,MAAA,GAAY,kBAAA,CAAmB,WAAW,IAAI,IAAA,CAAK,MAAA;AAClF,EAAA,MAAM,YAAA,GAAe,CAAC,GAAA,KAAwB;AAC5C,IAAA,IAAI,CAAC,OAAA,EAAS;AACd,IAAA,qBAAA,CAAsB,KAAK,MAAM,CAAA;AAAA,EACnC,CAAA;AAEA,EAAA,IAAI,eAAyB,EAAC;AAC9B,EAAA,IAAI,cAAwB,EAAC;AAE7B,EAAA,IAAI,KAAK,QAAA,EAAU;AACjB,IAAA,MAAM,WAAW,IAAA,CAAK,QAAA;AACtB,IAAA,MAAM,QAAA,uBAAe,GAAA,EAAuB;AAC5C,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAM,KAAA,GAAQ,eAAA,CAAgB,QAAA,EAAU,CAAC,CAAC,CAAC,CAAA;AAC3C,MAAA,IAAI,MAAA,GAAS,QAAA,CAAS,GAAA,CAAI,KAAK,CAAA;AAC/B,MAAA,IAAI,WAAW,MAAA,EAAW;AACxB,QAAA,MAAA,GAAS,EAAC;AACV,QAAA,QAAA,CAAS,GAAA,CAAI,OAAO,MAAM,CAAA;AAAA,MAC5B;AACA,MAAA,MAAA,CAAO,KAAK,CAAC,CAAA;AAAA,IACf;AAEA,IAAA,MAAM,MAAA,GAAS,MAAM,IAAA,CAAK,QAAA,CAAS,MAAM,CAAA,CAAE,KAAK,aAAa,CAAA;AAC7D,IAAA,MAAM,WAAW,MAAA,CAAO,MAAA;AACxB,IAAA,MAAM,UAAA,GAAa,MAAA,CAAO,GAAA,CAAI,CAAC,KAAA,KAAU,SAAS,GAAA,CAAI,KAAK,CAAA,EAAG,MAAA,IAAU,CAAC,CAAA;AACzE,IAAA,MAAM,eAAe,UAAA,CAAW,IAAA,CAAK,CAAC,IAAA,KAAS,OAAO,CAAC,CAAA;AAEvD,IAAA,IAAI,YAAA,IAAgB,OAAA,IAAW,WAAA,KAAgB,MAAA,EAAW;AACxD,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,gDAAA;AAAA,QACA,UAAA;AAAA,QACA;AAAA,OACF;AAAA,IACF;AAEA,IAAA,IAAI,IAAA,CAAK,SAAA,KAAc,MAAA,IAAa,MAAA,GAAS,QAAA,EAAU;AACrD,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,mEAAA;AAAA,QACA,WAAA;AAAA,QACA;AAAA,OACF;AAAA,IACF;AACA,IAAA,IAAI,QAAQ,QAAA,EAAU;AACpB,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,kEAAA;AAAA,QACA,UAAA;AAAA,QACA;AAAA,OACF;AAAA,IACF;AAEA,IAAA,MAAM,eAAe,KAAA,GAAQ,QAAA;AAC7B,IAAA,MAAM,wBAAwB,MAAA,GAAS,QAAA;AACvC,IAAA,MAAM,MAAA,GAAS,MAAA,CAAO,GAAA,CAAI,CAAC,KAAA,KAAU;AACnC,MAAA,MAAM,IAAA,GAAO,QAAA,CAAS,GAAA,CAAI,KAAK,GAAG,MAAA,IAAU,CAAA;AAC5C,MAAA,MAAM,QAAQ,IAAA,GAAO,YAAA;AACrB,MAAA,IAAI,SAAA,GAAY,IAAA,CAAK,KAAA,CAAM,KAAK,CAAA;AAChC,MAAA,IAAI,YAAY,KAAA,GAAQ,SAAA;AACxB,MAAA,IAAI,GAAA,GAAM,wBAAwB,CAAA,GAAI,CAAA;AACtC,MAAA,IAAI,GAAA,GAAM,qBAAA,GAAwB,IAAA,GAAO,IAAA,GAAO,CAAA;AAChD,MAAA,IAAI,OAAO,CAAA,EAAG;AACZ,QAAA,GAAA,GAAM,CAAA;AACN,QAAA,GAAA,GAAM,wBAAwB,IAAA,GAAO,CAAA;AACrC,QAAA,SAAA,GAAY,CAAA;AACZ,QAAA,SAAA,GAAY,CAAA;AAAA,MACd,CAAA,MAAO;AACL,QAAA,IAAI,SAAA,GAAY,KAAK,SAAA,GAAY,GAAA;AACjC,QAAA,IAAI,SAAA,GAAY,KAAK,SAAA,GAAY,GAAA;AAAA,MACnC;AACA,MAAA,OAAO,EAAE,KAAA,EAAO,IAAA,EAAM,SAAA,EAAW,SAAA,EAAW,KAAK,GAAA,EAAI;AAAA,IACvD,CAAC,CAAA;AAED,IAAA,IAAI,SAAA,GAAY,KAAA,GAAQ,MAAA,CAAO,MAAA,CAAO,CAAC,KAAK,CAAA,KAAM,GAAA,GAAM,CAAA,CAAE,SAAA,EAAW,CAAC,CAAA;AACtE,IAAA,IAAI,cAAc,CAAA,EAAG;AACnB,MAAA,MAAM,KAAA,GACJ,SAAA,GAAY,CAAA,GACR,CAAC,GAAG,MAAM,CAAA,CAAE,IAAA,CAAK,CAAC,CAAA,EAAG,CAAA,KAAM;AACzB,QAAA,IAAI,EAAE,SAAA,KAAc,CAAA,CAAE,WAAW,OAAO,CAAA,CAAE,YAAY,CAAA,CAAE,SAAA;AACxD,QAAA,OAAO,aAAA,CAAc,CAAA,CAAE,KAAA,EAAO,CAAA,CAAE,KAAK,CAAA;AAAA,MACvC,CAAC,IACD,CAAC,GAAG,MAAM,CAAA,CAAE,IAAA,CAAK,CAAC,CAAA,EAAG,CAAA,KAAM;AACzB,QAAA,IAAI,EAAE,SAAA,KAAc,CAAA,CAAE,WAAW,OAAO,CAAA,CAAE,YAAY,CAAA,CAAE,SAAA;AACxD,QAAA,OAAO,aAAA,CAAc,CAAA,CAAE,KAAA,EAAO,CAAA,CAAE,KAAK,CAAA;AAAA,MACvC,CAAC,CAAA;AAEP,MAAA,IAAI,KAAA,GAAQ,CAAA;AACZ,MAAA,OAAO,SAAA,KAAc,CAAA,IAAK,KAAA,GAAQ,MAAA,CAAO,SAAS,CAAA,EAAG;AACnD,QAAA,KAAA,MAAW,SAAS,KAAA,EAAO;AACzB,UAAA,IAAI,cAAc,CAAA,EAAG;AACrB,UAAA,IAAI,SAAA,GAAY,CAAA,IAAK,KAAA,CAAM,SAAA,GAAY,MAAM,GAAA,EAAK;AAChD,YAAA,KAAA,CAAM,SAAA,IAAa,CAAA;AACnB,YAAA,SAAA,IAAa,CAAA;AAAA,UACf,WAAW,SAAA,GAAY,CAAA,IAAK,KAAA,CAAM,SAAA,GAAY,MAAM,GAAA,EAAK;AACvD,YAAA,KAAA,CAAM,SAAA,IAAa,CAAA;AACnB,YAAA,SAAA,IAAa,CAAA;AAAA,UACf;AAAA,QACF;AACA,QAAA,KAAA,IAAS,CAAA;AAAA,MACX;AAEA,MAAA,IAAI,cAAc,CAAA,EAAG;AACnB,QAAA,MAAM,IAAI,aAAa,2DAA2D,CAAA;AAAA,MACpF;AAAA,IACF;AAEA,IAAA,MAAM,qBAA+B,EAAC;AACtC,IAAA,KAAA,MAAW,SAAS,MAAA,EAAQ;AAC1B,MAAA,MAAM,YAAA,GAAe,CAAC,GAAI,QAAA,CAAS,IAAI,KAAA,CAAM,KAAK,CAAA,IAAK,EAAG,CAAA;AAC1D,MAAA,YAAA,CAAa,YAAY,CAAA;AACzB,MAAA,WAAA,CAAY,KAAK,GAAG,YAAA,CAAa,MAAM,CAAA,EAAG,KAAA,CAAM,SAAS,CAAC,CAAA;AAC1D,MAAA,kBAAA,CAAmB,KAAK,GAAG,YAAA,CAAa,KAAA,CAAM,KAAA,CAAM,SAAS,CAAC,CAAA;AAAA,IAChE;AAEA,IAAA,YAAA,CAAa,WAAW,CAAA;AACxB,IAAA,YAAA,CAAa,kBAAkB,CAAA;AAC/B,IAAA,YAAA,GAAe,kBAAA,CAAmB,KAAA,CAAM,CAAA,EAAG,MAAM,CAAA;AAAA,EACnD,CAAA,MAAO;AACL,IAAA,YAAA,CAAa,OAAO,CAAA;AACpB,IAAA,YAAA,GAAe,OAAA,CAAQ,KAAA,CAAM,CAAA,EAAG,MAAM,CAAA;AACtC,IAAA,WAAA,GAAc,OAAA,CAAQ,KAAA,CAAM,MAAA,EAAQ,MAAA,GAAS,KAAK,CAAA;AAAA,EACpD;AAEA,EAAA,IAAI,YAAA,CAAa,MAAA,KAAW,MAAA,IAAU,WAAA,CAAY,WAAW,KAAA,EAAO;AAClE,IAAA,MAAM,IAAI,aAAa,qEAAqE,CAAA;AAAA,EAC9F;AAEA,EAAA,MAAM,MAAA,GAAS,UAAA,CAAW,CAAA,EAAG,YAAY,CAAA;AACzC,EAAA,MAAM,KAAA,GAAQ,UAAA,CAAW,CAAA,EAAG,WAAW,CAAA;AAEvC,EAAA,IAAI,CAAA,EAAG;AACL,IAAA,MAAM,MAAA,GAAS,UAAA,CAAW,CAAA,EAAG,YAAY,CAAA;AACzC,IAAA,MAAM,KAAA,GAAQ,UAAA,CAAW,CAAA,EAAG,WAAW,CAAA;AACvC,IAAA,OAAO,CAAC,MAAA,EAAQ,KAAA,EAAO,MAAA,EAAQ,KAAK,CAAA;AAAA,EACtC;AAEA,EAAA,OAAO,CAAC,QAAQ,KAAK,CAAA;AACvB;AASO,IAAM,QAAN,MAAY;AAAA,EACT,OAAA;AAAA,EACA,OAAA;AAAA,EACA,WAAA;AAAA,EAER,WAAA,CACE,OAAA,GAII,EAAC,EACL;AACA,IAAA,IAAA,CAAK,OAAA,GAAU,QAAQ,OAAA,IAAW,CAAA;AAClC,IAAA,IAAA,CAAK,OAAA,GAAU,QAAQ,OAAA,IAAW,KAAA;AAClC,IAAA,IAAA,CAAK,cAAc,OAAA,CAAQ,WAAA;AAAA,EAC7B;AAAA,EAEA,MAAM,CAAA,EAAwC;AAC5C,IAAA,MAAM,MAAA,GAAS,CAAA,CAAE,KAAA,CAAM,CAAC,CAAA;AACxB,IAAA,IAAI,WAAW,MAAA,EAAW;AACxB,MAAA,MAAM,IAAI,WAAW,4BAA4B,CAAA;AAAA,IACnD;AACA,IAAA,MAAM,QAAA,GAAW,MAAA;AACjB,IAAA,eAAA,CAAgB,KAAK,OAAO,CAAA;AAC5B,IAAA,IAAI,IAAA,CAAK,UAAU,QAAA,EAAU;AAC3B,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,oDAAA;AAAA,QACA,SAAA;AAAA,QACA,IAAA,CAAK;AAAA,OACP;AAAA,IACF;AACA,IAAA,MAAM,OAAA,GAAU,KAAA,CAAM,IAAA,CAAK,EAAE,MAAA,EAAQ,UAAS,EAAG,CAAC,CAAA,EAAG,CAAA,KAAM,CAAC,CAAA;AAE5D,IAAA,IAAI,KAAK,OAAA,EAAS;AAChB,MAAA,MAAM,MAAA,GACJ,KAAK,WAAA,KAAgB,MAAA,GAAY,mBAAmB,IAAA,CAAK,WAAW,IAAI,IAAA,CAAK,MAAA;AAC/E,MAAA,qBAAA,CAAsB,SAAS,MAAM,CAAA;AAAA,IACvC;AAEA,IAAA,MAAM,SAAsC,EAAC;AAC7C,IAAA,MAAM,SAAA,GAAY,aAAA,CAAc,QAAA,EAAU,IAAA,CAAK,OAAO,CAAA;AACtD,IAAA,IAAI,OAAA,GAAU,CAAA;AAEd,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,CAAK,SAAS,CAAA,EAAA,EAAK;AACrC,MAAA,MAAM,QAAA,GAAW,SAAA,CAAU,CAAC,CAAA,IAAK,CAAA;AACjC,MAAA,MAAM,SAAA,GAAY,OAAA;AAClB,MAAA,MAAM,UAAU,OAAA,GAAU,QAAA;AAE1B,MAAA,MAAM,WAAA,GAAc,OAAA,CAAQ,KAAA,CAAM,SAAA,EAAW,OAAO,CAAA;AACpD,MAAA,MAAM,YAAA,GAAe,CAAC,GAAG,OAAA,CAAQ,KAAA,CAAM,CAAA,EAAG,SAAS,CAAA,EAAG,GAAG,OAAA,CAAQ,KAAA,CAAM,OAAO,CAAC,CAAA;AAE/E,MAAA,MAAA,CAAO,IAAA,CAAK,CAAC,YAAA,EAAc,WAAW,CAAC,CAAA;AACvC,MAAA,OAAA,GAAU,OAAA;AAAA,IACZ;AAEA,IAAA,OAAO,MAAA;AAAA,EACT;AAAA,EAEA,UAAA,GAAqB;AACnB,IAAA,OAAO,IAAA,CAAK,OAAA;AAAA,EACd;AACF;AASO,IAAM,kBAAN,MAAsB;AAAA,EACnB,OAAA;AAAA,EACA,OAAA;AAAA,EACA,WAAA;AAAA,EAER,WAAA,CACE,OAAA,GAII,EAAC,EACL;AACA,IAAA,IAAA,CAAK,OAAA,GAAU,QAAQ,OAAA,IAAW,CAAA;AAClC,IAAA,IAAA,CAAK,OAAA,GAAU,QAAQ,OAAA,IAAW,KAAA;AAClC,IAAA,IAAA,CAAK,cAAc,OAAA,CAAQ,WAAA;AAAA,EAC7B;AAAA,EAEA,KAAA,CAAM,GAAW,CAAA,EAAwC;AACvD,IAAA,MAAM,MAAA,GAAS,CAAA,CAAE,KAAA,CAAM,CAAC,CAAA;AACxB,IAAA,IAAI,WAAW,MAAA,EAAW;AACxB,MAAA,MAAM,IAAI,WAAW,4BAA4B,CAAA;AAAA,IACnD;AACA,IAAA,MAAM,QAAA,GAAW,MAAA;AACjB,IAAA,eAAA,CAAgB,KAAK,OAAO,CAAA;AAC5B,IAAA,IAAI,IAAA,CAAK,UAAU,QAAA,EAAU;AAC3B,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,oDAAA;AAAA,QACA,SAAA;AAAA,QACA,IAAA,CAAK;AAAA,OACP;AAAA,IACF;AACA,IAAA,MAAM,OAAA,GAAU,CAAA,CAAE,KAAA,CAAM,CAAC,CAAA;AACzB,IAAA,IAAI,OAAA,KAAY,MAAA,IAAa,OAAA,KAAY,QAAA,EAAU;AACjD,MAAA,MAAM,IAAI,qBAAA,CAAsB,0CAAA,EAA4C,GAAA,EAAK,OAAO,CAAA;AAAA,IAC1F;AACA,IAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,MAAA,MAAM,IAAI,UAAA,CAAW,CAAA,2BAAA,EAA8B,CAAA,CAAE,IAAI,CAAA,CAAA,CAAG,CAAA;AAAA,IAC9D;AACA,IAAA,MAAM,QAAA,uBAAe,GAAA,EAAwC;AAC7D,IAAA,MAAM,MAAA,GACJ,KAAK,WAAA,KAAgB,MAAA,GAAY,mBAAmB,IAAA,CAAK,WAAW,IAAI,IAAA,CAAK,MAAA;AAE/E,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAM,KAAA,GAAQ,eAAA,CAAgB,CAAA,EAAG,CAAC,CAAC,CAAC,CAAA;AACpC,MAAA,IAAI,MAAA,GAAS,QAAA,CAAS,GAAA,CAAI,KAAK,CAAA;AAC/B,MAAA,IAAI,WAAW,MAAA,EAAW;AACxB,QAAA,MAAA,GAAS,EAAC;AACV,QAAA,QAAA,CAAS,GAAA,CAAI,OAAO,MAAM,CAAA;AAAA,MAC5B;AACA,MAAA,MAAA,CAAO,KAAK,CAAC,CAAA;AAAA,IACf;AAEA,IAAA,KAAA,MAAW,CAAC,KAAA,EAAO,OAAO,CAAA,IAAK,QAAA,CAAS,SAAQ,EAAG;AACjD,MAAA,IAAI,KAAK,OAAA,EAAS;AAChB,QAAA,qBAAA,CAAsB,SAAS,MAAM,CAAA;AAAA,MACvC;AACA,MAAA,IAAI,OAAA,CAAQ,MAAA,GAAS,IAAA,CAAK,OAAA,EAAS;AACjC,QAAA,MAAM,IAAI,qBAAA;AAAA,UACR,CAAA,qDAAA,EAAwD,KAAK,CAAA,KAAA,EAAQ,OAAA,CAAQ,MAAM,CAAA,CAAA;AAAA,UACnF,SAAA;AAAA,UACA,IAAA,CAAK;AAAA,SACP;AAAA,MACF;AAAA,IACF;AAEA,IAAA,MAAM,WAAA,GAA0B,KAAA,CAAM,IAAA,CAAK,EAAE,MAAA,EAAQ,KAAK,OAAA,EAAQ,EAAG,MAAM,EAAE,CAAA;AAE7E,IAAA,KAAA,MAAW,OAAA,IAAW,QAAA,CAAS,MAAA,EAAO,EAAG;AACvC,MAAA,MAAM,SAAA,GAAY,aAAA,CAAc,OAAA,CAAQ,MAAA,EAAQ,KAAK,OAAO,CAAA;AAC5D,MAAA,IAAI,KAAA,GAAQ,CAAA;AACZ,MAAA,KAAA,IAAS,IAAA,GAAO,CAAA,EAAG,IAAA,GAAO,IAAA,CAAK,SAAS,IAAA,EAAA,EAAQ;AAC9C,QAAA,MAAM,IAAA,GAAO,SAAA,CAAU,IAAI,CAAA,IAAK,CAAA;AAChC,QAAA,MAAM,MAAM,KAAA,GAAQ,IAAA;AACpB,QAAA,MAAM,MAAA,GAAS,YAAY,IAAI,CAAA;AAC/B,QAAA,IAAI,CAAC,MAAA,EAAQ;AACX,UAAA,MAAM,IAAI,aAAa,iDAAiD,CAAA;AAAA,QAC1E;AACA,QAAA,MAAA,CAAO,KAAK,GAAG,OAAA,CAAQ,KAAA,CAAM,KAAA,EAAO,GAAG,CAAC,CAAA;AACxC,QAAA,KAAA,GAAQ,GAAA;AAAA,MACV;AAAA,IACF;AAEA,IAAA,MAAM,SAAsC,EAAC;AAE7C,IAAA,KAAA,IAAS,IAAA,GAAO,CAAA,EAAG,IAAA,GAAO,IAAA,CAAK,SAAS,IAAA,EAAA,EAAQ;AAC9C,MAAA,MAAM,WAAA,GAAc,WAAA,CAAY,IAAI,CAAA,IAAK,EAAC;AAC1C,MAAA,MAAM,eAAyB,EAAC;AAChC,MAAA,KAAA,IAAS,KAAA,GAAQ,CAAA,EAAG,KAAA,GAAQ,IAAA,CAAK,SAAS,KAAA,EAAA,EAAS;AACjD,QAAA,IAAI,UAAU,IAAA,EAAM;AACpB,QAAA,YAAA,CAAa,KAAK,GAAI,WAAA,CAAY,KAAK,CAAA,IAAK,EAAG,CAAA;AAAA,MACjD;AACA,MAAA,MAAA,CAAO,IAAA,CAAK,CAAC,YAAA,EAAc,WAAW,CAAC,CAAA;AAAA,IACzC;AAEA,IAAA,OAAO,MAAA;AAAA,EACT;AAAA,EAEA,UAAA,GAAqB;AACnB,IAAA,OAAO,IAAA,CAAK,OAAA;AAAA,EACd;AACF;AASO,IAAM,aAAN,MAAiB;AAAA,EACd,OAAA;AAAA,EAER,WAAA,CAAY,OAAA,GAAgC,EAAC,EAAG;AAC9C,IAAA,IAAA,CAAK,OAAA,GAAU,QAAQ,OAAA,IAAW,CAAA;AAAA,EACpC;AAAA,EAEA,KAAA,CAAM,CAAA,EAAW,EAAA,EAAwB,MAAA,EAA6C;AACpF,IAAA,MAAM,MAAA,GAAS,CAAA,CAAE,KAAA,CAAM,CAAC,CAAA;AACxB,IAAA,IAAI,WAAW,MAAA,EAAW;AACxB,MAAA,MAAM,IAAI,WAAW,4BAA4B,CAAA;AAAA,IACnD;AACA,IAAA,MAAM,QAAA,GAAW,MAAA;AACjB,IAAA,eAAA,CAAgB,KAAK,OAAO,CAAA;AAC5B,IAAA,IAAI,MAAA,CAAO,SAAS,CAAA,EAAG;AACrB,MAAA,MAAM,IAAI,UAAA,CAAW,CAAA,gCAAA,EAAmC,MAAA,CAAO,IAAI,CAAA,CAAA,CAAG,CAAA;AAAA,IACxE;AACA,IAAA,MAAM,YAAA,GAAe,MAAA,CAAO,KAAA,CAAM,CAAC,CAAA;AACnC,IAAA,IAAI,YAAA,KAAiB,MAAA,IAAa,YAAA,KAAiB,QAAA,EAAU;AAC3D,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,+CAAA;AAAA,QACA,QAAA;AAAA,QACA;AAAA,OACF;AAAA,IACF;AACA,IAAA,MAAM,QAAA,uBAAe,GAAA,EAAwC;AAE7D,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAM,KAAA,GAAQ,eAAA,CAAgB,MAAA,EAAQ,CAAC,CAAC,CAAC,CAAA;AACzC,MAAA,IAAI,MAAA,GAAS,QAAA,CAAS,GAAA,CAAI,KAAK,CAAA;AAC/B,MAAA,IAAI,WAAW,MAAA,EAAW;AACxB,QAAA,MAAA,GAAS,EAAC;AACV,QAAA,QAAA,CAAS,GAAA,CAAI,OAAO,MAAM,CAAA;AAAA,MAC5B;AACA,MAAA,MAAA,CAAO,KAAK,CAAC,CAAA;AAAA,IACf;AAEA,IAAA,MAAM,YAAA,GAAe,KAAA,CAAM,IAAA,CAAK,QAAA,CAAS,OAAA,EAAS,CAAA,CAAE,GAAA,CAAI,CAAC,CAAC,KAAA,EAAO,OAAO,CAAA,MAAO;AAAA,MAC7E,KAAA;AAAA,MACA,OAAA;AAAA,MACA,MAAM,OAAA,CAAQ;AAAA,KAChB,CAAE,CAAA;AACF,IAAA,IAAI,IAAA,CAAK,OAAA,GAAU,YAAA,CAAa,MAAA,EAAQ;AACtC,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,2CAAA;AAAA,QACA,SAAA;AAAA,QACA,IAAA,CAAK;AAAA,OACP;AAAA,IACF;AACA,IAAA,YAAA,CAAa,IAAA,CAAK,CAAC,CAAA,EAAG,CAAA,KAAM;AAC1B,MAAA,IAAI,EAAE,IAAA,KAAS,CAAA,CAAE,MAAM,OAAO,CAAA,CAAE,OAAO,CAAA,CAAE,IAAA;AACzC,MAAA,OAAO,aAAA,CAAc,CAAA,CAAE,KAAA,EAAO,CAAA,CAAE,KAAK,CAAA;AAAA,IACvC,CAAC,CAAA;AAED,IAAA,MAAM,WAAA,GAA0B,KAAA,CAAM,IAAA,CAAK,EAAE,MAAA,EAAQ,KAAK,OAAA,EAAQ,EAAG,MAAM,EAAE,CAAA;AAC7E,IAAA,MAAM,YAAY,IAAI,KAAA,CAAc,KAAK,OAAO,CAAA,CAAE,KAAK,CAAC,CAAA;AAExD,IAAA,KAAA,MAAW,SAAS,YAAA,EAAc;AAChC,MAAA,IAAI,QAAA,GAAW,CAAA;AACf,MAAA,IAAI,QAAA,GAAW,SAAA,CAAU,CAAC,CAAA,IAAK,CAAA;AAC/B,MAAA,KAAA,IAAS,IAAA,GAAO,CAAA,EAAG,IAAA,GAAO,IAAA,CAAK,SAAS,IAAA,EAAA,EAAQ;AAC9C,QAAA,MAAM,IAAA,GAAO,SAAA,CAAU,IAAI,CAAA,IAAK,CAAA;AAChC,QAAA,IAAI,OAAO,QAAA,EAAU;AACnB,UAAA,QAAA,GAAW,IAAA;AACX,UAAA,QAAA,GAAW,IAAA;AAAA,QACb;AAAA,MACF;AACA,MAAA,MAAM,MAAA,GAAS,YAAY,QAAQ,CAAA;AACnC,MAAA,IAAI,CAAC,MAAA,EAAQ;AACX,QAAA,MAAM,IAAI,aAAa,4CAA4C,CAAA;AAAA,MACrE;AACA,MAAA,MAAA,CAAO,IAAA,CAAK,GAAG,KAAA,CAAM,OAAO,CAAA;AAC5B,MAAA,SAAA,CAAU,QAAQ,CAAA,GAAI,QAAA,GAAW,KAAA,CAAM,IAAA;AAAA,IACzC;AAEA,IAAA,MAAM,SAAsC,EAAC;AAC7C,IAAA,KAAA,IAAS,IAAA,GAAO,CAAA,EAAG,IAAA,GAAO,IAAA,CAAK,SAAS,IAAA,EAAA,EAAQ;AAC9C,MAAA,MAAM,WAAA,GAAc,WAAA,CAAY,IAAI,CAAA,IAAK,EAAC;AAC1C,MAAA,MAAM,eAAyB,EAAC;AAChC,MAAA,KAAA,IAAS,KAAA,GAAQ,CAAA,EAAG,KAAA,GAAQ,IAAA,CAAK,SAAS,KAAA,EAAA,EAAS;AACjD,QAAA,IAAI,UAAU,IAAA,EAAM;AACpB,QAAA,YAAA,CAAa,KAAK,GAAI,WAAA,CAAY,KAAK,CAAA,IAAK,EAAG,CAAA;AAAA,MACjD;AACA,MAAA,MAAA,CAAO,IAAA,CAAK,CAAC,YAAA,EAAc,WAAW,CAAC,CAAA;AAAA,IACzC;AAEA,IAAA,OAAO,MAAA;AAAA,EACT;AAAA,EAEA,UAAA,GAAqB;AACnB,IAAA,OAAO,IAAA,CAAK,OAAA;AAAA,EACd;AACF;AAOO,IAAM,cAAN,MAAkB;AAAA,EACvB,MAAM,CAAA,EAAwC;AAC5C,IAAA,MAAM,MAAA,GAAS,CAAA,CAAE,KAAA,CAAM,CAAC,CAAA;AACxB,IAAA,IAAI,WAAW,MAAA,EAAW;AACxB,MAAA,MAAM,IAAI,WAAW,4BAA4B,CAAA;AAAA,IACnD;AACA,IAAA,MAAM,QAAA,GAAW,MAAA;AACjB,IAAA,MAAM,SAAsC,EAAC;AAE7C,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACjC,MAAA,MAAM,YAAA,GAAe;AAAA,QACnB,GAAG,KAAA,CAAM,IAAA,CAAK,EAAE,MAAA,EAAQ,GAAE,EAAG,CAAC,CAAA,EAAG,CAAA,KAAM,CAAC,CAAA;AAAA,QACxC,GAAG,KAAA,CAAM,IAAA,CAAK,EAAE,QAAQ,QAAA,GAAW,CAAA,GAAI,CAAA,EAAE,EAAG,CAAC,CAAA,EAAG,CAAA,KAAM,CAAA,GAAI,IAAI,CAAC;AAAA,OACjE;AACA,MAAA,MAAM,WAAA,GAAc,CAAC,CAAC,CAAA;AACtB,MAAA,MAAA,CAAO,IAAA,CAAK,CAAC,YAAA,EAAc,WAAW,CAAC,CAAA;AAAA,IACzC;AAEA,IAAA,OAAO,MAAA;AAAA,EACT;AAAA,EAEA,WAAW,CAAA,EAAmB;AAC5B,IAAA,MAAM,MAAA,GAAS,CAAA,CAAE,KAAA,CAAM,CAAC,CAAA;AACxB,IAAA,IAAI,WAAW,MAAA,EAAW;AACxB,MAAA,MAAM,IAAI,WAAW,4BAA4B,CAAA;AAAA,IACnD;AACA,IAAA,OAAO,MAAA;AAAA,EACT;AACF;AAOO,IAAM,YAAN,MAAgB;AAAA,EACb,CAAA;AAAA,EAER,YAAY,CAAA,EAAW;AACrB,IAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,CAAC,CAAA,IAAK,CAAC,MAAA,CAAO,SAAA,CAAU,CAAC,CAAA,IAAK,CAAA,IAAK,CAAA,EAAG;AACzD,MAAA,MAAM,IAAI,qBAAA,CAAsB,8BAAA,EAAgC,GAAA,EAAK,CAAC,CAAA;AAAA,IACxE;AACA,IAAA,IAAA,CAAK,CAAA,GAAI,CAAA;AAAA,EACX;AAAA,EAEA,MAAM,CAAA,EAAwC;AAC5C,IAAA,MAAM,MAAA,GAAS,CAAA,CAAE,KAAA,CAAM,CAAC,CAAA;AACxB,IAAA,IAAI,WAAW,MAAA,EAAW;AACxB,MAAA,MAAM,IAAI,WAAW,4BAA4B,CAAA;AAAA,IACnD;AACA,IAAA,MAAM,QAAA,GAAW,MAAA;AACjB,IAAA,IAAI,IAAA,CAAK,IAAI,QAAA,EAAU;AACrB,MAAA,MAAM,IAAI,qBAAA,CAAsB,8CAAA,EAAgD,GAAA,EAAK,KAAK,CAAC,CAAA;AAAA,IAC7F;AAGA,IAAA,IAAI,OAAA,GAAU,CAAA;AACd,IAAA,MAAM,CAAA,GAAI,KAAK,CAAA,GAAI,QAAA,GAAW,IAAI,QAAA,GAAW,IAAA,CAAK,IAAI,IAAA,CAAK,CAAA;AAC3D,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,MAAA,OAAA,GAAW,OAAA,IAAW,QAAA,GAAW,CAAA,CAAA,IAAO,CAAA,GAAI,CAAA,CAAA;AAAA,IAC9C;AAIA,IAAA,IAAI,UAAU,GAAA,EAAQ;AACpB,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,CAAA,mBAAA,EAAsB,IAAA,CAAK,KAAA,CAAM,OAAO,CAAC,CAAA,qDAAA,CAAA;AAAA,QACzC,GAAA;AAAA,QACA,IAAA,CAAK;AAAA,OACP;AAAA,IACF;AAEA,IAAA,MAAM,SAAsC,EAAC;AAC7C,IAAA,MAAM,UAAA,GAAa,KAAA,CAAM,IAAA,CAAK,EAAE,MAAA,EAAQ,UAAS,EAAG,CAAC,CAAA,EAAG,CAAA,KAAM,CAAC,CAAA;AAG/D,IAAA,MAAM,OAAA,GAAU,CAAC,KAAA,EAAe,YAAA,KAA2B;AACzD,MAAA,IAAI,YAAA,CAAa,MAAA,KAAW,IAAA,CAAK,CAAA,EAAG;AAClC,QAAA,MAAM,OAAA,GAAU,IAAI,GAAA,CAAI,YAAY,CAAA;AACpC,QAAA,MAAM,WAAA,GAAc,CAAC,GAAG,YAAY,CAAA;AACpC,QAAA,MAAM,YAAA,GAAe,WAAW,MAAA,CAAO,CAAC,MAAM,CAAC,OAAA,CAAQ,GAAA,CAAI,CAAC,CAAC,CAAA;AAC7D,QAAA,MAAA,CAAO,IAAA,CAAK,CAAC,YAAA,EAAc,WAAW,CAAC,CAAA;AACvC,QAAA;AAAA,MACF;AACA,MAAA,KAAA,IAAS,CAAA,GAAI,KAAA,EAAO,CAAA,GAAI,QAAA,EAAU,CAAA,EAAA,EAAK;AACrC,QAAA,YAAA,CAAa,KAAK,CAAC,CAAA;AACnB,QAAA,OAAA,CAAQ,CAAA,GAAI,GAAG,YAAY,CAAA;AAC3B,QAAA,YAAA,CAAa,GAAA,EAAI;AAAA,MACnB;AAAA,IACF,CAAA;AAEA,IAAA,OAAA,CAAQ,CAAA,EAAG,EAAE,CAAA;AAEb,IAAA,OAAO,MAAA;AAAA,EACT;AAAA,EAEA,WAAW,CAAA,EAAmB;AAC5B,IAAA,MAAM,MAAA,GAAS,CAAA,CAAE,KAAA,CAAM,CAAC,CAAA;AACxB,IAAA,IAAI,WAAW,MAAA,EAAW;AACxB,MAAA,MAAM,IAAI,WAAW,4BAA4B,CAAA;AAAA,IACnD;AACA,IAAA,MAAM,CAAA,GAAI,MAAA;AACV,IAAA,IAAI,IAAA,CAAK,IAAI,CAAA,EAAG;AACd,MAAA,MAAM,IAAI,qBAAA,CAAsB,8CAAA,EAAgD,GAAA,EAAK,KAAK,CAAC,CAAA;AAAA,IAC7F;AAEA,IAAA,IAAI,MAAA,GAAS,CAAA;AACb,IAAA,MAAM,CAAA,GAAI,KAAK,CAAA,GAAI,CAAA,GAAI,IAAI,CAAA,GAAI,IAAA,CAAK,IAAI,IAAA,CAAK,CAAA;AAC7C,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,MAAA,MAAA,GAAU,MAAA,IAAU,CAAA,GAAI,CAAA,CAAA,IAAO,CAAA,GAAI,CAAA,CAAA;AAAA,IACrC;AACA,IAAA,OAAO,IAAA,CAAK,MAAM,MAAM,CAAA;AAAA,EAC1B;AACF","file":"chunk-AD436M45.js","sourcesContent":["// Encoders\nexport {\n  LabelBinarizer,\n  LabelEncoder,\n  MultiLabelBinarizer,\n  OneHotEncoder,\n  OrdinalEncoder,\n} from \"./encoders\";\n\n// Scalers\nexport {\n  MaxAbsScaler,\n  MinMaxScaler,\n  Normalizer,\n  PowerTransformer,\n  QuantileTransformer,\n  RobustScaler,\n  StandardScaler,\n} from \"./scalers\";\n\n// Splitting\nexport {\n  GroupKFold,\n  KFold,\n  LeaveOneOut,\n  LeavePOut,\n  StratifiedKFold,\n  trainTestSplit,\n} from \"./split\";\n","/**\n * Shared internal helpers for the preprocess module.\n *\n * These functions are used across scalers and splitting utilities\n * to handle tensor shape validation, seeded RNG, and index shuffling.\n *\n * @internal\n */\n\nimport { DeepboxError, DTypeError, InvalidParameterError, ShapeError } from \"../core/errors\";\nimport type { Tensor } from \"../ndarray\";\n\n/**\n * Assert that a tensor has numeric dtype (not string).\n *\n * @internal\n */\nexport function assertNumericTensor(X: Tensor, name: string): void {\n  if (X.dtype === \"string\") {\n    throw new DTypeError(`${name} must be numeric`);\n  }\n}\n\n/**\n * Assert that a tensor is 2-dimensional.\n *\n * @internal\n */\nexport function assert2D(X: Tensor, name: string): void {\n  if (X.ndim !== 2) {\n    throw new ShapeError(`${name} must be a 2D tensor, got ${X.ndim}D`);\n  }\n}\n\n/**\n * Extract and validate the shape of a 2D tensor.\n *\n * @internal\n */\nexport function getShape2D(X: Tensor): [number, number] {\n  if (X.ndim !== 2 || X.shape[0] === undefined || X.shape[1] === undefined) {\n    throw new ShapeError(`Expected 2D tensor with valid shape, got shape [${X.shape.join(\", \")}]`);\n  }\n  return [X.shape[0], X.shape[1]];\n}\n\n/**\n * Extract the stride of a 1D tensor.\n *\n * @internal\n */\nexport function getStride1D(X: Tensor): number {\n  const stride = X.strides[0];\n  if (stride === undefined) {\n    throw new DeepboxError(\"Internal error: missing stride for 1D tensor\");\n  }\n  return stride;\n}\n\n/**\n * Extract the strides of a 2D tensor.\n *\n * @internal\n */\nexport function getStrides2D(X: Tensor): [number, number] {\n  const stride0 = X.strides[0];\n  const stride1 = X.strides[1];\n  if (stride0 === undefined || stride1 === undefined) {\n    throw new DeepboxError(\"Internal error: missing strides for 2D tensor\");\n  }\n  return [stride0, stride1];\n}\n\n/**\n * Simple seeded random number generator using Linear Congruential Generator (LCG).\n * Provides reproducible pseudo-random sequences when given a seed.\n *\n * @param seed - Non-negative safe integer seed value\n * @returns Function that generates random numbers in [0, 1)\n *\n * @internal\n */\nexport function createSeededRandom(seed: number): () => number {\n  const a = 1103515245;\n  const c = 12345;\n  const m = 2 ** 31;\n\n  if (\n    !Number.isFinite(seed) ||\n    !Number.isInteger(seed) ||\n    !Number.isSafeInteger(seed) ||\n    seed < 0\n  ) {\n    throw new InvalidParameterError(\n      \"randomState must be a non-negative safe integer\",\n      \"randomState\",\n      seed\n    );\n  }\n\n  let state = seed % m;\n\n  return () => {\n    state = (a * state + c) % m;\n    return state / m;\n  };\n}\n\n/**\n * Shuffle an array of indices in-place using Fisher-Yates algorithm.\n *\n * @param indices - Array of integer indices to shuffle\n * @param random - Random number generator returning values in [0, 1)\n *\n * @internal\n */\nexport function shuffleIndicesInPlace(indices: number[], random: () => number): void {\n  for (let i = indices.length - 1; i > 0; i--) {\n    const j = Math.floor(random() * (i + 1));\n    const temp = indices[i];\n    if (temp === undefined) {\n      throw new DeepboxError(\"Internal error: shuffle source index missing\");\n    }\n    const swap = indices[j];\n    if (swap === undefined) {\n      throw new DeepboxError(\"Internal error: shuffle target index missing\");\n    }\n    indices[i] = swap;\n    indices[j] = temp;\n  }\n}\n","import {\n  DeepboxError,\n  DTypeError,\n  getConfig,\n  InvalidParameterError,\n  NotFittedError,\n  ShapeError,\n} from \"../core\";\nimport { CSRMatrix, empty, type Tensor, Tensor as TensorImpl, tensor, zeros } from \"../ndarray\";\nimport { assert2D, assertNumericTensor, getShape2D, getStride1D, getStrides2D } from \"./_internal\";\n\n/**\n * Type representing a category value that can be a string, number, or bigint.\n * Used for categorical encoding operations.\n */\ntype Category = number | string | bigint;\n\ntype CategoryType = \"string\" | \"number\" | \"bigint\";\n\ntype CategoriesOption = \"auto\" | ReadonlyArray<ReadonlyArray<Category>>;\n\ntype DropOption = \"first\" | \"if_binary\" | null;\n\nfunction getStringData(t: Tensor): string[] {\n  if (t.dtype !== \"string\") {\n    throw new DTypeError(\"Expected string tensor\");\n  }\n  if (!Array.isArray(t.data)) {\n    throw new DeepboxError(\"Internal error: invalid string tensor storage\");\n  }\n  return t.data;\n}\n\nfunction getNumericData(t: Tensor): ArrayLike<number | bigint> {\n  if (t.dtype === \"string\") {\n    throw new DTypeError(\"Expected numeric tensor\");\n  }\n  if (Array.isArray(t.data)) {\n    throw new DeepboxError(\"Internal error: invalid numeric tensor storage\");\n  }\n  return t.data;\n}\n\nfunction inferCategoryType(values: Category[], paramName: string): CategoryType {\n  let hasString = false;\n  let hasNumber = false;\n  let hasBigInt = false;\n\n  for (const value of values) {\n    if (typeof value === \"string\") {\n      hasString = true;\n    } else if (typeof value === \"number\") {\n      if (!Number.isFinite(value)) {\n        throw new InvalidParameterError(\"Category values must be finite numbers\", paramName, value);\n      }\n      hasNumber = true;\n    } else if (typeof value === \"bigint\") {\n      hasBigInt = true;\n    }\n  }\n\n  const typeCount = (hasString ? 1 : 0) + (hasNumber ? 1 : 0) + (hasBigInt ? 1 : 0);\n\n  if (typeCount === 0) {\n    return \"number\";\n  }\n  if (typeCount > 1) {\n    throw new InvalidParameterError(\"Mixed category types are not supported\", paramName);\n  }\n  if (hasString) return \"string\";\n  if (hasBigInt) return \"bigint\";\n  return \"number\";\n}\n\nfunction sortCategories(values: Iterable<Category>, paramName: string): Category[] {\n  const arr = Array.from(values);\n  if (arr.length === 0) return arr;\n\n  const categoryType = inferCategoryType(arr, paramName);\n\n  if (categoryType === \"string\") {\n    arr.sort((a, b) => {\n      if (typeof a !