@gmod/bam 7.8.0 → 7.8.1
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/dist/util.js +41 -1
- package/dist/util.js.map +1 -1
- package/esm/util.js +41 -1
- package/esm/util.js.map +1 -1
- package/package.json +1 -1
- package/src/util.ts +44 -1
package/dist/util.js
CHANGED
|
@@ -70,7 +70,47 @@ function optimizeChunks(chunks, lowest) {
|
|
|
70
70
|
lastMaxBlock = chunkMaxBlock;
|
|
71
71
|
}
|
|
72
72
|
}
|
|
73
|
-
return mergedChunks;
|
|
73
|
+
return makeDisjoint(mergedChunks);
|
|
74
|
+
}
|
|
75
|
+
/**
|
|
76
|
+
* Trim any chunk that starts before its predecessor ends, so the spans a query
|
|
77
|
+
* reads never overlap.
|
|
78
|
+
*
|
|
79
|
+
* Merging alone does not guarantee this. Two chunks from different bins can
|
|
80
|
+
* overlap inside a single BGZF block while the merge above still declines to
|
|
81
|
+
* join them, because the combined span would exceed its 5MB cap — on
|
|
82
|
+
* test/data/out.bam that leaves `3804:0-4977599:16404` next to
|
|
83
|
+
* `4977599:5843-9719917:27612`, whose 10561 shared bytes of block 4977599 get
|
|
84
|
+
* fetched, decompressed and decoded by both. Every record in the overlap is
|
|
85
|
+
* then returned TWICE from getRecordsForRange (5 of out.bam's 6551), which a
|
|
86
|
+
* consumer sees as a duplicated read: rendered twice, counted twice in
|
|
87
|
+
* coverage, and colliding on any id derived from fileOffset.
|
|
88
|
+
*
|
|
89
|
+
* Safe because a BAI chunk's `maxv` is the virtual offset just past its last
|
|
90
|
+
* record — a record boundary — so no record begins inside the trimmed span and
|
|
91
|
+
* the union of the chunks is unchanged. Verified by record counts: the
|
|
92
|
+
* duplicates disappear and the number of DISTINCT records is identical.
|
|
93
|
+
*/
|
|
94
|
+
function makeDisjoint(chunks) {
|
|
95
|
+
const out = [chunks[0]];
|
|
96
|
+
for (let i = 1; i < chunks.length; i++) {
|
|
97
|
+
const chunk = chunks[i];
|
|
98
|
+
const prevMax = out[out.length - 1].maxv;
|
|
99
|
+
const cmp = chunk.minv.blockPosition - prevMax.blockPosition ||
|
|
100
|
+
chunk.minv.dataPosition - prevMax.dataPosition;
|
|
101
|
+
if (cmp >= 0) {
|
|
102
|
+
out.push(chunk);
|
|
103
|
+
continue;
|
|
104
|
+
}
|
|
105
|
+
// starts inside the previous chunk: begin where that one ended, and drop it
|
|
106
|
+
// entirely if that leaves nothing
|
|
107
|
+
const stillHasData = chunk.maxv.blockPosition - prevMax.blockPosition ||
|
|
108
|
+
chunk.maxv.dataPosition - prevMax.dataPosition;
|
|
109
|
+
if (stillHasData > 0) {
|
|
110
|
+
out.push(new chunk_ts_1.default(prevMax, chunk.maxv, chunk.bin, chunk.endPosition));
|
|
111
|
+
}
|
|
112
|
+
}
|
|
113
|
+
return out;
|
|
74
114
|
}
|
|
75
115
|
// The pseudo-bin's mapped-record count, a little-endian uint64. Read as two
|
|
76
116
|
// 32-bit halves rather than via BigInt: exact up to Number.MAX_SAFE_INTEGER,
|
package/dist/util.js.map
CHANGED
