agentdb 2.0.0-alpha.2 → 2.0.0-alpha.2.1

package/dist/agentdb-advanced.js

@@ -1,2110 +0,0 @@
-
-/**
- * AgentDB v2.0.0-alpha.2 - Advanced Browser Bundle
- *
- * Complete feature set for browser environments:
- * - v1 API backward compatible
- * - v2 enhanced API (episodes, skills, causal_edges)
- * - Product Quantization (PQ8/PQ16/PQ32) - 4-32x compression
- * - HNSW Indexing - 10-20x faster search
- * - Graph Neural Networks - Graph attention & message passing
- * - MMR Diversity - Maximal marginal relevance ranking
- * - Tensor Compression - SVD dimension reduction
- * - Batch Operations - Optimized vector processing
- * - IndexedDB persistence
- * - Cross-tab synchronization
- *
- * Bundle Size: ~90 KB raw (~31 KB gzipped)
- * License: MIT
- */
-
-(function(global) {
-  'use strict';
-
-  // ============================================================================
-  // sql.js WASM Loader
-  // ============================================================================
-
-  
-// We are modularizing this manually because the current modularize setting in Emscripten has some issues:
-// https://github.com/kripken/emscripten/issues/5820
-// In addition, When you use emcc's modularization, it still expects to export a global object called `Module`,
-// which is able to be used/called before the WASM is loaded.
-// The modularization below exports a promise that loads and resolves to the actual sql.js module.
-// That way, this module can't be used before the WASM is finished loading.
-
-// We are going to define a function that a user will call to start loading initializing our Sql.js library
-// However, that function might be called multiple times, and on subsequent calls, we don't actually want it to instantiate a new instance of the Module
-// Instead, we want to return the previously loaded module
-
-// TODO: Make this not declare a global if used in the browser
-var initSqlJsPromise = undefined;
-
-var initSqlJs = function (moduleConfig) {
-
-    if (initSqlJsPromise){
-      return initSqlJsPromise;
-    }
-    // If we're here, we've never called this function before
-    initSqlJsPromise = new Promise(function (resolveModule, reject) {
-
-        // We are modularizing this manually because the current modularize setting in Emscripten has some issues:
-        // https://github.com/kripken/emscripten/issues/5820
-
-        // The way to affect the loading of emcc compiled modules is to create a variable called `Module` and add
-        // properties to it, like `preRun`, `postRun`, etc
-        // We are using that to get notified when the WASM has finished loading.
-        // Only then will we return our promise
-
-        // If they passed in a moduleConfig object, use that
-        // Otherwise, initialize Module to the empty object
-        var Module = typeof moduleConfig !== 'undefined' ? moduleConfig : {};
-
-        // EMCC only allows for a single onAbort function (not an array of functions)
-        // So if the user defined their own onAbort function, we remember it and call it
-        var originalOnAbortFunction = Module['onAbort'];
-        Module['onAbort'] = function (errorThatCausedAbort) {
-            reject(new Error(errorThatCausedAbort));
-            if (originalOnAbortFunction){
-              originalOnAbortFunction(errorThatCausedAbort);
-            }
-        };
-
-        Module['postRun'] = Module['postRun'] || [];
-        Module['postRun'].push(function () {
-            // When Emscripted calls postRun, this promise resolves with the built Module
-            resolveModule(Module);
-        });
-
-        // There is a section of code in the emcc-generated code below that looks like this:
-        // (Note that this is lowercase `module`)
-        // if (typeof module !== 'undefined') {
-        //     module['exports'] = Module;
-        // }
-        // When that runs, it's going to overwrite our own modularization export efforts in shell-post.js!
-        // The only way to tell emcc not to emit it is to pass the MODULARIZE=1 or MODULARIZE_INSTANCE=1 flags,
-        // but that carries with it additional unnecessary baggage/bugs we don't want either.
-        // So, we have three options:
-        // 1) We undefine `module`
-        // 2) We remember what `module['exports']` was at the beginning of this function and we restore it later
-        // 3) We write a script to remove those lines of code as part of the Make process.
-        //
-        // Since those are the only lines of code that care about module, we will undefine it. It's the most straightforward
-        // of the options, and has the side effect of reducing emcc's efforts to modify the module if its output were to change in the future.
-        // That's a nice side effect since we're handling the modularization efforts ourselves
-        module = undefined;
-
-        // The emcc-generated code and shell-post.js code goes below,
-        // meaning that all of it runs inside of this promise. If anything throws an exception, our promise will abort
-var f;f||=typeof Module != 'undefined' ? Module : {};var aa="object"==typeof window,ba="undefined"!=typeof WorkerGlobalScope,ca="object"==typeof process&&"object"==typeof process.versions&&"string"==typeof process.versions.node&&"renderer"!=process.type;"use strict";
-f.onRuntimeInitialized=function(){function a(g,l){switch(typeof l){case "boolean":dc(g,l?1:0);break;case "number":ec(g,l);break;case "string":fc(g,l,-1,-1);break;case "object":if(null===l)lb(g);else if(null!=l.length){var n=da(l,ea);gc(g,n,l.length,-1);fa(n)}else va(g,"Wrong API use : tried to return a value of an unknown type ("+l+").",-1);break;default:lb(g)}}function b(g,l){for(var n=[],r=0;r<g;r+=1){var t=m(l+4*r,"i32"),y=hc(t);if(1===y||2===y)t=ic(t);else if(3===y)t=jc(t);else if(4===y){y=t;
-t=kc(y);y=lc(y);for(var L=new Uint8Array(t),J=0;J<t;J+=1)L[J]=p[y+J];t=L}else t=null;n.push(t)}return n}function c(g,l){this.Qa=g;this.db=l;this.Oa=1;this.lb=[]}function d(g,l){this.db=l;l=ha(g)+1;this.eb=ia(l);if(null===this.eb)throw Error("Unable to allocate memory for the SQL string");u(g,x,this.eb,l);this.kb=this.eb;this.Za=this.pb=null}function e(g){this.filename="dbfile_"+(4294967295*Math.random()>>>0);if(null!=g){var l=this.filename,n="/",r=l;n&&(n="string"==typeof n?n:ja(n),r=l?ka(n+"/"+l):
-n);l=la(!0,!0);r=ma(r,l);if(g){if("string"==typeof g){n=Array(g.length);for(var t=0,y=g.length;t<y;++t)n[t]=g.charCodeAt(t);g=n}na(r,l|146);n=oa(r,577);pa(n,g,0,g.length,0);qa(n);na(r,l)}}this.handleError(q(this.filename,h));this.db=m(h,"i32");ob(this.db);this.fb={};this.Sa={}}var h=z(4),k=f.cwrap,q=k("sqlite3_open","number",["string","number"]),w=k("sqlite3_close_v2","number",["number"]),v=k("sqlite3_exec","number",["number","string","number","number","number"]),C=k("sqlite3_changes","number",["number"]),
-G=k("sqlite3_prepare_v2","number",["number","string","number","number","number"]),pb=k("sqlite3_sql","string",["number"]),nc=k("sqlite3_normalized_sql","string",["number"]),qb=k("sqlite3_prepare_v2","number",["number","number","number","number","number"]),oc=k("sqlite3_bind_text","number",["number","number","number","number","number"]),rb=k("sqlite3_bind_blob","number",["number","number","number","number","number"]),pc=k("sqlite3_bind_double","number",["number","number","number"]),qc=k("sqlite3_bind_int",
-"number",["number","number","number"]),rc=k("sqlite3_bind_parameter_index","number",["number","string"]),sc=k("sqlite3_step","number",["number"]),tc=k("sqlite3_errmsg","string",["number"]),uc=k("sqlite3_column_count","number",["number"]),vc=k("sqlite3_data_count","number",["number"]),wc=k("sqlite3_column_double","number",["number","number"]),sb=k("sqlite3_column_text","string",["number","number"]),xc=k("sqlite3_column_blob","number",["number","number"]),yc=k("sqlite3_column_bytes","number",["number",
-"number"]),zc=k("sqlite3_column_type","number",["number","number"]),Ac=k("sqlite3_column_name","string",["number","number"]),Bc=k("sqlite3_reset","number",["number"]),Cc=k("sqlite3_clear_bindings","number",["number"]),Dc=k("sqlite3_finalize","number",["number"]),tb=k("sqlite3_create_function_v2","number","number string number number number number number number number".split(" ")),hc=k("sqlite3_value_type","number",["number"]),kc=k("sqlite3_value_bytes","number",["number"]),jc=k("sqlite3_value_text",
-"string",["number"]),lc=k("sqlite3_value_blob","number",["number"]),ic=k("sqlite3_value_double","number",["number"]),ec=k("sqlite3_result_double","",["number","number"]),lb=k("sqlite3_result_null","",["number"]),fc=k("sqlite3_result_text","",["number","string","number","number"]),gc=k("sqlite3_result_blob","",["number","number","number","number"]),dc=k("sqlite3_result_int","",["number","number"]),va=k("sqlite3_result_error","",["number","string","number"]),ub=k("sqlite3_aggregate_context","number",
-["number","number"]),ob=k("RegisterExtensionFunctions","number",["number"]),vb=k("sqlite3_update_hook","number",["number","number","number"]);c.prototype.bind=function(g){if(!this.Qa)throw"Statement closed";this.reset();return Array.isArray(g)?this.Cb(g):null!=g&&"object"===typeof g?this.Db(g):!0};c.prototype.step=function(){if(!this.Qa)throw"Statement closed";this.Oa=1;var g=sc(this.Qa);switch(g){case 100:return!0;case 101:return!1;default:throw this.db.handleError(g);}};c.prototype.wb=function(g){null==
-g&&(g=this.Oa,this.Oa+=1);return wc(this.Qa,g)};c.prototype.Gb=function(g){null==g&&(g=this.Oa,this.Oa+=1);g=sb(this.Qa,g);if("function"!==typeof BigInt)throw Error("BigInt is not supported");return BigInt(g)};c.prototype.Hb=function(g){null==g&&(g=this.Oa,this.Oa+=1);return sb(this.Qa,g)};c.prototype.getBlob=function(g){null==g&&(g=this.Oa,this.Oa+=1);var l=yc(this.Qa,g);g=xc(this.Qa,g);for(var n=new Uint8Array(l),r=0;r<l;r+=1)n[r]=p[g+r];return n};c.prototype.get=function(g,l){l=l||{};null!=g&&
-this.bind(g)&&this.step();g=[];for(var n=vc(this.Qa),r=0;r<n;r+=1)switch(zc(this.Qa,r)){case 1:var t=l.useBigInt?this.Gb(r):this.wb(r);g.push(t);break;case 2:g.push(this.wb(r));break;case 3:g.push(this.Hb(r));break;case 4:g.push(this.getBlob(r));break;default:g.push(null)}return g};c.prototype.getColumnNames=function(){for(var g=[],l=uc(this.Qa),n=0;n<l;n+=1)g.push(Ac(this.Qa,n));return g};c.prototype.getAsObject=function(g,l){g=this.get(g,l);l=this.getColumnNames();for(var n={},r=0;r<l.length;r+=
-1)n[l[r]]=g[r];return n};c.prototype.getSQL=function(){return pb(this.Qa)};c.prototype.getNormalizedSQL=function(){return nc(this.Qa)};c.prototype.run=function(g){null!=g&&this.bind(g);this.step();return this.reset()};c.prototype.sb=function(g,l){null==l&&(l=this.Oa,this.Oa+=1);g=ra(g);var n=da(g,ea);this.lb.push(n);this.db.handleError(oc(this.Qa,l,n,g.length-1,0))};c.prototype.Bb=function(g,l){null==l&&(l=this.Oa,this.Oa+=1);var n=da(g,ea);this.lb.push(n);this.db.handleError(rb(this.Qa,l,n,g.length,
-0))};c.prototype.rb=function(g,l){null==l&&(l=this.Oa,this.Oa+=1);this.db.handleError((g===(g|0)?qc:pc)(this.Qa,l,g))};c.prototype.Eb=function(g){null==g&&(g=this.Oa,this.Oa+=1);rb(this.Qa,g,0,0,0)};c.prototype.tb=function(g,l){null==l&&(l=this.Oa,this.Oa+=1);switch(typeof g){case "string":this.sb(g,l);return;case "number":this.rb(g,l);return;case "bigint":this.sb(g.toString(),l);return;case "boolean":this.rb(g+0,l);return;case "object":if(null===g){this.Eb(l);return}if(null!=g.length){this.Bb(g,
-l);return}}throw"Wrong API use : tried to bind a value of an unknown type ("+g+").";};c.prototype.Db=function(g){var l=this;Object.keys(g).forEach(function(n){var r=rc(l.Qa,n);0!==r&&l.tb(g[n],r)});return!0};c.prototype.Cb=function(g){for(var l=0;l<g.length;l+=1)this.tb(g[l],l+1);return!0};c.prototype.reset=function(){this.freemem();return 0===Cc(this.Qa)&&0===Bc(this.Qa)};c.prototype.freemem=function(){for(var g;void 0!==(g=this.lb.pop());)fa(g)};c.prototype.free=function(){this.freemem();var g=
-0===Dc(this.Qa);delete this.db.fb[this.Qa];this.Qa=0;return g};d.prototype.next=function(){if(null===this.eb)return{done:!0};null!==this.Za&&(this.Za.free(),this.Za=null);if(!this.db.db)throw this.mb(),Error("Database closed");var g=sa(),l=z(4);ta(h);ta(l);try{this.db.handleError(qb(this.db.db,this.kb,-1,h,l));this.kb=m(l,"i32");var n=m(h,"i32");if(0===n)return this.mb(),{done:!0};this.Za=new c(n,this.db);this.db.fb[n]=this.Za;return{value:this.Za,done:!1}}catch(r){throw this.pb=ua(this.kb),this.mb(),
-r;}finally{wa(g)}};d.prototype.mb=function(){fa(this.eb);this.eb=null};d.prototype.getRemainingSQL=function(){return null!==this.pb?this.pb:ua(this.kb)};"function"===typeof Symbol&&"symbol"===typeof Symbol.iterator&&(d.prototype[Symbol.iterator]=function(){return this});e.prototype.run=function(g,l){if(!this.db)throw"Database closed";if(l){g=this.prepare(g,l);try{g.step()}finally{g.free()}}else this.handleError(v(this.db,g,0,0,h));return this};e.prototype.exec=function(g,l,n){if(!this.db)throw"Database closed";
-var r=sa(),t=null;try{var y=xa(g),L=z(4);for(g=[];0!==m(y,"i8");){ta(h);ta(L);this.handleError(qb(this.db,y,-1,h,L));var J=m(h,"i32");y=m(L,"i32");if(0!==J){var I=null;t=new c(J,this);for(null!=l&&t.bind(l);t.step();)null===I&&(I={columns:t.getColumnNames(),values:[]},g.push(I)),I.values.push(t.get(null,n));t.free()}}return g}catch(M){throw t&&t.free(),M;}finally{wa(r)}};e.prototype.each=function(g,l,n,r,t){"function"===typeof l&&(r=n,n=l,l=void 0);g=this.prepare(g,l);try{for(;g.step();)n(g.getAsObject(null,
-t))}finally{g.free()}if("function"===typeof r)return r()};e.prototype.prepare=function(g,l){ta(h);this.handleError(G(this.db,g,-1,h,0));g=m(h,"i32");if(0===g)throw"Nothing to prepare";var n=new c(g,this);null!=l&&n.bind(l);return this.fb[g]=n};e.prototype.iterateStatements=function(g){return new d(g,this)};e.prototype["export"]=function(){Object.values(this.fb).forEach(function(l){l.free()});Object.values(this.Sa).forEach(A);this.Sa={};this.handleError(w(this.db));var g=ya(this.filename);this.handleError(q(this.filename,
-h));this.db=m(h,"i32");ob(this.db);return g};e.prototype.close=function(){null!==this.db&&(Object.values(this.fb).forEach(function(g){g.free()}),Object.values(this.Sa).forEach(A),this.Sa={},this.Ya&&(A(this.Ya),this.Ya=void 0),this.handleError(w(this.db)),za("/"+this.filename),this.db=null)};e.prototype.handleError=function(g){if(0===g)return null;g=tc(this.db);throw Error(g);};e.prototype.getRowsModified=function(){return C(this.db)};e.prototype.create_function=function(g,l){Object.prototype.hasOwnProperty.call(this.Sa,
-g)&&(A(this.Sa[g]),delete this.Sa[g]);var n=Aa(function(r,t,y){t=b(t,y);try{var L=l.apply(null,t)}catch(J){va(r,J,-1);return}a(r,L)},"viii");this.Sa[g]=n;this.handleError(tb(this.db,g,l.length,1,0,n,0,0,0));return this};e.prototype.create_aggregate=function(g,l){var n=l.init||function(){return null},r=l.finalize||function(I){return I},t=l.step;if(!t)throw"An aggregate function must have a step function in "+g;var y={};Object.hasOwnProperty.call(this.Sa,g)&&(A(this.Sa[g]),delete this.Sa[g]);l=g+"__finalize";
-Object.hasOwnProperty.call(this.Sa,l)&&(A(this.Sa[l]),delete this.Sa[l]);var L=Aa(function(I,M,Ra){var X=ub(I,1);Object.hasOwnProperty.call(y,X)||(y[X]=n());M=b(M,Ra);M=[y[X]].concat(M);try{y[X]=t.apply(null,M)}catch(Fc){delete y[X],va(I,Fc,-1)}},"viii"),J=Aa(function(I){var M=ub(I,1);try{var Ra=r(y[M])}catch(X){delete y[M];va(I,X,-1);return}a(I,Ra);delete y[M]},"vi");this.Sa[g]=L;this.Sa[l]=J;this.handleError(tb(this.db,g,t.length-1,1,0,0,L,J,0));return this};e.prototype.updateHook=function(g){this.Ya&&
-(vb(this.db,0,0),A(this.Ya),this.Ya=void 0);g&&(this.Ya=Aa(function(l,n,r,t,y){switch(n){case 18:l="insert";break;case 23:l="update";break;case 9:l="delete";break;default:throw"unknown operationCode in updateHook callback: "+n;}r=r?B(x,r):"";t=t?B(x,t):"";if(y>Number.MAX_SAFE_INTEGER)throw"rowId too big to fit inside a Number";g(l,r,t,Number(y))},"viiiij"),vb(this.db,this.Ya,0))};f.Database=e};var Ba={...f},Ca="./this.program",Da=(a,b)=>{throw b;},D="",Ea,Fa;
-if(ca){var fs=require("fs");require("path");D=__dirname+"/";Fa=a=>{a=Ga(a)?new URL(a):a;return fs.readFileSync(a)};Ea=async a=>{a=Ga(a)?new URL(a):a;return fs.readFileSync(a,void 0)};!f.thisProgram&&1<process.argv.length&&(Ca=process.argv[1].replace(/\\/g,"/"));process.argv.slice(2);"undefined"!=typeof module&&(module.exports=f);Da=(a,b)=>{process.exitCode=a;throw b;}}else if(aa||ba)ba?D=self.location.href:"undefined"!=typeof document&&document.currentScript&&(D=document.currentScript.src),D=D.startsWith("blob:")?