== \"string\" || typeof b !== \"string\") {\n        throw new DeepboxError(\"Internal error: inconsistent category types\");\n      }\n      return a.localeCompare(b);\n    });\n    return arr;\n  }\n\n  if (categoryType === \"bigint\") {\n    arr.sort((a, b) => {\n      if (typeof a !== \"bigint\" || typeof b !== \"bigint\") {\n        throw new DeepboxError(\"Internal error: inconsistent category types\");\n      }\n      if (a < b) return -1;\n      if (a > b) return 1;\n      return 0;\n    });\n    return arr;\n  }\n\n  arr.sort((a, b) => {\n    if (typeof a !== \"number\" || typeof b !== \"number\") {\n      throw new DeepboxError(\"Internal error: inconsistent category types\");\n    }\n    return a - b;\n  });\n  return arr;\n}\n\nfunction validateCategoryValues(values: ReadonlyArray<Category>, paramName: string): Category[] {\n  if (values.length === 0) {\n    throw new InvalidParameterError(\"categories must contain at least one value\", paramName);\n  }\n  const arr = Array.from(values);\n  inferCategoryType(arr, paramName);\n\n  const seen = new Set<Category>();\n  for (const value of arr) {\n    if (seen.has(value)) {\n      throw new InvalidParameterError(\n        `categories must be unique; duplicate value ${String(value)}`,\n        paramName,\n        value\n      );\n    }\n    seen.add(value);\n  }\n  return arr;\n}\n\nfunction resolveCategoriesOption(\n  categoriesOption: CategoriesOption,\n  nFeatures: number,\n  paramName: string\n): ReadonlyArray<ReadonlyArray<Category>> | null {\n  if (categoriesOption === \"auto\") {\n    return null;\n  }\n  if (!Array.isArray(categoriesOption)) {\n    throw new InvalidParameterError(\n      \"categories must be 'auto' or an array of category arrays\",\n      paramName,\n      categoriesOption\n    );\n  }\n  if (categoriesOption.length !== nFeatures) {\n    throw new InvalidParameterError(\n      \"categories length must match number of features\",\n      paramName,\n      categoriesOption.length\n    );\n  }\n  return categoriesOption;\n}\n\n/**\n * Reads a single value from a 1D tensor at the specified index.\n * Handles both string and numeric dtypes safely.\n *\n * @param t - The tensor to read from\n * @param i - The index to read at\n * @returns The value as a string, number, or bigint\n */\nfunction read1DValue(t: Tensor, i: number): Category {\n  const stride = getStride1D(t);\n  const idx = t.offset + i * stride;\n  if (t.dtype === \"string\") {\n    const value = getStringData(t)[idx];\n    if (value === undefined) {\n      throw new DeepboxError(\"Internal error: string tensor access out of bounds\");\n    }\n    return value;\n  }\n  const value = getNumericData(t)[idx];\n  if (value === undefined) {\n    throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n  }\n  return typeof value === \"bigint\" ? value : Number(value);\n}\n\n/**\n * Reads a value from a 2D tensor at the specified row and column.\n * Handles both string and numeric dtypes safely.\n *\n * @param t - The tensor to read from\n * @param row - The row index\n * @param col - The column index\n * @returns The value as a string, number, or bigint\n */\nfunction read2DValue(t: Tensor, row: number, col: number): Category {\n  const [stride0, stride1] = getStrides2D(t);\n  const idx = t.offset + row * stride0 + col * stride1;\n  if (t.dtype === \"string\") {\n    const value = getStringData(t)[idx];\n    if (value === undefined) {\n      throw new DeepboxError(\"Internal error: string tensor access out of bounds\");\n    }\n    return value;\n  }\n  const value = getNumericData(t)[idx];\n  if (value === undefined) {\n    throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n  }\n  return typeof value === \"bigint\" ? value : Number(value);\n}\n\nfunction assert1D(t: Tensor, name: string): void {\n  if (t.ndim !== 1) {\n    throw new ShapeError(`${name} must be a 1D tensor`);\n  }\n}\n\nfunction categoryValueAt(values: Category[], index: number, context: string): Category {\n  const value = values[index];\n  if (value === undefined) {\n    throw new DeepboxError(`Internal error: missing category at index ${index} (${context})`);\n  }\n  return value;\n}\n\nfunction inferCategoryTypeFromRows(rows: Category[][], paramName: string): CategoryType {\n  const values: Category[] = [];\n  for (const row of rows) {\n    for (const value of row) {\n      values.push(value);\n    }\n  }\n  return inferCategoryType(values, paramName);\n}\n\nfunction emptyCategoryVectorFromClasses(classes: Category[], paramName: string): Tensor {\n  const categoryType = inferCategoryType(classes, paramName);\n  if (categoryType === \"string\") {\n    return empty([0], { dtype: \"string\" });\n  }\n  if (categoryType === \"bigint\") {\n    return empty([0], { dtype: \"int64\" });\n  }\n  return zeros([0], { dtype: \"float64\" });\n}\n\nfunction emptyCategoryMatrixFromCategories(\n  categories: Category[][],\n  nFeatures: number,\n  paramName: string\n): Tensor {\n  const categoryType = inferCategoryTypeFromRows(categories, paramName);\n  if (categoryType === \"string\") {\n    return empty([0, nFeatures], { dtype: \"string\" });\n  }\n  if (categoryType === \"bigint\") {\n    return empty([0, nFeatures], { dtype: \"int64\" });\n  }\n  return zeros([0, nFeatures], { dtype: \"float64\" });\n}\n\nfunction toCategoryVectorTensor(values: Category[], paramName = \"y\"): Tensor {\n  const categoryType = inferCategoryType(values, paramName);\n\n  if (categoryType === \"string\") {\n    const out = new Array<string>(values.length);\n    for (let i = 0; i < values.length; i++) {\n      const value = values[i];\n      if (typeof value !== \"string\") {\n        throw new DeepboxError(\"Internal error: expected string category value\");\n      }\n      out[i] = value;\n    }\n    return tensor(out);\n  }\n\n  if (categoryType === \"bigint\") {\n    const out = new BigInt64Array(values.length);\n    for (let i = 0; i < values.length; i++) {\n      const value = values[i];\n      if (typeof value !== \"bigint\") {\n        throw new DeepboxError(\"Internal error: expected bigint category value\");\n      }\n      out[i] = value;\n    }\n    return tensor(out);\n  }\n\n  const out = new Float64Array(values.length);\n  for (let i = 0; i < values.length; i++) {\n    const value = values[i];\n    if (value === undefined || typeof value !== \"number\") {\n      throw new DeepboxError(\"Internal error: expected numeric category value\");\n    }\n    out[i] = value;\n  }\n  return tensor(out);\n}\n\nfunction toCategoryMatrixTensor(values: Category[][], paramName = \"X\"): Tensor {\n  const rows = values.length;\n  const cols = rows > 0 ? (values[0]?.length ?? 0) : 0;\n\n  for (let i = 0; i < rows; i++) {\n    const row = values[i];\n    if (!row) {\n      throw new DeepboxError(\"Internal error: missing row in category matrix\");\n    }\n    if (row.length !== cols) {\n      throw new ShapeError(\"Ragged category matrix cannot be converted to tensor\");\n    }\n  }\n\n  const flat: Category[] = [];\n  for (const row of values) {\n    for (const value of row) {\n      flat.push(value);\n    }\n  }\n\n  const categoryType = inferCategoryType(flat, paramName);\n  if (categoryType === \"string\") {\n    const out = new Array<string[]>(rows);\n    for (let i = 0; i < rows; i++) {\n      const row = values[i];\n      if (!row) {\n        throw new DeepboxError(\"Internal error: missing row in category matrix\");\n      }\n      const outRow = new Array<string>(cols);\n      for (let j = 0; j < cols; j++) {\n        const value = row[j];\n        if (typeof value !== \"string\") {\n          throw new DeepboxError(\"Internal error: expected string category value\");\n        }\n        outRow[j] = value;\n      }\n      out[i] = outRow;\n    }\n    return tensor(out);\n  }\n  if (categoryType === \"number\") {\n    const out = new Array<number[]>(rows);\n    for (let i = 0; i < rows; i++) {\n      const row = values[i];\n      if (!row) {\n        throw new DeepboxError(\"Internal error: missing row in category matrix\");\n      }\n      const outRow = new Array<number>(cols);\n      for (let j = 0; j < cols; j++) {\n        const value = row[j];\n        if (typeof value !== \"number\") {\n          throw new DeepboxError(\"Internal error: expected numeric category value\");\n        }\n        outRow[j] = value;\n      }\n      out[i] = outRow;\n    }\n    return tensor(out, { dtype: \"float64\" });\n  }\n\n  const data = new BigInt64Array(rows * cols);\n  for (let i = 0; i < flat.length; i++) {\n    const value = flat[i];\n    if (typeof value !== \"bigint\") {\n      throw new DeepboxError(\"Internal error: expected bigint category value\");\n    }\n    data[i] = value;\n  }\n\n  const { defaultDevice } = getConfig();\n  return TensorImpl.fromTypedArray({\n    data,\n    shape: [rows, cols],\n    dtype: \"int64\",\n    device: defaultDevice,\n  });\n}\n\n/**\n * Encode target labels with value between 0 and n_classes-1.\n *\n * This transformer encodes categorical labels (strings or numbers) into integers\n * in the range [0, n_classes-1]. It maintains a mapping of unique classes to\n * their integer representations and can reverse the transformation.\n *\n * **Time Complexity:**\n * - fit: O(n) where n is the number of samples\n * - transform: O(n) with O(1) lookup per sample\n * - inverseTransform: O(n)\n *\n * **Space Complexity:** O(k) where k is the number of unique classes\n *\n * @example\n * ```js\n * import { LabelEncoder } from 'deepbox/preprocess';\n * import { tensor } from 'deepbox/ndarray';\n *\n * const y = tensor(['cat', 'dog', 'cat', 'bird']);\n * const encoder = new LabelEncoder();\n * encoder.fit(y);\n * const yEncoded = encoder.transform(y);  // [1, 2, 1, 0]\n * const yDecoded = encoder.inverseTransform(yEncoded); // ['cat', 'dog', 'cat', 'bird']\n * ```\n *\n * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.LabelEncoder.html | Scikit-learn LabelEncoder}\n */\nexport class LabelEncoder {\n  /** Indicates whether the encoder has been fitted to data */\n  private fitted = false;\n  /** Array of unique classes found during fitting, sorted for consistency */\n  private classes_?: Category[];\n  /** Map from class value to encoded integer index for O(1) lookup */\n  private classToIndex_?: Map<Category, number>;\n\n  /**\n   * Fit label encoder to a set of labels.\n   * Extracts unique classes and creates an index mapping.\n   *\n   * @param y - Target labels (1D tensor of strings or numbers)\n   * @returns this - Returns self for method chaining\n   * @throws {InvalidParameterError} If y is empty\n   */\n  fit(y: Tensor): this {\n    assert1D(y, \"y\");\n    if (y.size === 0) {\n      throw new InvalidParameterError(\"Cannot fit LabelEncoder on empty array\", \"y\");\n    }\n\n    // Collect unique classes using a Set for O(n) complexity\n    const uniqueSet = new Set<Category>();\n    for (let i = 0; i < y.size; i++) {\n      uniqueSet.add(read1DValue(y, i));\n    }\n\n    // Sort classes for consistent ordering across fits\n    this.classes_ = sortCategories(uniqueSet, \"y\");\n\n    // Build index map for O(1) transform lookups\n    this.classToIndex_ = new Map();\n    for (let i = 0; i < this.classes_.length; i++) {\n      this.classToIndex_.set(categoryValueAt(this.classes_, i, \"LabelEncoder.fit\"), i);\n    }\n\n    this.fitted = true;\n    return this;\n  }\n\n  /**\n   * Transform labels to normalized encoding.\n   * Each unique label is mapped to an integer in [0, n_classes-1].\n   *\n   * @param y - Target labels to encode (1D tensor)\n   * @returns Encoded labels as integer tensor\n   * @throws {NotFittedError} If encoder is not fitted\n   * @throws {InvalidParameterError} If y contains labels not seen during fit\n   */\n  transform(y: Tensor): Tensor {\n    if (!this.fitted) {\n      throw new NotFittedError(\"LabelEncoder must be fitted before transform\");\n    }\n    assert1D(y, \"y\");\n    if (y.size === 0) {\n      return tensor([]);\n    }\n\n    const lookup = this.classToIndex_;\n    if (!this.classes_ || !lookup) {\n      throw new DeepboxError(\"LabelEncoder internal error: missing fitted state\");\n    }\n\n    // Pre-allocate result array for better performance\n    const result = new Array<number>(y.size);\n\n    // Transform each label using O(1) map lookup\n    for (let i = 0; i < y.size; i++) {\n      const val = read1DValue(y, i);\n      const idx = lookup.get(val);\n      if (idx === undefined) {\n        throw new InvalidParameterError(\n          `Unknown label: ${String(val)}. Label must be present during fit.`,\n          \"y\",\n          val\n        );\n      }\n      result[i] = idx;\n    }\n\n    return tensor(result, { dtype: \"float64\" });\n  }\n\n  /**\n   * Fit label encoder and return encoded labels in one step.\n   * Convenience method equivalent to calling fit(y).transform(y).\n   *\n   * @param y - Target labels (1D tensor)\n   * @returns Encoded labels as integer tensor\n   */\n  fitTransform(y: Tensor): Tensor {\n    return this.fit(y).transform(y);\n  }\n\n  /**\n   * Transform integer labels back to original encoding.\n   * Reverses the encoding performed by transform().\n   *\n   * @param y - Encoded labels (1D integer tensor)\n   * @returns Original labels (strings or numbers)\n   * @throws {NotFittedError} If encoder is not fitted\n   * @throws {InvalidParameterError} If y contains invalid indices\n   */\n  inverseTransform(y: Tensor): Tensor {\n    if (!this.fitted) {\n      throw new NotFittedError(\"LabelEncoder must be fitted before inverse_transform\");\n    }\n    assert1D(y, \"y\");\n    assertNumericTensor(y, \"y\");\n    const classes = this.classes_;\n    if (!classes) {\n      throw new DeepboxError(\"LabelEncoder internal error: missing fitted state\");\n    }\n    if (y.size === 0) {\n      return emptyCategoryVectorFromClasses(classes, \"y\");\n    }\n\n    const classesLen = classes.length;\n\n    const result = new Array<Category>(y.size);\n    const stride = getStride1D(y);\n    const data = getNumericData(y);\n\n    // Map each encoded index back to its original class\n    for (let i = 0; i < y.size; i++) {\n      const raw = data[y.offset + i * stride];\n      if (raw === undefined) {\n        throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n      }\n      const idx = Number(raw);\n\n      // Validate index is in valid range\n      if (idx < 0 || idx >= classesLen || !Number.isInteger(idx)) {\n        throw new InvalidParameterError(\n          `Invalid label index: ${idx}. Must be integer in [0, ${classesLen - 1}]`,\n          \"y\",\n          idx\n        );\n      }\n\n      result[i] = categoryValueAt(classes, idx, \"LabelEncoder.inverseTransform\");\n    }\n\n    // Return tensor with appropriate dtype\n    return toCategoryVectorTensor(result, \"y\");\n  }\n}\n\n/**\n * Encode categorical features as one-hot numeric array.\n *\n * This encoder transforms categorical features into a binary one-hot encoding.\n * Each categorical feature with n unique values is transformed into n binary features,\n * with only one active (set to 1) per sample.\n *\n * **Time Complexity:**\n * - fit: O(n*m) where n is samples, m is features\n * - transform: O(n*m*k) where k is average categories per feature\n * - Sparse mode is more efficient for high-cardinality features\n *\n * **Space Complexity:**\n * - Dense: O(n * sum(k_i)) where k_i is unique categories for feature i\n * - Sparse: O(nnz) where nnz is number of non-zero elements\n *\n * @example\n * ```js\n * const X = tensor([['red', 'S'], ['blue', 'M'], ['red', 'L']]);\n * const encoder = new OneHotEncoder({ sparse: false });\n * encoder.fit(X);\n * const encoded = encoder.transform(X);\n * // Result: [[1,0,1,0,0], [0,1,0,1,0], [1,0,0,0,1]]\n * ```\n *\n * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OneHotEncoder.html | Scikit-learn OneHotEncoder}\n */\nexport class OneHotEncoder {\n  /** Indicates whether the encoder has been fitted to data */\n  private fitted = false;\n  /** Array of unique categories for each feature */\n  private categories_?: Category[][];\n  /** Maps from category value to index for each feature (for O(1) lookup) */\n  private categoryToIndex_?: Array<Map<Category, number>>;\n  /** Whether to return sparse matrix (CSR) or dense array */\n  private sparse: boolean;\n  /** How to handle unknown categories during transform */\n  private handleUnknown: \"error\" | \"ignore\";\n  /** Drop policy to avoid collinearity */\n  private drop: DropOption;\n  /** Per-feature dropped category index */\n  private dropIndices_?: Array<number | null>;\n  /** Categories configuration */\n  private categoriesOption: CategoriesOption;\n\n  /**\n   * Creates a new OneHotEncoder instance.\n   *\n   * @param options - Configuration options\n   * @param options.sparse - If true, returns CSRMatrix; if false, returns dense Tensor (default: false)\n   * @param options.sparseOutput - Alias for sparse (default: false)\n   * @param options.handleUnknown - How to handle unknown categories (default: \"error\")\n   * @param options.drop - If set, drops the first or binary category per feature\n   * @param options.categories - \"auto\" or explicit category list per feature\n   */\n  constructor(\n    options: {\n      sparse?: boolean;\n      sparseOutput?: boolean;\n      handleUnknown?: \"error\" | \"ignore\";\n      drop?: \"first\" | \"if_binary\" | null;\n      categories?: CategoriesOption;\n    } = {}\n  ) {\n    const sparseOption = options.sparse ?? options.sparseOutput ?? false;\n    if (options.sparse !== undefined && options.sparseOutput !== undefined) {\n      if (options.sparse !== options.sparseOutput) {\n        throw new InvalidParameterError(\n          \"sparse and sparseOutput must match when both are provided\",\n          \"sparse\",\n          options.sparse\n        );\n      }\n    }\n    this.sparse = sparseOption;\n    this.handleUnknown = options.handleUnknown ?? \"error\";\n    this.drop = options.drop ?? null;\n    this.categoriesOption = options.categories ?? \"auto\";\n\n    if (typeof this.sparse !== \"boolean\") {\n      throw new InvalidParameterError(\"sparse must be a boolean\", \"sparse\", this.sparse);\n    }\n    if (this.handleUnknown !== \"error\" && this.handleUnknown !== \"ignore\") {\n      throw new InvalidParameterError(\n        \"handleUnknown must be 'error' or 'ignore'\",\n        \"handleUnknown\",\n        this.handleUnknown\n      );\n    }\n    if (this.drop !== null && this.drop !== \"first\" && this.drop !== \"if_binary\") {\n      throw new InvalidParameterError(\n        \"drop must be 'first', 'if_binary', or null\",\n        \"drop\",\n        this.drop\n      );\n    }\n  }\n\n  /**\n   * Fit OneHotEncoder to X.\n   * Learns the unique categories for each feature.\n   *\n   * @param X - Training data (2D tensor of categorical features)\n   * @returns this - Returns self for method chaining\n   * @throws {ShapeError} If X is not a 2D tensor\n   * @throws {InvalidParameterError} If X is empty\n   */\n  fit(X: Tensor): this {\n    assert2D(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n\n    if (nSamples === 0 || nFeatures === 0) {\n      throw new InvalidParameterError(\"Cannot fit OneHotEncoder on empty array\", \"X\");\n    }\n\n    // Initialize storage for categories and lookup maps\n    this.categories_ = [];\n    this.categoryToIndex_ = [];\n\n    const explicitCategories = resolveCategoriesOption(\n      this.categoriesOption,\n      nFeatures,\n      \"categories\"\n    );\n\n    // For each feature, collect or validate categories\n    for (let j = 0; j < nFeatures; j++) {\n      let cats: Category[];\n\n      if (explicitCategories) {\n        const featureCats = explicitCategories[j];\n        if (!featureCats) {\n          throw new InvalidParameterError(\"Missing categories for feature\", \"categories\", j);\n        }\n        if (!Array.isArray(featureCats)) {\n          throw new InvalidParameterError(\n            \"categories must be an array of category arrays\",\n            \"categories\",\n            featureCats\n          );\n        }\n        cats = validateCategoryValues(featureCats, \"categories\");\n      } else {\n        const uniqueSet = new Set<Category>();\n\n        // Scan all samples to find unique values in this feature\n        for (let i = 0; i < nSamples; i++) {\n          uniqueSet.add(read2DValue(X, i, j));\n        }\n\n        // Sort categories for consistent ordering\n        cats = sortCategories(uniqueSet, \"X\");\n      }\n\n      if (cats.length === 0) {\n        throw new InvalidParameterError(\"Each feature must have at least one category\", \"X\", j);\n      }\n\n      this.categories_.push(cats);\n\n      // Build index map for O(1) transform lookups\n      const map = new Map<Category, number>();\n      for (let k = 0; k < cats.length; k++) {\n        map.set(categoryValueAt(cats, k, \"OneHotEncoder.fit\"), k);\n      }\n      this.categoryToIndex_.push(map);\n\n      // Validate training data against explicit categories\n      if (explicitCategories) {\n        for (let i = 0; i < nSamples; i++) {\n          const val = read2DValue(X, i, j);\n          if (!map.has(val)) {\n            throw new InvalidParameterError(\n              `Unknown category: ${String(val)} in feature ${j}`,\n              \"X\",\n              val\n            );\n          }\n        }\n      }\n    }\n\n    this.dropIndices_ = this.categories_.map((cats) => {\n      if (this.drop === null) return null;\n      if (this.drop === \"first\") return cats.length > 0 ? 0 : null;\n      if (this.drop === \"if_binary\") return cats.length === 2 ? 0 : null;\n      return null;\n    });\n\n    this.fitted = true;\n    return this;\n  }\n\n  /**\n   * Transform X using one-hot encoding.\n   * Each categorical value is converted to a binary vector.