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"util.js","sourceRoot":"","sources":["../src/util.ts"],"names":[],"mappings":";;;;;AA+BA,wCAsEC;
|
|
1
|
+
{"version":3,"file":"util.js","sourceRoot":"","sources":["../src/util.ts"],"names":[],"mappings":";;;;;AA+BA,wCAsEC;AAgDD,wCAcC;AASD,wCAkDC;AAKD,oCAuCC;AAoBD,4CA6BC;AAGD,wCAuBC;AAED,4CAYC;AAYD,sCAoBC;AAnYD,0DAA8B;AAC9B,yDAAkD;AA8BlD,SAAgB,cAAc,CAAC,MAAe,EAAE,MAAqB;IACnE,MAAM,CAAC,GAAG,MAAM,CAAC,MAAM,CAAA;IACvB,IAAI,CAAC,KAAK,CAAC,EAAE,CAAC;QACZ,OAAO,MAAM,CAAA;IACf,CAAC;IAED,0DAA0D;IAC1D,IAAI,QAAiB,CAAA;IACrB,IAAI,MAAM,EAAE,CAAC;QACX,MAAM,WAAW,GAAG,MAAM,CAAC,aAAa,CAAA;QACxC,MAAM,UAAU,GAAG,MAAM,CAAC,YAAY,CAAA;QACtC,QAAQ,GAAG,EAAE,CAAA;QACb,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC;YAC3B,MAAM,KAAK,GAAG,MAAM,CAAC,CAAC,CAAE,CAAA;YACxB,MAAM,IAAI,GAAG,KAAK,CAAC,IAAI,CAAA;YACvB,MAAM,GAAG,GACP,IAAI,CAAC,aAAa,GAAG,WAAW,IAAI,IAAI,CAAC,YAAY,GAAG,UAAU,CAAA;YACpE,IAAI,GAAG,GAAG,CAAC,EAAE,CAAC;gBACZ,QAAQ,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YACtB,CAAC;QACH,CAAC;QACD,IAAI,QAAQ,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;YAC1B,OAAO,QAAQ,CAAA;QACjB,CAAC;IACH,CAAC;SAAM,CAAC;QACN,QAAQ,GAAG,MAAM,CAAA;IACnB,CAAC;IAED,QAAQ,CAAC,IAAI,CAAC,CAAC,EAAE,EAAE,EAAE,EAAE,EAAE;QACvB,MAAM,GAAG,GAAG,EAAE,CAAC,IAAI,CAAC,aAAa,GAAG,EAAE,CAAC,IAAI,CAAC,aAAa,CAAA;QACzD,OAAO,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,CAAC,IAAI,CAAC,YAAY,GAAG,EAAE,CAAC,IAAI,CAAC,YAAY,CAAA;IACtE,CAAC,CAAC,CAAA;IAEF,yEAAyE;IACzE,uEAAuE;IACvE,MAAM,YAAY,GAAY,CAAC,QAAQ,CAAC,CAAC,CAAE,CAAC,CAAA;IAC5C,IAAI,YAAY,GAAG,QAAQ,CAAC,CAAC,CAAE,CAAC,IAAI,CAAC,aAAa,CAAA;IAClD,IAAI,YAAY,GAAG,QAAQ,CAAC,CAAC,CAAE,CAAC,IAAI,CAAC,aAAa,CAAA;IAElD,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,QAAQ,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACzC,MAAM,KAAK,GAAG,QAAQ,CAAC,CAAC,CAAE,CAAA;QAC1B,MAAM,aAAa,GAAG,KAAK,CAAC,IAAI,CAAC,aAAa,CAAA;QAC9C,MAAM,aAAa,GAAG,KAAK,CAAC,IAAI,CAAC,aAAa,CAAA;QAC9C,oEAAoE;QACpE,yEAAyE;QACzE,IACE,aAAa,GAAG,YAAY,GAAG,KAAK;YACpC,aAAa,GAAG,YAAY,GAAG,OAAO,EACtC,CAAC;YACD,MAAM,SAAS,GAAG,YAAY,CAAC,YAAY,CAAC,MAAM,GAAG,CAAC,CAAE,CAAA;YACxD,MAAM,GAAG,GACP,aAAa,GAAG,YAAY;gBAC5B,KAAK,CAAC,IAAI,CAAC,YAAY,GAAG,SAAS,CAAC,IAAI,CAAC,YAAY,CAAA;YACvD,IAAI,GAAG,GAAG,CAAC,EAAE,CAAC;gBACZ,YAAY,CAAC,YAAY,CAAC,MAAM,GAAG,CAAC,CAAC,GAAG,IAAI,kBAAK,CAC/C,SAAS,CAAC,IAAI,EACd,KAAK,CAAC,IAAI,EACV,SAAS,CAAC,GAAG,EACb,KAAK,CAAC,WAAW,CAClB,CAAA;gBACD,YAAY,GAAG,aAAa,CAAA;YAC9B,CAAC;QACH,CAAC;aAAM,CAAC;YACN,YAAY,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YACxB,YAAY,GAAG,aAAa,CAAA;YAC5B,YAAY,GAAG,aAAa,CAAA;QAC9B,CAAC;IACH,CAAC;IAED,OAAO,YAAY,CAAC,YAAY,CAAC,CAAA;AACnC,CAAC;AAED;;;;;;;;;;;;;;;;;;GAkBG;AACH,SAAS,YAAY,CAAC,MAAe;IACnC,MAAM,GAAG,GAAY,CAAC,MAAM,CAAC,CAAC,CAAE,CAAC,CAAA;IACjC,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,MAAM,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACvC,MAAM,KAAK,GAAG,MAAM,CAAC,CAAC,CAAE,CAAA;QACxB,MAAM,OAAO,GAAG,GAAG,CAAC,GAAG,CAAC,MAAM,GAAG,CAAC,CAAE,CAAC,IAAI,CAAA;QACzC,MAAM,GAAG,GACP,KAAK,CAAC,IAAI,CAAC,aAAa,GAAG