-"":D.slice(0,D.replace(/[?#].*/,"").lastIndexOf("/")+1),ba&&(Fa=a=>{var b=new XMLHttpRequest;b.open("GET",a,!1);b.responseType="arraybuffer";b.send(null);return new Uint8Array(b.response)}),Ea=async a=>{if(Ga(a))return new Promise((c,d)=>{var e=new XMLHttpRequest;e.open("GET",a,!0);e.responseType="arraybuffer";e.onload=()=>{200==e.status||0==e.status&&e.response?c(e.response):d(e.status)};e.onerror=d;e.send(null)});var b=await fetch(a,{credentials:"same-origin"});if(b.ok)return b.arrayBuffer();throw Error(b.status+
-" : "+b.url);};var Ha=f.print||console.log.bind(console),Ia=f.printErr||console.error.bind(console);Object.assign(f,Ba);Ba=null;f.thisProgram&&(Ca=f.thisProgram);var Ja=f.wasmBinary,Ka,La=!1,Ma,p,x,Na,E,F,Oa,H,Pa,Ga=a=>a.startsWith("file://");
-function Qa(){var a=Ka.buffer;f.HEAP8=p=new Int8Array(a);f.HEAP16=Na=new Int16Array(a);f.HEAPU8=x=new Uint8Array(a);f.HEAPU16=new Uint16Array(a);f.HEAP32=E=new Int32Array(a);f.HEAPU32=F=new Uint32Array(a);f.HEAPF32=Oa=new Float32Array(a);f.HEAPF64=Pa=new Float64Array(a);f.HEAP64=H=new BigInt64Array(a);f.HEAPU64=new BigUint64Array(a)}var K=0,Sa=null;function Ta(a){f.onAbort?.(a);a="Aborted("+a+")";Ia(a);La=!0;throw new WebAssembly.RuntimeError(a+". Build with -sASSERTIONS for more info.");}var Ua;
-async function Va(a){if(!Ja)try{var b=await Ea(a);return new Uint8Array(b)}catch{}if(a==Ua&&Ja)a=new Uint8Array(Ja);else if(Fa)a=Fa(a);else throw"both async and sync fetching of the wasm failed";return a}async function Wa(a,b){try{var c=await Va(a);return await WebAssembly.instantiate(c,b)}catch(d){Ia(`failed to asynchronously prepare wasm: ${d}`),Ta(d)}}
-async function Xa(a){var b=Ua;if(!Ja&&"function"==typeof WebAssembly.instantiateStreaming&&!Ga(b)&&!ca)try{var c=fetch(b,{credentials:"same-origin"});return await WebAssembly.instantiateStreaming(c,a)}catch(d){Ia(`wasm streaming compile failed: ${d}`),Ia("falling back to ArrayBuffer instantiation")}return Wa(b,a)}class Ya{name="ExitStatus";constructor(a){this.message=`Program terminated with exit(${a})`;this.status=a}}
-var Za=a=>{for(;0<a.length;)a.shift()(f)},$a=[],ab=[],bb=()=>{var a=f.preRun.shift();ab.unshift(a)};function m(a,b="i8"){b.endsWith("*")&&(b="*");switch(b){case "i1":return p[a];case "i8":return p[a];case "i16":return Na[a>>1];case "i32":return E[a>>2];case "i64":return H[a>>3];case "float":return Oa[a>>2];case "double":return Pa[a>>3];case "*":return F[a>>2];default:Ta(`invalid type for getValue: ${b}`)}}var cb=f.noExitRuntime||!0;
-function ta(a){var b="i32";b.endsWith("*")&&(b="*");switch(b){case "i1":p[a]=0;break;case "i8":p[a]=0;break;case "i16":Na[a>>1]=0;break;case "i32":E[a>>2]=0;break;case "i64":H[a>>3]=BigInt(0);break;case "float":Oa[a>>2]=0;break;case "double":Pa[a>>3]=0;break;case "*":F[a>>2]=0;break;default:Ta(`invalid type for setValue: ${b}`)}}
-var db="undefined"!=typeof TextDecoder?new TextDecoder:void 0,B=(a,b=0,c=NaN)=>{var d=b+c;for(c=b;a[c]&&!(c>=d);)++c;if(16<c-b&&a.buffer&&db)return db.decode(a.subarray(b,c));for(d="";b<c;){var e=a[b++];if(e&128){var h=a[b++]&63;if(192==(e&224))d+=String.fromCharCode((e&31)<<6|h);else{var k=a[b++]&63;e=224==(e&240)?(e&15)<<12|h<<6|k:(e&7)<<18|h<<12|k<<6|a[b++]&63;65536>e?d+=String.fromCharCode(e):(e-=65536,d+=String.fromCharCode(55296|e>>10,56320|e&1023))}}else d+=String.fromCharCode(e)}return d},
-ua=(a,b)=>a?B(x,a,b):"",eb=(a,b)=>{for(var c=0,d=a.length-1;0<=d;d--){var e=a[d];"."===e?a.splice(d,1):".."===e?(a.splice(d,1),c++):c&&(a.splice(d,1),c--)}if(b)for(;c;c--)a.unshift("..");return a},ka=a=>{var b="/"===a.charAt(0),c="/"===a.slice(-1);(a=eb(a.split("/").filter(d=>!!d),!b).join("/"))||b||(a=".");a&&c&&(a+="/");return(b?"/":"")+a},fb=a=>{var b=/^(\/?|)([\s\S]*?)((?:\.{1,2}|[^\/]+?|)(\.[^.\/]*|))(?:[\/]*)$/.exec(a).slice(1);a=b[0];b=b[1];if(!a&&!b)return".";b&&=b.slice(0,-1);return a+b},
-gb=a=>a&&a.match(/([^\/]+|\/)\/*$/)[1],hb=()=>{if(ca){var a=require("crypto");return b=>a.randomFillSync(b)}return b=>crypto.getRandomValues(b)},ib=a=>{(ib=hb())(a)},jb=(...a)=>{for(var b="",c=!1,d=a.length-1;-1<=d&&!c;d--){c=0<=d?a[d]:"/";if("string"!=typeof c)throw new TypeError("Arguments to path.resolve must be strings");if(!c)return"";b=c+"/"+b;c="/"===c.charAt(0)}b=eb(b.split("/").filter(e=>!!e),!c).join("/");return(c?"/":"")+b||"."},kb=[],ha=a=>{for(var b=0,c=0;c<a.length;++c){var d=a.charCodeAt(c);
-127>=d?b++:2047>=d?b+=2:55296<=d&&57343>=d?(b+=4,++c):b+=3}return b},u=(a,b,c,d)=>{if(!(0<d))return 0;var e=c;d=c+d-1;for(var h=0;h<a.length;++h){var k=a.charCodeAt(h);if(55296<=k&&57343>=k){var q=a.charCodeAt(++h);k=65536+((k&1023)<<10)|q&1023}if(127>=k){if(c>=d)break;b[c++]=k}else{if(2047>=k){if(c+1>=d)break;b[c++]=192|k>>6}else{if(65535>=k){if(c+2>=d)break;b[c++]=224|k>>12}else{if(c+3>=d)break;b[c++]=240|k>>18;b[c++]=128|k>>12&63}b[c++]=128|k>>6&63}b[c++]=128|k&63}}b[c]=0;return c-e},ra=(a,b)=>
-{var c=Array(ha(a)+1);a=u(a,c,0,c.length);b&&(c.length=a);return c},mb=[];function nb(a,b){mb[a]={input:[],output:[],cb:b};wb(a,xb)}
-var xb={open(a){var b=mb[a.node.rdev];if(!b)throw new N(43);a.tty=b;a.seekable=!1},close(a){a.tty.cb.fsync(a.tty)},fsync(a){a.tty.cb.fsync(a.tty)},read(a,b,c,d){if(!a.tty||!a.tty.cb.xb)throw new N(60);for(var e=0,h=0;h<d;h++){try{var k=a.tty.cb.xb(a.tty)}catch(q){throw new N(29);}if(void 0===k&&0===e)throw new N(6);if(null===k||void 0===k)break;e++;b[c+h]=k}e&&(a.node.atime=Date.now());return e},write(a,b,c,d){if(!a.tty||!a.tty.cb.qb)throw new N(60);try{for(var e=0;e<d;e++)a.tty.cb.qb(a.tty,b[c+e])}catch(h){throw new N(29);
-}d&&(a.node.mtime=a.node.ctime=Date.now());return e}},yb={xb(){a:{if(!kb.length){var a=null;if(ca){var b=Buffer.alloc(256),c=0,d=process.stdin.fd;try{c=fs.readSync(d,b,0,256)}catch(e){if(e.toString().includes("EOF"))c=0;else throw e;}0<c&&(a=b.slice(0,c).toString("utf-8"))}else"undefined"!=typeof window&&"function"==typeof window.prompt&&(a=window.prompt("Input: "),null!==a&&(a+="\n"));if(!a){a=null;break a}kb=ra(a,!0)}a=kb.shift()}return a},qb(a,b){null===b||10===b?(Ha(B(a.output)),a.output=[]):
-0!=b&&a.output.push(b)},fsync(a){0<a.output?.length&&(Ha(B(a.output)),a.output=[])},Tb(){return{Ob:25856,Qb:5,Nb:191,Pb:35387,Mb:[3,28,127,21,4,0,1,0,17,19,26,0,18,15,23,22,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]}},Ub(){return 0},Vb(){return[24,80]}},zb={qb(a,b){null===b||10===b?(Ia(B(a.output)),a.output=[]):0!=b&&a.output.push(b)},fsync(a){0<a.output?.length&&(Ia(B(a.output)),a.output=[])}},O={Wa:null,Xa(){return O.createNode(null,"/",16895,0)},createNode(a,b,c,d){if(24576===(c&61440)||4096===(c&61440))throw new N(63);
-O.Wa||(O.Wa={dir:{node:{Ta:O.La.Ta,Ua:O.La.Ua,lookup:O.La.lookup,hb:O.La.hb,rename:O.La.rename,unlink:O.La.unlink,rmdir:O.La.rmdir,readdir:O.La.readdir,symlink:O.La.symlink},stream:{Va:O.Ma.Va}},file:{node:{Ta:O.La.Ta,Ua:O.La.Ua},stream:{Va:O.Ma.Va,read:O.Ma.read,write:O.Ma.write,ib:O.Ma.ib,jb:O.Ma.jb}},link:{node:{Ta:O.La.Ta,Ua:O.La.Ua,readlink:O.La.readlink},stream:{}},ub:{node:{Ta:O.La.Ta,Ua:O.La.Ua},stream:Ab}});c=Bb(a,b,c,d);P(c.mode)?(c.La=O.Wa.dir.node,c.Ma=O.Wa.dir.stream,c.Na={}):32768===
-(c.mode&61440)?(c.La=O.Wa.file.node,c.Ma=O.Wa.file.stream,c.Ra=0,c.Na=null):40960===(c.mode&61440)?(c.La=O.Wa.link.node,c.Ma=O.Wa.link.stream):8192===(c.mode&61440)&&(c.La=O.Wa.ub.node,c.Ma=O.Wa.ub.stream);c.atime=c.mtime=c.ctime=Date.now();a&&(a.Na[b]=c,a.atime=a.mtime=a.ctime=c.atime);return c},Sb(a){return a.Na?a.Na.subarray?a.Na.subarray(0,a.Ra):new Uint8Array(a.Na):new Uint8Array(0)},La:{Ta(a){var b={};b.dev=8192===(a.mode&61440)?a.id:1;b.ino=a.id;b.mode=a.mode;b.nlink=1;b.uid=0;b.gid=0;b.rdev=
-a.rdev;P(a.mode)?b.size=4096:32768===(a.mode&61440)?b.size=a.Ra:40960===(a.mode&61440)?b.size=a.link.length:b.size=0;b.atime=new Date(a.atime);b.mtime=new Date(a.mtime);b.ctime=new Date(a.ctime);b.blksize=4096;b.blocks=Math.ceil(b.size/b.blksize);return b},Ua(a,b){for(var c of["mode","atime","mtime","ctime"])null!=b[c]&&(a[c]=b[c]);void 0!==b.size&&(b=b.size,a.Ra!=b&&(0==b?(a.Na=null,a.Ra=0):(c=a.Na,a.Na=new Uint8Array(b),c&&a.Na.set(c.subarray(0,Math.min(b,a.Ra))),a.Ra=b)))},lookup(){throw O.vb;
-},hb(a,b,c,d){return O.createNode(a,b,c,d)},rename(a,b,c){try{var d=Q(b,c)}catch(h){}if(d){if(P(a.mode))for(var e in d.Na)throw new N(55);Cb(d)}delete a.parent.Na[a.name];b.Na[c]=a;a.name=c;b.ctime=b.mtime=a.parent.ctime=a.parent.mtime=Date.now()},unlink(a,b){delete a.Na[b];a.ctime=a.mtime=Date.now()},rmdir(a,b){var c=Q(a,b),d;for(d in c.Na)throw new N(55);delete a.Na[b];a.ctime=a.mtime=Date.now()},readdir(a){return[".","..",...Object.keys(a.Na)]},symlink(a,b,c){a=O.createNode(a,b,41471,0);a.link=
-c;return a},readlink(a){if(40960!==(a.mode&61440))throw new N(28);return a.link}},Ma:{read(a,b,c,d,e){var h=a.node.Na;if(e>=a.node.Ra)return 0;a=Math.min(a.node.Ra-e,d);if(8<a&&h.subarray)b.set(h.subarray(e,e+a),c);else for(d=0;d<a;d++)b[c+d]=h[e+d];return a},write(a,b,c,d,e,h){b.buffer===p.buffer&&(h=!1);if(!d)return 0;a=a.node;a.mtime=a.ctime=Date.now();if(b.subarray&&(!a.Na||a.Na.subarray)){if(h)return a.Na=b.subarray(c,c+d),a.Ra=d;if(0===a.Ra&&0===e)return a.Na=b.slice(c,c+d),a.Ra=d;if(e+d<=a.Ra)return a.Na.set(b.subarray(c,
-c+d),e),d}h=e+d;var k=a.Na?a.Na.length:0;k>=h||(h=Math.max(h,k*(1048576>k?2:1.125)>>>0),0!=k&&(h=Math.max(h,256)),k=a.Na,a.Na=new Uint8Array(h),0<a.Ra&&a.Na.set(k.subarray(0,a.Ra),0));if(a.Na.subarray&&b.subarray)a.Na.set(b.subarray(c,c+d),e);else for(h=0;h<d;h++)a.Na[e+h]=b[c+h];a.Ra=Math.max(a.Ra,e+d);return d},Va(a,b,c){1===c?b+=a.position:2===c&&32768===(a.node.mode&61440)&&(b+=a.node.Ra);if(0>b)throw new N(28);return b},ib(a,b,c,d,e){if(32768!==(a.node.mode&61440))throw new N(43);a=a.node.Na;
-if(e&2||!a||a.buffer!==p.buffer){e=!0;d=65536*Math.ceil(b/65536);var h=Db(65536,d);h&&x.fill(0,h,h+d);d=h;if(!d)throw new N(48);if(a){if(0<c||c+b<a.length)a.subarray?a=a.subarray(c,c+b):a=Array.prototype.slice.call(a,c,c+b);p.set(a,d)}}else e=!1,d=a.byteOffset;return{Kb:d,Ab:e}},jb(a,b,c,d){O.Ma.write(a,b,0,d,c,!1);return 0}}},la=(a,b)=>{var c=0;a&&(c|=365);b&&(c|=146);return c},Eb=null,Fb={},Gb=[],Hb=1,R=null,Ib=!1,Jb=!0,Kb={},N=class{name="ErrnoError";constructor(a){this.Pa=a}},Lb=class{gb={};node=null;get flags(){return this.gb.flags}set flags(a){this.gb.flags=
-a}get position(){return this.gb.position}set position(a){this.gb.position=a}},Mb=class{La={};Ma={};ab=null;constructor(a,b,c,d){a||=this;this.parent=a;this.Xa=a.Xa;this.id=Hb++;this.name=b;this.mode=c;this.rdev=d;this.atime=this.mtime=this.ctime=Date.now()}get read(){return 365===(this.mode&365)}set read(a){a?this.mode|=365:this.mode&=-366}get write(){return 146===(this.mode&146)}set write(a){a?this.mode|=146:this.mode&=-147}};
-function S(a,b={}){if(!a)throw new N(44);b.nb??(b.nb=!0);"/"===a.charAt(0)||(a="//"+a);var c=0;a:for(;40>c;c++){a=a.split("/").filter(q=>!!q);for(var d=Eb,e="/",h=0;h<a.length;h++){var k=h===a.length-1;if(k&&b.parent)break;if("."!==a[h])if(".."===a[h])e=fb(e),d=d.parent;else{e=ka(e+"/"+a[h]);try{d=Q(d,a[h])}catch(q){if(44===q?.Pa&&k&&b.Jb)return{path:e};throw q;}!d.ab||k&&!b.nb||(d=d.ab.root);if(40960===(d.mode&61440)&&(!k||b.$a)){if(!d.La.readlink)throw new N(52);d=d.La.readlink(d);"/"===d.charAt(0)||