\n   *\n   * @param X - Data to transform (2D tensor)\n   * @returns Encoded data as dense Tensor or sparse CSRMatrix\n   * @throws {NotFittedError} If encoder is not fitted\n   * @throws {InvalidParameterError} If X contains unknown categories\n   */\n  transform(X: Tensor): Tensor | CSRMatrix {\n    if (!this.fitted) {\n      throw new NotFittedError(\"OneHotEncoder must be fitted before transform\");\n    }\n    assert2D(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n\n    const categories = this.categories_;\n    const categoryMaps = this.categoryToIndex_;\n    if (!categories || !categoryMaps) {\n      throw new DeepboxError(\"OneHotEncoder internal error: missing fitted state\");\n    }\n    const fittedFeatures = categories.length;\n    if (nFeatures !== fittedFeatures) {\n      throw new InvalidParameterError(\n        \"X has a different feature count than during fit\",\n        \"X\",\n        nFeatures\n      );\n    }\n\n    const dropIndices = this.dropIndices_ ?? categories.map(() => null);\n\n    // Calculate total output columns (sum of all category counts minus drops)\n    let totalCols = 0;\n    for (let j = 0; j < categories.length; j++) {\n      const cats = categories[j];\n      if (!cats) continue;\n      const dropIndex = dropIndices[j] ?? null;\n      totalCols += cats.length - (dropIndex === null ? 0 : 1);\n    }\n\n    if (nSamples === 0 || nFeatures === 0) {\n      return this.sparse\n        ? CSRMatrix.fromCOO({\n            rows: 0,\n            cols: totalCols,\n            rowIndices: new Int32Array(0),\n            colIndices: new Int32Array(0),\n            values: new Float64Array(0),\n          })\n        : zeros([0, totalCols], { dtype: \"float64\" });\n    }\n\n    if (this.sparse) {\n      const rowIdx: number[] = [];\n      const colIdx: number[] = [];\n      const vals: number[] = [];\n\n      for (let i = 0; i < nSamples; i++) {\n        let colOffset = 0;\n        for (let j = 0; j < nFeatures; j++) {\n          const cats = categories[j];\n          const map = categoryMaps[j];\n          const dropIndex = dropIndices[j] ?? null;\n          if (!cats || !map) {\n            throw new DeepboxError(\"OneHotEncoder internal error: missing fitted categories\");\n          }\n          const outSize = cats.length - (dropIndex === null ? 0 : 1);\n          const val = read2DValue(X, i, j);\n          const idx = map.get(val);\n          if (idx === undefined) {\n            if (this.handleUnknown === \"ignore\") {\n              colOffset += outSize;\n              continue;\n            }\n            throw new InvalidParameterError(`Unknown category: ${String(val)}`, \"X\", val);\n          }\n\n          if (dropIndex !== null && idx === dropIndex) {\n            colOffset += outSize;\n            continue;\n          }\n\n          const adjusted = dropIndex !== null && idx > dropIndex ? idx - 1 : idx;\n          rowIdx.push(i);\n          colIdx.push(colOffset + adjusted);\n          vals.push(1);\n          colOffset += outSize;\n        }\n      }\n\n      return CSRMatrix.fromCOO({\n        rows: nSamples,\n        cols: totalCols,\n        rowIndices: Int32Array.from(rowIdx),\n        colIndices: Int32Array.from(colIdx),\n        values: Float64Array.from(vals),\n      });\n    }\n\n    const result = Array.from({ length: nSamples }, () => new Array<number>(totalCols).fill(0));\n\n    for (let i = 0; i < nSamples; i++) {\n      let colOffset = 0;\n      for (let j = 0; j < nFeatures; j++) {\n        const cats = categories[j];\n        const map = categoryMaps[j];\n        const dropIndex = dropIndices[j] ?? null;\n        if (!cats || !map) {\n          throw new DeepboxError(\"OneHotEncoder internal error: missing fitted categories\");\n        }\n        const outSize = cats.length - (dropIndex === null ? 0 : 1);\n        const val = read2DValue(X, i, j);\n        const idx = map.get(val);\n        if (idx === undefined) {\n          if (this.handleUnknown === \"ignore\") {\n            colOffset += outSize;\n            continue;\n          }\n          throw new InvalidParameterError(`Unknown category: ${String(val)}`, \"X\", val);\n        }\n        if (dropIndex !== null && idx === dropIndex) {\n          colOffset += outSize;\n          continue;\n        }\n        const row = result[i];\n        if (row === undefined) {\n          throw new DeepboxError(\"Internal error: result row access failed\");\n        }\n        const adjusted = dropIndex !== null && idx > dropIndex ? idx - 1 : idx;\n        row[colOffset + adjusted] = 1;\n        colOffset += outSize;\n      }\n    }\n\n    return tensor(result, { dtype: \"float64\", device: X.device });\n  }\n\n  fitTransform(X: Tensor): Tensor | CSRMatrix {\n    return this.fit(X).transform(X);\n  }\n\n  inverseTransform(X: Tensor | CSRMatrix): Tensor {\n    if (!this.fitted) {\n      throw new NotFittedError(\"OneHotEncoder must be fitted before inverse_transform\");\n    }\n    const dense = X instanceof CSRMatrix ? X.toDense() : X;\n    assert2D(dense, \"X\");\n    assertNumericTensor(dense, \"X\");\n    const [nSamples, nCols] = getShape2D(dense);\n    const categories = this.categories_;\n    if (!categories) {\n      throw new DeepboxError(\"OneHotEncoder internal error: missing fitted categories\");\n    }\n    const nFeatures = categories.length;\n    const dropIndices = this.dropIndices_ ?? categories.map(() => null);\n    const totalCols = categories.reduce((sum, cats, idx) => {\n      const dropIndex = dropIndices[idx] ?? null;\n      return sum + cats.length - (dropIndex === null ? 0 : 1);\n    }, 0);\n    if (nCols !== totalCols) {\n      throw new InvalidParameterError(\"column count does not match fitted categories\", \"X\", nCols);\n    }\n    if (nSamples === 0) {\n      return emptyCategoryMatrixFromCategories(categories, nFeatures, \"X\");\n    }\n\n    const result = new Array<Category[]>(nSamples);\n    for (let i = 0; i < nSamples; i++) {\n      result[i] = new Array<Category>(nFeatures);\n    }\n    const denseData = getNumericData(dense);\n    const [stride0, stride1] = getStrides2D(dense);\n\n    for (let i = 0; i < nSamples; i++) {\n      let colOffset = 0;\n      for (let j = 0; j < nFeatures; j++) {\n        const cats = categories[j];\n        const dropIndex = dropIndices[j] ?? null;\n        if (!cats) {\n          throw new DeepboxError(\"OneHotEncoder internal error: missing fitted categories\");\n        }\n        const outSize = cats.length - (dropIndex === null ? 0 : 1);\n        if (outSize === 0) {\n          const row = result[i];\n          if (!row) {\n            throw new DeepboxError(\"Internal error: result row access failed\");\n          }\n          row[j] = categoryValueAt(cats, dropIndex ?? 0, \"OneHotEncoder.inverseTransform\");\n          continue;\n        }\n\n        let maxIdx = 0;\n        const rowBase = dense.offset + i * stride0 + colOffset * stride1;\n        const first = denseData[rowBase];\n        if (first === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        let maxVal = Number(first);\n        let hasPositive = maxVal > 0;\n\n        for (let k = 1; k < outSize; k++) {\n          const raw = denseData[rowBase + k * stride1];\n          if (raw === undefined) {\n            throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n          }\n          const val = Number(raw);\n          if (val > maxVal) {\n            maxVal = val;\n            maxIdx = k;\n          }\n          if (val > 0) {\n            hasPositive = true;\n          }\n        }\n\n        const row = result[i];\n        if (row === undefined) {\n          throw new DeepboxError(\"Internal error: result row access failed\");\n        }\n        if (!hasPositive) {\n          if (dropIndex !== null) {\n            row[j] = categoryValueAt(cats, dropIndex, \"OneHotEncoder.inverseTransform\");\n          } else if (this.handleUnknown === \"ignore\") {\n            throw new InvalidParameterError(\n              \"Cannot inverse-transform: sample contains no active category (all zeros). This may happen if unknown categories were ignored during transform.\",\n              \"X\"\n            );\n          } else {\n            throw new InvalidParameterError(\"Invalid one-hot encoding: all zeros\", \"X\");\n          }\n        } else {\n          const actualIdx = dropIndex !== null && maxIdx >= dropIndex ? maxIdx + 1 : maxIdx;\n          row[j] = categoryValueAt(cats, actualIdx, \"OneHotEncoder.inverseTransform\");\n        }\n\n        colOffset += outSize;\n      }\n    }\n\n    return toCategoryMatrixTensor(result, \"X\");\n  }\n}\n\n/**\n * Encode categorical features as integer array.\n *\n * This encoder transforms categorical features into ordinal integers.\n * Each feature's categories are mapped to integers [0, n_categories-1]\n * based on their sorted order. Unlike OneHotEncoder, this maintains\n * a single column per feature.\n *\n * **Time Complexity:**\n * - fit: O(n*m*log(k)) where n=samples, m=features, k=avg categories\n * - transform: O(n*m*log(k)) due to indexOf lookup\n *\n * **Space Complexity:** O(m*k) where m=features, k=avg categories per feature\n *\n * @example\n * ```js\n * const X = tensor([['low', 'red'], ['high', 'blue'], ['medium', 'red']]);\n * const encoder = new OrdinalEncoder();\n * encoder.fit(X);\n * const encoded = encoder.transform(X);\n * // Result: [[1, 1], [0, 0], [2, 1]] (alphabetically sorted)\n * ```\n *\n * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OrdinalEncoder.html | Scikit-learn OrdinalEncoder}\n */\nexport class OrdinalEncoder {\n  /** Indicates whether the encoder has been fitted to data */\n  private fitted = false;\n  /** Array of unique categories for each feature, sorted */\n  private categories_?: Category[][];\n  /** Maps from category value to index for each feature (for O(1) lookup) */\n  private categoryToIndex_?: Array<Map<Category, number>>;\n  /** How to handle unknown categories during transform */\n  private handleUnknown: \"error\" | \"useEncodedValue\";\n  /** Value used for unknown categories when handleUnknown = \"useEncodedValue\" */\n  private unknownValue: number;\n  /** Categories configuration */\n  private categoriesOption: CategoriesOption;\n\n  /**\n   * Creates a new OrdinalEncoder instance.\n   *\n   * @param options - Configuration options\n   * @param options.handleUnknown - How to handle unknown categories\n   * @param options.unknownValue - Encoded value for unknown categories when handleUnknown=\"useEncodedValue\"\n   * @param options.categories - \"auto\" or explicit categories per feature\n   */\n  constructor(\n    options: {\n      handleUnknown?: \"error\" | \"useEncodedValue\";\n      unknownValue?: number;\n      categories?: CategoriesOption;\n    } = {}\n  ) {\n    this.handleUnknown = options.handleUnknown ?? \"error\";\n    this.unknownValue = options.unknownValue ?? -1;\n    this.categoriesOption = options.categories ?? \"auto\";\n\n    if (this.handleUnknown !== \"error\" && this.handleUnknown !== \"useEncodedValue\") {\n      throw new InvalidParameterError(\n        \"handleUnknown must be 'error' or 'useEncodedValue'\",\n        \"handleUnknown\",\n        this.handleUnknown\n      );\n    }\n    if (!Number.isFinite(this.unknownValue) && !Number.isNaN(this.unknownValue)) {\n      throw new InvalidParameterError(\n        \"unknownValue must be a finite number or NaN\",\n        \"unknownValue\",\n        this.unknownValue\n      );\n    }\n    if (Number.isFinite(this.unknownValue) && !Number.isInteger(this.unknownValue)) {\n      throw new InvalidParameterError(\n        \"unknownValue must be an integer when finite\",\n        \"unknownValue\",\n        this.unknownValue\n      );\n    }\n  }\n\n  /**\n   * Fit OrdinalEncoder to X.\n   * Learns the unique categories for each feature and their ordering.\n   *\n   * @param X - Training data (2D tensor of categorical features)\n   * @returns this - Returns self for method chaining\n   * @throws {InvalidParameterError} If X is empty\n   */\n  fit(X: Tensor): this {\n    assert2D(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n\n    if (nSamples === 0) {\n      throw new InvalidParameterError(\"Cannot fit OrdinalEncoder on empty array\", \"X\");\n    }\n\n    this.categories_ = [];\n    this.categoryToIndex_ = [];\n\n    const explicitCategories = resolveCategoriesOption(\n      this.categoriesOption,\n      nFeatures,\n      \"categories\"\n    );\n\n    // For each feature, collect and sort unique categories\n    for (let j = 0; j < nFeatures; j++) {\n      let sorted: Category[];\n\n      if (explicitCategories) {\n        const featureCats = explicitCategories[j];\n        if (!featureCats) {\n          throw new InvalidParameterError(\"Missing categories for feature\", \"categories\", j);\n        }\n        if (!Array.isArray(featureCats)) {\n          throw new InvalidParameterError(\n            \"categories must be an array of category arrays\",\n            \"categories\",\n            featureCats\n          );\n        }\n        sorted = validateCategoryValues(featureCats, \"categories\");\n      } else {\n        const uniqueSet = new Set<Category>();\n\n        // Collect all unique values in this feature\n        for (let i = 0; i < nSamples; i++) {\n          uniqueSet.add(read2DValue(X, i, j));\n        }\n\n        // Sort categories for consistent ordering\n        sorted = sortCategories(uniqueSet, \"X\");\n      }\n\n      if (sorted.length === 0) {\n        throw new InvalidParameterError(\"Each feature must have at least one category\", \"X\", j);\n      }\n\n      this.categories_.push(sorted);\n\n      // Build index map for O(1) transform lookups\n      const map = new Map<Category, number>();\n      for (let k = 0; k < sorted.length; k++) {\n        map.set(categoryValueAt(sorted, k, \"OrdinalEncoder.fit\"), k);\n      }\n      this.categoryToIndex_.push(map);\n\n      if (explicitCategories) {\n        for (let i = 0; i < nSamples; i++) {\n          const val = read2DValue(X, i, j);\n          if (!map.has(val)) {\n            throw new InvalidParameterError(\n              `Unknown category: ${String(val)} in feature ${j}`,\n              \"X\",\n              val\n            );\n          }\n        }\n      }\n\n      if (this.handleUnknown === \"useEncodedValue\") {\n        if (\n          Number.isFinite(this.unknownValue) &&\n          this.unknownValue >= 0 &&\n          this.unknownValue < sorted.length\n        ) {\n          throw new InvalidParameterError(\n            \"unknownValue must be outside the range of encoded categories\",\n            \"unknownValue\",\n            this.unknownValue\n          );\n        }\n      }\n    }\n\n    this.fitted = true;\n    return this;\n  }\n\n  /**\n   * Transform X using ordinal encoding.\n   * Each category is mapped to its index in the sorted categories array.\n   *\n   * @param X - Data to transform (2D tensor)\n   * @returns Encoded data with integer values\n   * @throws {NotFittedError} If encoder is not fitted\n   * @throws {InvalidParameterError} If X contains unknown categories\n   */\n  transform(X: Tensor): Tensor {\n    if (!this.fitted) {\n      throw new NotFittedError(\"OrdinalEncoder must be fitted before transform\");\n    }\n    assert2D(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const fittedFeatures = this.categories_?.length ?? 0;\n    if (nFeatures !== fittedFeatures) {\n      throw new InvalidParameterError(\n        \"X has a different feature count than during fit\",\n        \"X\",\n        nFeatures\n      );\n    }\n\n    if (nSamples === 0) {\n      return zeros([0, nFeatures], { dtype: \"float64\" });\n    }\n\n    // Pre-allocate result array\n    const result = new Array<number[]>(nSamples);\n    for (let i = 0; i < nSamples; i++) {\n      result[i] = new Array<number>(nFeatures);\n    }\n\n    // Transform each value to its ordinal index using O(1) map lookup\n    for (let i = 0; i < nSamples; i++) {\n      for (let j = 0; j < nFeatures; j++) {\n        const val = read2DValue(X, i, j);\n        const map = this.categoryToIndex_?.[j];\n        if (!map) {\n          throw new DeepboxError(\"OrdinalEncoder internal error: missing fitted categories\");\n        }\n\n        // Use O(1) map lookup instead of O(n) indexOf\n        const idx = map.get(val);\n        const row = result[i];\n        if (!row) {\n          throw new DeepboxError(\"Internal error: result row access failed\");\n        }\n        if (idx === undefined) {\n          if (this.handleUnknown === \"useEncodedValue\") {\n            row[j] = this.unknownValue;\n            continue;\n          }\n          throw new InvalidParameterError(\n            `Unknown category: ${String(val)} in feature ${j}`,\n            \"X\",\n            val\n          );\n        }\n\n        row[j] = idx;\n      }\n    }\n\n    return tensor(result, { dtype: \"float64\" });\n  }\n\n  /**\n   * Fit encoder and transform X in one step.\n   * Convenience method equivalent to calling fit(X).transform(X).\n   *\n   * @param X - Training data (2D tensor)\n   * @returns Encoded data\n   */\n  fitTransform(X: Tensor): Tensor {\n    return this.fit(X).transform(X);\n  }\n\n  /**\n   * Transform ordinal integers back to original categories.\n   * Reverses the encoding performed by transform().\n   *\n   * @param X - Encoded data (2D integer tensor)\n   * @returns Original categorical data\n   * @throws {NotFittedError} If encoder is not fitted\n   * @throws {InvalidParameterError} If X contains invalid indices\n   */\n  inverseTransform(X: Tensor): Tensor {\n    if (!this.fitted) {\n      throw new NotFittedError(\"OrdinalEncoder must be fitted before inverse_transform\");\n    }\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const fittedFeatures = this.categories_?.length ?? 0;\n    if (nFeatures !== fittedFeatures) {\n      throw new InvalidParameterError(\n        \"X has a different feature count than during fit\",\n        \"X\",\n        nFeatures\n      );\n    }\n\n    if (nSamples === 0 || nFeatures === 0) {\n      const categoryRows = this.categories_ ?? [];\n      const categoryType = inferCategoryTypeFromRows(categoryRows, \"X\");\n      if (categoryType === \"string\") {\n        return empty([0, nFeatures], { dtype: \"string\" });\n      }\n      if (categoryType === \"bigint\") {\n        return empty([0, nFeatures], { dtype: \"int64\" });\n      }\n      return zeros([0, nFeatures], { dtype: \"float64\" });\n    }\n\n    // Pre-allocate result array\n    const result = new Array<Category[]>(nSamples);\n    for (let i = 0; i < nSamples; i++) {\n      result[i] = new Array<Category>(nFeatures);\n    }\n\n    // Map each ordinal index back to its original category\n    const [stride0, stride1] = getStrides2D(X);\n    const data = getNumericData(X);\n    for (let i = 0; i < nSamples; i++) {\n      for (let j = 0; j < nFeatures; j++) {\n        const raw = data[X.offset + i * stride0 + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        const idx = Number(raw);\n        const isUnknownValue =\n          this.handleUnknown === \"useEncodedValue\" &&\n          (Number.isNaN(idx) ? Number.isNaN(this.unknownValue) : idx === this.unknownValue);\n        if (isUnknownValue) {\n          throw new InvalidParameterError(\n            \"Cannot inverse-transform unknown encoded value\",\n            \"X\",\n            idx\n          );\n        }\n        const cats = this.categories_?.[j];\n\n        // Validate index is in valid range\n        if (!cats || idx < 0 || idx >= cats.length || !Number.isInteger(idx)) {\n          throw new InvalidParameterError(\n            `Invalid encoded value: ${idx} for feature ${j}. Must be integer in [0, ${(cats?.length ?? 0) - 1}]`,\n            \"X\",\n            idx\n          );\n        }\n\n        const row = result[i];\n        if (!row) {\n          throw new DeepboxError(\"Internal error: result row access failed\");\n        }\n        const catVal = cats[idx];\n        if (catVal === undefined) {\n          throw new DeepboxError(\"Internal error: category value missing\");\n        }\n        row[j] = catVal;\n      }\n    }\n\n    return toCategoryMatrixTensor(result, \"X\");\n  }\n}\n\n/**\n * Binarize labels in a one-vs-all fashion.\n *\n * This transformer creates a binary matrix representation of labels where\n * each class gets its own column. For multi-class problems, this creates\n * a one-hot encoding of the labels.\n *\n * **Time Complexity:**\n * - fit: O(n) where n is the number of samples\n * - transform: O(n*k) where k is the number of classes\n *\n * **Space Complexity:** O(n*k) for the output matrix\n *\n * @example\n * ```js\n * const y = tensor([0, 1, 2, 0, 1]);\n * const binarizer = new LabelBinarizer();\n * const yBin = binarizer.fitTransform(y);\n * // Result shape: [5, 3] with one-hot encoding\n * ```\n *\n * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.LabelBinarizer.html | Scikit-learn LabelBinarizer}\n */\nexport class LabelBinarizer {\n  /** Indicates whether the binarizer has been fitted to data */\n  private fitted = false;\n  /** Array of unique classes found during fitting, sorted */\n  private classes_?: Category[];\n  /** Map from class value to index for O(1) lookups */\n  private classToIndex_?: Map<Category, number>;\n  /** Value used for positive class */\n  private posLabel: number;\n  /** Value used for negative class */\n  private negLabel: number;\n  /** Whether to return sparse matrix output */\n  private sparse: boolean;\n\n  /**\n   * Creates a new LabelBinarizer instance.\n   *\n   * @param options - Configuration options\n   * @param options.posLabel - Value for positive class (default: 1)\n   * @param options.negLabel - Value for negative class (default: 0)\n   * @param options.sparse - If true, returns CSRMatrix (default: false)\n   * @param options.sparseOutput - Alias for sparse (default: false)\n   */\n  constructor(\n    options: {\n      posLabel?: number;\n      negLabel?: number;\n      sparse?: boolean;\n      sparseOutput?: boolean;\n    } = {}\n  ) {\n    this.posLabel = options.posLabel ?? 1;\n    this.negLabel = options.negLabel ?? 