,OAAO,CAAC,aAAa;YAChD,KAAK,CAAC,IAAI,CAAC,YAAY,GAAG,OAAO,CAAC,YAAY,CAAA;QAChD,IAAI,GAAG,IAAI,CAAC,EAAE,CAAC;YACb,GAAG,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YACf,SAAQ;QACV,CAAC;QACD,4EAA4E;QAC5E,kCAAkC;QAClC,MAAM,YAAY,GAChB,KAAK,CAAC,IAAI,CAAC,aAAa,GAAG,OAAO,CAAC,aAAa;YAChD,KAAK,CAAC,IAAI,CAAC,YAAY,GAAG,OAAO,CAAC,YAAY,CAAA;QAChD,IAAI,YAAY,GAAG,CAAC,EAAE,CAAC;YACrB,GAAG,CAAC,IAAI,CAAC,IAAI,kBAAK,CAAC,OAAO,EAAE,KAAK,CAAC,IAAI,EAAE,KAAK,CAAC,GAAG,EAAE,KAAK,CAAC,WAAW,CAAC,CAAC,CAAA;QACxE,CAAC;IACH,CAAC;IACD,OAAO,GAAG,CAAA;AACZ,CAAC;AAED,4EAA4E;AAC5E,6EAA6E;AAC7E,kEAAkE;AAClE,SAAgB,cAAc,CAAC,KAAiB,EAAE,MAAc;IAC9D,MAAM,GAAG,GACP,KAAK,CAAC,MAAM,CAAE;QACd,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,CAAE,IAAI,CAAC,CAAC;QACzB,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,CAAE,IAAI,EAAE,CAAC;QAC1B,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,CAAE,IAAI,EAAE,CAAC,CAAA;IAC5B,MAAM,IAAI,GACR,KAAK,CAAC,MAAM,GAAG,CAAC,CAAE;QAClB,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,CAAE,IAAI,CAAC,CAAC;QACzB,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,CAAE,IAAI,EAAE,CAAC;QAC1B,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,CAAE,IAAI,EAAE,CAAC,CAAA;IAC5B,OAAO;QACL,SAAS,EAAE,CAAC,IAAI,KAAK,CAAC,CAAC,GAAG,CAAC,IAAI,EAAE,GAAG,CAAC,GAAG,KAAK,CAAC,CAAC;KAChD,CAAA;AACH,CAAC;AAED,6EAA6E;AAC7E,kFAAkF;AAClF,kFAAkF;AAClF,iFAAiF;AACjF,kFAAkF;AAClF,iFAAiF;AACjF,yEAAyE;AACzE,SAAgB,cAAc,CAC5B,MAAe,EACf,kBAAqC,EAAE;IAEvC,IAAI,MAAM,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;QACxB,OAAM;IACR,CAAC;IACD,0EAA0E;IAC1E,8EAA8E;IAC9E,6EAA6E;IAC7E,2EAA2E;IAC3E,6EAA6E;IAC7E,2EAA2E;IAC3E,6DAA6D;IAC7D,4EAA4E;IAC5E,sEAAsE;IACtE,wEAAwE;IACxE,qEAAqE;IACrE,kEAAkE;IAClE,kEAAkE;IAClE,MAAM,UAAU,GAAG,IAAI,KAAK,CAC1B,eAAe,CAAC,MAAM,GAAG,MAAM,CAAC,MAAM,GAAG,CAAC,CAC3C,CAAA;IACD,IAAI,CAAC,GAAG,CAAC,CAAA;IACT,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,eAAe,CAAC,MAAM,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC;QACvD,UAAU,CAAC,CAAC,EAAE,CAAC,GAAG,eAAe,CAAC,CAAC,CAAE,CAAA;IACvC,CAAC;IACD,KAAK,MAAM,CAAC,IAAI,MAAM,EAAE,CAAC;QACvB,UAAU,CAAC,CAAC,EAAE,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,aAAa,CAAA;QACtC,UAAU,CAAC,CAAC,EAAE,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,aAAa,CAAA;IACxC,CAAC;IACD,UAAU,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,GAAG,CAAC,CAAC,CAAA;IAEhC,KAAK,MAAM,CAAC,IAAI,MAAM,EAAE,CAAC;QACvB,MAAM,GAAG,GAAG,CAAC,CAAC,IAAI,CAAC,aAAa,CAAA;QAChC,2CAA2C;QAC3C,IAAI,EAAE,GAAG,CAAC,CAAA;QACV,IAAI,EAAE,GAAG,UAAU,CAAC,MAAM,CAAA;QAC1B,OAAO,EAAE,GAAG,EAAE,EAAE,CAAC;YACf,MAAM,GAAG,GAAG,CAAC,EAAE,GAAG,EAAE,CAAC,KAAK,CAAC,CAAA;YAC3B,IAAI,UAAU,CAAC,GAAG,CAAE,GAAG,GAAG,EAAE,CAAC;gBAC3B,EAAE,GAAG,GAAG,CAAA;YACV,CAAC;iBAAM,CAAC;gBACN,EAAE,GAAG,GAAG,GAAG,CAAC,CAAA;YACd,CAAC;QACH,CAAC;QACD,IAAI,EAAE,GAAG,UAAU,CAAC,MAAM,EAAE,CAAC;YAC3B,CAAC,CAAC,WAAW,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,WAAW,EAAE,UAAU,CAAC,EAAE,CAAE,CAAC,CAAA;QAC1D,CAAC;IACH,CAAC;AACH,CAAC;AAED,6EAA6E