-(d=fb(e)+"/"+d);a=d+"/"+a.slice(h+1).join("/");continue a}}}return{path:e,node:d}}throw new N(32);}function ja(a){for(var b;;){if(a===a.parent)return a=a.Xa.zb,b?"/"!==a[a.length-1]?`${a}/${b}`:a+b:a;b=b?`${a.name}/${b}`:a.name;a=a.parent}}function Nb(a,b){for(var c=0,d=0;d<b.length;d++)c=(c<<5)-c+b.charCodeAt(d)|0;return(a+c>>>0)%R.length}function Cb(a){var b=Nb(a.parent.id,a.name);if(R[b]===a)R[b]=a.bb;else for(b=R[b];b;){if(b.bb===a){b.bb=a.bb;break}b=b.bb}}
-function Q(a,b){var c=P(a.mode)?(c=Ob(a,"x"))?c:a.La.lookup?0:2:54;if(c)throw new N(c);for(c=R[Nb(a.id,b)];c;c=c.bb){var d=c.name;if(c.parent.id===a.id&&d===b)return c}return a.La.lookup(a,b)}function Bb(a,b,c,d){a=new Mb(a,b,c,d);b=Nb(a.parent.id,a.name);a.bb=R[b];return R[b]=a}function P(a){return 16384===(a&61440)}function Pb(a){var b=["r","w","rw"][a&3];a&512&&(b+="w");return b}
-function Ob(a,b){if(Jb)return 0;if(!b.includes("r")||a.mode&292){if(b.includes("w")&&!(a.mode&146)||b.includes("x")&&!(a.mode&73))return 2}else return 2;return 0}function Qb(a,b){if(!P(a.mode))return 54;try{return Q(a,b),20}catch(c){}return Ob(a,"wx")}function Rb(a,b,c){try{var d=Q(a,b)}catch(e){return e.Pa}if(a=Ob(a,"wx"))return a;if(c){if(!P(d.mode))return 54;if(d===d.parent||"/"===ja(d))return 10}else if(P(d.mode))return 31;return 0}function Sb(a){if(!a)throw new N(63);return a}
-function T(a){a=Gb[a];if(!a)throw new N(8);return a}function Tb(a,b=-1){a=Object.assign(new Lb,a);if(-1==b)a:{for(b=0;4096>=b;b++)if(!Gb[b])break a;throw new N(33);}a.fd=b;return Gb[b]=a}function Ub(a,b=-1){a=Tb(a,b);a.Ma?.Rb?.(a);return a}function Vb(a,b,c){var d=a?.Ma.Ua;a=d?a:b;d??=b.La.Ua;Sb(d);d(a,c)}var Ab={open(a){a.Ma=Fb[a.node.rdev].Ma;a.Ma.open?.(a)},Va(){throw new N(70);}};function wb(a,b){Fb[a]={Ma:b}}
-function Wb(a,b){var c="/"===b;if(c&&Eb)throw new N(10);if(!c&&b){var d=S(b,{nb:!1});b=d.path;d=d.node;if(d.ab)throw new N(10);if(!P(d.mode))throw new N(54);}b={type:a,Wb:{},zb:b,Ib:[]};a=a.Xa(b);a.Xa=b;b.root=a;c?Eb=a:d&&(d.ab=b,d.Xa&&d.Xa.Ib.push(b))}function Xb(a,b,c){var d=S(a,{parent:!0}).node;a=gb(a);if(!a)throw new N(28);if("."===a||".."===a)throw new N(20);var e=Qb(d,a);if(e)throw new N(e);if(!d.La.hb)throw new N(63);return d.La.hb(d,a,b,c)}
-function ma(a,b=438){return Xb(a,b&4095|32768,0)}function U(a,b=511){return Xb(a,b&1023|16384,0)}function Yb(a,b,c){"undefined"==typeof c&&(c=b,b=438);Xb(a,b|8192,c)}function Zb(a,b){if(!jb(a))throw new N(44);var c=S(b,{parent:!0}).node;if(!c)throw new N(44);b=gb(b);var d=Qb(c,b);if(d)throw new N(d);if(!c.La.symlink)throw new N(63);c.La.symlink(c,b,a)}
-function $b(a){var b=S(a,{parent:!0}).node;a=gb(a);var c=Q(b,a),d=Rb(b,a,!0);if(d)throw new N(d);if(!b.La.rmdir)throw new N(63);if(c.ab)throw new N(10);b.La.rmdir(b,a);Cb(c)}function za(a){var b=S(a,{parent:!0}).node;if(!b)throw new N(44);a=gb(a);var c=Q(b,a),d=Rb(b,a,!1);if(d)throw new N(d);if(!b.La.unlink)throw new N(63);if(c.ab)throw new N(10);b.La.unlink(b,a);Cb(c)}function ac(a,b){a=S(a,{$a:!b}).node;return Sb(a.La.Ta)(a)}
-function bc(a,b,c,d){Vb(a,b,{mode:c&4095|b.mode&-4096,ctime:Date.now(),Fb:d})}function na(a,b){a="string"==typeof a?S(a,{$a:!0}).node:a;bc(null,a,b)}function cc(a,b,c){if(P(b.mode))throw new N(31);if(32768!==(b.mode&61440))throw new N(28);var d=Ob(b,"w");if(d)throw new N(d);Vb(a,b,{size:c,timestamp:Date.now()})}
-function oa(a,b,c=438){if(""===a)throw new N(44);if("string"==typeof b){var d={r:0,"r+":2,w:577,"w+":578,a:1089,"a+":1090}[b];if("undefined"==typeof d)throw Error(`Unknown file open mode: ${b}`);b=d}c=b&64?c&4095|32768:0;if("object"==typeof a)d=a;else{var e=a.endsWith("/");a=S(a,{$a:!(b&131072),Jb:!0});d=a.node;a=a.path}var h=!1;if(b&64)if(d){if(b&128)throw new N(20);}else{if(e)throw new N(31);d=Xb(a,c|511,0);h=!0}if(!d)throw new N(44);8192===(d.mode&61440)&&(b&=-513);if(b&65536&&!P(d.mode))throw new N(54);
-if(!h&&(e=d?40960===(d.mode&61440)?32:P(d.mode)&&("r"!==Pb(b)||b&576)?31:Ob(d,Pb(b)):44))throw new N(e);b&512&&!h&&(e=d,e="string"==typeof e?S(e,{$a:!0}).node:e,cc(null,e,0));b&=-131713;e=Tb({node:d,path:ja(d),flags:b,seekable:!0,position:0,Ma:d.Ma,Lb:[],error:!1});e.Ma.open&&e.Ma.open(e);h&&na(d,c&511);!f.logReadFiles||b&1||a in Kb||(Kb[a]=1);return e}function qa(a){if(null===a.fd)throw new N(8);a.ob&&(a.ob=null);try{a.Ma.close&&a.Ma.close(a)}catch(b){throw b;}finally{Gb[a.fd]=null}a.fd=null}
-function mc(a,b,c){if(null===a.fd)throw new N(8);if(!a.seekable||!a.Ma.Va)throw new N(70);if(0!=c&&1!=c&&2!=c)throw new N(28);a.position=a.Ma.Va(a,b,c);a.Lb=[]}function Ec(a,b,c,d,e){if(0>d||0>e)throw new N(28);if(null===a.fd)throw new N(8);if(1===(a.flags&2097155))throw new N(8);if(P(a.node.mode))throw new N(31);if(!a.Ma.read)throw new N(28);var h="undefined"!=typeof e;if(!h)e=a.position;else if(!a.seekable)throw new N(70);b=a.Ma.read(a,b,c,d,e);h||(a.position+=b);return b}
-function pa(a,b,c,d,e){if(0>d||0>e)throw new N(28);if(null===a.fd)throw new N(8);if(0===(a.flags&2097155))throw new N(8);if(P(a.node.mode))throw new N(31);if(!a.Ma.write)throw new N(28);a.seekable&&a.flags&1024&&mc(a,0,2);var h="undefined"!=typeof e;if(!h)e=a.position;else if(!a.seekable)throw new N(70);b=a.Ma.write(a,b,c,d,e,void 0);h||(a.position+=b);return b}
-function ya(a){var b="binary";if("utf8"!==b&&"binary"!==b)throw Error(`Invalid encoding type "${b}"`);var c;var d=oa(a,d||0);a=ac(a).size;var e=new Uint8Array(a);Ec(d,e,0,a,0);"utf8"===b?c=B(e):"binary"===b&&(c=e);qa(d);return c}
-function V(a,b,c){a=ka("/dev/"+a);var d=la(!!b,!!c);V.yb??(V.yb=64);var e=V.yb++<<8|0;wb(e,{open(h){h.seekable=!1},close(){c?.buffer?.length&&c(10)},read(h,k,q,w){for(var v=0,C=0;C<w;C++){try{var G=b()}catch(pb){throw new N(29);}if(void 0===G&&0===v)throw new N(6);if(null===G||void 0===G)break;v++;k[q+C]=G}v&&(h.node.atime=Date.now());return v},write(h,k,q,w){for(var v=0;v<w;v++)try{c(k[q+v])}catch(C){throw new N(29);}w&&(h.node.mtime=h.node.ctime=Date.now());return v}});Yb(a,d,e)}var W={};
-function Gc(a,b,c){if("/"===b.charAt(0))return b;a=-100===a?"/":T(a).path;if(0==b.length){if(!c)throw new N(44);return a}return a+"/"+b}
-function Hc(a,b){E[a>>2]=b.dev;E[a+4>>2]=b.mode;F[a+8>>2]=b.nlink;E[a+12>>2]=b.uid;E[a+16>>2]=b.gid;E[a+20>>2]=b.rdev;H[a+24>>3]=BigInt(b.size);E[a+32>>2]=4096;E[a+36>>2]=b.blocks;var c=b.atime.getTime(),d=b.mtime.getTime(),e=b.ctime.getTime();H[a+40>>3]=BigInt(Math.floor(c/1E3));F[a+48>>2]=c%1E3*1E6;H[a+56>>3]=BigInt(Math.floor(d/1E3));F[a+64>>2]=d%1E3*1E6;H[a+72>>3]=BigInt(Math.floor(e/1E3));F[a+80>>2]=e%1E3*1E6;H[a+88>>3]=BigInt(b.ino);return 0}
-var Ic=void 0,Jc=()=>{var a=E[+Ic>>2];Ic+=4;return a},Kc=0,Lc=[0,31,60,91,121,152,182,213,244,274,305,335],Mc=[0,31,59,90,120,151,181,212,243,273,304,334],Nc={},Oc=a=>{Ma=a;cb||0<Kc||(f.onExit?.(a),La=!0);Da(a,new Ya(a))},Pc=a=>{if(!La)try{if(a(),!(cb||0<Kc))try{Ma=a=Ma,Oc(a)}catch(b){b instanceof Ya||"unwind"==b||Da(1,b)}}catch(b){b instanceof Ya||"unwind"==b||Da(1,b)}},Qc={},Sc=()=>{if(!Rc){var a={USER:"web_user",LOGNAME:"web_user",PATH:"/",PWD:"/",HOME:"/home/web_user",LANG:("object"==typeof navigator&&
-navigator.languages&&navigator.languages[0]||"C").replace("-","_")+".UTF-8",_:Ca||"./this.program"},b;for(b in Qc)void 0===Qc[b]?delete a[b]:a[b]=Qc[b];var c=[];for(b in a)c.push(`${b}=${a[b]}`);Rc=c}return Rc},Rc,xa=a=>{var b=ha(a)+1,c=z(b);u(a,x,c,b);return c},Tc=(a,b,c,d)=>{var e={string:v=>{var C=0;null!==v&&void 0!==v&&0!==v&&(C=xa(v));return C},array:v=>{var C=z(v.length);p.set(v,C);return C}};a=f["_"+a];var h=[],k=0;if(d)for(var q=0;q<d.length;q++){var w=e[c[q]];w?(0===k&&(k=sa()),h[q]=w(d[q])):
-h[q]=d[q]}c=a(...h);return c=function(v){0!==k&&wa(k);return"string"===b?v?B(x,v):"":"boolean"===b?!!v:v}(c)},ea=0,da=(a,b)=>{b=1==b?z(a.length):ia(a.length);a.subarray||a.slice||(a=new Uint8Array(a));x.set(a,b);return b},Uc,Vc=[],Y,A=a=>{Uc.delete(Y.get(a));Y.set(a,null);Vc.push(a)},Aa=(a,b)=>{if(!Uc){Uc=new WeakMap;var c=Y.length;if(Uc)for(var d=0;d<0+c;d++){var e=Y.get(d);e&&Uc.set(e,d)}}if(c=Uc.get(a)||0)return c;if(Vc.length)c=Vc.pop();else{try{Y.grow(1)}catch(w){if(!(w instanceof RangeError))throw w;
-throw"Unable to grow wasm table. Set ALLOW_TABLE_GROWTH.";}c=Y.length-1}try{Y.set(c,a)}catch(w){if(!(w instanceof TypeError))throw w;if("function"==typeof WebAssembly.Function){var h=WebAssembly.Function;d={i:"i32",j:"i64",f:"f32",d:"f64",e:"externref",p:"i32"};e={parameters:[],results:"v"==b[0]?[]:[d[b[0]]]};for(var k=1;k<b.length;++k)e.parameters.push(d[b[k]]);b=new h(e,a)}else{d=[1];e=b.slice(0,1);b=b.slice(1);k={i:127,p:127,j:126,f:125,d:124,e:111};d.push(96);var q=b.length;128>q?d.push(q):d.push(q%
-128|128,q>>7);for(h of b)d.push(k[h]);"v"==e?d.push(0):d.push(1,k[e]);b=[0,97,115,109,1,0,0,0,1];h=d.length;128>h?b.push(h):b.push(h%128|128,h>>7);b.push(...d);b.push(2,7,1,1,101,1,102,0,0,7,5,1,1,102,0,0);b=new WebAssembly.Module(new Uint8Array(b));b=(new WebAssembly.Instance(b,{e:{f:a}})).exports.f}Y.set(c,b)}Uc.set(a,c);return c};R=Array(4096);Wb(O,"/");U("/tmp");U("/home");U("/home/web_user");
-(function(){U("/dev");wb(259,{read:()=>0,write:(d,e,h,k)=>k,Va:()=>0});Yb("/dev/null",259);nb(1280,yb);nb(1536,zb);Yb("/dev/tty",1280);Yb("/dev/tty1",1536);var a=new Uint8Array(1024),b=0,c=()=>{0===b&&(ib(a),b=a.byteLength);return a[--b]};V("random",c);V("urandom",c);U("/dev/shm");U("/dev/shm/tmp")})();
-(function(){U("/proc");var a=U("/proc/self");U("/proc/self/fd");Wb({Xa(){var b=Bb(a,"fd",16895,73);b.Ma={Va:O.Ma.Va};b.La={lookup(c,d){c=+d;var e=T(c);c={parent:null,Xa:{zb:"fake"},La:{readlink:()=>e.path},id:c+1};return c.parent=c},readdir(){return Array.from(Gb.entries()).filter(([,c])=>c).map(([c])=>c.toString())}};return b}},"/proc/self/fd")})();O.vb=new N(44);O.vb.stack="<generic error, no stack>";
-var Xc={a:(a,b,c,d)=>Ta(`Assertion failed: ${a?B(x,a):""}, at: `+[b?b?B(x,b):"":"unknown filename",c,d?d?B(x,d):"":"unknown function"]),i:function(a,b){try{return a=a?B(x,a):"",na(a,b),0}catch(c){if("undefined"==typeof W||"ErrnoError"!==c.name)throw c;return-c.Pa}},L:function(a,b,c){try{b=b?B(x,b):"";b=Gc(a,b);if(c&-8)return-28;var d=S(b,{$a:!0}).node;if(!d)return-44;a="";c&4&&(a+="r");c&2&&(a+="w");c&1&&(a+="x");return a&&Ob(d,a)?-2:0}catch(e){if("undefined"==typeof W||"ErrnoError"!==e.name)throw e;
-return-e.Pa}},j:function(a,b){try{var c=T(a);bc(c,c.node,b,!1);return 0}catch(d){if("undefined"==typeof W||"ErrnoError"!==d.name)throw d;return-d.Pa}},h:function(a){try{var b=T(a);Vb(b,b.node,{timestamp:Date.now(),Fb:!1});return 0}catch(c){if("undefined"==typeof W||"ErrnoError"!==c.name)throw c;return-c.Pa}},b:function(a,b,c){Ic=c;try{var d=T(a);switch(b){case 0:var e=Jc();if(0>e)break;for(;Gb[e];)e++;return Ub(d,e).fd;case 1:case 2:return 0;case 3:return d.flags;case 4:return e=Jc(),d.flags|=e,0;
-case 12:return e=Jc(),Na[e+0>>1]=2,0;case 13:case 14:return 0}return-28}catch(h){if("undefined"==typeof W||"ErrnoError"!==h.name)throw h;return-h.Pa}},g:function(a,b){try{var c=T(a),d=c.node,e=c.Ma.Ta;a=e?c:d;e??=d.La.Ta;Sb(e);var h=e(a);return Hc(b,h)}catch(k){if("undefined"==typeof W||"ErrnoError"!==k.name)throw k;return-k.Pa}},H:function(a,b){b=-9007199254740992>b||9007199254740992<b?NaN:Number(b);try{if(isNaN(b))return 61;var c=T(a);if(0>b||0===(c.flags&2097155))throw new N(28);cc(c,c.node,b);