0;\n    const sparseOption = options.sparse ?? options.sparseOutput ?? false;\n    if (!Number.isFinite(this.posLabel) || !Number.isFinite(this.negLabel)) {\n      throw new InvalidParameterError(\"posLabel and negLabel must be finite numbers\", \"posLabel\");\n    }\n    if (this.posLabel <= this.negLabel) {\n      throw new InvalidParameterError(\n        \"posLabel must be greater than negLabel\",\n        \"posLabel\",\n        this.posLabel\n      );\n    }\n    if (options.sparse !== undefined && options.sparseOutput !== undefined) {\n      if (options.sparse !== options.sparseOutput) {\n        throw new InvalidParameterError(\n          \"sparse and sparseOutput must match when both are provided\",\n          \"sparse\",\n          options.sparse\n        );\n      }\n    }\n    if (typeof sparseOption !== \"boolean\") {\n      throw new InvalidParameterError(\"sparse must be a boolean\", \"sparse\", sparseOption);\n    }\n    if (sparseOption && this.negLabel !== 0) {\n      throw new InvalidParameterError(\n        \"sparse output requires negLabel to be 0\",\n        \"negLabel\",\n        this.negLabel\n      );\n    }\n    this.sparse = sparseOption;\n  }\n\n  /**\n   * Fit label binarizer to a set of labels.\n   * Learns the unique classes present in the data.\n   *\n   * @param y - Target labels (1D tensor)\n   * @returns this - Returns self for method chaining\n   * @throws {InvalidParameterError} If y is empty\n   */\n  fit(y: Tensor): this {\n    assert1D(y, \"y\");\n    if (y.size === 0) {\n      throw new InvalidParameterError(\"Cannot fit LabelBinarizer on empty array\", \"y\");\n    }\n\n    // Collect unique classes\n    const uniqueSet = new Set<Category>();\n    for (let i = 0; i < y.size; i++) {\n      uniqueSet.add(read1DValue(y, i));\n    }\n\n    // Sort classes for consistent ordering\n    this.classes_ = sortCategories(uniqueSet, \"y\");\n    this.classToIndex_ = new Map();\n    for (let i = 0; i < this.classes_.length; i++) {\n      this.classToIndex_.set(categoryValueAt(this.classes_, i, \"LabelBinarizer.fit\"), i);\n    }\n\n    this.fitted = true;\n    return this;\n  }\n\n  /**\n   * Transform labels to binary matrix.\n   * Each label is converted to a binary vector with a single 1.\n   *\n   * @param y - Labels to transform (1D tensor)\n   * @returns Binary matrix (Tensor or CSRMatrix) with shape [n_samples, n_classes]\n   * @throws {NotFittedError} If binarizer is not fitted\n   * @throws {InvalidParameterError} If y contains unknown labels\n   */\n  transform(y: Tensor): Tensor | CSRMatrix {\n    if (!this.fitted) {\n      throw new NotFittedError(\"LabelBinarizer must be fitted before transform\");\n    }\n    assert1D(y, \"y\");\n    if (y.size === 0) {\n      const nClasses = this.classes_?.length ?? 0;\n      return this.sparse\n        ? CSRMatrix.fromCOO({\n            rows: 0,\n            cols: nClasses,\n            rowIndices: new Int32Array(0),\n            colIndices: new Int32Array(0),\n            values: new Float64Array(0),\n          })\n        : zeros([0, nClasses], { dtype: \"float64\" });\n    }\n\n    const nSamples = y.size;\n    const nClasses = this.classes_?.length ?? 0;\n    const lookup = this.classToIndex_;\n    if (!lookup) {\n      throw new DeepboxError(\"LabelBinarizer internal error: missing fitted lookup\");\n    }\n\n    if (this.sparse) {\n      const rowIdx: number[] = [];\n      const colIdx: number[] = [];\n      const vals: number[] = [];\n\n      for (let i = 0; i < nSamples; i++) {\n        const val = read1DValue(y, i);\n        const idx = lookup.get(val);\n\n        if (idx === undefined) {\n          throw new InvalidParameterError(\n            `Unknown label: ${String(val)}. Label must be present during fit.`,\n            \"y\",\n            val\n          );\n        }\n\n        rowIdx.push(i);\n        colIdx.push(idx);\n        vals.push(this.posLabel);\n      }\n\n      return CSRMatrix.fromCOO({\n        rows: nSamples,\n        cols: nClasses,\n        rowIndices: Int32Array.from(rowIdx),\n        colIndices: Int32Array.from(colIdx),\n        values: Float64Array.from(vals),\n      });\n    }\n\n    // Pre-allocate binary matrix\n    const result = new Array<number[]>(nSamples);\n    for (let i = 0; i < nSamples; i++) {\n      result[i] = new Array<number>(nClasses).fill(this.negLabel);\n    }\n\n    // Set appropriate bit for each label\n    for (let i = 0; i < nSamples; i++) {\n      const val = read1DValue(y, i);\n      const idx = lookup.get(val);\n\n      if (idx === undefined) {\n        throw new InvalidParameterError(\n          `Unknown label: ${String(val)}. Label must be present during fit.`,\n          \"y\",\n          val\n        );\n      }\n\n      const row = result[i];\n      if (!row) {\n        throw new DeepboxError(\"Internal error: result row access failed\");\n      }\n      row[idx] = this.posLabel;\n    }\n\n    return tensor(result, { dtype: \"float64\" });\n  }\n\n  /**\n   * Fit binarizer and transform labels in one step.\n   * Convenience method equivalent to calling fit(y).transform(y).\n   *\n   * @param y - Target labels (1D tensor)\n   * @returns Binary matrix (Tensor or CSRMatrix)\n   */\n  fitTransform(y: Tensor): Tensor | CSRMatrix {\n    return this.fit(y).transform(y);\n  }\n\n  /**\n   * Transform binary matrix back to labels.\n   * Finds the column with maximum value for each row.\n   *\n   * @param Y - Binary matrix (2D tensor or CSRMatrix)\n   * @returns Original labels (1D tensor)\n   * @throws {NotFittedError} If binarizer is not fitted\n   * @throws {InvalidParameterError} If Y has invalid shape\n   */\n  inverseTransform(Y: Tensor | CSRMatrix): Tensor {\n    if (!this.fitted) {\n      throw new NotFittedError(\"LabelBinarizer must be fitted before inverse_transform\");\n    }\n    if (Y instanceof CSRMatrix) {\n      if (this.negLabel !== 0) {\n        throw new InvalidParameterError(\n          \"Sparse inverse transform requires negLabel to be 0\",\n          \"negLabel\",\n          this.negLabel\n        );\n      }\n      const [rows, cols] = Y.shape;\n      if (rows === undefined || cols === undefined) {\n        throw new ShapeError(\"Y must have valid shape\");\n      }\n      const nClasses = this.classes_?.length ?? 0;\n      if (cols !== nClasses) {\n        throw new InvalidParameterError(\"column count does not match number of classes\", \"Y\", cols);\n      }\n      const classes = this.classes_;\n      if (!classes) {\n        throw new DeepboxError(\"LabelBinarizer internal error: missing fitted classes\");\n      }\n      if (rows === 0) {\n        return emptyCategoryVectorFromClasses(classes, \"y\");\n      }\n\n      const result = new Array<Category>(rows);\n      for (let i = 0; i < rows; i++) {\n        let maxIdx = 0;\n        let maxVal = this.negLabel;\n        const start = Y.indptr[i] ?? 0;\n        const end = Y.indptr[i + 1] ?? start;\n        for (let p = start; p < end; p++) {\n          const col = Y.indices[p];\n          if (col === undefined) {\n            throw new DeepboxError(\"Internal error: sparse column index missing\");\n          }\n          if (col < 0 || col >= nClasses) {\n            throw new InvalidParameterError(\n              \"column index out of bounds for fitted classes\",\n              \"Y\",\n              col\n            );\n          }\n          const raw = Y.data[p];\n          if (raw === undefined) {\n            throw new DeepboxError(\"Internal error: sparse value missing\");\n          }\n          const val = Number(raw);\n          if (val > maxVal) {\n            maxVal = val;\n            maxIdx = col;\n          }\n        }\n\n        if (maxVal <= this.negLabel) {\n          throw new InvalidParameterError(\n            `No active label found for sample ${i}. LabelBinarizer expects exactly one active label.`,\n            \"Y\"\n          );\n        }\n        result[i] = categoryValueAt(classes, maxIdx, \"LabelBinarizer.inverseTransform\");\n      }\n      return toCategoryVectorTensor(result, \"y\");\n    }\n\n    assert2D(Y, \"Y\");\n    assertNumericTensor(Y, \"Y\");\n    const [nSamples, nCols] = getShape2D(Y);\n\n    const nClasses = this.classes_?.length ?? 0;\n    if (nCols !== nClasses) {\n      throw new InvalidParameterError(\"column count does not match number of classes\", \"Y\", nCols);\n    }\n    const classes = this.classes_;\n    if (!classes) {\n      throw new DeepboxError(\"LabelBinarizer internal error: missing fitted classes\");\n    }\n    if (nSamples === 0) {\n      return emptyCategoryVectorFromClasses(classes, \"y\");\n    }\n    const result = new Array<Category>(nSamples);\n    const [stride0, stride1] = getStrides2D(Y);\n    const data = getNumericData(Y);\n\n    // For each sample, find the class with maximum activation\n    for (let i = 0; i < nSamples; i++) {\n      let maxIdx = 0;\n      const rowBase = Y.offset + i * stride0;\n      const first = data[rowBase];\n      if (first === undefined) {\n        throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n      }\n      let maxVal = Number(first);\n\n      // Find column with highest value\n      for (let j = 1; j < nCols; j++) {\n        const raw = data[rowBase + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        const val = Number(raw);\n        if (val > maxVal) {\n          maxVal = val;\n          maxIdx = j;\n        }\n      }\n\n      if (maxVal <= this.negLabel) {\n        throw new InvalidParameterError(\n          `No active label found for sample ${i}. LabelBinarizer expects exactly one active label.`,\n          \"Y\"\n        );\n      }\n\n      result[i] = categoryValueAt(classes, maxIdx, \"LabelBinarizer.inverseTransform\");\n    }\n\n    // Return tensor with appropriate dtype\n    return toCategoryVectorTensor(result, \"y\");\n  }\n}\n\n/**\n * Transform multi-label classification data to binary format.\n *\n * This transformer handles multi-label classification where each sample\n * can belong to multiple classes simultaneously. It creates a binary\n * matrix where each column represents a class and multiple columns can\n * be active (set to 1) for a single sample.\n *\n * **Time Complexity:**\n * - fit: O(n*k) where n is samples, k is avg labels per sample\n * - transform: O(n*k*c) where c is total unique classes\n *\n * **Space Complexity:** O(n*c) for the output matrix\n *\n * @example\n * ```js\n * const y = [['sci-fi', 'action'], ['comedy'], ['action', 'drama']];\n * const binarizer = new MultiLabelBinarizer();\n * const yBin = binarizer.fitTransform(y);\n * // Each row can have multiple 1s\n * ```\n *\n * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.MultiLabelBinarizer.html | Scikit-learn MultiLabelBinarizer}\n */\nexport class MultiLabelBinarizer {\n  /** Indicates whether the binarizer has been fitted to data */\n  private fitted = false;\n  /** Array of all unique classes found across all samples, sorted */\n  private classes_?: Category[];\n  /** Map from class value to index for O(1) lookups */\n  private classToIndex_?: Map<Category, number>;\n  /** Whether to return sparse matrix (CSR) or dense array */\n  private sparse: boolean;\n  /** Optional explicit class ordering */\n  private classesOption?: Category[];\n\n  /**\n   * Creates a new MultiLabelBinarizer instance.\n   *\n   * @param options - Configuration options\n   * @param options.sparse - If true, returns CSRMatrix; if false, returns dense Tensor (default: false)\n   * @param options.sparseOutput - Alias for sparse (default: false)\n   * @param options.classes - Explicit class ordering to use instead of sorting\n   */\n  constructor(\n    options: {\n      sparse?: boolean;\n      sparseOutput?: boolean;\n      classes?: ReadonlyArray<Category>;\n    } = {}\n  ) {\n    const sparseOption = options.sparse ?? options.sparseOutput ?? false;\n    if (options.sparse !== undefined && options.sparseOutput !== undefined) {\n      if (options.sparse !== options.sparseOutput) {\n        throw new InvalidParameterError(\n          \"sparse and sparseOutput must match when both are provided\",\n          \"sparse\",\n          options.sparse\n        );\n      }\n    }\n    this.sparse = sparseOption;\n    if (typeof this.sparse !== \"boolean\") {\n      throw new InvalidParameterError(\"sparse must be a boolean\", \"sparse\", this.sparse);\n    }\n    if (options.classes !== undefined) {\n      this.classesOption = validateCategoryValues(options.classes, \"classes\");\n    }\n  }\n\n  /**\n   * Fit multi-label binarizer to label sets.\n   * Learns all unique classes present across all samples.\n   *\n   * @param y - Array of label sets, where each element is an array of string/number/bigint labels\n   * @returns this - Returns self for method chaining\n   * @throws {InvalidParameterError} If y is empty\n   */\n  fit(y: ReadonlyArray<ReadonlyArray<Category>>): this {\n    if (y.length === 0) {\n      throw new InvalidParameterError(\"Cannot fit MultiLabelBinarizer on empty array\", \"y\");\n    }\n    for (const labels of y) {\n      if (!Array.isArray(labels)) {\n        throw new InvalidParameterError(\"MultiLabelBinarizer expects label arrays\", \"y\", labels);\n      }\n      for (const label of labels) {\n        if (typeof label !== \"string\" && typeof label !== \"number\" && typeof label !== \"bigint\") {\n          throw new InvalidParameterError(\n            \"MultiLabelBinarizer labels must be strings, numbers, or bigints\",\n            \"y\",\n            label\n          );\n        }\n      }\n    }\n\n    if (this.classesOption && this.classesOption.length === 0) {\n      throw new InvalidParameterError(\"classes must contain at least one value\", \"classes\");\n    }\n\n    if (this.classesOption) {\n      this.classes_ = Array.from(this.classesOption);\n    } else {\n      // Collect all unique labels across all samples\n      const uniqueSet = new Set<Category>();\n      for (const labels of y) {\n        for (const label of labels) {\n          uniqueSet.add(label);\n        }\n      }\n\n      // Sort classes for consistent ordering\n      this.classes_ = sortCategories(uniqueSet, \"y\");\n    }\n    this.classToIndex_ = new Map();\n    for (let i = 0; i < this.classes_.length; i++) {\n      this.classToIndex_.set(categoryValueAt(this.classes_, i, \"MultiLabelBinarizer.fit\"), i);\n    }\n    if (this.classesOption) {\n      for (const labels of y) {\n        for (const label of labels) {\n          if (!this.classToIndex_.has(label)) {\n            throw new InvalidParameterError(\n              `Unknown label: ${String(label)}. Label must be present in classes.`,\n              \"y\",\n              label\n            );\n          }\n        }\n      }\n    }\n\n    this.fitted = true;\n    return this;\n  }\n\n  /**\n   * Transform label sets to binary matrix.\n   * Each sample can have multiple active (1) columns.\n   *\n   * @param y - Array of label sets to transform (string/number/bigint labels)\n   * @returns Binary matrix (Tensor or CSRMatrix) with shape [n_samples, n_classes]\n   * @throws {NotFittedError} If binarizer is not fitted\n   * @throws {InvalidParameterError} If y contains unknown labels\n   */\n  transform(y: ReadonlyArray<ReadonlyArray<Category>>): Tensor | CSRMatrix {\n    if (!this.fitted) {\n      throw new NotFittedError(\"MultiLabelBinarizer must be fitted before transform\");\n    }\n    for (const labels of y) {\n      if (!Array.isArray(labels)) {\n        throw new InvalidParameterError(\"MultiLabelBinarizer expects label arrays\", \"y\", labels);\n      }\n      for (const label of labels) {\n        if (typeof label !== \"string\" && typeof label !== \"number\" && typeof label !== \"bigint\") {\n          throw new InvalidParameterError(\n            \"MultiLabelBinarizer labels must be strings, numbers, or bigints\",\n            \"y\",\n            label\n          );\n        }\n      }\n    }\n    if (y.length === 0) {\n      const nClasses = this.classes_?.length ?? 0;\n      return this.sparse\n        ? CSRMatrix.fromCOO({\n            rows: 0,\n            cols: nClasses,\n            rowIndices: new Int32Array(0),\n            colIndices: new Int32Array(0),\n            values: new Float64Array(0),\n          })\n        : zeros([0, nClasses], { dtype: \"float64\" });\n    }\n\n    const nSamples = y.length;\n    const nClasses = this.classes_?.length ?? 0;\n    const lookup = this.classToIndex_;\n    if (!lookup) {\n      throw new DeepboxError(\"MultiLabelBinarizer internal error: missing fitted lookup\");\n    }\n\n    if (this.sparse) {\n      const rowIdx: number[] = [];\n      const colIdx: number[] = [];\n      const vals: number[] = [];\n\n      for (let i = 0; i < nSamples; i++) {\n        const yRow = y[i];\n        if (!yRow) continue;\n        const seen = new Set<number>();\n        for (const label of yRow) {\n          const idx = lookup.get(label);\n          if (idx === undefined) {\n            throw new InvalidParameterError(\n              `Unknown label: ${String(label)}. Label must be present during fit.`,\n              \"y\",\n              label\n            );\n          }\n          if (seen.has(idx)) continue;\n          seen.add(idx);\n          rowIdx.push(i);\n          colIdx.push(idx);\n          vals.push(1);\n        }\n      }\n\n      return CSRMatrix.fromCOO({\n        rows: nSamples,\n        cols: nClasses,\n        rowIndices: Int32Array.from(rowIdx),\n        colIndices: Int32Array.from(colIdx),\n        values: Float64Array.from(vals),\n      });\n    }\n\n    // Pre-allocate binary matrix\n    const result = new Array<number[]>(nSamples);\n    for (let i = 0; i < nSamples; i++) {\n      result[i] = new Array<number>(nClasses).fill(0);\n    }\n\n    // Set bits for all labels in each sample\n    for (let i = 0; i < nSamples; i++) {\n      const yRow = y[i];\n      if (!yRow) continue;\n\n      for (const label of yRow) {\n        const idx = lookup.get(label);\n        if (idx === undefined) {\n          throw new InvalidParameterError(\n            `Unknown label: ${String(label)}. Label must be present during fit.`,\n            \"y\",\n            label\n          );\n        }\n\n        const row = result[i];\n        if (!row) {\n          throw new DeepboxError(\"Internal error: result row access failed\");\n        }\n        row[idx] = 1;\n      }\n    }\n\n    return tensor(result, { dtype: \"float64\" });\n  }\n\n  /**\n   * Fit binarizer and transform label sets in one step.\n   * Convenience method equivalent to calling fit(y).transform(y).\n   *\n   * @param y - Array of label sets (string/number/bigint labels)\n   * @returns Binary matrix (Tensor or CSRMatrix)\n   */\n  fitTransform(y: ReadonlyArray<ReadonlyArray<Category>>): Tensor | CSRMatrix {\n    return this.fit(y).transform(y);\n  }\n\n  /**\n   * Transform binary matrix back to label sets.\n   * Finds all active (1) columns for each row.\n   *\n   * @param Y - Binary matrix (Tensor or CSRMatrix)\n   * @returns Array of label sets, one per sample\n   * @throws {NotFittedError} If binarizer is not fitted\n   * @throws {InvalidParameterError} If Y has invalid shape\n   */\n  inverseTransform(Y: Tensor | CSRMatrix): Category[][] {\n    if (!this.fitted) {\n      throw new NotFittedError(\"MultiLabelBinarizer must be fitted before inverse_transform\");\n    }\n    if (Y instanceof CSRMatrix) {\n      const [rows, cols] = Y.shape;\n      if (rows === undefined || cols === undefined) {\n        throw new ShapeError(\"Y must have valid shape\");\n      }\n      const fittedClasses = this.classes_?.length ?? 0;\n      if (cols !== fittedClasses) {\n        throw new InvalidParameterError(\"column count does not match number of classes\", \"Y\", cols);\n      }\n      if (rows === 0) {\n        return [];\n      }\n\n      const classes = this.classes_;\n      if (!classes) {\n        throw new DeepboxError(\"MultiLabelBinarizer internal error: missing fitted classes\");\n      }\n\n      const result: Category[][] = [];\n      for (let i = 0; i < rows; i++) {\n        const labels: Category[] = [];\n        const start = Y.indptr[i] ?? 0;\n        const end = Y.indptr[i + 1] ?? start;\n        for (let p = start; p < end; p++) {\n          const col = Y.indices[p];\n          if (col === undefined) {\n            throw new DeepboxError(\"Internal error: sparse column index missing\");\n          }\n          if (col < 0 || col >= fittedClasses) {\n            throw new InvalidParameterError(\n              \"column index out of bounds for fitted classes\",\n              \"Y\",\n              col\n            );\n          }\n          const raw = Y.data[p];\n          if (raw === undefined) {\n            throw new DeepboxError(\"Internal error: sparse value missing\");\n          }\n          const value = Number(raw);\n          if (value > 0) {\n            labels.push(categoryValueAt(classes, col, \"MultiLabelBinarizer.inverseTransform\"));\n          }\n        }\n        result.push(labels);\n      }\n      return result;\n    }\n\n    assert2D(Y, \"Y\");\n    assertNumericTensor(Y, \"Y\");\n    const nSamples = Y.shape[0] ?? 0;\n    const nClasses = Y.shape[1] ?? 0;\n    const fittedClasses = this.classes_?.length ?? 0;\n    if (nClasses !