;AAC7E,4EAA4E;AAC5E,eAAe;AACf,SAAgB,YAAY,CAC1B,KAAiB,EACjB,KAAa,EACb,YAAmC;IAEnC,IAAI,KAAK,GAAG,CAAC,GAAG,KAAK,CAAC,MAAM,EAAE,CAAC;QAC7B,OAAO,SAAS,CAAA;IAClB,CAAC;IACD,MAAM,QAAQ,GAAG,IAAI,QAAQ,CAC3B,KAAK,CAAC,MAAM,EACZ,KAAK,CAAC,UAAU,EAChB,KAAK,CAAC,UAAU,CACjB,CAAA;IACD,MAAM,IAAI,GAAG,QAAQ,CAAC,QAAQ,CAAC,KAAK,EAAE,IAAI,CAAC,CAAA;IAC3C,kEAAkE;IAClE,uDAAuD;IACvD,MAAM,UAAU,GAA2B,MAAM,CAAC,MAAM,CAAC,IAAI,CAAC,CAAA;IAC9D,MAAM,UAAU,GAA0C,EAAE,CAAA;IAC5D,MAAM,OAAO,GAAG,IAAI,WAAW,CAAC,MAAM,CAAC,CAAA;IAEvC,IAAI,CAAC,GAAG,KAAK,GAAG,CAAC,CAAA;IACjB,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,IAAI,EAAE,CAAC,EAAE,EAAE,CAAC;QAC9B,IAAI,CAAC,GAAG,CAAC,GAAG,KAAK,CAAC,MAAM,EAAE,CAAC;YACzB,OAAO,SAAS,CAAA;QAClB,CAAC;QACD,MAAM,KAAK,GAAG,QAAQ,CAAC,QAAQ,CAAC,CAAC,EAAE,IAAI,CAAC,CAAA;QACxC,IAAI,CAAC,GAAG,CAAC,GAAG,KAAK,GAAG,KAAK,CAAC,MAAM,EAAE,CAAC;YACjC,OAAO,SAAS,CAAA;QAClB,CAAC;QACD,MAAM,OAAO,GAAG,YAAY,CAC1B,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,QAAQ,CAAC,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,GAAG,KAAK,GAAG,CAAC,CAAC,CAAC,CACzD,CAAA;QACD,MAAM,IAAI,GAAG,QAAQ,CAAC,QAAQ,CAAC,CAAC,GAAG,KAAK,GAAG,CAAC,EAAE,IAAI,CAAC,CAAA;QACnD,UAAU,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;QACvB,UAAU,CAAC,IAAI,CAAC,EAAE,OAAO,EAAE,MAAM,EAAE,IAAI,EAAE,CAAC,CAAA;QAC1C,CAAC,IAAI,CAAC,GAAG,KAAK,CAAA;IAChB,CAAC;IACD,6EAA6E;IAC7E,OAAO,EAAE,UAAU,EAAE,UAAU,EAAE,GAAG,EAAE,CAAC,EAAE,CAAA;AAC3C,CAAC;AAED,2EAA2E;AAC3E,+EAA+E;AAC/E,4CAA4C;AAC5C;;;;;;;;;;;;;;GAcG;AACH,SAAgB,gBAAgB,CAC9B,KAAiB,EACjB,MAAc,EACd,KAAa,EACb,OAAkC;IAElC,IAAI,QAAQ,GAAG,OAAO,CAAC,CAAC,CAAC,OAAO,CAAC,aAAa,CAAC,CAAC,CAAC,QAAQ,CAAA;IACzD,IAAI,OAAO,GAAG,OAAO,CAAC,CAAC,CAAC,OAAO,CAAC,YAAY,CAAC,CAAC,CAAC,CAAC,CAAA;IAChD,IAAI,KAAK,GAAG,KAAK,CAAA;IACjB,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,KAAK,EAAE,CAAC,EAAE,EAAE,CAAC;QAC/B,MAAM,CAAC,GAAG,MAAM,GAAG,CAAC,GAAG,CAAC,CAAA;QACxB,MAAM,KAAK,GACT,KAAK,CAAC,CAAC,GAAG,CAAC,CAAE,GAAG,aAAa;YAC7B,KAAK,CAAC,CAAC,GAAG,CAAC,CAAE,GAAG,WAAW;YAC3B,KAAK,CAAC,CAAC,GAAG,CAAC,CAAE,GAAG,SAAS;YACzB,KAAK,CAAC,CAAC,GAAG,CAAC,CAAE,GAAG,OAAO;YACvB,KAAK,CAAC,CAAC,GAAG,CAAC,CAAE,GAAG,KAAK;YACrB,KAAK,CAAC,CAAC,GAAG,CAAC,CAAE,CAAA;QACf,MAAM,IAAI,GAAG,CAAC,KAAK,CAAC,CAAC,GAAG,CAAC,CAAE,IAAI,CAAC,CAAC,GAAG,KAAK,CAAC,CAAC,CAAE,CAAA;QAC7C,IACE,CAAC,KAAK,KAAK,CAAC,IAAI,IAAI,KAAK,CAAC,CAAC;YAC3B,CAAC,KAAK,GAAG,QAAQ,IAAI,CAAC,KAAK,KAAK,QAAQ,IAAI,IAAI,GAAG,OAAO,CAAC,CAAC,EAC5D,CAAC;YACD,QAAQ,GAAG,KAAK,CAAA;YAChB,OAAO,GAAG,IAAI,CAAA;YACd,KAAK,GAAG,IAAI,CAAA;QACd,CAAC;IACH,CAAC;IACD,OAAO,KAAK,CAAC,CAAC,CAAC,IAAI,gCAAa,CAAC,QAAQ,EAAE,OAAO,CAAC,CAAC,CAAC,CAAC,OAAO,CAAA;AAC/D,CAAC;AAED,4FAA4F;AAC5F,SAAgB,cAAc,CAC5B,UAAsB,EACtB,eAAwC,CAAC,CAAC,EAAE,CAAC,CAAC;IAE9C,MAAM,OAAO,GAAG,IAAI,WAAW,EAAE,CAAA;IACjC,IAAI,SAAS,GAAG,CAAC,CAAA;IACjB,IAAI,aAAa,GAAG,CAAC,CAAA;IACrB,MAAM,WAAW,GAAa,EAAE,CAAA;IAChC,MAAM,WAAW,GAA2B,MAAM,CAAC,MAAM,CAAC,IAAI,CAAC,CAAA;IAC/D,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,UAAU,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