-return 0}catch(d){if("undefined"==typeof W||"ErrnoError"!==d.name)throw d;return-d.Pa}},G:function(a,b){try{if(0===b)return-28;var c=ha("/")+1;if(b<c)return-68;u("/",x,a,b);return c}catch(d){if("undefined"==typeof W||"ErrnoError"!==d.name)throw d;return-d.Pa}},K:function(a,b){try{return a=a?B(x,a):"",Hc(b,ac(a,!0))}catch(c){if("undefined"==typeof W||"ErrnoError"!==c.name)throw c;return-c.Pa}},C:function(a,b,c){try{return b=b?B(x,b):"",b=Gc(a,b),U(b,c),0}catch(d){if("undefined"==typeof W||"ErrnoError"!==
-d.name)throw d;return-d.Pa}},J:function(a,b,c,d){try{b=b?B(x,b):"";var e=d&256;b=Gc(a,b,d&4096);return Hc(c,e?ac(b,!0):ac(b))}catch(h){if("undefined"==typeof W||"ErrnoError"!==h.name)throw h;return-h.Pa}},x:function(a,b,c,d){Ic=d;try{b=b?B(x,b):"";b=Gc(a,b);var e=d?Jc():0;return oa(b,c,e).fd}catch(h){if("undefined"==typeof W||"ErrnoError"!==h.name)throw h;return-h.Pa}},v:function(a,b,c,d){try{b=b?B(x,b):"";b=Gc(a,b);if(0>=d)return-28;var e=S(b).node;if(!e)throw new N(44);if(!e.La.readlink)throw new N(28);
-var h=e.La.readlink(e);var k=Math.min(d,ha(h)),q=p[c+k];u(h,x,c,d+1);p[c+k]=q;return k}catch(w){if("undefined"==typeof W||"ErrnoError"!==w.name)throw w;return-w.Pa}},u:function(a){try{return a=a?B(x,a):"",$b(a),0}catch(b){if("undefined"==typeof W||"ErrnoError"!==b.name)throw b;return-b.Pa}},f:function(a,b){try{return a=a?B(x,a):"",Hc(b,ac(a))}catch(c){if("undefined"==typeof W||"ErrnoError"!==c.name)throw c;return-c.Pa}},r:function(a,b,c){try{return b=b?B(x,b):"",b=Gc(a,b),0===c?za(b):512===c?$b(b):
-Ta("Invalid flags passed to unlinkat"),0}catch(d){if("undefined"==typeof W||"ErrnoError"!==d.name)throw d;return-d.Pa}},q:function(a,b,c){try{b=b?B(x,b):"";b=Gc(a,b,!0);var d=Date.now(),e,h;if(c){var k=F[c>>2]+4294967296*E[c+4>>2],q=E[c+8>>2];1073741823==q?e=d:1073741822==q?e=null:e=1E3*k+q/1E6;c+=16;k=F[c>>2]+4294967296*E[c+4>>2];q=E[c+8>>2];1073741823==q?h=d:1073741822==q?h=null:h=1E3*k+q/1E6}else h=e=d;if(null!==(h??e)){a=e;var w=S(b,{$a:!0}).node;Sb(w.La.Ua)(w,{atime:a,mtime:h})}return 0}catch(v){if("undefined"==
-typeof W||"ErrnoError"!==v.name)throw v;return-v.Pa}},m:()=>Ta(""),l:()=>{cb=!1;Kc=0},A:function(a,b){a=-9007199254740992>a||9007199254740992<a?NaN:Number(a);a=new Date(1E3*a);E[b>>2]=a.getSeconds();E[b+4>>2]=a.getMinutes();E[b+8>>2]=a.getHours();E[b+12>>2]=a.getDate();E[b+16>>2]=a.getMonth();E[b+20>>2]=a.getFullYear()-1900;E[b+24>>2]=a.getDay();var c=a.getFullYear();E[b+28>>2]=(0!==c%4||0===c%100&&0!==c%400?Mc:Lc)[a.getMonth()]+a.getDate()-1|0;E[b+36>>2]=-(60*a.getTimezoneOffset());c=(new Date(a.getFullYear(),
-6,1)).getTimezoneOffset();var d=(new Date(a.getFullYear(),0,1)).getTimezoneOffset();E[b+32>>2]=(c!=d&&a.getTimezoneOffset()==Math.min(d,c))|0},y:function(a,b,c,d,e,h,k){e=-9007199254740992>e||9007199254740992<e?NaN:Number(e);try{if(isNaN(e))return 61;var q=T(d);if(0!==(b&2)&&0===(c&2)&&2!==(q.flags&2097155))throw new N(2);if(1===(q.flags&2097155))throw new N(2);if(!q.Ma.ib)throw new N(43);if(!a)throw new N(28);var w=q.Ma.ib(q,a,e,b,c);var v=w.Kb;E[h>>2]=w.Ab;F[k>>2]=v;return 0}catch(C){if("undefined"==
-typeof W||"ErrnoError"!==C.name)throw C;return-C.Pa}},z:function(a,b,c,d,e,h){h=-9007199254740992>h||9007199254740992<h?NaN:Number(h);try{var k=T(e);if(c&2){c=h;if(32768!==(k.node.mode&61440))throw new N(43);if(!(d&2)){var q=x.slice(a,a+b);k.Ma.jb&&k.Ma.jb(k,q,c,b,d)}}}catch(w){if("undefined"==typeof W||"ErrnoError"!==w.name)throw w;return-w.Pa}},n:(a,b)=>{Nc[a]&&(clearTimeout(Nc[a].id),delete Nc[a]);if(!b)return 0;var c=setTimeout(()=>{delete Nc[a];Pc(()=>Wc(a,performance.now()))},b);Nc[a]={id:c,
-Xb:b};return 0},B:(a,b,c,d)=>{var e=(new Date).getFullYear(),h=(new Date(e,0,1)).getTimezoneOffset();e=(new Date(e,6,1)).getTimezoneOffset();F[a>>2]=60*Math.max(h,e);E[b>>2]=Number(h!=e);b=k=>{var q=Math.abs(k);return`UTC${0<=k?"-":"+"}${String(Math.floor(q/60)).padStart(2,"0")}${String(q%60).padStart(2,"0")}`};a=b(h);b=b(e);e<h?(u(a,x,c,17),u(b,x,d,17)):(u(a,x,d,17),u(b,x,c,17))},d:()=>Date.now(),s:()=>2147483648,c:()=>performance.now(),o:a=>{var b=x.length;a>>>=0;if(2147483648<a)return!1;for(var c=
-1;4>=c;c*=2){var d=b*(1+.2/c);d=Math.min(d,a+100663296);a:{d=(Math.min(2147483648,65536*Math.ceil(Math.max(a,d)/65536))-Ka.buffer.byteLength+65535)/65536|0;try{Ka.grow(d);Qa();var e=1;break a}catch(h){}e=void 0}if(e)return!0}return!1},E:(a,b)=>{var c=0;Sc().forEach((d,e)=>{var h=b+c;e=F[a+4*e>>2]=h;for(h=0;h<d.length;++h)p[e++]=d.charCodeAt(h);p[e]=0;c+=d.length+1});return 0},F:(a,b)=>{var c=Sc();F[a>>2]=c.length;var d=0;c.forEach(e=>d+=e.length+1);F[b>>2]=d;return 0},e:function(a){try{var b=T(a);
-qa(b);return 0}catch(c){if("undefined"==typeof W||"ErrnoError"!==c.name)throw c;return c.Pa}},p:function(a,b){try{var c=T(a);p[b]=c.tty?2:P(c.mode)?3:40960===(c.mode&61440)?7:4;Na[b+2>>1]=0;H[b+8>>3]=BigInt(0);H[b+16>>3]=BigInt(0);return 0}catch(d){if("undefined"==typeof W||"ErrnoError"!==d.name)throw d;return d.Pa}},w:function(a,b,c,d){try{a:{var e=T(a);a=b;for(var h,k=b=0;k<c;k++){var q=F[a>>2],w=F[a+4>>2];a+=8;var v=Ec(e,p,q,w,h);if(0>v){var C=-1;break a}b+=v;if(v<w)break;"undefined"!=typeof h&&
-(h+=v)}C=b}F[d>>2]=C;return 0}catch(G){if("undefined"==typeof W||"ErrnoError"!==G.name)throw G;return G.Pa}},D:function(a,b,c,d){b=-9007199254740992>b||9007199254740992<b?NaN:Number(b);try{if(isNaN(b))return 61;var e=T(a);mc(e,b,c);H[d>>3]=BigInt(e.position);e.ob&&0===b&&0===c&&(e.ob=null);return 0}catch(h){if("undefined"==typeof W||"ErrnoError"!==h.name)throw h;return h.Pa}},I:function(a){try{var b=T(a);return b.Ma?.fsync?b.Ma.fsync(b):0}catch(c){if("undefined"==typeof W||"ErrnoError"!==c.name)throw c;
-return c.Pa}},t:function(a,b,c,d){try{a:{var e=T(a);a=b;for(var h,k=b=0;k<c;k++){var q=F[a>>2],w=F[a+4>>2];a+=8;var v=pa(e,p,q,w,h);if(0>v){var C=-1;break a}b+=v;if(v<w)break;"undefined"!=typeof h&&(h+=v)}C=b}F[d>>2]=C;return 0}catch(G){if("undefined"==typeof W||"ErrnoError"!==G.name)throw G;return G.Pa}},k:Oc},Z;
-(async function(){function a(c){Z=c.exports;Ka=Z.M;Qa();Y=Z.O;K--;f.monitorRunDependencies?.(K);0==K&&Sa&&(c=Sa,Sa=null,c());return Z}K++;f.monitorRunDependencies?.(K);var b={a:Xc};if(f.instantiateWasm)return new Promise(c=>{f.instantiateWasm(b,(d,e)=>{a(d,e);c(d.exports)})});Ua??=f.locateFile?f.locateFile("sql-wasm.wasm",D):D+"sql-wasm.wasm";return a((await Xa(b)).instance)})();f._sqlite3_free=a=>(f._sqlite3_free=Z.P)(a);f._sqlite3_value_text=a=>(f._sqlite3_value_text=Z.Q)(a);
-f._sqlite3_prepare_v2=(a,b,c,d,e)=>(f._sqlite3_prepare_v2=Z.R)(a,b,c,d,e);f._sqlite3_step=a=>(f._sqlite3_step=Z.S)(a);f._sqlite3_reset=a=>(f._sqlite3_reset=Z.T)(a);f._sqlite3_exec=(a,b,c,d,e)=>(f._sqlite3_exec=Z.U)(a,b,c,d,e);f._sqlite3_finalize=a=>(f._sqlite3_finalize=Z.V)(a);f._sqlite3_column_name=(a,b)=>(f._sqlite3_column_name=Z.W)(a,b);f._sqlite3_column_text=(a,b)=>(f._sqlite3_column_text=Z.X)(a,b);f._sqlite3_column_type=(a,b)=>(f._sqlite3_column_type=Z.Y)(a,b);
-f._sqlite3_errmsg=a=>(f._sqlite3_errmsg=Z.Z)(a);f._sqlite3_clear_bindings=a=>(f._sqlite3_clear_bindings=Z._)(a);f._sqlite3_value_blob=a=>(f._sqlite3_value_blob=Z.$)(a);f._sqlite3_value_bytes=a=>(f._sqlite3_value_bytes=Z.aa)(a);f._sqlite3_value_double=a=>(f._sqlite3_value_double=Z.ba)(a);f._sqlite3_value_int=a=>(f._sqlite3_value_int=Z.ca)(a);f._sqlite3_value_type=a=>(f._sqlite3_value_type=Z.da)(a);f._sqlite3_result_blob=(a,b,c,d)=>(f._sqlite3_result_blob=Z.ea)(a,b,c,d);
-f._sqlite3_result_double=(a,b)=>(f._sqlite3_result_double=Z.fa)(a,b);f._sqlite3_result_error=(a,b,c)=>(f._sqlite3_result_error=Z.ga)(a,b,c);f._sqlite3_result_int=(a,b)=>(f._sqlite3_result_int=Z.ha)(a,b);f._sqlite3_result_int64=(a,b)=>(f._sqlite3_result_int64=Z.ia)(a,b);f._sqlite3_result_null=a=>(f._sqlite3_result_null=Z.ja)(a);f._sqlite3_result_text=(a,b,c,d)=>(f._sqlite3_result_text=Z.ka)(a,b,c,d);f._sqlite3_aggregate_context=(a,b)=>(f._sqlite3_aggregate_context=Z.la)(a,b);
-f._sqlite3_column_count=a=>(f._sqlite3_column_count=Z.ma)(a);f._sqlite3_data_count=a=>(f._sqlite3_data_count=Z.na)(a);f._sqlite3_column_blob=(a,b)=>(f._sqlite3_column_blob=Z.oa)(a,b);f._sqlite3_column_bytes=(a,b)=>(f._sqlite3_column_bytes=Z.pa)(a,b);f._sqlite3_column_double=(a,b)=>(f._sqlite3_column_double=Z.qa)(a,b);f._sqlite3_bind_blob=(a,b,c,d,e)=>(f._sqlite3_bind_blob=Z.ra)(a,b,c,d,e);f._sqlite3_bind_double=(a,b,c)=>(f._sqlite3_bind_double=Z.sa)(a,b,c);
-f._sqlite3_bind_int=(a,b,c)=>(f._sqlite3_bind_int=Z.ta)(a,b,c);f._sqlite3_bind_text=(a,b,c,d,e)=>(f._sqlite3_bind_text=Z.ua)(a,b,c,d,e);f._sqlite3_bind_parameter_index=(a,b)=>(f._sqlite3_bind_parameter_index=Z.va)(a,b);f._sqlite3_sql=a=>(f._sqlite3_sql=Z.wa)(a);f._sqlite3_normalized_sql=a=>(f._sqlite3_normalized_sql=Z.xa)(a);f._sqlite3_changes=a=>(f._sqlite3_changes=Z.ya)(a);f._sqlite3_close_v2=a=>(f._sqlite3_close_v2=Z.za)(a);
-f._sqlite3_create_function_v2=(a,b,c,d,e,h,k,q,w)=>(f._sqlite3_create_function_v2=Z.Aa)(a,b,c,d,e,h,k,q,w);f._sqlite3_update_hook=(a,b,c)=>(f._sqlite3_update_hook=Z.Ba)(a,b,c);f._sqlite3_open=(a,b)=>(f._sqlite3_open=Z.Ca)(a,b);var ia=f._malloc=a=>(ia=f._malloc=Z.Da)(a),fa=f._free=a=>(fa=f._free=Z.Ea)(a);f._RegisterExtensionFunctions=a=>(f._RegisterExtensionFunctions=Z.Fa)(a);var Db=(a,b)=>(Db=Z.Ga)(a,b),Wc=(a,b)=>(Wc=Z.Ha)(a,b),wa=a=>(wa=Z.Ia)(a),z=a=>(z=Z.Ja)(a),sa=()=>(sa=Z.Ka)();
-f.stackSave=()=>sa();f.stackRestore=a=>wa(a);f.stackAlloc=a=>z(a);f.cwrap=(a,b,c,d)=>{var e=!c||c.every(h=>"number"===h||"boolean"===h);return"string"!==b&&e&&!d?f["_"+a]:(...h)=>Tc(a,b,c,h)};f.addFunction=Aa;f.removeFunction=A;f.UTF8ToString=ua;f.ALLOC_NORMAL=ea;f.allocate=da;f.allocateUTF8OnStack=xa;
-function Yc(){function a(){f.calledRun=!0;if(!La){if(!f.noFSInit&&!Ib){var b,c;Ib=!0;d??=f.stdin;b??=f.stdout;c??=f.stderr;d?V("stdin",d):Zb("/dev/tty","/dev/stdin");b?V("stdout",null,b):Zb("/dev/tty","/dev/stdout");c?V("stderr",null,c):Zb("/dev/tty1","/dev/stderr");oa("/dev/stdin",0);oa("/dev/stdout",1);oa("/dev/stderr",1)}Z.N();Jb=!1;f.onRuntimeInitialized?.();if(f.postRun)for("function"==typeof f.postRun&&(f.postRun=[f.postRun]);f.postRun.length;){var d=f.postRun.shift();$a.unshift(d)}Za($a)}}
-if(0<K)Sa=Yc;else{if(f.preRun)for("function"==typeof f.preRun&&(f.preRun=[f.preRun]);f.preRun.length;)bb();Za(ab);0<K?Sa=Yc:f.setStatus?(f.setStatus("Running..."),setTimeout(()=>{setTimeout(()=>f.setStatus(""),1);a()},1)):a()}}if(f.preInit)for("function"==typeof f.preInit&&(f.preInit=[f.preInit]);0<f.preInit.length;)f.preInit.pop()();Yc();
-
-
-        // The shell-pre.js and emcc-generated code goes above
-        return Module;
-    }); // The end of the promise being returned
-
-  return initSqlJsPromise;
-} // The end of our initSqlJs function
-
-// This bit below is copied almost exactly from what you get when you use the MODULARIZE=1 flag with emcc
-// However, we don't want to use the emcc modularization. See shell-pre.js
-if (typeof exports === 'object' && typeof module === 'object'){
-    module.exports = initSqlJs;
-    // This will allow the module to be used in ES6 or CommonJS
-    module.exports.default = initSqlJs;
-}
-else if (typeof define === 'function' && define['amd']) {
-    define([], function() { return initSqlJs; });
-}
-else if (typeof exports === 'object'){
-    exports["Module"] = initSqlJs;
-}
-
-
-  // ============================================================================
-  // Advanced Features
-  // ============================================================================
-
-  /**
- * Product Quantization for Browser
- *
- * Compresses high-dimensional vectors using product quantization.