== fittedClasses) {\n      throw new InvalidParameterError(\n        \"column count does not match number of classes\",\n        \"Y\",\n        nClasses\n      );\n    }\n\n    if (nSamples === 0) {\n      return [];\n    }\n\n    const classes = this.classes_;\n    if (!classes) {\n      throw new DeepboxError(\"MultiLabelBinarizer internal error: missing fitted classes\");\n    }\n\n    const result: Category[][] = [];\n    const [stride0, stride1] = getStrides2D(Y);\n    const data = getNumericData(Y);\n\n    // For each sample, collect all active classes\n    for (let i = 0; i < nSamples; i++) {\n      const labels: Category[] = [];\n      const rowBase = Y.offset + i * stride0;\n\n      // Check each class column\n      for (let j = 0; j < nClasses; j++) {\n        const raw = data[rowBase + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        const val = Number(raw);\n        // If this class is active (typically 1, but allow any positive value)\n        if (val > 0) {\n          labels.push(categoryValueAt(classes, j, \"MultiLabelBinarizer.inverseTransform\"));\n        }\n      }\n\n      result.push(labels);\n    }\n\n    return result;\n  }\n}\n","import {\n  DataValidationError,\n  DeepboxError,\n  DTypeError,\n  InvalidParameterError,\n  NotFittedError,\n} from \"../core/errors\";\nimport { type Tensor, tensor, zeros } from \"../ndarray\";\nimport {\n  assert2D,\n  assertNumericTensor,\n  createSeededRandom,\n  getShape2D,\n  getStride1D,\n  getStrides2D,\n  shuffleIndicesInPlace,\n} from \"./_internal\";\n\nfunction getNumericData(X: Tensor, name: string): ArrayLike<number | bigint> {\n  if (X.dtype === \"string\") {\n    throw new DTypeError(`${name} must be numeric`);\n  }\n  if (Array.isArray(X.data)) {\n    throw new DeepboxError(\"Internal error: invalid numeric tensor storage\");\n  }\n  return X.data;\n}\n\nfunction parseBooleanOption(value: unknown, name: string, defaultValue: boolean): boolean {\n  if (value === undefined) {\n    return defaultValue;\n  }\n  if (typeof value !== \"boolean\") {\n    throw new InvalidParameterError(`${name} must be a boolean`, name, value);\n  }\n  return value;\n}\n\nfunction validateFiniteData(X: Tensor, name: string): void {\n  const [nSamples, nFeatures] = getShape2D(X);\n  const data = getNumericData(X, name);\n  const [stride0, stride1] = getStrides2D(X);\n  let flatIndex = 0;\n\n  for (let i = 0; i < nSamples; i++) {\n    const rowBase = X.offset + i * stride0;\n    for (let j = 0; j < nFeatures; j++) {\n      const raw = data[rowBase + j * stride1];\n      if (raw === undefined) {\n        throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n      }\n      const val = Number(raw);\n      if (!Number.isFinite(val)) {\n        throw new DataValidationError(`${name} contains NaN or Infinity at index ${flatIndex}`);\n      }\n      flatIndex += 1;\n    }\n  }\n}\n\nfunction snapInverseValue(value: number): number {\n  if (!Number.isFinite(value)) return value;\n  const rounded = Math.round(value);\n  if (Math.abs(value - rounded) < 1e-12) return rounded;\n  const scaled = Math.round(value * 1e12) / 1e12;\n  if (Math.abs(value - scaled) < 1e-12) return scaled;\n  return value;\n}\n\nfunction normalQuantile(p: number): number {\n  if (!Number.isFinite(p) || p <= 0 || p >= 1) {\n    throw new InvalidParameterError(\n      \"normalQuantile requires p in the open interval (0, 1)\",\n      \"p\",\n      p\n    );\n  }\n  // Acklam's inverse normal CDF approximation.\n  const a1 = -3.969683028665376e1;\n  const a2 = 2.209460984245205e2;\n  const a3 = -2.759285104469687e2;\n  const a4 = 1.38357751867269e2;\n  const a5 = -3.066479806614716e1;\n  const a6 = 2.506628277459239;\n\n  const b1 = -5.447609879822406e1;\n  const b2 = 1.615858368580409e2;\n  const b3 = -1.556989798598866e2;\n  const b4 = 6.680131188771972e1;\n  const b5 = -1.328068155288572e1;\n\n  const c1 = -7.784894002430293e-3;\n  const c2 = -3.223964580411365e-1;\n  const c3 = -2.400758277161838;\n  const c4 = -2.549732539343734;\n  const c5 = 4.374664141464968;\n  const c6 = 2.938163982698783;\n\n  const d1 = 7.784695709041462e-3;\n  const d2 = 3.224671290700398e-1;\n  const d3 = 2.445134137142996;\n  const d4 = 3.754408661907416;\n\n  const plow = 0.02425;\n  const phigh = 1 - plow;\n\n  if (p < plow) {\n    const q = Math.sqrt(-2 * Math.log(p));\n    return (\n      (((((c1 * q + c2) * q + c3) * q + c4) * q + c5) * q + c6) /\n      ((((d1 * q + d2) * q + d3) * q + d4) * q + 1)\n    );\n  }\n  if (p > phigh) {\n    const q = Math.sqrt(-2 * Math.log(1 - p));\n    return -(\n      (((((c1 * q + c2) * q + c3) * q + c4) * q + c5) * q + c6) /\n      ((((d1 * q + d2) * q + d3) * q + d4) * q + 1)\n    );\n  }\n\n  const q = p - 0.5;\n  const r = q * q;\n  return (\n    ((((((a1 * r + a2) * r + a3) * r + a4) * r + a5) * r + a6) * q) /\n    (((((b1 * r + b2) * r + b3) * r + b4) * r + b5) * r + 1)\n  );\n}\n\n/**\n * Standardize features by removing mean and scaling to unit variance.\n *\n * **Formula**: z = (x - μ) / σ\n *\n * **Attributes** (after fitting):\n * - `mean_`: Mean of each feature\n * - `scale_`: Standard deviation of each feature\n *\n * @example\n * ```js\n * import { StandardScaler } from 'deepbox/preprocess';\n * import { tensor } from 'deepbox/ndarray';\n *\n * const X = tensor([[1, 2], [3, 4], [5, 6]]);\n * const scaler = new StandardScaler();\n * scaler.fit(X);\n * const XScaled = scaler.transform(X);\n * ```\n *\n * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html | Scikit-learn StandardScaler}\n */\nexport class StandardScaler {\n  private fitted = false;\n  private mean_: Tensor | undefined;\n  private scale_: Tensor | undefined;\n  private withMean: boolean;\n  private withStd: boolean;\n\n  /**\n   * Creates a new StandardScaler.\n   *\n   * @param options - Configuration options\n   * @param options.withMean - Center data before scaling (default: true)\n   * @param options.withStd - Scale data to unit variance (default: true)\n   * @param options.copy - Accepted for API parity; transforms are always out-of-place (default: true)\n   */\n  constructor(options: { withMean?: boolean; withStd?: boolean; copy?: boolean } = {}) {\n    this.withMean = parseBooleanOption(options.withMean, \"withMean\", true);\n    this.withStd = parseBooleanOption(options.withStd, \"withStd\", true);\n    parseBooleanOption(options.copy, \"copy\", true);\n  }\n\n  fit(X: Tensor): this {\n    if (X.size === 0) {\n      throw new InvalidParameterError(\"X must contain at least one sample\", \"X\");\n    }\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    validateFiniteData(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const data = getNumericData(X, \"X\");\n    const [stride0, stride1] = getStrides2D(X);\n\n    let means: number[] | undefined;\n    if (this.withMean || this.withStd) {\n      means = new Array<number>(nFeatures).fill(0);\n      for (let j = 0; j < nFeatures; j++) {\n        let sum = 0;\n        for (let i = 0; i < nSamples; i++) {\n          const raw = data[X.offset + i * stride0 + j * stride1];\n          if (raw === undefined) {\n            throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n          }\n          sum += Number(raw);\n        }\n        if (means) {\n          means[j] = sum / nSamples;\n        }\n      }\n    }\n\n    if (this.withStd) {\n      const stds = new Array<number>(nFeatures).fill(0);\n      for (let j = 0; j < nFeatures; j++) {\n        const mean = means ? (means[j] ?? 0) : 0;\n        let sumSq = 0;\n        for (let i = 0; i < nSamples; i++) {\n          const raw = data[X.offset + i * stride0 + j * stride1];\n          if (raw === undefined) {\n            throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n          }\n          const val = Number(raw) - mean;\n          sumSq += val * val;\n        }\n        stds[j] = Math.sqrt(sumSq / nSamples);\n      }\n      this.scale_ = tensor(stds, { dtype: \"float64\" });\n    } else {\n      this.scale_ = undefined;\n    }\n\n    this.mean_ = this.withMean && means ? tensor(means, { dtype: \"float64\" }) : undefined;\n\n    this.fitted = true;\n    return this;\n  }\n\n  transform(X: Tensor): Tensor {\n    if (!this.fitted) {\n      throw new NotFittedError(\"StandardScaler must be fitted before transform\");\n    }\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    validateFiniteData(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const data = getNumericData(X, \"X\");\n    const [stride0, stride1] = getStrides2D(X);\n    const mean = this.mean_;\n    const scale = this.scale_;\n    const meanData = mean ? getNumericData(mean, \"mean_\") : undefined;\n    const scaleData = scale ? getNumericData(scale, \"scale_\") : undefined;\n    const meanStride = mean ? getStride1D(mean) : 0;\n    const scaleStride = scale ? getStride1D(scale) : 0;\n\n    if (this.withMean && !mean) {\n      throw new DeepboxError(\"StandardScaler internal error: missing mean_\");\n    }\n    if (this.withStd && !scale) {\n      throw new DeepboxError(\"StandardScaler internal error: missing scale_\");\n    }\n\n    const result = Array.from({ length: nSamples }, () => new Array<number>(nFeatures).fill(0));\n\n    for (let i = 0; i < nSamples; i++) {\n      const rowBase = X.offset + i * stride0;\n      for (let j = 0; j < nFeatures; j++) {\n        const raw = data[rowBase + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        let val = Number(raw);\n\n        if (this.withMean && mean && meanData) {\n          const meanValue = meanData[mean.offset + j * meanStride];\n          if (meanValue === undefined) {\n            throw new DeepboxError(\"Internal error: mean tensor access out of bounds\");\n          }\n          val -= Number(meanValue);\n        }\n\n        if (this.withStd && scale && scaleData) {\n          const rawScale = scaleData[scale.offset + j * scaleStride];\n          if (rawScale === undefined) {\n            throw new DeepboxError(\"Internal error: scale tensor access out of bounds\");\n          }\n          const std = Number(rawScale);\n          const safeStd = std === 0 ? 1 : std;\n          val /= safeStd;\n        }\n\n        const row = result[i];\n        if (row === undefined) {\n          throw new DeepboxError(\"Internal error: result row access failed\");\n        }\n        row[j] = val;\n      }\n    }\n\n    return tensor(result, { dtype: \"float64\", device: X.device });\n  }\n\n  fitTransform(X: Tensor): Tensor {\n    return this.fit(X).transform(X);\n  }\n\n  inverseTransform(X: Tensor): Tensor {\n    if (!this.fitted) {\n      throw new NotFittedError(\"StandardScaler must be fitted before inverse_transform\");\n    }\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const data = getNumericData(X, \"X\");\n    const [stride0, stride1] = getStrides2D(X);\n    const mean = this.mean_;\n    const scale = this.scale_;\n    const meanData = mean ? getNumericData(mean, \"mean_\") : undefined;\n    const scaleData = scale ? getNumericData(scale, \"scale_\") : undefined;\n    const meanStride = mean ? getStride1D(mean) : 0;\n    const scaleStride = scale ? getStride1D(scale) : 0;\n\n    if (this.withMean && !mean) {\n      throw new DeepboxError(\"StandardScaler internal error: missing mean_\");\n    }\n    if (this.withStd && !scale) {\n      throw new DeepboxError(\"StandardScaler internal error: missing scale_\");\n    }\n\n    const result = Array.from({ length: nSamples }, () => new Array<number>(nFeatures).fill(0));\n\n    for (let i = 0; i < nSamples; i++) {\n      const rowBase = X.offset + i * stride0;\n      for (let j = 0; j < nFeatures; j++) {\n        const raw = data[rowBase + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        let val = Number(raw);\n\n        if (this.withStd && scale && scaleData) {\n          const rawScale = scaleData[scale.offset + j * scaleStride];\n          if (rawScale === undefined) {\n            throw new DeepboxError(\"Internal error: scale tensor access out of bounds\");\n          }\n          const std = Number(rawScale);\n          const safeStd = std === 0 ? 1 : std;\n          val *= safeStd;\n        }\n\n        if (this.withMean && mean && meanData) {\n          const meanValue = meanData[mean.offset + j * meanStride];\n          if (meanValue === undefined) {\n            throw new DeepboxError(\"Internal error: mean tensor access out of bounds\");\n          }\n          val += Number(meanValue);\n        }\n\n        const resultRow = result[i];\n        if (resultRow === undefined) {\n          throw new DeepboxError(\"Internal error: result row access failed\");\n        }\n        resultRow[j] = snapInverseValue(val);\n      }\n    }\n\n    return tensor(result, { dtype: \"float64\", device: X.device });\n  }\n}\n\n/**\n * Scale features to a range [min, max].\n *\n * **Formula**: X_scaled = (X - X.min) / (X.max - X.min) * (max - min) + min\n *\n * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.MinMaxScaler.html | Scikit-learn MinMaxScaler}\n */\nexport class MinMaxScaler {\n  private fitted = false;\n  private dataMin_?: Tensor;\n  private dataMax_?: Tensor;\n  private featureRange: [number, number];\n  private clip: boolean;\n\n  /**\n   * Creates a new MinMaxScaler.\n   *\n   * @param options - Configuration options\n   * @param options.featureRange - Desired feature range [min, max] (default: [0, 1])\n   * @param options.clip - Clip transformed values to featureRange (default: false)\n   * @param options.copy - Accepted for API parity; transforms are always out-of-place (default: true)\n   */\n  constructor(\n    options: {\n      featureRange?: [number, number];\n      clip?: boolean;\n      copy?: boolean;\n    } = {}\n  ) {\n    this.featureRange = options.featureRange ?? [0, 1];\n    this.clip = parseBooleanOption(options.clip, \"clip\", false);\n    parseBooleanOption(options.copy, \"copy\", true);\n    const [minRange, maxRange] = this.featureRange;\n    if (!Number.isFinite(minRange) || !Number.isFinite(maxRange) || minRange >= maxRange) {\n      throw new InvalidParameterError(\n        \"featureRange must be [min, max] with min < max\",\n        \"featureRange\",\n        this.featureRange\n      );\n    }\n  }\n\n  fit(X: Tensor): this {\n    if (X.size === 0) {\n      throw new InvalidParameterError(\"X must contain at least one sample\", \"X\");\n    }\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    validateFiniteData(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const data = getNumericData(X, \"X\");\n    const [stride0, stride1] = getStrides2D(X);\n\n    const mins = new Array<number>(nFeatures).fill(Number.POSITIVE_INFINITY);\n    const maxs = new Array<number>(nFeatures).fill(Number.NEGATIVE_INFINITY);\n\n    for (let j = 0; j < nFeatures; j++) {\n      for (let i = 0; i < nSamples; i++) {\n        const raw = data[X.offset + i * stride0 + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        const val = Number(raw);\n        const currentMin = mins[j];\n        const currentMax = maxs[j];\n        if (currentMin === undefined || currentMax === undefined) {\n          throw new DeepboxError(\"Internal error: min/max array access failed\");\n        }\n        mins[j] = Math.min(currentMin, val);\n        maxs[j] = Math.max(currentMax, val);\n      }\n    }\n\n    this.dataMin_ = tensor(mins, { dtype: \"float64\" });\n    this.dataMax_ = tensor(maxs, { dtype: \"float64\" });\n    this.fitted = true;\n    return this;\n  }\n\n  transform(X: Tensor): Tensor {\n    if (!this.fitted) {\n      throw new NotFittedError(\"MinMaxScaler must be fitted before transform\");\n    }\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    validateFiniteData(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const [minRange, maxRange] = this.featureRange;\n    const data = getNumericData(X, \"X\");\n    const [stride0, stride1] = getStrides2D(X);\n    const dataMin = this.dataMin_;\n    const dataMax = this.dataMax_;\n\n    if (!dataMin || !dataMax) {\n      throw new DeepboxError(\"MinMaxScaler internal error: missing fitted min/max\");\n    }\n    const minData = getNumericData(dataMin, \"dataMin_\");\n    const maxData = getNumericData(dataMax, \"dataMax_\");\n    const minStride = getStride1D(dataMin);\n    const maxStride = getStride1D(dataMax);\n\n    const result = Array.from({ length: nSamples }, () => new Array<number>(nFeatures).fill(0));\n\n    for (let i = 0; i < nSamples; i++) {\n      const rowBase = X.offset + i * stride0;\n      for (let j = 0; j < nFeatures; j++) {\n        const raw = data[rowBase + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        const val = Number(raw);\n        const rawMin = minData[dataMin.offset + j * minStride];\n        const rawMax = maxData[dataMax.offset + j * maxStride];\n        if (rawMin === undefined || rawMax === undefined) {\n          throw new DeepboxError(\"Internal error: min/max tensor access out of bounds\");\n        }\n        const min = Number(rawMin);\n        const max = Number(rawMax);\n        const range = max - min;\n\n        const row = result[i];\n        if (row === undefined) {\n          throw new DeepboxError(\"Internal error: result row access failed\");\n        }\n        let scaled =\n          range !== 0 ? ((val - min) / range) * (maxRange - minRange) + minRange : minRange;\n        if (this.clip) {\n          scaled = Math.max(minRange, Math.min(maxRange, scaled));\n        }\n        row[j] = scaled;\n      }\n    }\n\n    return tensor(result, { dtype: \"float64\", device: X.device });\n  }\n\n  fitTransform(X: Tensor): Tensor {\n    return this.fit(X).transform(X);\n  }\n\n  inverseTransform(X: Tensor): Tensor {\n    if (!this.fitted) {\n      throw new NotFittedError(\"MinMaxScaler must be fitted before inverse_transform\");\n    }\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const [minRange, maxRange] = this.featureRange;\n    const data = getNumericData(X, \"X\");\n    const [stride0, stride1] = getStrides2D(X);\n    const dataMin = this.dataMin_;\n    const dataMax = this.dataMax_;\n\n    if (!dataMin || !dataMax) {\n      throw new DeepboxError(\"MinMaxScaler internal error: missing fitted min/max\");\n    }\n    const minData = getNumericData(dataMin, \"dataMin_\");\n    const maxData = getNumericData(dataMax, \"dataMax_\");\n    const minStride = getStride1D(dataMin);\n    const maxStride = getStride1D(dataMax);\n\n    const result = Array.from({ length: nSamples }, () => new Array<number>(nFeatures).fill(0));\n\n    for (let i = 0; i < nSamples; i++) {\n      const rowBase = X.offset + i * stride0;\n      for (let j = 0; j < nFeatures; j++) {\n        const raw = data[rowBase + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        const val = Number(raw);\n        const rawMin = minData[dataMin.offset + j * minStride];\n        const rawMax = maxData[dataMax.offset + j * maxStride];\n        if (rawMin === undefined || rawMax === undefined) {\n          throw new DeepboxError(\"Internal error: min/max tensor access out of bounds\");\n        }\n        const min = Number(rawMin);\n        const max = Number(rawMax);\n        const range = max - min;\n\n        const row = result[i];\n        if (row === undefined) {\n          throw new DeepboxError(\"Internal error: result row access failed\");\n        }\n        row[j] = ((val - minRange) / (maxRange - minRange)) * range + min;\n      }\n    }\n\n    return tensor(result, { dtype: \"float64\", device: X.device });\n  }\n}\n\n/**\n * Scale features by maximum absolute value.\n *\n * Scales to range [-1, 1]. Suitable for data that is already centered at zero.\n *\n * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.MaxAbsScaler.html | Scikit-learn MaxAbsScaler}\n */\nexport class MaxAbsScaler {\n  private fitted = false;\n  private maxAbs_?: Tensor;\n\n  /**\n   * Creates a new MaxAbsScaler.\n   *\n   * @param options - Configuration options\n   * @param options.copy - Accepted for API parity; transforms are always out-of-place (default: true)\n   */\n  constructor(options: { copy?: boolean } = {}) {\n    parseBooleanOption(options.copy, \"copy\", true);\n  }\n\n  fit(X: Tensor): this {\n    if (X.size === 0) {\n      throw new InvalidParameterError(\"X must contain at least one sample\", \"X\");\n    }\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    validateFiniteData(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const data = getNumericData(X, \"X\");\n    const [stride0, stride1] = getStrides2D(X);\n\n    const maxAbs = new Array<number>(nFeatures).fill(0);\n\n    for (let j = 0; j < nFeatures; j++) {\n      for (let i = 0; i < nSamples; i++) {\n        const raw = data[X.offset + i * stride0 + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        const currentMax = maxAbs[j];\n        if (currentMax === undefined) {\n          throw new DeepboxError(\"Internal error: maxAbs array access failed\");\n        }\n        maxAbs[j] = Math.max(currentMax, Math.abs(Number(raw)));\n      }\n    }\n\n    this.maxAbs_ = tensor(maxAbs, { dtype: \"float64\" });\n    this.fitted = true;\n    return this;\n  }\n\n  transform(X: Tensor): Tensor {\n    if (!this.fitted) {\n      throw new NotFittedError(\"MaxAbsScaler must be fitted before transform\");\n    }\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    validateFiniteData(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const data = getNumericData(X, \"X\");\n    const [stride0, stride1] = getStrides2D(X);\n    const maxAbs = this.maxAbs_;\n    if (!maxAbs) {\n      throw new DeepboxError(\"MaxAbsScaler internal error: missing fitted maxAbs\");\n    }\n    const maxData = getNumericData(maxAbs, \"maxAbs_\");\n    const maxStride = getStride1D(maxAbs);\n\n    const result = Array.from({ length: nSamples }, () => new Array<number>(nFeatures).fill(0));\n\n    for (let i = 0; i < nSamples; i++) {\n      const rowBase = X.offset + i * stride0;\n      for (let j = 0; j < nFeatures; j++) {\n        const raw = data[rowBase + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        const val = Number(raw);\n        const rawScale = maxData[maxAbs.offset + j * maxStride];\n        if (rawScale === undefined) {\n          throw new DeepboxError(\"Internal error: maxAbs tensor access out of bounds\");\n        }\n        const scale = Number(rawScale);\n        const safeScale = scale === 0 ? 