QAC3C,IAAI,CAAC,UAAU,CAAC,CAAC,CAAC,EAAE,CAAC;YACnB,IAAI,aAAa,GAAG,CAAC,EAAE,CAAC;gBACtB,MAAM,OAAO,GAAG,YAAY,CAC1B,OAAO,CAAC,MAAM,CAAC,UAAU,CAAC,QAAQ,CAAC,aAAa,EAAE,CAAC,CAAC,CAAC,CACtD,CAAA;gBACD,WAAW,CAAC,SAAS,CAAC,GAAG,OAAO,CAAA;gBAChC,WAAW,CAAC,OAAO,CAAC,GAAG,SAAS,CAAA;YAClC,CAAC;YACD,aAAa,GAAG,CAAC,GAAG,CAAC,CAAA;YACrB,SAAS,EAAE,CAAA;QACb,CAAC;IACH,CAAC;IACD,OAAO,EAAE,WAAW,EAAE,WAAW,EAAE,CAAA;AACrC,CAAC;AAED,SAAgB,gBAAgB,CAAC,IAAkB;IACjD,IAAI,WAAW,GAAG,CAAC,CAAA;IACnB,KAAK,MAAM,KAAK,IAAI,IAAI,EAAE,CAAC;QACzB,WAAW,IAAI,KAAK,CAAC,MAAM,CAAA;IAC7B,CAAC;IACD,MAAM,WAAW,GAAG,IAAI,UAAU,CAAC,WAAW,CAAC,CAAA;IAC/C,IAAI,MAAM,GAAG,CAAC,CAAA;IACd,KAAK,MAAM,KAAK,IAAI,IAAI,EAAE,CAAC;QACzB,WAAW,CAAC,GAAG,CAAC,KAAK,EAAE,MAAM,CAAC,CAAA;QAC9B,MAAM,IAAI,KAAK,CAAC,MAAM,CAAA;IACxB,CAAC;IACD,OAAO,WAAW,CAAA;AACpB,CAAC;AAQD,0EAA0E;AAC1E,4EAA4E;AAC5E,6EAA6E;AAC7E,iDAAiD;AACjD,SAAgB,aAAa,CAC3B,OAAY,EACZ,KAAa,EACb,GAAW,EACX,GAAW,EACX,MAAW,EAAE;IAEb,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,OAAO,CAAC,MAAM,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC;QAC/C,MAAM,CAAC,GAAG,OAAO,CAAC,CAAC,CAAE,CAAA;QACrB,IAAI,CAAC,CAAC,MAAM,KAAK,KAAK,EAAE,CAAC;YACvB,IAAI,CAAC,CAAC,KAAK,IAAI,GAAG,EAAE,CAAC;gBACnB,MAAK;YACP,CAAC;iBAAM,IAAI,CAAC,CAAC,GAAG,IAAI,GAAG,EAAE,CAAC;gBACxB,GAAG,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;YACb,CAAC;QACH,CAAC;aAAM,IAAI,CAAC,CAAC,MAAM,GAAG,KAAK,EAAE,CAAC;YAC5B,MAAK;QACP,CAAC;IACH,CAAC;IACD,OAAO,GAAG,CAAA;AACZ,CAAC"}
|
package/esm/util.js
CHANGED
|
@@ -57,7 +57,47 @@ export function optimizeChunks(chunks, lowest) {
|
|
|
57
57
|
lastMaxBlock = chunkMaxBlock;
|
|
58
58
|
}
|
|
59
59
|
}
|
|
60
|
-
return mergedChunks;
|
|
60
|
+
return makeDisjoint(mergedChunks);
|
|
61
|
+
}
|
|
62
|
+
/**
|
|
63
|
+
* Trim any chunk that starts before its predecessor ends, so the spans a query
|
|
64
|
+
* reads never overlap.
|
|
65
|
+
*
|
|
66
|
+
* Merging alone does not guarantee this. Two chunks from different bins can
|
|
67
|
+
* overlap inside a single BGZF block while the merge above still declines to
|
|
68
|
+
* join them, because the combined span would exceed its 5MB cap — on
|
|
69
|
+
* test/data/out.bam that leaves `3804:0-4977599:16404` next to
|
|
70
|
+
* `4977599:5843-9719917:27612`, whose 10561 shared bytes of block 4977599 get
|
|
71
|
+
* fetched, decompressed and decoded by both. Every record in the overlap is
|
|
72
|
+
* then returned TWICE from getRecordsForRange (5 of out.bam's 6551), which a
|
|
73
|
+
* consumer sees as a duplicated read: rendered twice, counted twice in
|
|
74
|
+
* coverage, and colliding on any id derived from fileOffset.
|
|
75
|
+
*
|
|
76
|
+
* Safe because a BAI chunk's `maxv` is the virtual offset just past its last
|
|
77
|
+
* record — a record boundary — so no record begins inside the trimmed span and
|
|
78
|
+
* the union of the chunks is unchanged. Verified by record counts: the
|
|
79
|
+
* duplicates disappear and the number of DISTINCT records is identical.