- * Achieves 4-32x memory reduction with minimal accuracy loss.
- *
- * Features:
- * - PQ8: 8 subvectors, 256 centroids each (4x compression)
- * - PQ16: 16 subvectors, 256 centroids each (8x compression)
- * - Asymmetric distance computation (ADC)
- * - K-means clustering for codebook training
- *
- * Performance:
- * - Memory: Float32 (4 bytes) → uint8 (1 byte) per subvector
- * - Speed: ~1.5x slower search vs uncompressed
- * - Accuracy: 95-99% recall@10
- */
-var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
-    function adopt(value) { return value instanceof P ? value : new P(function (resolve) { resolve(value); }); }
-    return new (P || (P = Promise))(function (resolve, reject) {
-        function fulfilled(value) { try { step(generator.next(value)); } catch (e) { reject(e); } }
-        function rejected(value) { try { step(generator["throw"](value)); } catch (e) { reject(e); } }
-        function step(result) { result.done ? resolve(result.value) : adopt(result.value).then(fulfilled, rejected); }
-        step((generator = generator.apply(thisArg, _arguments || [])).next());
-    });
-};
-class ProductQuantization {
-    constructor(config) {
-        this.codebook = null;
-        this.trained = false;
-        this.config = {
-            dimension: config.dimension,
-            numSubvectors: config.numSubvectors,
-            numCentroids: config.numCentroids,
-            maxIterations: config.maxIterations || 50,
-            convergenceThreshold: config.convergenceThreshold || 1e-4
-        };
-        // Validate config
-        if (this.config.dimension % this.config.numSubvectors !== 0) {
-            throw new Error(`Dimension ${this.config.dimension} must be divisible by numSubvectors ${this.config.numSubvectors}`);
-        }
-    }
-    /**
-     * Train codebook using k-means on training vectors
-     */
-    train(vectors) {
-        return __awaiter(this, void 0, void 0, function* () {
-            if (vectors.length === 0) {
-                throw new Error('Training requires at least one vector');
-            }
-            const subvectorDim = this.config.dimension / this.config.numSubvectors;
-            const centroids = [];
-            console.log(`[PQ] Training ${this.config.numSubvectors} subvectors with ${this.config.numCentroids} centroids each...`);
-            // Train each subvector independently
-            for (let s = 0; s < this.config.numSubvectors; s++) {
-                const startDim = s * subvectorDim;
-                const endDim = startDim + subvectorDim;
-                // Extract subvectors
-                const subvectors = vectors.map(v => v.slice(startDim, endDim));
-                // Run k-means
-                const subCentroids = yield this.kMeans(subvectors, this.config.numCentroids);
-                centroids.push(...subCentroids);
-                if ((s + 1) % 4 === 0 || s === this.config.numSubvectors - 1) {
-                    console.log(`[PQ] Trained ${s + 1}/${this.config.numSubvectors} subvectors`);
-                }
-            }
-            this.codebook = {
-                subvectorDim,
-                numSubvectors: this.config.numSubvectors,
-                numCentroids: this.config.numCentroids,
-                centroids
-            };
-            this.trained = true;
-            console.log('[PQ] Training complete');
-        });
-    }
-    /**
-     * K-means clustering for centroids
-     */
-    kMeans(vectors, k) {
-        return __awaiter(this, void 0, void 0, function* () {
-            const dim = vectors[0].length;
-            const n = vectors.length;
-            // Initialize centroids with k-means++
-            const centroids = this.kMeansPlusPlus(vectors, k);
-            const assignments = new Uint32Array(n);
-            let prevInertia = Infinity;
-            for (let iter = 0; iter < this.config.maxIterations; iter++) {
-                // Assign vectors to nearest centroid
-                let inertia = 0;
-                for (let i = 0; i < n; i++) {
-                    let minDist = Infinity;
-                    let minIdx = 0;
-                    for (let j = 0; j < k; j++) {
-                        const dist = this.squaredDistance(vectors[i], centroids[j]);
-                        if (dist < minDist) {
-                            minDist = dist;
-                            minIdx = j;
-                        }
-                    }
-                    assignments[i] = minIdx;
-                    inertia += minDist;
-                }
-                // Check convergence
-                if (Math.abs(prevInertia - inertia) < this.config.convergenceThreshold) {
-                    break;
-                }
-                prevInertia = inertia;
-                // Update centroids
-                const counts = new Uint32Array(k);
-                const sums = Array.from({ length: k }, () => new Float32Array(dim));
-                for (let i = 0; i < n; i++) {
-                    const cluster = assignments[i];
-                    counts[cluster]++;
-                    for (let d = 0; d < dim; d++) {
-                        sums[cluster][d] += vectors[i][d];
-                    }
-                }
-                for (let j = 0; j < k; j++) {
-                    if (counts[j] > 0) {
-                        for (let d = 0; d < dim; d++) {
-                            centroids[j][d] = sums[j][d] / counts[j];
-                        }
-                    }
-                }
-            }
-            return centroids;
-        });
-    }
-    /**
-     * K-means++ initialization for better centroid selection
-     */
-    kMeansPlusPlus(vectors, k) {
-        const n = vectors.length;
-        const dim = vectors[0].length;
-        const centroids = [];
-        // Choose first centroid randomly
-        const firstIdx = Math.floor(Math.random() * n);
-        centroids.push(new Float32Array(vectors[firstIdx]));
-        // Choose remaining centroids
-        for (let i = 1; i < k; i++) {
-            const distances = new Float32Array(n);
-            let sumDistances = 0;
-            // Calculate distances to nearest centroid
-            for (let j = 0; j < n; j++) {
-                let minDist = Infinity;
-                for (const centroid of centroids) {
-                    const dist = this.squaredDistance(vectors[j], centroid);
-                    minDist = Math.min(minDist, dist);
-                }
-                distances[j] = minDist;
-                sumDistances += minDist;
-            }
-            // Choose next centroid with probability proportional to distance²
-            let r = Math.random() * sumDistances;
-            for (let j = 0; j < n; j++) {
-                r -= distances[j];
-                if (r <= 0) {
-                    centroids.push(new Float32Array(vectors[j]));
-                    break;
-                }
-            }
-        }
-        return centroids;
-    }
-    /**
-     * Compress a vector using trained codebook
-     */
-    compress(vector) {
-        if (!this.trained || !this.codebook) {
-            throw new Error('Codebook must be trained before compression');
-        }
-        const codes = new Uint8Array(this.config.numSubvectors);
-        const subvectorDim = this.codebook.subvectorDim;
-        // Compute norm for later reconstruction
-        let norm = 0;
-        for (let i = 0; i < vector.length; i++) {
-            norm += vector[i] * vector[i];
-        }
-        norm = Math.sqrt(norm);
-        // Encode each subvector
-        for (let s = 0; s < this.config.numSubvectors; s++) {
-            const startDim = s * subvectorDim;
-            const subvector = vector.slice(startDim, startDim + subvectorDim);
-            // Find nearest centroid
-            let minDist = Infinity;
-            let minIdx = 0;
-            const centroidOffset = s * this.config.numCentroids;
-            for (let c = 0; c < this.config.numCentroids; c++) {
-                const centroid = this.codebook.centroids[centroidOffset + c];
-                const dist = this.squaredDistance(subvector, centroid);
-                if (dist < minDist) {
-                    minDist = dist;
-                    minIdx = c;
-                }
-            }
-            codes[s] = minIdx;
-        }
-        return { codes, norm };
-    }
-    /**
-     * Decompress a vector (approximate reconstruction)
-     */
-    decompress(compressed) {
-        if (!this.codebook) {
-            throw new Error('Codebook not available');
-        }
-        const vector = new Float32Array(this.config.dimension);
-        const subvectorDim = this.codebook.subvectorDim;
-        for (let s = 0; s < this.config.numSubvectors; s++) {
-            const code = compressed.codes[s];
-            const centroidOffset = s * this.config.numCentroids;
-            const centroid = this.codebook.centroids[centroidOffset + code];
-            const startDim = s * subvectorDim;
-            for (let d = 0; d < subvectorDim; d++) {
-                vector[startDim + d] = centroid[d];
-            }
-        }
-        return vector;
-    }
-    /**
-     * Asymmetric Distance Computation (ADC)
-     * Computes distance from query vector to compressed vector
-     */
-    asymmetricDistance(query, compressed) {
-        if (!this.codebook) {
-            throw new Error('Codebook not available');
-        }
-        let distance = 0;
-        const subvectorDim = this.codebook.subvectorDim;
-        for (let s = 0; s < this.config.numSubvectors; s++) {
-            const code = compressed.codes[s];
-            const centroidOffset = s * this.config.numCentroids;
-            const centroid = this.codebook.centroids[centroidOffset + code];
-            const startDim = s * subvectorDim;
-            const querySubvector = query.slice(startDim, startDim + subvectorDim);
-            distance += this.squaredDistance(querySubvector, centroid);
-        }
-        return Math.sqrt(distance);
-    }
-    /**
-     * Batch compression for multiple vectors
-     */
-    batchCompress(vectors) {
-        return vectors.map(v => this.compress(v));
-    }
-    /**
-     * Get memory savings
-     */
-    getCompressionRatio() {
-        // Original: dimension * 4 bytes (Float32)
-        // Compressed: numSubvectors * 1 byte (Uint8) + 4 bytes (norm)
-        const originalBytes = this.config.dimension * 4;
-        const compressedBytes = this.config.numSubvectors + 4;
-        return originalBytes / compressedBytes;
-    }
-    /**
-     * Export codebook for persistence
-     */
-    exportCodebook() {
-        if (!this.codebook) {
-            throw new Error('No codebook to export');
-        }
-        return JSON.stringify({
-            config: this.config,
-            codebook: {
-                subvectorDim: this.codebook.subvectorDim,
-                numSubvectors: this.codebook.numSubvectors,
-                numCentroids: this.codebook.numCentroids,
-                centroids: this.codebook.centroids.map(c => Array.from(c))
-            }
-        });
-    }
-    /**
-     * Import codebook
-     */
-    importCodebook(json) {
-        const data = JSON.parse(json);
-        this.config = data.config;
-        this.codebook = {
-            subvectorDim: data.codebook.subvectorDim,
-            numSubvectors: data.codebook.numSubvectors,
-            numCentroids: data.codebook.numCentroids,
-            centroids: data.codebook.centroids.map((c) => new Float32Array(c))
-        };
-        this.trained = true;
-    }
-    /**
-     * Utility: Squared Euclidean distance
-     */
-    squaredDistance(a, b) {
-        let sum = 0;
-        for (let i = 0; i < a.length; i++) {
-            const diff = a[i] - b[i];
-            sum += diff * diff;
-        }
-        return sum;
-    }
-    /**
-     * Get statistics
-     */
-    getStats() {
-        const compressionRatio = this.getCompressionRatio();
-        const memoryPerVector = this.config.numSubvectors + 4; // codes + norm
-        const codebookSize = this.codebook
-            ? this.config.numSubvectors * this.config.numCentroids * (this.config.dimension / this.config.numSubvectors) * 4
-            : 0;
-        return {
-            trained: this.trained,
-            compressionRatio,
-            memoryPerVector,
-            codebookSize
-        };
-    }
-}
-/**
- * Helper function to create PQ8 (8 subvectors, 4x compression)
- */
-function createPQ8(dimension) {
-    return new ProductQuantization({
-        dimension,
-        numSubvectors: 8,
-        numCentroids: 256,
-        maxIterations: 50
-    });
-}
-/**
- * Helper function to create PQ16 (16 subvectors, 8x compression)
- */
-function createPQ16(dimension) {
-    return new ProductQuantization({
-        dimension,
-        numSubvectors: 16,
-        numCentroids: 256,
-        maxIterations: 50
-    });
-}
-/**
- * Helper function to create PQ32 (32 subvectors, 16x compression)
- */
-function createPQ32(dimension) {
-    return new ProductQuantization({
-        dimension,
-        numSubvectors: 32,
-        numCentroids: 256,
-        maxIterations: 50
-    });
-}
-
-
-  /**
- * HNSW (Hierarchical Navigable Small World) Index for Browser
- *
- * JavaScript implementation of HNSW algorithm for fast approximate nearest neighbor search.
- * Achieves O(log n) search complexity vs O(n) for linear scan.