1 : scale;\n        const row = result[i];\n        if (row === undefined) {\n          throw new DeepboxError(\"Internal error: result row access failed\");\n        }\n        row[j] = val / safeScale;\n      }\n    }\n\n    return tensor(result, { dtype: \"float64\", device: X.device });\n  }\n\n  fitTransform(X: Tensor): Tensor {\n    return this.fit(X).transform(X);\n  }\n\n  inverseTransform(X: Tensor): Tensor {\n    if (!this.fitted) {\n      throw new NotFittedError(\"MaxAbsScaler must be fitted before inverse_transform\");\n    }\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const data = getNumericData(X, \"X\");\n    const [stride0, stride1] = getStrides2D(X);\n    const maxAbs = this.maxAbs_;\n    if (!maxAbs) {\n      throw new DeepboxError(\"MaxAbsScaler internal error: missing fitted maxAbs\");\n    }\n    const maxData = getNumericData(maxAbs, \"maxAbs_\");\n    const maxStride = getStride1D(maxAbs);\n\n    const result = Array.from({ length: nSamples }, () => new Array<number>(nFeatures).fill(0));\n\n    for (let i = 0; i < nSamples; i++) {\n      const rowBase = X.offset + i * stride0;\n      for (let j = 0; j < nFeatures; j++) {\n        const raw = data[rowBase + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        const val = Number(raw);\n        const rawScale = maxData[maxAbs.offset + j * maxStride];\n        if (rawScale === undefined) {\n          throw new DeepboxError(\"Internal error: maxAbs tensor access out of bounds\");\n        }\n        const scale = Number(rawScale);\n        const row = result[i];\n        if (row === undefined) {\n          throw new DeepboxError(\"Internal error: result row access failed\");\n        }\n        row[j] = val * scale;\n      }\n    }\n\n    return tensor(result, { dtype: \"float64\", device: X.device });\n  }\n}\n\n/**\n * Robust scaler using median and IQR.\n *\n * Robust to outliers.\n *\n * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.RobustScaler.html | Scikit-learn RobustScaler}\n */\nexport class RobustScaler {\n  private fitted = false;\n  private center_: Tensor | undefined;\n  private scale_: Tensor | undefined;\n  private withCentering: boolean;\n  private withScaling: boolean;\n  private quantileRange: [number, number];\n  private unitVariance: boolean;\n\n  /**\n   * Creates a new RobustScaler.\n   *\n   * @param options - Configuration options\n   * @param options.withCentering - Center data using median (default: true)\n   * @param options.withScaling - Scale data using IQR (default: true)\n   * @param options.quantileRange - Quantile range for IQR as percentiles (default: [25, 75])\n   * @param options.unitVariance - Scale so that features have unit variance under normality (default: false)\n   * @param options.copy - Accepted for API parity; transforms are always out-of-place (default: true)\n   */\n  constructor(\n    options: {\n      withCentering?: boolean;\n      withScaling?: boolean;\n      quantileRange?: [number, number];\n      unitVariance?: boolean;\n      copy?: boolean;\n    } = {}\n  ) {\n    this.withCentering = parseBooleanOption(options.withCentering, \"withCentering\", true);\n    this.withScaling = parseBooleanOption(options.withScaling, \"withScaling\", true);\n    this.quantileRange = options.quantileRange ?? [25, 75];\n    this.unitVariance = parseBooleanOption(options.unitVariance, \"unitVariance\", false);\n    parseBooleanOption(options.copy, \"copy\", true);\n    const [lower, upper] = this.quantileRange;\n    if (\n      !Number.isFinite(lower) ||\n      !Number.isFinite(upper) ||\n      lower < 0 ||\n      upper > 100 ||\n      lower >= upper\n    ) {\n      throw new InvalidParameterError(\n        \"quantileRange must be a valid ascending percentile range\",\n        \"quantileRange\",\n        this.quantileRange\n      );\n    }\n  }\n\n  fit(X: Tensor): this {\n    if (X.size === 0) {\n      throw new InvalidParameterError(\"X must contain at least one sample\", \"X\");\n    }\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    validateFiniteData(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const data = getNumericData(X, \"X\");\n    const [stride0, stride1] = getStrides2D(X);\n\n    const centers = new Array<number>(nFeatures).fill(0);\n    const scales = new Array<number>(nFeatures).fill(0);\n\n    // Convert quantile range from percentiles to fractions\n    const [lowerPercentile, upperPercentile] = this.quantileRange;\n    const lowerFraction = lowerPercentile / 100;\n    const upperFraction = upperPercentile / 100;\n    const normalizer = this.unitVariance\n      ? normalQuantile(upperFraction) - normalQuantile(lowerFraction)\n      : 1;\n    if (this.unitVariance && (!Number.isFinite(normalizer) || normalizer <= 0)) {\n      throw new DeepboxError(\"RobustScaler internal error: invalid unit variance normalizer\");\n    }\n\n    for (let j = 0; j < nFeatures; j++) {\n      const values: number[] = [];\n      for (let i = 0; i < nSamples; i++) {\n        const raw = data[X.offset + i * stride0 + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        values.push(Number(raw));\n      }\n      values.sort((a, b) => a - b);\n\n      const interpolate = (q: number): number => {\n        if (values.length === 0) {\n          throw new DeepboxError(\"Internal error: cannot interpolate empty values\");\n        }\n        if (values.length === 1) {\n          const only = values[0];\n          if (only === undefined) {\n            throw new DeepboxError(\"Internal error: missing sorted value\");\n          }\n          return only;\n        }\n\n        const position = q * (values.length - 1);\n        const lower = Math.floor(position);\n        const upper = Math.ceil(position);\n        const lowerValue = values[lower];\n        const upperValue = values[upper];\n        if (lowerValue === undefined || upperValue === undefined) {\n          throw new DeepboxError(\"Internal error: quantile interpolation index out of bounds\");\n        }\n        if (upper === lower) {\n          return lowerValue;\n        }\n        const weight = position - lower;\n        return lowerValue * (1 - weight) + upperValue * weight;\n      };\n\n      // Median for centering and IQR for scaling\n      centers[j] = interpolate(0.5);\n      const qLower = interpolate(lowerFraction);\n      const qUpper = interpolate(upperFraction);\n      const iqr = qUpper - qLower;\n      scales[j] = this.unitVariance ? iqr / normalizer : iqr;\n    }\n\n    this.center_ = this.withCentering ? tensor(centers, { dtype: \"float64\" }) : undefined;\n    this.scale_ = this.withScaling ? tensor(scales, { dtype: \"float64\" }) : undefined;\n    this.fitted = true;\n    return this;\n  }\n\n  transform(X: Tensor): Tensor {\n    if (!this.fitted) {\n      throw new NotFittedError(\"RobustScaler must be fitted before transform\");\n    }\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    validateFiniteData(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const data = getNumericData(X, \"X\");\n    const [stride0, stride1] = getStrides2D(X);\n    const center = this.center_;\n    const scale = this.scale_;\n    const centerData = center ? getNumericData(center, \"center_\") : undefined;\n    const scaleData = scale ? getNumericData(scale, \"scale_\") : undefined;\n    const centerStride = center ? getStride1D(center) : 0;\n    const scaleStride = scale ? getStride1D(scale) : 0;\n\n    if (this.withCentering && !center) {\n      throw new DeepboxError(\"RobustScaler internal error: missing center_\");\n    }\n    if (this.withScaling && !scale) {\n      throw new DeepboxError(\"RobustScaler internal error: missing scale_\");\n    }\n\n    const result = Array.from({ length: nSamples }, () => new Array<number>(nFeatures).fill(0));\n\n    for (let i = 0; i < nSamples; i++) {\n      const rowBase = X.offset + i * stride0;\n      for (let j = 0; j < nFeatures; j++) {\n        const raw = data[rowBase + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        let val = Number(raw);\n\n        if (this.withCentering && center && centerData) {\n          const rawCenter = centerData[center.offset + j * centerStride];\n          if (rawCenter === undefined) {\n            throw new DeepboxError(\"Internal error: center tensor access out of bounds\");\n          }\n          val -= Number(rawCenter);\n        }\n\n        if (this.withScaling && scale && scaleData) {\n          const rawScale = scaleData[scale.offset + j * scaleStride];\n          if (rawScale === undefined) {\n            throw new DeepboxError(\"Internal error: scale tensor access out of bounds\");\n          }\n          const scaleValue = Number(rawScale);\n          const safeScale = scaleValue === 0 ? 1 : scaleValue;\n          val /= safeScale;\n        }\n\n        const resultRow = result[i];\n        if (resultRow === undefined) {\n          throw new DeepboxError(\"Internal error: result row access failed\");\n        }\n        resultRow[j] = val;\n      }\n    }\n\n    return tensor(result, { dtype: \"float64\", device: X.device });\n  }\n\n  fitTransform(X: Tensor): Tensor {\n    return this.fit(X).transform(X);\n  }\n\n  inverseTransform(X: Tensor): Tensor {\n    if (!this.fitted) {\n      throw new NotFittedError(\"RobustScaler must be fitted before inverse_transform\");\n    }\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const data = getNumericData(X, \"X\");\n    const [stride0, stride1] = getStrides2D(X);\n    const center = this.center_;\n    const scale = this.scale_;\n    const centerData = center ? getNumericData(center, \"center_\") : undefined;\n    const scaleData = scale ? getNumericData(scale, \"scale_\") : undefined;\n    const centerStride = center ? getStride1D(center) : 0;\n    const scaleStride = scale ? getStride1D(scale) : 0;\n\n    if (this.withCentering && !center) {\n      throw new DeepboxError(\"RobustScaler internal error: missing center_\");\n    }\n    if (this.withScaling && !scale) {\n      throw new DeepboxError(\"RobustScaler internal error: missing scale_\");\n    }\n\n    const result = Array.from({ length: nSamples }, () => new Array<number>(nFeatures).fill(0));\n\n    for (let i = 0; i < nSamples; i++) {\n      const rowBase = X.offset + i * stride0;\n      for (let j = 0; j < nFeatures; j++) {\n        const raw = data[rowBase + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        let val = Number(raw);\n\n        if (this.withScaling && scale && scaleData) {\n          const rawScale = scaleData[scale.offset + j * scaleStride];\n          if (rawScale === undefined) {\n            throw new DeepboxError(\"Internal error: scale tensor access out of bounds\");\n          }\n          const scaleValue = Number(rawScale);\n          const safeScale = scaleValue === 0 ? 1 : scaleValue;\n          val *= safeScale;\n        }\n\n        if (this.withCentering && center && centerData) {\n          const rawCenter = centerData[center.offset + j * centerStride];\n          if (rawCenter === undefined) {\n            throw new DeepboxError(\"Internal error: center tensor access out of bounds\");\n          }\n          val += Number(rawCenter);\n        }\n\n        const resultRow = result[i];\n        if (resultRow === undefined) {\n          throw new DeepboxError(\"Internal error: result row access failed\");\n        }\n        resultRow[j] = val;\n      }\n    }\n\n    return tensor(result, { dtype: \"float64\", device: X.device });\n  }\n}\n\n/**\n * Normalize samples to unit norm.\n *\n * Scales each sample (row) to have unit norm.\n *\n * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.Normalizer.html | Scikit-learn Normalizer}\n */\nexport class Normalizer {\n  private norm: \"l1\" | \"l2\" | \"max\";\n\n  /**\n   * Creates a new Normalizer.\n   *\n   * @param options - Configuration options\n   * @param options.norm - Norm to use (default: \"l2\")\n   * @param options.copy - Accepted for API parity; transforms are always out-of-place (default: true)\n   */\n  constructor(options: { norm?: \"l1\" | \"l2\" | \"max\"; copy?: boolean } = {}) {\n    this.norm = options.norm ?? \"l2\";\n    if (this.norm !== \"l1\" && this.norm !== \"l2\" && this.norm !== \"max\") {\n      throw new InvalidParameterError(\"norm must be one of: l1, l2, max\", \"norm\", this.norm);\n    }\n    parseBooleanOption(options.copy, \"copy\", true);\n  }\n\n  fit(_X: Tensor): this {\n    return this;\n  }\n\n  transform(X: Tensor): Tensor {\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    validateFiniteData(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const data = getNumericData(X, \"X\");\n    const [stride0, stride1] = getStrides2D(X);\n\n    const result = Array.from({ length: nSamples }, () => new Array<number>(nFeatures).fill(0));\n\n    for (let i = 0; i < nSamples; i++) {\n      let norm = 0;\n      const rowBase = X.offset + i * stride0;\n\n      if (this.norm === \"l2\") {\n        for (let j = 0; j < nFeatures; j++) {\n          const raw = data[rowBase + j * stride1];\n          if (raw === undefined) {\n            throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n          }\n          const val = Number(raw);\n          norm += val * val;\n        }\n        norm = Math.sqrt(norm);\n      } else if (this.norm === \"l1\") {\n        for (let j = 0; j < nFeatures; j++) {\n          const raw = data[rowBase + j * stride1];\n          if (raw === undefined) {\n            throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n          }\n          norm += Math.abs(Number(raw));\n        }\n      } else if (this.norm === \"max\") {\n        for (let j = 0; j < nFeatures; j++) {\n          const raw = data[rowBase + j * stride1];\n          if (raw === undefined) {\n            throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n          }\n          norm = Math.max(norm, Math.abs(Number(raw)));\n        }\n      }\n\n      for (let j = 0; j < nFeatures; j++) {\n        const raw = data[rowBase + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        const val = Number(raw);\n        const row = result[i];\n        if (row === undefined) {\n          throw new DeepboxError(\"Internal error: result row access failed\");\n        }\n        row[j] = norm === 0 ? val : val / norm;\n      }\n    }\n\n    return tensor(result, { dtype: \"float64\", device: X.device });\n  }\n\n  fitTransform(X: Tensor): Tensor {\n    return this.transform(X);\n  }\n}\n\n/**\n * Transform features using quantiles.\n *\n * Maps to uniform or normal distribution.\n *\n * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.QuantileTransformer.html | Scikit-learn QuantileTransformer}\n */\nexport class QuantileTransformer {\n  private fitted = false;\n  private nQuantiles: number;\n  private outputDistribution: \"uniform\" | \"normal\";\n  private quantiles_?: Map<number, { quantiles: number[]; references: number[] }>;\n  private subsample: number | undefined;\n  private randomState: number | undefined;\n\n  /**\n   * Creates a new QuantileTransformer.\n   *\n   * @param options - Configuration options\n   * @param options.nQuantiles - Number of quantiles to use (default: 1000)\n   * @param options.outputDistribution - \"uniform\" or \"normal\" (default: \"uniform\")\n   * @param options.subsample - Subsample size for quantile estimation (default: use all samples)\n   * @param options.randomState - Seed for subsampling reproducibility\n   * @param options.copy - Accepted for API parity; transforms are always out-of-place (default: true)\n   */\n  constructor(\n    options: {\n      nQuantiles?: number;\n      outputDistribution?: \"uniform\" | \"normal\";\n      subsample?: number;\n      randomState?: number;\n      copy?: boolean;\n    } = {}\n  ) {\n    this.nQuantiles = options.nQuantiles ?? 1000;\n    this.outputDistribution = options.outputDistribution ?? \"uniform\";\n    this.subsample = options.subsample;\n    this.randomState = options.randomState;\n    parseBooleanOption(options.copy, \"copy\", true);\n    if (\n      !Number.isFinite(this.nQuantiles) ||\n      !Number.isInteger(this.nQuantiles) ||\n      this.nQuantiles < 2\n    ) {\n      throw new InvalidParameterError(\n        \"nQuantiles must be at least 2\",\n        \"nQuantiles\",\n        this.nQuantiles\n      );\n    }\n    if (this.outputDistribution !== \"uniform\" && this.outputDistribution !== \"normal\") {\n      throw new InvalidParameterError(\n        \"outputDistribution must be 'uniform' or 'normal'\",\n        \"outputDistribution\",\n        this.outputDistribution\n      );\n    }\n    if (this.subsample !== undefined) {\n      if (\n        !Number.isFinite(this.subsample) ||\n        !Number.isInteger(this.subsample) ||\n        this.subsample < 2\n      ) {\n        throw new InvalidParameterError(\n          \"subsample must be an integer >= 2\",\n          \"subsample\",\n          this.subsample\n        );\n      }\n    }\n  }\n\n  fit(X: Tensor): this {\n    if (X.size === 0) {\n      throw new InvalidParameterError(\"X must contain at least one sample\", \"X\");\n    }\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    validateFiniteData(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const data = getNumericData(X, \"X\");\n    const [stride0, stride1] = getStrides2D(X);\n\n    this.quantiles_ = new Map();\n    const sampleCount =\n      this.subsample !== undefined ? Math.min(this.subsample, nSamples) : nSamples;\n    const nQuantilesEffective = Math.min(this.nQuantiles, sampleCount);\n    const references =\n      nQuantilesEffective <= 1\n        ? [0.5]\n        : Array.from({ length: nQuantilesEffective }, (_, i) => i / (nQuantilesEffective - 1));\n\n    let sampleIndices: number[] | undefined;\n    if (sampleCount < nSamples) {\n      sampleIndices = Array.from({ length: nSamples }, (_, i) => i);\n      const random =\n        this.randomState !== undefined ? createSeededRandom(this.randomState) : Math.random;\n      shuffleIndicesInPlace(sampleIndices, random);\n      sampleIndices = sampleIndices.slice(0, sampleCount);\n    }\n\n    for (let j = 0; j < nFeatures; j++) {\n      const values: number[] = [];\n      if (sampleIndices) {\n        for (const idx of sampleIndices) {\n          const raw = data[X.offset + idx * stride0 + j * stride1];\n          if (raw === undefined) {\n            throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n          }\n          values.push(Number(raw));\n        }\n      } else {\n        for (let i = 0; i < nSamples; i++) {\n          const raw = data[X.offset + i * stride0 + j * stride1];\n          if (raw === undefined) {\n            throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n          }\n          values.push(Number(raw));\n        }\n      }\n      const sorted = [...values].sort((a, b) => a - b);\n      const quantiles = references.map((q) => this.interpolateFromSorted(sorted, q));\n      this.quantiles_.set(j, { quantiles, references });\n    }\n\n    this.fitted = true;\n    return this;\n  }\n\n  transform(X: Tensor): Tensor {\n    if (!this.fitted || !this.quantiles_) {\n      throw new NotFittedError(\"QuantileTransformer must be fitted before transform\");\n    }\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    validateFiniteData(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const data = getNumericData(X, \"X\");\n    const [stride0, stride1] = getStrides2D(X);\n\n    if (nSamples === 0) {\n      return zeros([0, nFeatures], { dtype: \"float64\" });\n    }\n\n    const result = new Array<number[]>(nSamples);\n    for (let i = 0; i < nSamples; i++) {\n      result[i] = new Array<number>(nFeatures);\n    }\n\n    for (let j = 0; j < nFeatures; j++) {\n      const feature = this.quantiles_.get(j);\n      if (!feature) {\n        throw new DeepboxError(`Internal error: missing fitted quantiles for feature ${j}`);\n      }\n\n      for (let i = 0; i < nSamples; i++) {\n        const raw = data[X.offset + i * stride0 + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        const val = Number(raw);\n        const quantile = this.mapValueToQuantile(val, feature.quantiles, feature.references);\n\n        const row = result[i];\n        if (!row) {\n          throw new DeepboxError(\"Internal error: result row access failed\");\n        }\n\n        if (this.outputDistribution === \"uniform\") {\n          row[j] = quantile;\n        } else {\n          // Transform to normal distribution using inverse error function\n          // Clamp quantile to avoid numerical issues at boundaries\n          const clampedQuantile = Math.max(1e-7, Math.min(1 - 1e-7, quantile));\n          const z = Math.sqrt(2) * this.erfInv(2 * clampedQuantile - 1);\n          row[j] = z;\n        }\n      }\n    }\n\n    return tensor(result, { dtype: \"float64\", device: X.device });\n  }\n\n  /**\n   * Inverse transform data back to the original feature space.\n   *\n   * If `outputDistribution=\"normal\"`, values are first mapped back to uniform\n   * quantiles before being projected into the original data distribution.