|
|
80
|
+
*/
|
|
81
|
+
function makeDisjoint(chunks) {
|
|
82
|
+
const out = [chunks[0]];
|
|
83
|
+
for (let i = 1; i < chunks.length; i++) {
|
|
84
|
+
const chunk = chunks[i];
|
|
85
|
+
const prevMax = out[out.length - 1].maxv;
|
|
86
|
+
const cmp = chunk.minv.blockPosition - prevMax.blockPosition ||
|
|
87
|
+
chunk.minv.dataPosition - prevMax.dataPosition;
|
|
88
|
+
if (cmp >= 0) {
|
|
89
|
+
out.push(chunk);
|
|
90
|
+
continue;
|
|
91
|
+
}
|
|
92
|
+
// starts inside the previous chunk: begin where that one ended, and drop it
|
|
93
|
+
// entirely if that leaves nothing
|
|
94
|
+
const stillHasData = chunk.maxv.blockPosition - prevMax.blockPosition ||
|
|
95
|
+
chunk.maxv.dataPosition - prevMax.dataPosition;
|
|
96
|
+
if (stillHasData > 0) {
|
|
97
|
+
out.push(new Chunk(prevMax, chunk.maxv, chunk.bin, chunk.endPosition));
|
|
98
|
+
}
|
|
99
|
+
}
|
|
100
|
+
return out;
|
|
61
101
|
}
|
|
62
102
|
// The pseudo-bin's mapped-record count, a little-endian uint64. Read as two
|
|
63
103
|
// 32-bit halves rather than via BigInt: exact up to Number.MAX_SAFE_INTEGER,
|
package/esm/util.js.map
CHANGED
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"util.js","sourceRoot":"","sources":["../src/util.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,MAAM,YAAY,CAAA;AAC9B,OAAO,EAAE,aAAa,EAAE,MAAM,oBAAoB,CAAA;AA8BlD,MAAM,UAAU,cAAc,CAAC,MAAe,EAAE,MAAqB;IACnE,MAAM,CAAC,GAAG,MAAM,CAAC,MAAM,CAAA;IACvB,IAAI,CAAC,KAAK,CAAC,EAAE,CAAC;QACZ,OAAO,MAAM,CAAA;IACf,CAAC;IAED,0DAA0D;IAC1D,IAAI,QAAiB,CAAA;IACrB,IAAI,MAAM,EAAE,CAAC;QACX,MAAM,WAAW,GAAG,MAAM,CAAC,aAAa,CAAA;QACxC,MAAM,UAAU,GAAG,MAAM,CAAC,YAAY,CAAA;QACtC,QAAQ,GAAG,EAAE,CAAA;QACb,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC;YAC3B,MAAM,KAAK,GAAG,MAAM,CAAC,CAAC,CAAE,CAAA;YACxB,MAAM,IAAI,GAAG,KAAK,CAAC,IAAI,CAAA;YACvB,MAAM,GAAG,GACP,IAAI,CAAC,aAAa,GAAG,WAAW,IAAI,IAAI,CAAC,YAAY,GAAG,UAAU,CAAA;YACpE,IAAI,GAAG,GAAG,CAAC,EAAE,CAAC;gBACZ,QAAQ,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YACtB,CAAC;QACH,CAAC;QACD,IAAI,QAAQ,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;YAC1B,OAAO,QAAQ,CAAA;QACjB,CAAC;IACH,CAAC;SAAM,CAAC;QACN,QAAQ,GAAG,MAAM,CAAA;IACnB,CAAC;IAED,QAAQ,CAAC,IAAI,CAAC,CAAC,EAAE,EAAE,EAAE,EAAE,EAAE;QACvB,MAAM,GAAG,GAAG,EAAE,CAAC,IAAI,CAAC,aAAa,GAAG,EAAE,CAAC,IAAI,CAAC,aAAa,CAAA;QACzD,OAAO,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,CAAC,IAAI,CAAC,YAAY,GAAG,EAAE,CAAC,IAAI,CAAC,YAAY,CAAA;IACtE,CAAC,CAAC,CAAA;IAEF,yEAAyE;IACzE,uEAAuE;IACvE,MAAM,YAAY,GAAY,CAAC,QAAQ,CAAC,CAAC,CAAE,CAAC,CAAA;IAC5C,IAAI,YAAY,GAAG,QAAQ,CAAC,CAAC,CAAE,CAAC,IAAI,CAAC,aAAa,CAAA;IAClD,IAAI,YAAY,GAAG,QAAQ,CAAC,CAAC,CAAE,CAAC,IAAI,CAAC,aAAa,CAAA;IAElD,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,QAAQ,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACzC,MAAM,KAAK,GAAG,QAAQ,CAAC,CAAC,CAAE,CAAA;QAC1B,MAAM,aAAa,GAAG,KAAK,CAAC,IAAI,CAAC,aAAa,CAAA;QAC9C,MAAM,aAAa,GAAG,KAAK,CAAC,IAAI,CAAC,aAAa,CAAA;QAC9C,oEAAoE;QACpE,yEAAyE;