- *
- * Features:
- * - Multi-layer graph structure
- * - Probabilistic layer assignment
- * - Greedy search algorithm
- * - Dynamic insertion
- * - Configurable M (connections per node)
- * - Configurable efConstruction and efSearch
- *
- * Performance:
- * - 10-20x faster than linear scan (vs 150x for native HNSW)
- * - Memory: ~16 bytes per edge + vector storage
- * - Suitable for datasets up to 100K vectors in browser
- */
-class MinHeap {
-    constructor() {
-        this.items = [];
-    }
-    push(item, priority) {
-        this.items.push({ item, priority });
-        this.bubbleUp(this.items.length - 1);
-    }
-    pop() {
-        if (this.items.length === 0)
-            return undefined;
-        const result = this.items[0].item;
-        const last = this.items.pop();
-        if (this.items.length > 0) {
-            this.items[0] = last;
-            this.bubbleDown(0);
-        }
-        return result;
-    }
-    peek() {
-        var _a;
-        return (_a = this.items[0]) === null || _a === void 0 ? void 0 : _a.item;
-    }
-    size() {
-        return this.items.length;
-    }
-    bubbleUp(index) {
-        while (index > 0) {
-            const parentIndex = Math.floor((index - 1) / 2);
-            if (this.items[index].priority >= this.items[parentIndex].priority)
-                break;
-            [this.items[index], this.items[parentIndex]] = [this.items[parentIndex], this.items[index]];
-            index = parentIndex;
-        }
-    }
-    bubbleDown(index) {
-        while (true) {
-            const leftChild = 2 * index + 1;
-            const rightChild = 2 * index + 2;
-            let smallest = index;
-            if (leftChild < this.items.length && this.items[leftChild].priority < this.items[smallest].priority) {
-                smallest = leftChild;
-            }
-            if (rightChild < this.items.length && this.items[rightChild].priority < this.items[smallest].priority) {
-                smallest = rightChild;
-            }
-            if (smallest === index)
-                break;
-            [this.items[index], this.items[smallest]] = [this.items[smallest], this.items[index]];
-            index = smallest;
-        }
-    }
-}
-class HNSWIndex {
-    constructor(config = {}) {
-        this.nodes = new Map();
-        this.entryPoint = null;
-        this.currentId = 0;
-        this.config = {
-            dimension: config.dimension || 384,
-            M: config.M || 16,
-            efConstruction: config.efConstruction || 200,
-            efSearch: config.efSearch || 50,
-            ml: config.ml || 1 / Math.log(2),
-            maxLayers: config.maxLayers || 16,
-            distanceFunction: config.distanceFunction || 'cosine'
-        };
-        this.ml = this.config.ml;
-    }
-    /**
-     * Add vector to index
-     */
-    add(vector, id) {
-        const nodeId = id !== undefined ? id : this.currentId++;
-        const level = this.randomLevel();
-        const node = {
-            id: nodeId,
-            vector,
-            level,
-            connections: new Map()
-        };
-        // Initialize connections for each layer
-        for (let l = 0; l <= level; l++) {
-            node.connections.set(l, []);
-        }
-        if (this.entryPoint === null) {
-            // First node
-            this.entryPoint = nodeId;
-            this.nodes.set(nodeId, node);
-            return nodeId;
-        }
-        // Find nearest neighbors at each layer
-        const ep = this.entryPoint;
-        let nearest = ep;
-        // Search from top layer to target layer + 1
-        for (let lc = this.nodes.get(ep).level; lc > level; lc--) {
-            nearest = this.searchLayer(vector, nearest, 1, lc)[0];
-        }
-        // Insert node at layers 0 to level
-        for (let lc = Math.min(level, this.nodes.get(ep).level); lc >= 0; lc--) {
-            const candidates = this.searchLayer(vector, nearest, this.config.efConstruction, lc);
-            // Select M neighbors
-            const M = lc === 0 ? this.config.M * 2 : this.config.M;
-            const neighbors = this.selectNeighbors(vector, candidates, M);
-            // Add bidirectional connections
-            for (const neighbor of neighbors) {
-                this.connect(nodeId, neighbor, lc);
-                this.connect(neighbor, nodeId, lc);
-                // Prune connections if necessary
-                const neighborNode = this.nodes.get(neighbor);
-                const neighborConnections = neighborNode.connections.get(lc);
-                if (neighborConnections.length > M) {
-                    const newNeighbors = this.selectNeighbors(neighborNode.vector, neighborConnections, M);
-                    neighborNode.connections.set(lc, newNeighbors);
-                }
-            }
-            nearest = candidates[0];
-        }
-        // Update entry point if necessary
-        if (level > this.nodes.get(this.entryPoint).level) {
-            this.entryPoint = nodeId;
-        }
-        this.nodes.set(nodeId, node);
-        return nodeId;
-    }
-    /**
-     * Search for k nearest neighbors
-     */
-    search(query, k, ef) {
-        if (this.entryPoint === null)
-            return [];
-        ef = ef || Math.max(this.config.efSearch, k);
-        let ep = this.entryPoint;
-        let nearest = ep;
-        // Search from top to layer 1
-        for (let lc = this.nodes.get(ep).level; lc > 0; lc--) {
-            nearest = this.searchLayer(query, nearest, 1, lc)[0];
-        }
-        // Search at layer 0
-        const candidates = this.searchLayer(query, nearest, ef, 0);
-        // Convert to SearchResult and return top k
-        return candidates
-            .slice(0, k)
-            .map(id => ({
-            id,
-            distance: this.distance(query, this.nodes.get(id).vector),
-            vector: this.nodes.get(id).vector
-        }));
-    }
-    /**
-     * Search at specific layer
-     */
-    searchLayer(query, ep, ef, layer) {
-        const visited = new Set();
-        const candidates = new MinHeap();
-        const w = new MinHeap();
-        const dist = this.distance(query, this.nodes.get(ep).vector);
-        candidates.push(ep, dist);
-        w.push(ep, -dist); // Max heap (negate for min heap)
-        visited.add(ep);
-        while (candidates.size() > 0) {
-            const c = candidates.pop();
-            const fDist = -w.peek(); // Furthest point distance
-            const cDist = this.distance(query, this.nodes.get(c).vector);
-            if (cDist > fDist)
-                break;
-            const neighbors = this.nodes.get(c).connections.get(layer) || [];
-            for (const e of neighbors) {
-                if (visited.has(e))
-                    continue;
-                visited.add(e);
-                const eDist = this.distance(query, this.nodes.get(e).vector);
-                const fDist = -w.peek();
-                if (eDist < fDist || w.size() < ef) {
-                    candidates.push(e, eDist);
-                    w.push(e, -eDist);
-                    if (w.size() > ef) {
-                        w.pop();
-                    }
-                }
-            }
-        }
-        // Return ef nearest neighbors
-        const result = [];
-        while (w.size() > 0) {
-            result.unshift(w.pop());
-        }
-        return result;
-    }
-    /**
-     * Select best neighbors using heuristic
-     */
-    selectNeighbors(base, candidates, M) {
-        if (candidates.length <= M)
-            return candidates;
-        // Sort by distance
-        const sorted = candidates
-            .map(id => ({
-            id,
-            distance: this.distance(base, this.nodes.get(id).vector)
-        }))
-            .sort((a, b) => a.distance - b.distance);
-        return sorted.slice(0, M).map(x => x.id);
-    }
-    /**
-     * Connect two nodes at layer
-     */
-    connect(from, to, layer) {
-        const node = this.nodes.get(from);
-        const connections = node.connections.get(layer);
-        if (!connections.includes(to)) {
-            connections.push(to);
-        }
-    }
-    /**
-     * Random level assignment
-     */
-    randomLevel() {
-        let level = 0;
-        while (Math.random() < this.ml && level < this.config.maxLayers - 1) {
-            level++;
-        }
-        return level;
-    }
-    /**
-     * Distance function
-     */
-    distance(a, b) {
-        switch (this.config.distanceFunction) {
-            case 'cosine':
-                return 1 - this.cosineSimilarity(a, b);
-            case 'euclidean':
-                return this.euclideanDistance(a, b);
-            case 'manhattan':
-                return this.manhattanDistance(a, b);
-            default:
-                return 1 - this.cosineSimilarity(a, b);
-        }
-    }
-    cosineSimilarity(a, b) {
-        let dotProduct = 0;
-        let normA = 0;
-        let normB = 0;
-        for (let i = 0; i < a.length; i++) {
-            dotProduct += a[i] * b[i];
-            normA += a[i] * a[i];
-            normB += b[i] * b[i];
-        }
-        return dotProduct / (Math.sqrt(normA) * Math.sqrt(normB));
-    }
-    euclideanDistance(a, b) {
-        let sum = 0;
-        for (let i = 0; i < a.length; i++) {
-            const diff = a[i] - b[i];
-            sum += diff * diff;
-        }
-        return Math.sqrt(sum);
-    }
-    manhattanDistance(a, b) {
-        let sum = 0;
-        for (let i = 0; i < a.length; i++) {
-            sum += Math.abs(a[i] - b[i]);
-        }
-        return sum;
-    }
-    /**
-     * Get index statistics
-     */
-    getStats() {
-        if (this.nodes.size === 0) {
-            return {
-                numNodes: 0,
-                numLayers: 0,
-                avgConnections: 0,
-                entryPointLevel: 0,
-                memoryBytes: 0
-            };
-        }
-        const maxLevel = Math.max(...Array.from(this.nodes.values()).map(n => n.level));
-        let totalConnections = 0;
-        for (const node of this.nodes.values()) {
-            for (const connections of node.connections.values()) {
-                totalConnections += connections.length;
-            }
-        }
-        const avgConnections = totalConnections / this.nodes.size;
-        // Estimate memory: vector + connections + metadata
-        const vectorBytes = this.config.dimension * 4; // Float32Array
-        const connectionBytes = avgConnections * 4; // number array
-        const metadataBytes = 100; // rough estimate for node object
-        const memoryBytes = this.nodes.size * (vectorBytes + connectionBytes + metadataBytes);
-        return {
-            numNodes: this.nodes.size,
-            numLayers: maxLevel + 1,
-            avgConnections,
-            entryPointLevel: this.entryPoint ? this.nodes.get(this.entryPoint).level : 0,
-            memoryBytes
-        };
-    }
-    /**
-     * Export index for persistence
-     */
-    export() {
-        const data = {
-            config: this.config,
-            entryPoint: this.entryPoint,
-            currentId: this.currentId,
-            nodes: Array.from(this.nodes.entries()).map(([id, node]) => ({
-                id,
-                vector: Array.from(node.vector),
-                level: node.level,
-                connections: Array.from(node.connections.entries())
-            }))
-        };
-        return JSON.stringify(data);
-    }
-    /**
-     * Import index from JSON
-     */
-    import(json) {
-        const data = JSON.parse(json);
-        this.config = data.config;
-        this.entryPoint = data.entryPoint;
-        this.currentId = data.currentId;
-        this.nodes.clear();
-        for (const nodeData of data.nodes) {
-            const node = {
-                id: nodeData.id,
-                vector: new Float32Array(nodeData.vector),
-                level: nodeData.level,
-                connections: new Map(nodeData.connections)
-            };
-            this.nodes.set(nodeData.id, node);
-        }
-    }
-    /**
-     * Clear index
-     */
-    clear() {
-        this.nodes.clear();
-        this.entryPoint = null;
-        this.currentId = 0;
-    }
-    /**
-     * Get number of nodes
-     */
-    size() {
-        return this.nodes.size;
-    }
-}
-/**
- * Helper function to create HNSW index with default settings
- */
-function createHNSW(dimension) {
-    return new HNSWIndex({
-        dimension,
-        M: 16,
-        efConstruction: 200,
-        efSearch: 50
-    });
-}
-/**
- * Helper function to create fast HNSW (lower quality, faster build)
- */
-function createFastHNSW(dimension) {
-    return new HNSWIndex({
-        dimension,
-        M: 8,
-        efConstruction: 100,
-        efSearch: 30
-    });
-}
-/**
- * Helper function to create accurate HNSW (higher quality, slower build)
- */
-function createAccurateHNSW(dimension) {
-    return new HNSWIndex({
-        dimension,
-        M: 32,
-        efConstruction: 400,
-        efSearch: 100
-    });
-}
-
-
-  /**
- * Advanced Features for AgentDB Browser
- *
- * Includes:
- * - GNN (Graph Neural Networks) - Graph attention and message passing
- * - MMR (Maximal Marginal Relevance) - Diversity ranking
- * - SVD (Singular Value Decomposition) - Tensor compression
- * - Batch operations and utilities
- */
-/**
- * Graph Neural Network with attention mechanism
- */
-class GraphNeuralNetwork {
-    constructor(config = {}) {
-        this.nodes = new Map();
-        this.edges = [];
-        this.attentionWeights = new Map();
-        this.config = {
-            hiddenDim: config.hiddenDim || 64,
-            numHeads: config.numHeads || 4,
-            dropout: config.dropout || 0.1,
-            learningRate: config.learningRate || 0.01,
-            attentionType: config.attentionType || 'gat'
-        };
-    }
-    /**
-     * Add node to graph
-     */
-    addNode(id, features) {
-        this.nodes.set(id, {
-            id,
-            features,
-            neighbors: []
-        });
-    }
-    /**
-     * Add edge to graph
-     */
-    addEdge(from, to, weight = 1.0) {
-        this.edges.push({ from, to, weight });
-        // Update neighbor lists
-        const fromNode = this.nodes.get(from);
-        const toNode = this.nodes.get(to);
-        if (fromNode && !fromNode.neighbors.includes(to)) {
-            fromNode.neighbors.push(to);
-        }
-        if (toNode && !toNode.neighbors.includes(from)) {
-            toNode.neighbors.push(from);
-        }
-    }
-    /**
-     * Graph Attention Network (GAT) message passing
-     */
-    graphAttention(nodeId) {
-        const node = this.nodes.get(nodeId);
-        if (!node)
-            throw new Error(`Node ${nodeId} not found`);
-        const neighbors = node.neighbors;
-        if (neighbors.length === 0) {
-            return node.features;
-        }
-        // Multi-head attention
-        const headDim = Math.floor(this.config.hiddenDim / this.config.numHeads);
-        const aggregated = new Float32Array(this.config.hiddenDim);
-        for (let h = 0; h < this.config.numHeads; h++) {
-            let attentionSum = 0;
-            const headOutput = new Float32Array(headDim);
-            // Compute attention scores for each neighbor
-            for (const neighborId of neighbors) {
-                const neighbor = this.nodes.get(neighborId);
-                // Attention score: similarity between node and neighbor
-                const score = this.computeAttentionScore(node.features, neighbor.features, h);
-                attentionSum += score;
-                // Aggregate neighbor features weighted by attention
-                for (let i = 0; i < headDim && i < neighbor.features.length; i++) {
-                    headOutput[i] += score * neighbor.features[i];
-                }
-            }
-            // Normalize by attention sum
-            if (attentionSum > 0) {
-                for (let i = 0; i < headDim; i++) {
-                    headOutput[i] /= attentionSum;
-                }
-            }
-            // Concatenate head outputs
-            const offset = h * headDim;
-            for (let i = 0; i < headDim; i++) {
-                aggregated[offset + i] = headOutput[i];
-            }
-        }
-        // Apply non-linearity (LeakyReLU)
-        for (let i = 0; i < aggregated.length; i++) {
-            aggregated[i] = aggregated[i] > 0 ? aggregated[i] : 0.01 * aggregated[i];
-        }
-        return aggregated;
-    }
-    /**
-     * Compute attention score between two nodes
-     */
-    computeAttentionScore(features1, features2, head) {
-        // Simple dot-product attention
-        let score = 0;
-        const len = Math.min(features1.length, features2.length);
-        for (let i = 0; i < len; i++) {
-            score += features1[i] * features2[i];
-        }
-        // Apply softmax-like normalization
-        return Math.exp(score / Math.sqrt(len));
-    }
-    /**
-     * Message passing for all nodes
-     */
-    messagePass() {
-        const newFeatures = new Map();
-        for (const [nodeId] of this.nodes) {
-            newFeatures.set(nodeId, this.graphAttention(nodeId));
-        }
-        return newFeatures;
-    }
-    /**
-     * Update node features after message passing
-     */
-    update(newFeatures) {
-        for (const [nodeId, features] of newFeatures) {
-            const node = this.nodes.get(nodeId);
-            if (node) {