\n   *\n   * @param X - Transformed data (2D tensor)\n   * @returns Data in the original feature space\n   * @throws {NotFittedError} If transformer is not fitted\n   */\n  inverseTransform(X: Tensor): Tensor {\n    if (!this.fitted || !this.quantiles_) {\n      throw new NotFittedError(\"QuantileTransformer must be fitted before inverse_transform\");\n    }\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    validateFiniteData(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const data = getNumericData(X, \"X\");\n    const [stride0, stride1] = getStrides2D(X);\n\n    if (nSamples === 0) {\n      return zeros([0, nFeatures], { dtype: \"float64\" });\n    }\n\n    const result = new Array<number[]>(nSamples);\n    for (let i = 0; i < nSamples; i++) {\n      result[i] = new Array<number>(nFeatures);\n    }\n\n    for (let j = 0; j < nFeatures; j++) {\n      const feature = this.quantiles_.get(j);\n      if (!feature) {\n        throw new DeepboxError(`Internal error: missing fitted quantiles for feature ${j}`);\n      }\n\n      for (let i = 0; i < nSamples; i++) {\n        const raw = data[X.offset + i * stride0 + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        const value = Number(raw);\n        let quantile = this.outputDistribution === \"normal\" ? this.normalCdf(value) : value;\n\n        quantile = Math.max(0, Math.min(1, quantile));\n        const row = result[i];\n        if (!row) {\n          throw new DeepboxError(\"Internal error: result row access failed\");\n        }\n        row[j] = this.mapQuantileToValue(quantile, feature.quantiles, feature.references);\n      }\n    }\n\n    return tensor(result, { dtype: \"float64\", device: X.device });\n  }\n\n  private erf(x: number): number {\n    // Abramowitz and Stegun approximation\n    const sign = x < 0 ? -1 : 1;\n    const absX = Math.abs(x);\n    const t = 1 / (1 + 0.3275911 * absX);\n    const a1 = 0.254829592;\n    const a2 = -0.284496736;\n    const a3 = 1.421413741;\n    const a4 = -1.453152027;\n    const a5 = 1.061405429;\n    const poly = ((((a5 * t + a4) * t + a3) * t + a2) * t + a1) * t;\n    return sign * (1 - poly * Math.exp(-absX * absX));\n  }\n\n  private normalCdf(z: number): number {\n    return 0.5 * (1 + this.erf(z / Math.sqrt(2)));\n  }\n\n  private erfInv(x: number): number {\n    const a = 0.147;\n    const b = 2 / (Math.PI * a) + Math.log(1 - x * x) / 2;\n    const sign = x < 0 ? -1 : 1;\n    return sign * Math.sqrt(Math.sqrt(b * b - Math.log(1 - x * x) / a) - b);\n  }\n\n  private interpolateFromSorted(sorted: number[], q: number): number {\n    if (sorted.length === 0) {\n      throw new DeepboxError(\"Internal error: cannot interpolate empty sorted values\");\n    }\n    if (sorted.length === 1) {\n      const only = sorted[0];\n      if (only === undefined) {\n        throw new DeepboxError(\"Internal error: missing sorted value\");\n      }\n      return only;\n    }\n\n    const position = q * (sorted.length - 1);\n    const lower = Math.floor(position);\n    const upper = Math.ceil(position);\n    const lowerValue = sorted[lower];\n    const upperValue = sorted[upper];\n    if (lowerValue === undefined || upperValue === undefined) {\n      throw new DeepboxError(\"Internal error: quantile interpolation index out of bounds\");\n    }\n\n    if (upper === lower) {\n      return lowerValue;\n    }\n\n    const weight = position - lower;\n    return lowerValue * (1 - weight) + upperValue * weight;\n  }\n\n  private mapValueToQuantile(value: number, quantiles: number[], references: number[]): number {\n    const n = quantiles.length;\n    if (n === 0) {\n      return 0;\n    }\n    if (n === 1) {\n      const onlyReference = references[0];\n      if (onlyReference === undefined) {\n        throw new DeepboxError(\"Internal error: missing quantile reference\");\n      }\n      return onlyReference;\n    }\n\n    const firstQuantile = quantiles[0];\n    const lastQuantile = quantiles[n - 1];\n    if (firstQuantile === undefined || lastQuantile === undefined) {\n      throw new DeepboxError(\"Internal error: missing quantile endpoints\");\n    }\n    if (value <= firstQuantile) {\n      return 0;\n    }\n    if (value >= lastQuantile) {\n      return 1;\n    }\n\n    let left = 0;\n    let right = n - 1;\n    while (left + 1 < right) {\n      const mid = Math.floor((left + right) / 2);\n      const midValue = quantiles[mid];\n      if (midValue === undefined) {\n        throw new DeepboxError(\"Internal error: missing quantile midpoint\");\n      }\n      if (midValue <= value) {\n        left = mid;\n      } else {\n        right = mid;\n      }\n    }\n\n    const qLeft = quantiles[left];\n    const qRight = quantiles[right];\n    const rLeft = references[left];\n    const rRight = references[right];\n    if (\n      qLeft === undefined ||\n      qRight === undefined ||\n      rLeft === undefined ||\n      rRight === undefined\n    ) {\n      throw new DeepboxError(\"Internal error: missing quantile interpolation points\");\n    }\n    if (qRight <= qLeft) {\n      return (rLeft + rRight) / 2;\n    }\n\n    const ratio = (value - qLeft) / (qRight - qLeft);\n    return rLeft + ratio * (rRight - rLeft);\n  }\n\n  private mapQuantileToValue(quantile: number, quantiles: number[], references: number[]): number {\n    const n = references.length;\n    if (n === 0) {\n      return 0;\n    }\n    if (n === 1) {\n      const onlyQuantile = quantiles[0];\n      if (onlyQuantile === undefined) {\n        throw new DeepboxError(\"Internal error: missing quantile value\");\n      }\n      return onlyQuantile;\n    }\n\n    const firstRef = references[0];\n    const lastRef = references[n - 1];\n    if (firstRef === undefined || lastRef === undefined) {\n      throw new DeepboxError(\"Internal error: missing reference endpoints\");\n    }\n    if (quantile <= firstRef) {\n      const firstQuantile = quantiles[0];\n      if (firstQuantile === undefined) {\n        throw new DeepboxError(\"Internal error: missing quantile endpoints\");\n      }\n      return firstQuantile;\n    }\n    if (quantile >= lastRef) {\n      const lastQuantile = quantiles[n - 1];\n      if (lastQuantile === undefined) {\n        throw new DeepboxError(\"Internal error: missing quantile endpoints\");\n      }\n      return lastQuantile;\n    }\n\n    let left = 0;\n    let right = n - 1;\n    while (left + 1 < right) {\n      const mid = Math.floor((left + right) / 2);\n      const midRef = references[mid];\n      if (midRef === undefined) {\n        throw new DeepboxError(\"Internal error: missing quantile reference\");\n      }\n      if (midRef <= quantile) {\n        left = mid;\n      } else {\n        right = mid;\n      }\n    }\n\n    const rLeft = references[left];\n    const rRight = references[right];\n    const qLeft = quantiles[left];\n    const qRight = quantiles[right];\n    if (\n      rLeft === undefined ||\n      rRight === undefined ||\n      qLeft === undefined ||\n      qRight === undefined\n    ) {\n      throw new DeepboxError(\"Internal error: missing quantile interpolation points\");\n    }\n    if (rRight <= rLeft) {\n      return (qLeft + qRight) / 2;\n    }\n    const ratio = (quantile - rLeft) / (rRight - rLeft);\n    return qLeft + ratio * (qRight - qLeft);\n  }\n\n  fitTransform(X: Tensor): Tensor {\n    return this.fit(X).transform(X);\n  }\n}\n\n/**\n * Apply power transform to make data more Gaussian-like.\n *\n * Supports Box-Cox and Yeo-Johnson transforms, with optional standardization.\n *\n * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.PowerTransformer.html | Scikit-learn PowerTransformer}\n */\nexport class PowerTransformer {\n  private fitted = false;\n  private method: \"box-cox\" | \"yeo-johnson\";\n  private lambdas_: number[] | undefined;\n  private standardize: boolean;\n  private mean_: number[] | undefined;\n  private scale_: number[] | undefined;\n\n  /**\n   * Creates a new PowerTransformer.\n   *\n   * @param options - Configuration options\n   * @param options.method - \"box-cox\" or \"yeo-johnson\" (default: \"yeo-johnson\")\n   * @param options.standardize - Whether to standardize transformed features (default: false)\n   * @param options.copy - Accepted for API parity; transforms are always out-of-place (default: true)\n   */\n  constructor(\n    options: {\n      method?: \"box-cox\" | \"yeo-johnson\";\n      standardize?: boolean;\n      copy?: boolean;\n    } = {}\n  ) {\n    this.method = options.method ?? \"yeo-johnson\";\n    if (this.method !== \"box-cox\" && this.method !== \"yeo-johnson\") {\n      throw new InvalidParameterError(\n        \"method must be 'box-cox' or 'yeo-johnson'\",\n        \"method\",\n        this.method\n      );\n    }\n    this.standardize = parseBooleanOption(options.standardize, \"standardize\", false);\n    parseBooleanOption(options.copy, \"copy\", true);\n  }\n\n  fit(X: Tensor): this {\n    if (X.size === 0) {\n      throw new InvalidParameterError(\"X must contain at least one sample\", \"X\");\n    }\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    validateFiniteData(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const data = getNumericData(X, \"X\");\n    const [stride0, stride1] = getStrides2D(X);\n\n    const lambdas = new Array<number>(nFeatures);\n    const means = this.standardize ? new Array<number>(nFeatures).fill(0) : undefined;\n    const scales = this.standardize ? new Array<number>(nFeatures).fill(0) : undefined;\n\n    for (let j = 0; j < nFeatures; j++) {\n      const featureValues = new Array<number>(nSamples);\n      for (let i = 0; i < nSamples; i++) {\n        const raw = data[X.offset + i * stride0 + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        const value = Number(raw);\n        if (this.method === \"box-cox\" && value <= 0) {\n          throw new InvalidParameterError(\n            `Box-Cox requires strictly positive values in fit data (feature ${j})`,\n            \"X\",\n            value\n          );\n        }\n        featureValues[i] = value;\n      }\n      const lambda = this.optimizeLambda(featureValues);\n      lambdas[j] = lambda;\n\n      if (this.standardize && means && scales) {\n        let sum = 0;\n        // Pass 1: Mean\n        for (const value of featureValues) {\n          const transformed =\n            this.method === \"box-cox\"\n              ? this.boxCoxTransformValue(value, lambda)\n              : this.yeoJohnsonTransformValue(value, lambda);\n          sum += transformed;\n        }\n        const mean = sum / nSamples;\n        means[j] = mean;\n\n        // Pass 2: Variance (stable)\n        let sumSqDiff = 0;\n        for (const value of featureValues) {\n          const transformed =\n            this.method === \"box-cox\"\n              ? this.boxCoxTransformValue(value, lambda)\n              : this.yeoJohnsonTransformValue(value, lambda);\n          const diff = transformed - mean;\n          sumSqDiff += diff * diff;\n        }\n        const variance = sumSqDiff / nSamples;\n        const std = Math.sqrt(Math.max(variance, 0));\n        scales[j] = std === 0 ? 1 : std;\n      }\n    }\n\n    this.lambdas_ = lambdas;\n    this.mean_ = this.standardize ? means : undefined;\n    this.scale_ = this.standardize ? scales : undefined;\n    this.fitted = true;\n    return this;\n  }\n\n  transform(X: Tensor): Tensor {\n    if (!this.fitted || !this.lambdas_) {\n      throw new NotFittedError(\"PowerTransformer must be fitted before transform\");\n    }\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    validateFiniteData(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const data = getNumericData(X, \"X\");\n    const [stride0, stride1] = getStrides2D(X);\n\n    if (this.standardize && (!this.mean_ || !this.scale_)) {\n      throw new DeepboxError(\"PowerTransformer internal error: missing standardization stats\");\n    }\n\n    const result = Array.from({ length: nSamples }, () => new Array<number>(nFeatures).fill(0));\n\n    for (let i = 0; i < nSamples; i++) {\n      const rowBase = X.offset + i * stride0;\n      for (let j = 0; j < nFeatures; j++) {\n        const raw = data[rowBase + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        const val = Number(raw);\n        const lambda = this.lambdas_[j];\n        if (lambda === undefined) {\n          throw new DeepboxError(`Internal error: missing fitted lambda for feature ${j}`);\n        }\n\n        let transformed: number;\n        if (this.method === \"box-cox\") {\n          if (val <= 0) {\n            throw new InvalidParameterError(\"Box-Cox requires strictly positive values\", \"X\", val);\n          }\n          transformed = this.boxCoxTransformValue(val, lambda);\n        } else {\n          transformed = this.yeoJohnsonTransformValue(val, lambda);\n        }\n\n        if (this.standardize && this.mean_ && this.scale_) {\n          const mean = this.mean_[j] ?? 0;\n          const scale = this.scale_[j] ?? 1;\n          transformed = (transformed - mean) / scale;\n        }\n\n        const row = result[i];\n        if (row === undefined) {\n          throw new DeepboxError(\"Internal error: result row access failed\");\n        }\n        row[j] = transformed;\n      }\n    }\n\n    return tensor(result, { dtype: \"float64\", device: X.device });\n  }\n\n  /**\n   * Inverse transform data back to the original feature space.\n   * If `standardize=true`, de-standardizes before applying the inverse power transform.\n   *\n   * @param X - Transformed data (2D tensor)\n   * @returns Data in the original feature space\n   * @throws {NotFittedError} If transformer is not fitted\n   */\n  inverseTransform(X: Tensor): Tensor {\n    if (!this.fitted || !this.lambdas_) {\n      throw new NotFittedError(\"PowerTransformer must be fitted before inverse_transform\");\n    }\n    assert2D(X, \"X\");\n    assertNumericTensor(X, \"X\");\n    validateFiniteData(X, \"X\");\n    const [nSamples, nFeatures] = getShape2D(X);\n    const data = getNumericData(X, \"X\");\n    const [stride0, stride1] = getStrides2D(X);\n\n    if (this.standardize && (!this.mean_ || !this.scale_)) {\n      throw new DeepboxError(\"PowerTransformer internal error: missing standardization stats\");\n    }\n\n    const result = Array.from({ length: nSamples }, () => new Array<number>(nFeatures).fill(0));\n\n    for (let i = 0; i < nSamples; i++) {\n      const rowBase = X.offset + i * stride0;\n      for (let j = 0; j < nFeatures; j++) {\n        const raw = data[rowBase + j * stride1];\n        if (raw === undefined) {\n          throw new DeepboxError(\"Internal error: numeric tensor access out of bounds\");\n        }\n        let val = Number(raw);\n\n        if (this.standardize && this.mean_ && this.scale_) {\n          const mean = this.mean_[j] ?? 0;\n          const scale = this.scale_[j] ?? 1;\n          val = val * scale + mean;\n        }\n\n        const lambda = this.lambdas_[j];\n        if (lambda === undefined) {\n          throw new DeepboxError(`Internal error: missing fitted lambda for feature ${j}`);\n        }\n\n        let inverted: number;\n        if (this.method === \"box-cox\") {\n          inverted = this.boxCoxInverseValue(val, lambda);\n        } else {\n          inverted = this.yeoJohnsonInverseValue(val, lambda);\n        }\n\n        const row = result[i];\n        if (row === undefined) {\n          throw new DeepboxError(\"Internal error: result row access failed\");\n        }\n        row[j] = inverted;\n      }\n    }\n\n    return tensor(result, { dtype: \"float64\", device: X.device });\n  }\n\n  private boxCoxTransformValue(value: number, lambda: number): number {\n    return Math.abs(lambda) < 1e-12 ? Math.log(value) : (value ** lambda - 1) / lambda;\n  }\n\n  private yeoJohnsonTransformValue(value: number, lambda: number): number {\n    if (value >= 0) {\n      return Math.abs(lambda) < 1e-12 ? Math.log(value + 1) : ((value + 1) ** lambda - 1) / lambda;\n    }\n    const twoMinusLambda = 2 - lambda;\n    return Math.abs(twoMinusLambda) < 1e-12\n      ? -Math.log(1 - value)\n      : -((1 - value) ** twoMinusLambda - 1) / twoMinusLambda;\n  }\n\n  private boxCoxInverseValue(value: number, lambda: number): number {\n    if (Math.abs(lambda) < 1e-12) {\n      return Math.exp(value);\n    }\n    const base = value * lambda + 1;\n    if (base <= 0) {\n      throw new InvalidParameterError(\"Box-Cox inverse encountered invalid value\", \"X\", value);\n    }\n    return base ** (1 / lambda);\n  }\n\n  private yeoJohnsonInverseValue(value: number, lambda: number): number {\n    if (value >= 0) {\n      if (Math.abs(lambda) < 1e-12) {\n        return Math.exp(value) - 1;\n      }\n      const base = value * lambda + 1;\n      if (base <= 0) {\n        throw new InvalidParameterError(\n          \"Yeo-Johnson inverse encountered invalid value\",\n          \"X\",\n          value\n        );\n      }\n      return base ** (1 / lambda) - 1;\n    }\n\n    const twoMinusLambda = 2 - lambda;\n    if (Math.abs(twoMinusLambda) < 1e-12) {\n      return 1 - Math.exp(-value);\n    }\n    const base = 1 - value * twoMinusLambda;\n    if (base <= 0) {\n      throw new InvalidParameterError(\"Yeo-Johnson inverse encountered invalid value\", \"X\", value);\n    }\n    return 1 - base ** (1 / twoMinusLambda);\n  }\n\n  private logLikelihood(values: readonly number[], lambda: number): number {\n    const transformed = new Array<number>(values.length);\n    let jacobian = 0;\n\n    for (let i = 0; i < values.length; i++) {\n      const value = values[i];\n      if (value === undefined) {\n        throw new DeepboxError(\"Internal error: missing feature value during optimization\");\n      }\n\n      let transformedValue: number;\n      if (this.method === \"box-cox\") {\n        if (value <= 0) {\n          return Number.NEGATIVE_INFINITY;\n        }\n        transformedValue = this.boxCoxTransformValue(value, lambda);\n        jacobian += (lambda - 1) * Math.log(value);\n      } else {\n        transformedValue = this.yeoJohnsonTransformValue(value, lambda);\n        jacobian +=\n          value >= 0 ? (lambda - 1) * Math.log(value + 1) : (1 - lambda) * Math.log(1 - value);\n      }\n\n      if (!Number.isFinite(transformedValue)) {\n        return Number.NEGATIVE_INFINITY;\n      }\n      transformed[i] = transformedValue;\n    }\n\n    let sum = 0;\n    for (const value of transformed) {\n      sum += value;\n    }\n    const mean = sum / transformed.length;\n\n    let varianceSum = 0;\n    for (const value of transformed) {\n      const delta = value - mean;\n      varianceSum += delta * delta;\n    }\n    const variance = varianceSum / transformed.length;\n    if (!Number.isFinite(variance) || variance <= 1e-15) {\n      return Number.NEGATIVE_INFINITY;\n    }\n\n    return -0.5 * transformed.length * Math.log(variance) + jacobian;\n  }\n\n  private optimizeLambda(values: readonly number[]): number {\n    if (values.length < 2) {\n      return 1;\n    }\n\n    let minValue = Number.POSITIVE_INFINITY;\n    let maxValue = Number.NEGATIVE_INFINITY;\n    for (const value of values) {\n      if (value < minValue) minValue = value;\n      if (value > maxValue) maxValue = value;\n    }\n\n    if (!Number.isFinite(minValue) || !Number.isFinite(maxValue) || maxValue - minValue <= 1e-15) {\n      return 1;\n    }\n\n    let left = -5;\n    let right = 5;\n    const phi = (Math.sqrt(5) - 1) / 2;\n    let c = right - phi * (right - left);\n    let d = left + phi * (right - left);\n    let fc = this.logLikelihood(values, c);\n    let fd = this.logLikelihood(values, d);\n\n    for (let iter = 0; iter < 80; iter++) {\n      if (Math.abs(right - left) < 1e-6) break;\n      if (fc > fd) {\n        right = d;\n        d = c;\n        fd = fc;\n        c = right - phi * (right - left);\n        fc = this.logLikelihood(values, c);\n      } else {\n        left = c;\n        c = d;\n        fc = fd;\n        d = left + phi * (right - left);\n        fd = this.logLikelihood(values, d);\n      }\n    }\n\n    const candidates = [left, right, (left + right) / 2, 0, 1, 2, -2];\n    let bestLambda = 1;\n    let bestScore = Number.NEGATIVE_INFINITY;\n    for (const lambda of candidates) {\n      const score = this.logLikelihood(values, lambda);\n      if (score > bestScore) {\n        bestScore = score;\n        bestLambda = lambda;\n      }\n    }\n\n    return Number.isFinite(bestLambda) ? bestLambda : 1;\n  }\n\n  fitTransform(X: Tensor): Tensor {\n    return this.fit(X).transform(X);\n  }\n}\n","import { DeepboxError, InvalidParameterError, ShapeError } from \"../core/errors\";\nimport { type Tensor, zeros } from \"../ndarray\";\nimport { createSeededRandom, getShape2D, shuffleIndicesInPlace } from \"./_internal\";\n\nfunction validateNSplits(nSplits: number): void {\n  if (!Number.isFinite(nSplits) || !Number.isInteger(nSplits) || nSplits < 2) {\n    throw new InvalidParameterError(\"nSplits must be an integer at least 2\", \"nSplits\", nSplits);\n  }\n}\n\ntype SplitSpec = {\n  kind: \"fraction\" | \"count\";\n  value: number;\n};\n\nfunction parseSplitSpec(value: number | undefined, name: string): SplitSpec | undefined {\n  if (value === undefined) {\n    return undefined;\n  }\n  if (!Number.isFinite(value) || value <= 0) {\n    throw new InvalidParameterError(`${name} must be a positive number`, name, value);\n  }\n  if (value < 1) {\n    return { kind: \"fraction\", value };\n  }\n  if (!Number.isInteger(value)) {\n    throw new InvalidParameterError(\n      `${name} must be an integer when provided as an absolute size`,\n      name,\n      value\n    );\n  }\n  return { kind: \"count\", value };\n}\n\nfunction resolveSplitCount(spec: SplitSpec, nSamples: number, isTrain: boolean): number {\n  if (spec.kind === \"count\") {\n    return spec.value;\n  }\n  const exact = nSamples * spec.value;\n  return isTrain ? Math.floor(exact) : Math.ceil(exact);\n}\n\nfunction resolveTrainTestCounts(\n  nSamples: number,\n  trainSize: number | undefined,\n  testSize: number | undefined\n): [number, number] {\n  const defaultTestSize = trainSize === undefined && testSize === undefined ? 0.25 : testSize;\n  const trainSpec = parseSplitSpec(trainSize, \"trainSize\");\n  const testSpec = parseSplitSpec(defaultTestSize, \"testSize\");\n\n  if (trainSpec?.kind === \"count\" && trainSpec.value > nSamples) {\n    throw new InvalidParameterError(\n      \"trainSize must not exceed number of samples\",\n      \"trainSize\",\n      trainSpec.value\n    );\n  }\n  if (testSpec?.kind === \"count\" && testSpec.value > nSamples) {\n    throw new InvalidParameterError(\n      \"testSize must not exceed number of samples\",\n      \"testSize\",\n      testSpec.value\n    );\n  }\n\n  if (\n    trainSpec?.kind === \"fraction\" &&\n    testSpec?.kind === \"fraction\" &&\n    trainSpec.value + testSpec.value > 1\n  ) {\n    throw new InvalidParameterError(\n      \"trainSize and testSize fractions must sum to at most 1\",\n      \"trainSize\",\n      trainSpec.value\n    );\n  }\n\n  let nTrain = trainSpec === undefined ? undefined : resolveSplitCount(trainSpec, nSamples, true);\n  let nTest = testSpec === undefined ? undefined : resolveSplitCount(testSpec, nSamples, false);\n\n  if (nTrain === undefined && nTest === undefined) {\n    throw new DeepboxError(\"Internal error: failed to resolve split sizes\");\n  }\n\n  if (nTrain === undefined) {\n    nTrain = nSamples - (nTest ?? 