QACzE,IACE,aAAa,GAAG,YAAY,GAAG,KAAK;YACpC,aAAa,GAAG,YAAY,GAAG,OAAO,EACtC,CAAC;YACD,MAAM,SAAS,GAAG,YAAY,CAAC,YAAY,CAAC,MAAM,GAAG,CAAC,CAAE,CAAA;YACxD,MAAM,GAAG,GACP,aAAa,GAAG,YAAY;gBAC5B,KAAK,CAAC,IAAI,CAAC,YAAY,GAAG,SAAS,CAAC,IAAI,CAAC,YAAY,CAAA;YACvD,IAAI,GAAG,GAAG,CAAC,EAAE,CAAC;gBACZ,YAAY,CAAC,YAAY,CAAC,MAAM,GAAG,CAAC,CAAC,GAAG,IAAI,KAAK,CAC/C,SAAS,CAAC,IAAI,EACd,KAAK,CAAC,IAAI,EACV,SAAS,CAAC,GAAG,EACb,KAAK,CAAC,WAAW,CAClB,CAAA;gBACD,YAAY,GAAG,aAAa,CAAA;YAC9B,CAAC;QACH,CAAC;aAAM,CAAC;YACN,YAAY,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YACxB,YAAY,GAAG,aAAa,CAAA;YAC5B,YAAY,GAAG,aAAa,CAAA;QAC9B,CAAC;IACH,CAAC;IAED,OAAO,YAAY,CAAA;
|
|
1
|
+
{"version":3,"file":"util.js","sourceRoot":"","sources":["../src/util.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,MAAM,YAAY,CAAA;AAC9B,OAAO,EAAE,aAAa,EAAE,MAAM,oBAAoB,CAAA;AA8BlD,MAAM,UAAU,cAAc,CAAC,MAAe,EAAE,MAAqB;IACnE,MAAM,CAAC,GAAG,MAAM,CAAC,MAAM,CAAA;IACvB,IAAI,CAAC,KAAK,CAAC,EAAE,CAAC;QACZ,OAAO,MAAM,CAAA;IACf,CAAC;IAED,0DAA0D;IAC1D,IAAI,QAAiB,CAAA;IACrB,IAAI,MAAM,EAAE,CAAC;QACX,MAAM,WAAW,GAAG,MAAM,CAAC,aAAa,CAAA;QACxC,MAAM,UAAU,GAAG,MAAM,CAAC,YAAY,CAAA;QACtC,QAAQ,GAAG,EAAE,CAAA;QACb,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC;YAC3B,MAAM,KAAK,GAAG,MAAM,CAAC,CAAC,CAAE,CAAA;YACxB,MAAM,IAAI,GAAG,KAAK,CAAC,IAAI,CAAA;YACvB,MAAM,GAAG,GACP,IAAI,CAAC,aAAa,GAAG,WAAW,IAAI,IAAI,CAAC,YAAY,GAAG,UAAU,CAAA;YACpE,IAAI,GAAG,GAAG,CAAC,EAAE,CAAC;gBACZ,QAAQ,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YACtB,CAAC;QACH,CAAC;QACD,IAAI,QAAQ,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;YAC1B,OAAO,QAAQ,CAAA;QACjB,CAAC;IACH,CAAC;SAAM,CAAC;QACN,QAAQ,GAAG,MAAM,CAAA;IACnB,CAAC;IAED,QAAQ,CAAC,IAAI,CAAC,CAAC,EAAE,EAAE,EAAE,EAAE,EAAE;QACvB,MAAM,GAAG,GAAG,EAAE,CAAC,IAAI,CAAC,aAAa,GAAG,EAAE,CAAC,IAAI,CAAC,aAAa,CAAA;QACzD,OAAO,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,CAAC,IAAI,CAAC,YAAY,GAAG,EAAE,CAAC,IAAI,CAAC,YAAY,CAAA;IACtE,CAAC,CAAC,CAAA;IAEF,yEAAyE;IACzE,uEAAuE;IACvE,MAAM,YAAY,GAAY,CAAC,QAAQ,CAAC,CAAC,CAAE,CAAC,CAAA;IAC5C,IAAI,YAAY,GAAG,QAAQ,CAAC,CAAC,CAAE,CAAC,IAAI,CAAC,aAAa,CAAA;IAClD,IAAI,YAAY,GAAG,QAAQ,CAAC,CAAC,CAAE,CAAC,IAAI,CAAC,aAAa,CAAA;IAElD,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,QAAQ,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACzC,MAAM,KAAK,GAAG,QAAQ,CAAC,CAAC,CAAE,CAAA;QAC1B,MAAM,aAAa,GAAG,KAAK,CAAC,IAAI,CAAC,aAAa,CAAA;QAC9C,MAAM,aAAa,GAAG,KAAK,CAAC,IAAI,CAAC,aAAa,CAAA;QAC9C,oEAAoE;QACpE,yEAAyE;QACzE,IACE,aAAa,GAAG,YAAY,GAAG,KAAK;YACpC,aAAa,GAAG,YAAY,GAAG,OAAO,EACtC,CAAC;YACD,MAAM,SAAS,GAAG,YAAY,CAAC,YAAY,CAAC,MAAM,GAAG,CAAC,CAAE,CAAA;YACxD,MAAM,GAAG,GACP,aAAa,GAAG,YAAY;gBAC5B,KAAK,CAAC,IAAI,CAAC,YAAY,GAAG,SAAS,CAAC,IAAI,CAAC,YAAY,CAAA;YACvD,IAAI,GAAG,GAAG,CAAC,EAAE,CAAC;gBACZ,YAAY,CAAC,YAAY,CAAC,MAAM,GAAG,CAAC,CAAC,GAAG,IAAI,KAAK,CAC/C,SAAS,CAAC,IAAI,EACd,KAAK,CAAC,IAAI,EACV,SAAS,CAAC,GAAG,EACb,KAAK,CAAC,WAAW,CAClB,CAAA;gBACD,YAAY,GAAG,aAAa,CAAA;YAC9B,CAAC;QACH,CAAC;aAAM,CAAC;YACN,YAAY,