-                node.features = features;
-            }
-        }
-    }
-    /**
-     * Compute graph embeddings for query enhancement
-     */
-    computeGraphEmbedding(nodeId, hops = 2) {
-        const features = new Map();
-        features.set(nodeId, this.nodes.get(nodeId).features);
-        // Multi-hop message passing
-        for (let h = 0; h < hops; h++) {
-            const newFeatures = this.messagePass();
-            this.update(newFeatures);
-        }
-        return this.nodes.get(nodeId).features;
-    }
-    /**
-     * Get statistics
-     */
-    getStats() {
-        return {
-            numNodes: this.nodes.size,
-            numEdges: this.edges.length,
-            avgDegree: this.edges.length / Math.max(this.nodes.size, 1),
-            config: this.config
-        };
-    }
-}
-/**
- * Maximal Marginal Relevance for diversity ranking
- */
-class MaximalMarginalRelevance {
-    constructor(config = {}) {
-        this.config = {
-            lambda: config.lambda || 0.7,
-            metric: config.metric || 'cosine'
-        };
-    }
-    /**
-     * Rerank results for diversity
-     * @param query Query vector
-     * @param candidates Candidate vectors with scores
-     * @param k Number of results to return
-     * @returns Reranked indices
-     */
-    rerank(query, candidates, k) {
-        if (candidates.length === 0)
-            return [];
-        const selected = [];
-        const remaining = new Set(candidates.map((_, i) => i));
-        // Select first result (highest relevance)
-        let bestIdx = 0;
-        let bestScore = -Infinity;
-        for (let i = 0; i < candidates.length; i++) {
-            if (candidates[i].score > bestScore) {
-                bestScore = candidates[i].score;
-                bestIdx = i;
-            }
-        }
-        selected.push(candidates[bestIdx].id);
-        remaining.delete(bestIdx);
-        // Iteratively select remaining results
-        while (selected.length < k && remaining.size > 0) {
-            let bestMMR = -Infinity;
-            let bestCandidate = -1;
-            for (const idx of remaining) {
-                const candidate = candidates[idx];
-                // Relevance to query
-                const relevance = this.similarity(query, candidate.vector);
-                // Maximum similarity to already selected
-                let maxSimilarity = -Infinity;
-                for (const selectedId of selected) {
-                    const selectedCandidate = candidates.find(c => c.id === selectedId);
-                    const sim = this.similarity(candidate.vector, selectedCandidate.vector);
-                    maxSimilarity = Math.max(maxSimilarity, sim);
-                }
-                // MMR score
-                const mmr = this.config.lambda * relevance -
-                    (1 - this.config.lambda) * maxSimilarity;
-                if (mmr > bestMMR) {
-                    bestMMR = mmr;
-                    bestCandidate = idx;
-                }
-            }
-            if (bestCandidate !== -1) {
-                selected.push(candidates[bestCandidate].id);
-                remaining.delete(bestCandidate);
-            }
-            else {
-                break;
-            }
-        }
-        return selected;
-    }
-    /**
-     * Similarity computation
-     */
-    similarity(a, b) {
-        if (this.config.metric === 'cosine') {
-            return this.cosineSimilarity(a, b);
-        }
-        else {
-            // Euclidean distance converted to similarity
-            const dist = this.euclideanDistance(a, b);
-            return 1 / (1 + dist);
-        }
-    }
-    cosineSimilarity(a, b) {
-        let dotProduct = 0;
-        let normA = 0;
-        let normB = 0;
-        for (let i = 0; i < a.length; i++) {
-            dotProduct += a[i] * b[i];
-            normA += a[i] * a[i];
-            normB += b[i] * b[i];
-        }
-        return dotProduct / (Math.sqrt(normA) * Math.sqrt(normB));
-    }
-    euclideanDistance(a, b) {
-        let sum = 0;
-        for (let i = 0; i < a.length; i++) {
-            const diff = a[i] - b[i];
-            sum += diff * diff;
-        }
-        return Math.sqrt(sum);
-    }
-    /**
-     * Set lambda (relevance vs diversity trade-off)
-     */
-    setLambda(lambda) {
-        this.config.lambda = Math.max(0, Math.min(1, lambda));
-    }
-}
-// ============================================================================
-// SVD (Singular Value Decomposition) for Tensor Compression
-// ============================================================================
-/**
- * Simple SVD implementation for dimension reduction
- */
-class TensorCompression {
-    /**
-     * Reduce dimensionality using truncated SVD
-     * @param vectors Array of vectors to compress
-     * @param targetDim Target dimension
-     * @returns Compressed vectors
-     */
-    static compress(vectors, targetDim) {
-        if (vectors.length === 0)
-            return [];
-        const originalDim = vectors[0].length;
-        if (targetDim >= originalDim)
-            return vectors;
-        // Create matrix (vectors as rows)
-        const matrix = vectors.map(v => Array.from(v));
-        // Center the data (subtract mean)
-        const mean = this.computeMean(matrix);
-        const centered = matrix.map(row => row.map((val, i) => val - mean[i]));
-        // Compute covariance matrix
-        const cov = this.computeCovariance(centered);
-        // Compute top k eigenvectors using power iteration
-        const eigenvectors = this.powerIteration(cov, targetDim);
-        // Project vectors onto eigenvectors
-        const compressed = centered.map(row => {
-            const projected = new Float32Array(targetDim);
-            for (let i = 0; i < targetDim; i++) {
-                let sum = 0;
-                for (let j = 0; j < originalDim; j++) {
-                    sum += row[j] * eigenvectors[i][j];
-                }
-                projected[i] = sum;
-            }
-            return projected;
-        });
-        return compressed;
-    }
-    /**
-     * Compute mean vector
-     */
-    static computeMean(matrix) {
-        const n = matrix.length;
-        const dim = matrix[0].length;
-        const mean = new Array(dim).fill(0);
-        for (const row of matrix) {
-            for (let i = 0; i < dim; i++) {
-                mean[i] += row[i];
-            }
-        }
-        return mean.map(v => v / n);
-    }
-    /**
-     * Compute covariance matrix
-     */
-    static computeCovariance(matrix) {
-        const n = matrix.length;
-        const dim = matrix[0].length;
-        const cov = Array.from({ length: dim }, () => new Array(dim).fill(0));
-        for (let i = 0; i < dim; i++) {
-            for (let j = 0; j <= i; j++) {
-                let sum = 0;
-                for (const row of matrix) {
-                    sum += row[i] * row[j];
-                }
-                cov[i][j] = cov[j][i] = sum / n;
-            }
-        }
-        return cov;
-    }
-    /**
-     * Power iteration for computing top eigenvectors
-     */
-    static powerIteration(matrix, k, iterations = 100) {
-        const dim = matrix.length;
-        const eigenvectors = [];
-        for (let i = 0; i < k; i++) {
-            // Random initialization
-            let v = new Array(dim).fill(0).map(() => Math.random() - 0.5);
-            // Power iteration
-            for (let iter = 0; iter < iterations; iter++) {
-                // Multiply by matrix
-                const newV = new Array(dim).fill(0);
-                for (let r = 0; r < dim; r++) {
-                    for (let c = 0; c < dim; c++) {
-                        newV[r] += matrix[r][c] * v[c];
-                    }
-                }
-                // Orthogonalize against previous eigenvectors
-                for (const prev of eigenvectors) {
-                    let dot = 0;
-                    for (let j = 0; j < dim; j++) {
-                        dot += newV[j] * prev[j];
-                    }
-                    for (let j = 0; j < dim; j++) {
-                        newV[j] -= dot * prev[j];
-                    }
-                }
-                // Normalize
-                let norm = 0;
-                for (const val of newV) {
-                    norm += val * val;
-                }
-                norm = Math.sqrt(norm);
-                if (norm < 1e-10)
-                    break;
-                v = newV.map(val => val / norm);
-            }
-            eigenvectors.push(v);
-        }
-        return eigenvectors;
-    }
-}
-// ============================================================================
-// Batch Operations
-// ============================================================================
-/**
- * Efficient batch processing utilities
- */
-class BatchProcessor {
-    /**
-     * Batch cosine similarity computation
-     */
-    static batchCosineSimilarity(query, vectors) {
-        const similarities = new Float32Array(vectors.length);
-        // Precompute query norm
-        let queryNorm = 0;
-        for (let i = 0; i < query.length; i++) {
-            queryNorm += query[i] * query[i];
-        }
-        queryNorm = Math.sqrt(queryNorm);
-        // Compute similarities
-        for (let v = 0; v < vectors.length; v++) {
-            const vector = vectors[v];
-            let dotProduct = 0;
-            let vectorNorm = 0;
-            for (let i = 0; i < query.length; i++) {
-                dotProduct += query[i] * vector[i];
-                vectorNorm += vector[i] * vector[i];
-            }
-            vectorNorm = Math.sqrt(vectorNorm);
-            similarities[v] = dotProduct / (queryNorm * vectorNorm);
-        }
-        return similarities;
-    }
-    /**
-     * Batch vector normalization
-     */
-    static batchNormalize(vectors) {
-        return vectors.map(v => {
-            let norm = 0;
-            for (let i = 0; i < v.length; i++) {
-                norm += v[i] * v[i];
-            }
-            norm = Math.sqrt(norm);
-            const normalized = new Float32Array(v.length);
-            for (let i = 0; i < v.length; i++) {
-                normalized[i] = v[i] / norm;
-            }
-            return normalized;
-        });
-    }
-}
-
-
-  // ============================================================================
-  // Feature Index & Utilities
-  // ============================================================================
-
-  /**
- * AgentDB Browser Advanced Features
- *
- * Unified for all browser-compatible advanced features.
- *
- * Features:
- * - Product Quantization (PQ8/PQ16/PQ32) - 4-32x memory compression
- * - HNSW Indexing - 10-20x faster approximate search
- * - Graph Neural Networks - Graph attention and message passing
- * - MMR Diversity - Maximal marginal relevance ranking
- * - Tensor Compression - SVD dimension reduction
- * - Batch Operations - Optimized vector processing
- *
- * Bundle Size: ~35 KB minified (~12 KB gzipped)
- */
-var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
-    function adopt(value) { return value instanceof P ? value : new P(function (resolve) { resolve(value); }); }
-    return new (P || (P = Promise))(function (resolve, reject) {
-        function fulfilled(value) { try { step(generator.next(value)); } catch (e) { reject(e); } }
-        function rejected(value) { try { step(generator["throw"](value)); } catch (e) { reject(e); } }
-        function step(result) { result.done ? resolve(result.value) : adopt(result.value).then(fulfilled, rejected); }
-        step((generator = generator.apply(thisArg, _arguments || [])).next());
-    });
-};
-// ============================================================================
-// Product Quantization
-// ============================================================================
-// ============================================================================
-// HNSW Indexing
-// ============================================================================
-// ============================================================================
-// Advanced Features
-// ============================================================================
-// ============================================================================
-// Feature Detection
-// ============================================================================
-/**
- * Detect available browser features
- */
-function detectFeatures() {
-    return {
-        indexedDB: 'indexedDB' in globalThis,
-        broadcastChannel: 'BroadcastChannel' in globalThis,
-        webWorkers: typeof Worker !== 'undefined',
-        wasmSIMD: detectWasmSIMD(),
-        sharedArrayBuffer: typeof SharedArrayBuffer !== 'undefined'
-    };
-}
-/**
- * Detect WASM SIMD support
- */
-function detectWasmSIMD() {
-    return __awaiter(this, void 0, void 0, function* () {
-        try {
-            // WASM SIMD detection via feature test
-            const simdTest = new Uint8Array([
-                0x00, 0x61, 0x73, 0x6d, 0x01, 0x00, 0x00, 0x00,
-                0x01, 0x05, 0x01, 0x60, 0x00, 0x01, 0x7b, 0x03,
-                0x02, 0x01, 0x00, 0x0a, 0x0a, 0x01, 0x08, 0x00,
-                0xfd, 0x0c, 0xfd, 0x0c, 0xfd, 0x54, 0x0b
-            ]);
-            const module = yield WebAssembly.instantiate(simdTest);
-            return module instanceof WebAssembly.Instance;
-        }
-        catch (_a) {
-            return false;
-        }
-    });
-}
-// ============================================================================
-// Configuration Presets
-// ============================================================================
-/**
- * Recommended configuration for small datasets (<1K vectors)
- */
-const SMALL_DATASET_CONFIG = {
-    pq: { enabled: false },
-    hnsw: { enabled: false },
-    gnn: { enabled: true, numHeads: 2 },
-    mmr: { enabled: true, lambda: 0.7 },
-    svd: { enabled: false }
-};
-/**
- * Recommended configuration for medium datasets (1K-10K vectors)
- */
-const MEDIUM_DATASET_CONFIG = {
-    pq: { enabled: true, subvectors: 8 },
-    hnsw: { enabled: true, M: 16 },
-    gnn: { enabled: true, numHeads: 4 },
-    mmr: { enabled: true, lambda: 0.7 },
-    svd: { enabled: false }
-};
-/**
- * Recommended configuration for large datasets (10K-100K vectors)
- */
-const LARGE_DATASET_CONFIG = {
-    pq: { enabled: true, subvectors: 16 },
-    hnsw: { enabled: true, M: 32 },
-    gnn: { enabled: true, numHeads: 4 },
-    mmr: { enabled: true, lambda: 0.7 },
-    svd: { enabled: true, targetDim: 128 }
-};
-/**
- * Memory-optimized configuration (minimal memory usage)
- */
-const MEMORY_OPTIMIZED_CONFIG = {
-    pq: { enabled: true, subvectors: 32 }, // 16x compression
-    hnsw: { enabled: true, M: 8 }, // Fewer connections
-    gnn: { enabled: false },
-    mmr: { enabled: false },
-    svd: { enabled: true, targetDim: 64 } // Aggressive dimension reduction
-};
-/**
- * Speed-optimized configuration (fastest search)
- */
-const SPEED_OPTIMIZED_CONFIG = {
-    pq: { enabled: false }, // No compression overhead
-    hnsw: { enabled: true, M: 32, efSearch: 100 }, // Maximum HNSW quality
-    gnn: { enabled: false },
-    mmr: { enabled: false },
-    svd: { enabled: false }
-};
-/**
- * Quality-optimized configuration (best result quality)
- */
-const QUALITY_OPTIMIZED_CONFIG = {
-    pq: { enabled: false }, // No compression
-    hnsw: { enabled: true, M: 48, efConstruction: 400 }, // Highest quality
-    gnn: { enabled: true, numHeads: 8 }, // More attention heads
-    mmr: { enabled: true, lambda: 0.8 }, // More diversity
-    svd: { enabled: false } // No dimension loss
-};
-// ============================================================================
-// Version Information
-// ============================================================================
-const VERSION = {
-    major: 2,
-    minor: 0,
-    patch: 0,
-    prerelease: 'alpha.2',
-    features: 'advanced',
-    full: '2.0.0-alpha.2+advanced'
-};
-// ============================================================================
-// Utility Functions
-// ============================================================================
-/**
- * Estimate memory usage for configuration
- */
-function estimateMemoryUsage(numVectors, dimension, config) {
-    var _a, _b, _c;
-    let vectorBytes = numVectors * dimension * 4; // Float32Array
-    // Apply PQ compression
-    if ((_a = config.pq) === null || _a === void 0 ? void 0 : _a.enabled) {
-        const subvectors = config.pq.subvectors || 8;
-        vectorBytes = numVectors * (subvectors + 4); // codes + norm
-    }
-    // Apply SVD compression
-    if ((_b = config.svd) === null || _b === void 0 ? void 0 : _b.enabled) {
-        const targetDim = config.svd.targetDim || dimension / 2;
-        vectorBytes = numVectors * targetDim * 4;
-    }
-    // HNSW index overhead
-    let indexBytes = 0;
-    if ((_c = config.hnsw) === null || _c === void 0 ? void 0 : _c.enabled) {
-        const M = config.hnsw.M || 16;
-        const avgConnections = M * 1.5; // Estimate
-        indexBytes = numVectors * avgConnections * 4; // Connection IDs
-    }
-    const total = vectorBytes + indexBytes;
-    return {
-        vectors: vectorBytes,
-        index: indexBytes,
-        total,
-        totalMB: total / (1024 * 1024)
-    };
-}
-/**
- * Recommend configuration based on dataset size
- */
-function recommendConfig(numVectors, dimension) {