0);\n  }\n  if (nTest === undefined) {\n    nTest = nSamples - nTrain;\n  }\n\n  if (nTrain + nTest > nSamples) {\n    throw new InvalidParameterError(\n      \"trainSize and testSize exceed number of samples\",\n      \"trainSize\",\n      trainSize\n    );\n  }\n\n  if (nTrain < 1) {\n    throw new InvalidParameterError(\"trainSize must be at least 1 sample\", \"trainSize\", trainSize);\n  }\n  if (nTest < 1) {\n    throw new InvalidParameterError(\"testSize must be at least 1 sample\", \"testSize\", testSize);\n  }\n\n  return [nTrain, nTest];\n}\n\nfunction compareLabels(a: unknown, b: unknown): number {\n  if (typeof a === \"number\" && typeof b === \"number\") return a - b;\n  if (typeof a === \"bigint\" && typeof b === \"bigint\") {\n    if (a < b) return -1;\n    if (a > b) return 1;\n    return 0;\n  }\n  return String(a).localeCompare(String(b));\n}\n\nfunction makeFoldSizes(total: number, nSplits: number): number[] {\n  const base = Math.floor(total / nSplits);\n  const remainder = total % nSplits;\n  return Array.from({ length: nSplits }, (_, i) => base + (i < remainder ? 1 : 0));\n}\n\nfunction readTensorValue(t: Tensor, indices: number[]): string | number | bigint {\n  const value = t.at(...indices);\n  if (typeof value === \"string\" || typeof value === \"number\" || typeof value === \"bigint\") {\n    return value;\n  }\n  throw new DeepboxError(\"Internal error: unsupported tensor value type\");\n}\n\nfunction writeTensorValue(t: Tensor, flatIndex: number, value: string | number | bigint): void {\n  if (t.dtype === \"string\") {\n    if (typeof value !== \"string\") {\n      throw new DeepboxError(\"Internal error: expected string value for string tensor\");\n    }\n    t.data[flatIndex] = value;\n    return;\n  }\n\n  if (typeof value === \"string\") {\n    throw new DeepboxError(\"Internal error: encountered string value in numeric tensor\");\n  }\n\n  if (t.data instanceof BigInt64Array) {\n    t.data[flatIndex] = typeof value === \"bigint\" ? value : BigInt(value);\n    return;\n  }\n\n  t.data[flatIndex] = Number(value);\n}\n\nfunction takeRows2D(X: Tensor, sampleIndices: number[]): Tensor {\n  const [, nFeatures] = getShape2D(X);\n  const out = zeros([sampleIndices.length, nFeatures], { dtype: X.dtype });\n\n  for (let i = 0; i < sampleIndices.length; i++) {\n    const sampleIndex = sampleIndices[i];\n    if (sampleIndex === undefined) {\n      throw new DeepboxError(\"Internal error: sample index access failed\");\n    }\n    for (let j = 0; j < nFeatures; j++) {\n      const value = readTensorValue(X, [sampleIndex, j]);\n      writeTensorValue(out, out.offset + i * nFeatures + j, value);\n    }\n  }\n\n  return out;\n}\n\nfunction takeVector(y: Tensor, sampleIndices: number[]): Tensor {\n  if (y.ndim !== 1) {\n    throw new ShapeError(`y must be a 1D tensor, got ${y.ndim}D`);\n  }\n  const out = zeros([sampleIndices.length], { dtype: y.dtype });\n\n  for (let i = 0; i < sampleIndices.length; i++) {\n    const sampleIndex = sampleIndices[i];\n    if (sampleIndex === undefined) {\n      throw new DeepboxError(\"Internal error: sample index access failed\");\n    }\n    const value = readTensorValue(y, [sampleIndex]);\n    writeTensorValue(out, out.offset + i, value);\n  }\n\n  return out;\n}\n\n/**\n * Split arrays into random train and test subsets.\n *\n * @param X - Feature matrix (2D tensor)\n * @param y - Optional target labels (1D tensor)\n * @param options - Split configuration options\n * @param options.testSize - Proportion or absolute number of test samples\n * @param options.trainSize - Proportion or absolute number of train samples\n * @param options.randomState - Random seed\n * @param options.shuffle - Whether to shuffle data before splitting\n * @param options.stratify - If not undefined, data is split in stratified fashion using this as class labels\n *\n * @example\n * ```js\n * import { trainTestSplit } from 'deepbox/preprocess';\n * import { tensor } from 'deepbox/ndarray';\n *\n * const X = tensor([[1, 2], [3, 4], [5, 6], [7, 8]]);\n * const y = tensor([0, 1, 0, 1]);\n * const [XTrain, XTest, yTrain, yTest] = trainTestSplit(X, y, { testSize: 0.25 });\n * ```\n *\n * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.train_test_split.html | Scikit-learn train_test_split}\n */\nexport function trainTestSplit(\n  X: Tensor,\n  y?: Tensor,\n  options?: {\n    testSize?: number;\n    trainSize?: number;\n    randomState?: number;\n    shuffle?: boolean;\n    stratify?: Tensor;\n  }\n): Tensor[] {\n  const opts = options ?? {};\n  const shuffle = opts.shuffle ?? true;\n  const randomState = opts.randomState;\n\n  const [nSamples] = getShape2D(X);\n\n  if (nSamples === 0) {\n    throw new InvalidParameterError(\"Cannot split empty array\", \"X\");\n  }\n\n  if (y) {\n    const yShape0 = y.shape[0];\n    if (yShape0 === undefined || yShape0 !== nSamples) {\n      throw new InvalidParameterError(\"X and y must have same number of samples\", \"y\", yShape0);\n    }\n  }\n\n  if (opts.stratify) {\n    if (opts.stratify.ndim !== 1) {\n      throw new ShapeError(`stratify must be a 1D tensor, got ${opts.stratify.ndim}D`);\n    }\n    const stratifyShape0 = opts.stratify.shape[0];\n    if (stratifyShape0 === undefined || stratifyShape0 !== nSamples) {\n      throw new InvalidParameterError(\n        \"stratify must have same number of samples as X\",\n        \"stratify\",\n        stratifyShape0\n      );\n    }\n  }\n\n  const [nTrain, nTest] = resolveTrainTestCounts(nSamples, opts.trainSize, opts.testSize);\n\n  const indices = Array.from({ length: nSamples }, (_, i) => i);\n  const random = randomState !== undefined ? createSeededRandom(randomState) : Math.random;\n  const maybeShuffle = (arr: number[]): void => {\n    if (!shuffle) return;\n    shuffleIndicesInPlace(arr, random);\n  };\n\n  let trainIndices: number[] = [];\n  let testIndices: number[] = [];\n\n  if (opts.stratify) {\n    const stratify = opts.stratify;\n    const labelMap = new Map<unknown, number[]>();\n    for (let i = 0; i < nSamples; i++) {\n      const label = readTensorValue(stratify, [i]);\n      let bucket = labelMap.get(label);\n      if (bucket === undefined) {\n        bucket = [];\n        labelMap.set(label, bucket);\n      }\n      bucket.push(i);\n    }\n\n    const labels = Array.from(labelMap.keys()).sort(compareLabels);\n    const nClasses = labels.length;\n    const classSizes = labels.map((label) => labelMap.get(label)?.length ?? 0);\n    const hasSingleton = classSizes.some((size) => size < 2);\n\n    if (hasSingleton && shuffle && randomState === undefined) {\n      throw new InvalidParameterError(\n        \"stratify requires at least 2 samples per class\",\n        \"stratify\",\n        classSizes\n      );\n    }\n\n    if (opts.trainSize !== undefined && nTrain < nClasses) {\n      throw new InvalidParameterError(\n        \"trainSize must be at least the number of classes when stratifying\",\n        \"trainSize\",\n        nTrain\n      );\n    }\n    if (nTest < nClasses) {\n      throw new InvalidParameterError(\n        \"testSize must be at least the number of classes when stratifying\",\n        \"testSize\",\n        nTest\n      );\n    }\n\n    const testFraction = nTest / nSamples;\n    const allowEmptyClassSplits = nTrain < nClasses;\n    const counts = labels.map((label) => {\n      const size = labelMap.get(label)?.length ?? 0;\n      const exact = size * testFraction;\n      let testCount = Math.floor(exact);\n      let remainder = exact - testCount;\n      let min = allowEmptyClassSplits ? 0 : 1;\n      let max = allowEmptyClassSplits ? size : size - 1;\n      if (size < 2) {\n        min = 0;\n        max = allowEmptyClassSplits ? size : 0;\n        testCount = 0;\n        remainder = 0;\n      } else {\n        if (testCount < min) testCount = min;\n        if (testCount > max) testCount = max;\n      }\n      return { label, size, testCount, remainder, min, max };\n    });\n\n    let remaining = nTest - counts.reduce((sum, c) => sum + c.testCount, 0);\n    if (remaining !== 0) {\n      const order =\n        remaining > 0\n          ? [...counts].sort((a, b) => {\n              if (b.remainder !== a.remainder) return b.remainder - a.remainder;\n              return compareLabels(a.label, b.label);\n            })\n          : [...counts].sort((a, b) => {\n              if (a.remainder !== b.remainder) return a.remainder - b.remainder;\n              return compareLabels(a.label, b.label);\n            });\n\n      let guard = 0;\n      while (remaining !== 0 && guard < counts.length * 2) {\n        for (const entry of order) {\n          if (remaining === 0) break;\n          if (remaining > 0 && entry.testCount < entry.max) {\n            entry.testCount += 1;\n            remaining -= 1;\n          } else if (remaining < 0 && entry.testCount > entry.min) {\n            entry.testCount -= 1;\n            remaining += 1;\n          }\n        }\n        guard += 1;\n      }\n\n      if (remaining !== 0) {\n        throw new DeepboxError(\"Internal error: unable to allocate stratified split sizes\");\n      }\n    }\n\n    const remainingTrainPool: number[] = [];\n    for (const entry of counts) {\n      const labelIndices = [...(labelMap.get(entry.label) ?? [])];\n      maybeShuffle(labelIndices);\n      testIndices.push(...labelIndices.slice(0, entry.testCount));\n      remainingTrainPool.push(...labelIndices.slice(entry.testCount));\n    }\n\n    maybeShuffle(testIndices);\n    maybeShuffle(remainingTrainPool);\n    trainIndices = remainingTrainPool.slice(0, nTrain);\n  } else {\n    maybeShuffle(indices);\n    trainIndices = indices.slice(0, nTrain);\n    testIndices = indices.slice(nTrain, nTrain + nTest);\n  }\n\n  if (trainIndices.length !== nTrain || testIndices.length !== nTest) {\n    throw new DeepboxError(\"Internal error: resolved split indices do not match requested sizes\");\n  }\n\n  const XTrain = takeRows2D(X, trainIndices);\n  const XTest = takeRows2D(X, testIndices);\n\n  if (y) {\n    const yTrain = takeVector(y, trainIndices);\n    const yTest = takeVector(y, testIndices);\n    return [XTrain, XTest, yTrain, yTest];\n  }\n\n  return [XTrain, XTest];\n}\n\n/**\n * K-Folds cross-validator.\n *\n * Provides train/test indices to split data in train/test sets.\n *\n * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.KFold.html | Scikit-learn KFold}\n */\nexport class KFold {\n  private nSplits: number;\n  private shuffle: boolean;\n  private randomState: number | undefined;\n\n  constructor(\n    options: {\n      nSplits?: number;\n      shuffle?: boolean;\n      randomState?: number;\n    } = {}\n  ) {\n    this.nSplits = options.nSplits ?? 5;\n    this.shuffle = options.shuffle ?? false;\n    this.randomState = options.randomState;\n  }\n\n  split(X: Tensor): Array<[number[], number[]]> {\n    const shape0 = X.shape[0];\n    if (shape0 === undefined) {\n      throw new ShapeError(\"X must have valid shape[0]\");\n    }\n    const nSamples = shape0;\n    validateNSplits(this.nSplits);\n    if (this.nSplits > nSamples) {\n      throw new InvalidParameterError(\n        \"nSplits must not be greater than number of samples\",\n        \"nSplits\",\n        this.nSplits\n      );\n    }\n    const indices = Array.from({ length: nSamples }, (_, i) => i);\n\n    if (this.shuffle) {\n      const random =\n        this.randomState !== undefined ? createSeededRandom(this.randomState) : Math.random;\n      shuffleIndicesInPlace(indices, random);\n    }\n\n    const splits: Array<[number[], number[]]> = [];\n    const foldSizes = makeFoldSizes(nSamples, this.nSplits);\n    let current = 0;\n\n    for (let i = 0; i < this.nSplits; i++) {\n      const foldSize = foldSizes[i] ?? 0;\n      const testStart = current;\n      const testEnd = current + foldSize;\n\n      const testIndices = indices.slice(testStart, testEnd);\n      const trainIndices = [...indices.slice(0, testStart), ...indices.slice(testEnd)];\n\n      splits.push([trainIndices, testIndices]);\n      current = testEnd;\n    }\n\n    return splits;\n  }\n\n  getNSplits(): number {\n    return this.nSplits;\n  }\n}\n\n/**\n * Stratified K-Folds cross-validator.\n *\n * Provides train/test indices while preserving class distribution.\n *\n * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.StratifiedKFold.html | Scikit-learn StratifiedKFold}\n */\nexport class StratifiedKFold {\n  private nSplits: number;\n  private shuffle: boolean;\n  private randomState: number | undefined;\n\n  constructor(\n    options: {\n      nSplits?: number;\n      shuffle?: boolean;\n      randomState?: number;\n    } = {}\n  ) {\n    this.nSplits = options.nSplits ?? 5;\n    this.shuffle = options.shuffle ?? false;\n    this.randomState = options.randomState;\n  }\n\n  split(X: Tensor, y: Tensor): Array<[number[], number[]]> {\n    const shape0 = X.shape[0];\n    if (shape0 === undefined) {\n      throw new ShapeError(\"X must have valid shape[0]\");\n    }\n    const nSamples = shape0;\n    validateNSplits(this.nSplits);\n    if (this.nSplits > nSamples) {\n      throw new InvalidParameterError(\n        \"nSplits must not be greater than number of samples\",\n        \"nSplits\",\n        this.nSplits\n      );\n    }\n    const yShape0 = y.shape[0];\n    if (yShape0 === undefined || yShape0 !== nSamples) {\n      throw new InvalidParameterError(\"X and y must have same number of samples\", \"y\", yShape0);\n    }\n    if (y.ndim !== 1) {\n      throw new ShapeError(`y must be a 1D tensor, got ${y.ndim}D`);\n    }\n    const labelMap = new Map<string | number | bigint, number[]>();\n    const random =\n      this.randomState !== undefined ? createSeededRandom(this.randomState) : Math.random;\n\n    for (let i = 0; i < nSamples; i++) {\n      const label = readTensorValue(y, [i]);\n      let bucket = labelMap.get(label);\n      if (bucket === undefined) {\n        bucket = [];\n        labelMap.set(label, bucket);\n      }\n      bucket.push(i);\n    }\n\n    for (const [label, indices] of labelMap.entries()) {\n      if (this.shuffle) {\n        shuffleIndicesInPlace(indices, random);\n      }\n      if (indices.length < this.nSplits) {\n        throw new InvalidParameterError(\n          `Each class must have at least nSplits samples; class ${label} has ${indices.length}`,\n          \"nSplits\",\n          this.nSplits\n        );\n      }\n    }\n\n    const foldIndices: number[][] = Array.from({ length: this.nSplits }, () => []);\n\n    for (const indices of labelMap.values()) {\n      const foldSizes = makeFoldSizes(indices.length, this.nSplits);\n      let start = 0;\n      for (let fold = 0; fold < this.nSplits; fold++) {\n        const size = foldSizes[fold] ?? 0;\n        const end = start + size;\n        const target = foldIndices[fold];\n        if (!target) {\n          throw new DeepboxError(\"Internal error: stratified fold storage missing\");\n        }\n        target.push(...indices.slice(start, end));\n        start = end;\n      }\n    }\n\n    const splits: Array<[number[], number[]]> = [];\n\n    for (let fold = 0; fold < this.nSplits; fold++) {\n      const testIndices = foldIndices[fold] ?? [];\n      const trainIndices: number[] = [];\n      for (let other = 0; other < this.nSplits; other++) {\n        if (other === fold) continue;\n        trainIndices.push(...(foldIndices[other] ?? []));\n      }\n      splits.push([trainIndices, testIndices]);\n    }\n\n    return splits;\n  }\n\n  getNSplits(): number {\n    return this.nSplits;\n  }\n}\n\n/**\n * Group K-Fold cross-validator.\n *\n * Ensures same group is not in both train and test.\n *\n * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.GroupKFold.html | Scikit-learn GroupKFold}\n */\nexport class GroupKFold {\n  private nSplits: number;\n\n  constructor(options: { nSplits?: number } = {}) {\n    this.nSplits = options.nSplits ?? 5;\n  }\n\n  split(X: Tensor, _y: Tensor | undefined, groups: Tensor): Array<[number[], number[]]> {\n    const shape0 = X.shape[0];\n    if (shape0 === undefined) {\n      throw new ShapeError(\"X must have valid shape[0]\");\n    }\n    const nSamples = shape0;\n    validateNSplits(this.nSplits);\n    if (groups.ndim !== 1) {\n      throw new ShapeError(`groups must be a 1D tensor, got ${groups.ndim}D`);\n    }\n    const groupsShape0 = groups.shape[0];\n    if (groupsShape0 === undefined || groupsShape0 !== nSamples) {\n      throw new InvalidParameterError(\n        \"X and groups must have same number of samples\",\n        \"groups\",\n        groupsShape0\n      );\n    }\n    const groupMap = new Map<string | number | bigint, number[]>();\n\n    for (let i = 0; i < nSamples; i++) {\n      const group = readTensorValue(groups, [i]);\n      let bucket = groupMap.get(group);\n      if (bucket === undefined) {\n        bucket = [];\n        groupMap.set(group, bucket);\n      }\n      bucket.push(i);\n    }\n\n    const groupEntries = Array.from(groupMap.entries()).map(([group, indices]) => ({\n      group,\n      indices,\n      size: indices.length,\n    }));\n    if (this.nSplits > groupEntries.length) {\n      throw new InvalidParameterError(\n        \"Number of groups must be at least nSplits\",\n        \"nSplits\",\n        this.nSplits\n      );\n    }\n    groupEntries.sort((a, b) => {\n      if (b.size !== a.size) return b.size - a.size;\n      return compareLabels(a.group, b.group);\n    });\n\n    const foldIndices: number[][] = Array.from({ length: this.nSplits }, () => []);\n    const foldSizes = new Array<number>(this.nSplits).fill(0);\n\n    for (const entry of groupEntries) {\n      let bestFold = 0;\n      let bestSize = foldSizes[0] ?? 0;\n      for (let fold = 1; fold < this.nSplits; fold++) {\n        const size = foldSizes[fold] ?? 0;\n        if (size < bestSize) {\n          bestSize = size;\n          bestFold = fold;\n        }\n      }\n      const target = foldIndices[bestFold];\n      if (!target) {\n        throw new DeepboxError(\"Internal error: group fold storage missing\");\n      }\n      target.push(...entry.indices);\n      foldSizes[bestFold] = bestSize + entry.size;\n    }\n\n    const splits: Array<[number[], number[]]> = [];\n    for (let fold = 0; fold < this.nSplits; fold++) {\n      const testIndices = foldIndices[fold] ?? [];\n      const trainIndices: number[] = [];\n      for (let other = 0; other < this.nSplits; other++) {\n        if (other === fold) continue;\n        trainIndices.push(...(foldIndices[other] ?? []));\n      }\n      splits.push([trainIndices, testIndices]);\n    }\n\n    return splits;\n  }\n\n  getNSplits(): number {\n    return this.nSplits;\n  }\n}\n\n/**\n * Leave-One-Out cross-validator.\n *\n * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.LeaveOneOut.html | Scikit-learn LeaveOneOut}\n */\nexport class LeaveOneOut {\n  split(X: Tensor): Array<[number[], number[]]> {\n    const shape0 = X.shape[0];\n    if (shape0 === undefined) {\n      throw new ShapeError(\"X must have valid shape[0]\");\n    }\n    const nSamples = shape0;\n    const splits: Array<[number[], number[]]> = [];\n\n    for (let i = 0; i < nSamples; i++) {\n      const trainIndices = [\n        ...Array.from({ length: i }, (_, j) => j),\n        ...Array.from({ length: nSamples - i - 1 }, (_, j) => i + 1 + j),\n      ];\n      const testIndices = [i];\n      splits.push([trainIndices, testIndices]);\n    }\n\n    return splits;\n  }\n\n  getNSplits(X: Tensor): number {\n    const shape0 = X.shape[0];\n    if (shape0 === undefined) {\n      throw new ShapeError(\"X must have valid shape[0]\");\n    }\n    return shape0;\n  }\n}\n\n/**\n * Leave-P-Out cross-validator.\n *\n * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.LeavePOut.html | Scikit-learn LeavePOut}\n */\nexport class LeavePOut {\n  private p: number;\n\n  constructor(p: number) {\n    if (!Number.isFinite(p) || !Number.isInteger(p) || p <= 0) {\n      throw new InvalidParameterError(\"p must be a positive integer\", \"p\", p);\n    }\n    this.p = p;\n  }\n\n  split(X: Tensor): Array<[number[], number[]]> {\n    const shape0 = X.shape[0];\n    if (shape0 === undefined) {\n      throw new ShapeError(\"X must have valid shape[0]\");\n    }\n    const nSamples = shape0;\n    if (this.p > nSamples) {\n      throw new InvalidParameterError(\"p must not be greater than number of samples\", \"p\", this.p);\n    }\n\n    // Calculate number of combinations to prevent memory explosion\n    let nCombos = 1;\n    const k = this.p > nSamples / 2 ? nSamples - this.p : this.p;\n    for (let i = 0; i < k; i++) {\n      nCombos = (nCombos * (nSamples - i)) / (i + 1);\n    }\n\n    // Safety limit: 100,000 splits is generous for in-memory JS arrays\n    // For larger splits, a generator approach would be needed, but split() returns Array.\n    if (nCombos > 100000) {\n      throw new InvalidParameterError(\n        `LeavePOut produces ${Math.floor(nCombos)} splits, which exceeds memory safety limit of 100,000`,\n        \"p\",\n        this.p\n      );\n    }\n\n    const splits: Array<[number[], number[]]> = [];\n    const allIndices = Array.from({ length: nSamples }, (_, i) => i);\n\n    // Iterative combination generator\n    const combine = (start: number, currentCombo: number[]) => {\n      if (currentCombo.length === this.p) {\n        const testSet = new Set(currentCombo);\n        const testIndices = [...currentCombo];\n        const trainIndices = allIndices.filter((i) => !testSet.has(i));\n        splits.push([trainIndices, testIndices]);\n        return;\n      }\n      for (let i = start; i < nSamples; i++) {\n        currentCombo.push(i);\n        combine(i + 1, currentCombo);\n        currentCombo.pop();\n      }\n    };\n\n    combine(0, []);\n\n    return splits;\n  }\n\n  getNSplits(X: Tensor): number {\n    const shape0 = X.shape[0];\n    if (shape0 === undefined) {\n      throw new ShapeError(\"X must have valid shape[0]\");\n    }\n    const n = shape0;\n    if (this.p > n) {\n      throw new InvalidParameterError(\"p must not be greater than number of samples\", \"p\", this.p);\n    }\n    // C(n, p) = n! / (p! * (n-p)!)\n    let result = 1;\n    const k = this.p > n / 2 ? n - this.p : this.p;\n    for (let i = 0; i < k; i++) {\n      result = (result * (n - i)) / (i + 1);\n    }\n    return Math.round(result);\n  }\n}\n"]}