-    if (numVectors < 1000) {
-        return {
-            name: 'SMALL_DATASET',
-            config: SMALL_DATASET_CONFIG,
-            reason: 'Small dataset, linear search is fast enough'
-        };
-    }
-    else if (numVectors < 10000) {
-        return {
-            name: 'MEDIUM_DATASET',
-            config: MEDIUM_DATASET_CONFIG,
-            reason: 'Medium dataset, HNSW + PQ8 recommended'
-        };
-    }
-    else {
-        return {
-            name: 'LARGE_DATASET',
-            config: LARGE_DATASET_CONFIG,
-            reason: 'Large dataset, aggressive compression + HNSW recommended'
-        };
-    }
-}
-/**
- * Benchmark search performance
- */
-function benchmarkSearch(searchFn_1) {
-    return __awaiter(this, arguments, void 0, function* (searchFn, numQueries = 100, k = 10, dimension = 384) {
-        const times = [];
-        for (let i = 0; i < numQueries; i++) {
-            const query = new Float32Array(dimension);
-            for (let d = 0; d < dimension; d++) {
-                query[d] = Math.random() - 0.5;
-            }
-            const start = performance.now();
-            searchFn(query, k);
-            const end = performance.now();
-            times.push(end - start);
-        }
-        times.sort((a, b) => a - b);
-        return {
-            avgTimeMs: times.reduce((a, b) => a + b, 0) / times.length,
-            minTimeMs: times[0],
-            maxTimeMs: times[times.length - 1],
-            p50Ms: times[Math.floor(times.length * 0.5)],
-            p95Ms: times[Math.floor(times.length * 0.95)],
-            p99Ms: times[Math.floor(times.length * 0.99)]
-        };
-    });
-}
-
-
-  // ============================================================================
-  // Create Advanced Features Namespace
-  // ============================================================================
-
-  // Create global namespace for advanced features
-  const AgentDBAdvanced = {
-    // Product Quantization
-    ProductQuantization: ProductQuantization,
-    createPQ8: createPQ8,
-    createPQ16: createPQ16,
-    createPQ32: createPQ32,
-
-    // HNSW Index
-    HNSWIndex: HNSWIndex,
-    createHNSW: createHNSW,
-    createFastHNSW: createFastHNSW,
-    createAccurateHNSW: createAccurateHNSW,
-
-    // Advanced Features
-    GraphNeuralNetwork: GraphNeuralNetwork,
-    MaximalMarginalRelevance: MaximalMarginalRelevance,
-    TensorCompression: TensorCompression,
-    BatchProcessor: BatchProcessor,
-
-    // Utilities
-    detectFeatures: detectFeatures,
-    estimateMemoryUsage: estimateMemoryUsage,
-    recommendConfig: recommendConfig,
-    benchmarkSearch: benchmarkSearch,
-
-    // Configuration Presets
-    SMALL_DATASET_CONFIG: SMALL_DATASET_CONFIG,
-    MEDIUM_DATASET_CONFIG: MEDIUM_DATASET_CONFIG,
-    LARGE_DATASET_CONFIG: LARGE_DATASET_CONFIG,
-    MEMORY_OPTIMIZED_CONFIG: MEMORY_OPTIMIZED_CONFIG,
-    SPEED_OPTIMIZED_CONFIG: SPEED_OPTIMIZED_CONFIG,
-    QUALITY_OPTIMIZED_CONFIG: QUALITY_OPTIMIZED_CONFIG,
-
-    // Version
-    VERSION: VERSION
-  };
-
-  // Expose to global for use in AgentDB initialization
-  global.AgentDBAdvanced = AgentDBAdvanced;
-
-  // ============================================================================
-  // AgentDB v2 API with Advanced Features
-  // ============================================================================
-
-  const AgentDB = {};
-
-  // SQL.js database wrapper with advanced features
-  AgentDB.SQLiteVectorDB = function(config) {
-    config = config || {};
-    const self = this;
-
-    // Configuration
-    const enablePQ = config.enablePQ !== false;
-    const enableHNSW = config.enableHNSW !== false;
-    const enableGNN = config.enableGNN !== false;
-    const enableMMR = config.enableMMR !== false;
-    const enableSVD = config.enableSVD || false;
-    const enableIndexedDB = config.enableIndexedDB !== false && ('indexedDB' in global);
-    const enableCrossTab = config.enableCrossTab !== false && ('BroadcastChannel' in global);
-
-    // Initialize sql.js
-    let db = null;
-
-    // Advanced feature instances
-    let pqInstance = null;
-    let hnswInstance = null;
-    let gnnInstance = null;
-    let mmrInstance = null;
-
-    // Mock embeddings function (deterministic hash)
-    function mockEmbed(text) {
-      const embedding = new Float32Array(384);
-      let hash = 0;
-      for (let i = 0; i < text.length; i++) {
-        hash = ((hash << 5) - hash) + text.charCodeAt(i);
-        hash = hash & hash;
-      }
-      for (let i = 0; i < 384; i++) {
-        const seed = hash + i * 997;
-        embedding[i] = (Math.sin(seed) + Math.cos(seed * 1.618)) * 0.5;
-      }
-      const norm = Math.sqrt(embedding.reduce((sum, val) => sum + val * val, 0));
-      for (let i = 0; i < 384; i++) {
-        embedding[i] /= norm;
-      }
-      return embedding;
-    }
-
-    // Initialize database
-    this.initializeAsync = function() {
-      return new Promise(function(resolve, reject) {
-        global.initSqlJs().then(function(SQL) {
-          db = new SQL.Database();
-
-          // Create v2 schema
-          db.run(`CREATE TABLE IF NOT EXISTS episodes (
-            id INTEGER PRIMARY KEY AUTOINCREMENT,
-            session_id TEXT NOT NULL,
-            task TEXT NOT NULL,
-            input TEXT,
-            output TEXT,
-            critique TEXT,
-            reward REAL NOT NULL,
-            tokens_used INTEGER,
-            latency_ms REAL,
-            success BOOLEAN NOT NULL,
-            timestamp DATETIME DEFAULT CURRENT_TIMESTAMP
-          )`);
-
-          db.run(`CREATE TABLE IF NOT EXISTS episode_embeddings (
-            id INTEGER PRIMARY KEY AUTOINCREMENT,
-            episode_id INTEGER NOT NULL,
-            embedding BLOB NOT NULL,
-            FOREIGN KEY (episode_id) REFERENCES episodes(id) ON DELETE CASCADE
-          )`);
-
-          db.run(`CREATE TABLE IF NOT EXISTS skills (
-            id INTEGER PRIMARY KEY AUTOINCREMENT,
-            name TEXT UNIQUE NOT NULL,
-            code TEXT NOT NULL,
-            signature TEXT NOT NULL,
-            success_rate REAL DEFAULT 0.0,
-            uses INTEGER DEFAULT 0,
-            timestamp DATETIME DEFAULT CURRENT_TIMESTAMP
-          )`);
-
-          db.run(`CREATE TABLE IF NOT EXISTS skill_embeddings (
-            id INTEGER PRIMARY KEY AUTOINCREMENT,
-            skill_id INTEGER NOT NULL,
-            embedding BLOB NOT NULL,
-            FOREIGN KEY (skill_id) REFERENCES skills(id) ON DELETE CASCADE
-          )`);
-
-          db.run(`CREATE TABLE IF NOT EXISTS causal_edges (
-            id INTEGER PRIMARY KEY AUTOINCREMENT,
-            from_memory_id INTEGER NOT NULL,
-            to_memory_id INTEGER NOT NULL,
-            similarity REAL NOT NULL,
-            timestamp DATETIME DEFAULT CURRENT_TIMESTAMP,
-            FOREIGN KEY (from_memory_id) REFERENCES episodes(id) ON DELETE CASCADE,
-            FOREIGN KEY (to_memory_id) REFERENCES episodes(id) ON DELETE CASCADE
-          )`);
-
-          // Legacy v1 tables
-          db.run(`CREATE TABLE IF NOT EXISTS vectors (
-            id INTEGER PRIMARY KEY AUTOINCREMENT,
-            content TEXT,
-            embedding BLOB
-          )`);
-
-          db.run(`CREATE TABLE IF NOT EXISTS patterns (
-            id INTEGER PRIMARY KEY AUTOINCREMENT,
-            pattern TEXT,
-            metadata TEXT
-          )`);
-
-          // Initialize advanced features
-          if (enablePQ) {
-            pqInstance = global.AgentDBAdvanced.createPQ8(384);
-            console.log('[AgentDB] Product Quantization (PQ8) enabled');
-          }
-
-          if (enableHNSW) {
-            hnswInstance = global.AgentDBAdvanced.createHNSW(384);
-            console.log('[AgentDB] HNSW Indexing enabled');
-          }
-
-          if (enableGNN) {
-            gnnInstance = new global.AgentDBAdvanced.GraphNeuralNetwork({ numHeads: 4 });
-            console.log('[AgentDB] Graph Neural Networks enabled');
-          }
-
-          if (enableMMR) {
-            mmrInstance = new global.AgentDBAdvanced.MaximalMarginalRelevance({ lambda: 0.7 });
-            console.log('[AgentDB] MMR Diversity enabled');
-          }
-
-          console.log('[AgentDB] Initialized with advanced features');
-          resolve(self);
-        }).catch(reject);
-      });
-    };
-
-    // v2 Episodes API
-    this.episodes = {
-      store: async function(episodeData) {
-        const stmt = db.prepare(`
-          INSERT INTO episodes (session_id, task, input, output, critique, reward, tokens_used, latency_ms, success)
-          VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?)
-        `);
-        stmt.run([
-          episodeData.sessionId || 'default',
-          episodeData.task,
-          episodeData.input || '',
-          episodeData.output || '',
-          episodeData.critique || '',
-          episodeData.reward,
-          episodeData.tokensUsed || 0,
-          episodeData.latencyMs || 0,
-          episodeData.success ? 1 : 0
-        ]);
-        const id = db.exec('SELECT last_insert_rowid()')[0].values[0][0];
-
-        const embedding = mockEmbed(episodeData.task);
-        const embeddingBlob = new Uint8Array(embedding.buffer);
-
-        const embStmt = db.prepare('INSERT INTO episode_embeddings (episode_id, embedding) VALUES (?, ?)');
-        embStmt.run([id, embeddingBlob]);
-
-        // Add to HNSW index if enabled
-        if (enableHNSW && hnswInstance) {
-          hnswInstance.add(embedding, id);
-        }
-
-        // Add to GNN if enabled
-        if (enableGNN && gnnInstance) {
-          gnnInstance.addNode(id, embedding);
-        }
-
-        return id;
-      },
-
-      search: async function(options) {
-        const query = options.task;
-        const k = options.k || 10;
-        const minReward = options.minReward;
-        const onlySuccesses = options.onlySuccesses;
-        const diversify = options.diversify && enableMMR;
-
-        const queryEmbedding = mockEmbed(query);
-
-        let results;
-
-        if (enableHNSW && hnswInstance && hnswInstance.size() > 0) {
-          // Use HNSW for fast search
-          results = hnswInstance.search(queryEmbedding, k * 2);  // Get more for filtering
-        } else {
-          // Fallback to linear scan
-          results = self._linearSearch(queryEmbedding, k * 2, 'episodes');
-        }
-
-        // Filter by criteria
-        let filtered = results.map(r => {
-          const episode = db.exec(`SELECT * FROM episodes WHERE id = ${r.id}`)[0];
-          if (!episode || !episode.values.length) return null;
-
-          const row = episode.values[0];
-          const episodeObj = {
-            id: row[0],
-            session_id: row[1],
-            task: row[2],
-            reward: row[5],
-            success: row[8] === 1,
-            distance: r.distance,
-            similarity: 1 - r.distance
-          };
-
-          // Apply filters
-          if (minReward && episodeObj.reward < minReward) return null;
-          if (onlySuccesses && !episodeObj.success) return null;
-
-          return episodeObj;
-        }).filter(e => e !== null);
-
-        // Apply MMR diversity if enabled
-        if (diversify && mmrInstance && filtered.length > k) {
-          const candidates = filtered.map(e => ({
-            id: e.id,
-            vector: results.find(r => r.id === e.id).vector,
-            score: e.similarity
-          }));
-          const diverseIds = mmrInstance.rerank(queryEmbedding, candidates, k);
-          filtered = diverseIds.map(id => filtered.find(e => e.id === id));
-        } else {
-          filtered = filtered.slice(0, k);
-        }
-
-        return filtered;
-      }
-    };
-
-    // v2 Skills API
-    this.skills = {
-      store: async function(skillData) {
-        const stmt = db.prepare(`
-          INSERT INTO skills (name, code, signature, success_rate, uses)
-          VALUES (?, ?, ?, ?, ?)
-          ON CONFLICT(name) DO UPDATE SET
-            code = excluded.code,
-            signature = excluded.signature,
-            success_rate = excluded.success_rate,
-            uses = excluded.uses
-        `);
-        stmt.run([
-          skillData.name,
-          skillData.code,
-          skillData.signature,
-          skillData.successRate || 0.0,
-          skillData.uses || 0
-        ]);
-        const id = db.exec('SELECT last_insert_rowid()')[0].values[0][0];
-
-        const embedding = mockEmbed(skillData.name + ' ' + skillData.signature);
-        const embeddingBlob = new Uint8Array(embedding.buffer);
-
-        const embStmt = db.prepare('INSERT INTO skill_embeddings (skill_id, embedding) VALUES (?, ?)');
-        embStmt.run([id, embeddingBlob]);
-
-        return id;
-      },
-
-      search: async function(options) {
-        const query = options.query;
-        const k = options.k || 10;
-
-        const queryEmbedding = mockEmbed(query);
-        const results = self._linearSearch(queryEmbedding, k, 'skills');
-
-        return results.map(r => {
-          const skill = db.exec(`SELECT * FROM skills WHERE id = ${r.id}`)[0];
-          if (!skill || !skill.values.length) return null;
-
-          const row = skill.values[0];
-          return {
-            id: row[0],
-            name: row[1],
-            code: row[2],
-            signature: row[3],
-            success_rate: row[4],
-            uses: row[5],
-            similarity: 1 - r.distance
-          };
-        }).filter(s => s !== null);
-      }
-    };
-
-    // v2 Causal Edges API
-    this.causal_edges = {
-      add: async function(edge) {
-        const stmt = db.prepare(`
-          INSERT INTO causal_edges (from_memory_id, to_memory_id, similarity)
-          VALUES (?, ?, ?)
-        `);
-        stmt.run([edge.fromMemoryId, edge.toMemoryId, edge.similarity]);
-
-        // Add edge to GNN if enabled
-        if (enableGNN && gnnInstance) {
-          gnnInstance.addEdge(edge.fromMemoryId, edge.toMemoryId, edge.similarity);
-        }
-
-        return db.exec('SELECT last_insert_rowid()')[0].values[0][0];
-      },
-
-      get: async function(memoryId) {
-        const results = db.exec(`
-          SELECT * FROM causal_edges
-          WHERE from_memory_id = ${memoryId} OR to_memory_id = ${memoryId}
-        `);
-
-        if (!results.length || !results[0].values.length) return [];
-
-        return results[0].values.map(row => ({
-          id: row[0],
-          from_memory_id: row[1],
-          to_memory_id: row[2],
-          similarity: row[3]
-        }));
-      }
-    };
-
-    // Linear search fallback
-    this._linearSearch = function(queryEmbedding, k, table) {
-      const embeddingTable = table === 'episodes' ? 'episode_embeddings' : 'skill_embeddings';
-      const idColumn = table === 'episodes' ? 'episode_id' : 'skill_id';
-
-      const results = db.exec(`SELECT id, ${idColumn}, embedding FROM ${embeddingTable}`);
-      if (!results.length || !results[0].values.length) return [];
-
-      const distances = results[0].values.map(row => {
-        const id = row[1];
-        const embeddingBlob = row[2];
-        const embedding = new Float32Array(embeddingBlob.buffer);
-
-        let distance = 0;
-        for (let i = 0; i < queryEmbedding.length; i++) {
-          const diff = queryEmbedding[i] - embedding[i];
-          distance += diff * diff;
-        }
-        distance = Math.sqrt(distance);
-
-        return { id, distance, vector: embedding };
-      });
-
-      distances.sort((a, b) => a.distance - b.distance);
-      return distances.slice(0, k);
-    };
-
-    // v1 API (backward compatible)
-    this.storeVector = function(content, embedding) {
-      const embeddingBlob = new Uint8Array(embedding.buffer);
-      const stmt = db.prepare('INSERT INTO vectors (content, embedding) VALUES (?, ?)');
-      stmt.run([content, embeddingBlob]);
-      return db.exec('SELECT last_insert_rowid()')[0].values[0][0];
-    };
-
-    this.storePattern = function(pattern, metadata) {
-      const stmt = db.prepare('INSERT INTO patterns (pattern, metadata) VALUES (?, ?)');
-      stmt.run([pattern, JSON.stringify(metadata)]);
-      return db.exec('SELECT last_insert_rowid()')[0].values[0][0];
-    };
-
-    this.searchVectors = function(queryEmbedding, k) {
-      return self._linearSearch(queryEmbedding, k, 'vectors');
-    };
-
-    // Advanced Features API
-    this.advanced = {
-      getPQ: () => pqInstance,
-      getHNSW: () => hnswInstance,
-      getGNN: () => gnnInstance,
-      getMMR: () => mmrInstance,
-
-      stats: function() {
-        return {
-          pq: pqInstance ? pqInstance.getStats() : null,
-          hnsw: hnswInstance ? hnswInstance.getStats() : null,
-          gnn: gnnInstance ? gnnInstance.getStats() : null
-        };
-      }
-    };
-
-    // Export database
-    this.export = function() {
-      return db.export();
-    };
-
-    // Close database
-    this.close = function() {
-      if (db) {
-        db.close();
-        db = null;
-      }
-    };
-
-    // Run SQL
-    this.run = function(sql, params) {
-      return db.run(sql, params);
-    };
-
-    this.exec = function(sql) {
-      return db.exec(sql);
-    };
-  };
-
-  // Export advanced features namespace
-  AgentDB.Advanced = global.AgentDBAdvanced;
-
-  // Export to global
-  if (typeof module !== 'undefined' && module.exports) {
-    module.exports = AgentDB;
-  } else {
-    global.AgentDB = AgentDB;
-  }
-
-})(typeof window !== 'undefined' ? window : global);