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YACxB,YAAY,GAAG,aAAa,CAAA;YAC5B,YAAY,GAAG,aAAa,CAAA;QAC9B,CAAC;IACH,CAAC;IAED,OAAO,YAAY,CAAC,YAAY,CAAC,CAAA;AACnC,CAAC;AAED;;;;;;;;;;;;;;;;;;GAkBG;AACH,SAAS,YAAY,CAAC,MAAe;IACnC,MAAM,GAAG,GAAY,CAAC,MAAM,CAAC,CAAC,CAAE,CAAC,CAAA;IACjC,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,MAAM,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACvC,MAAM,KAAK,GAAG,MAAM,CAAC,CAAC,CAAE,CAAA;QACxB,MAAM,OAAO,GAAG,GAAG,CAAC,GAAG,CAAC,MAAM,GAAG,CAAC,CAAE,CAAC,IAAI,CAAA;QACzC,MAAM,GAAG,GACP,KAAK,CAAC,IAAI,CAAC,aAAa,GAAG,OAAO,CAAC,aAAa;YAChD,KAAK,CAAC,IAAI,CAAC,YAAY,GAAG,OAAO,CAAC,YAAY,CAAA;QAChD,IAAI,GAAG,IAAI,CAAC,EAAE,CAAC;YACb,GAAG,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YACf,SAAQ;QACV,CAAC;QACD,4EAA4E;QAC5E,kCAAkC;QAClC,MAAM,YAAY,GAChB,KAAK,CAAC,IAAI,CAAC,aAAa,GAAG,OAAO,CAAC,aAAa;YAChD,KAAK,CAAC,IAAI,CAAC,YAAY,GAAG,OAAO,CAAC,YAAY,CAAA;QAChD,IAAI,YAAY,GAAG,CAAC,EAAE,CAAC;YACrB,GAAG,CAAC,IAAI,CAAC,IAAI,KAAK,CAAC,OAAO,EAAE,KAAK,CAAC,IAAI,EAAE,KAAK,CAAC,GAAG,EAAE,KAAK,CAAC,WAAW,CAAC,CAAC,CAAA;QACxE,CAAC;IACH,CAAC;IACD,OAAO,GAAG,CAAA;AACZ,CAAC;AAED,4EAA4E;AAC5E,6EAA6E;AAC7E,kEAAkE;AAClE,MAAM,UAAU,cAAc,CAAC,KAAiB,EAAE,MAAc;IAC9D,MAAM,GAAG,GACP,KAAK,CAAC,MAAM,CAAE;QACd,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,CAAE,IAAI,CAAC,CAAC;QACzB,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,CAAE,IAAI,EAAE,CAAC;QAC1B,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,CAAE,IAAI,EAAE,CAAC,CAAA;IAC5B,MAAM,IAAI,GACR,KAAK,CAAC,MAAM,GAAG,CAAC,CAAE;QAClB,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,CAAE,IAAI,CAAC,CAAC;QACzB,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,CAAE,IAAI,EAAE,CAAC;QAC1B,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,CAAE,IAAI,EAAE,CAAC,CAAA;IAC5B,OAAO;QACL,SAAS,EAAE,CAAC,IAAI,KAAK,CAAC,CAAC,GAAG,CAAC,IAAI,EAAE,GAAG,CAAC,GAAG,KAAK,CAAC,CAAC;KAChD,CAAA;AACH,CAAC;AAED,6EAA6E;AAC7E,kFAAkF;AAClF,kFAAkF;AAClF,iFAAiF;AACjF,kFAAkF;AAClF,iFAAiF;AACjF,yEAAyE;AACzE,MAAM,UAAU,cAAc,CAC5B,MAAe,EACf,kBAAqC,EAAE;IAEvC,IAAI,MAAM,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;QACxB,OAAM;IACR,CAAC;IACD,0EAA0E;IAC1E,8EAA8E;IAC9E,6EAA6E;IAC7E,2EAA2E;IAC3E,6EAA6E;IAC7E,2EAA2E;IAC3E,6DAA6D;IAC7D,4EAA4E;IAC5E,sEAAsE;IACtE,wEAAwE;IACxE,qEAAqE;IACrE,kEAAkE;IAClE,kEAAkE;IAClE,MAAM,UAAU,GAAG,IAAI,KAAK,CAC1B,eAAe,CAAC,MAAM,GAAG,MAAM,CAAC,MAAM,GAAG,CAAC,CAC3C,CAAA;IACD,IAAI,CAAC,GAAG,CAAC,CAAA;IACT,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,eAAe,CAAC,MAAM,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC;QACvD,UAAU,CAAC,CAAC,EAAE,CAAC,GAAG,eAAe,CAAC,CAAC,CAAE,CAAA;IACvC,CAAC;IACD,KAAK,MAAM,CAAC,IAAI,MAAM,EAAE,CAAC;QACvB,UAAU,CAAC,CAAC,EAAE,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,aAAa,CAAA;QACtC,UAAU,CAAC,CAAC,EAAE,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,aAAa,CAAA;IACxC,CAAC;IACD,UAAU,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,GAAG,CAAC,CAAC,CAAA;IAEhC,KAAK,MAAM,CAAC,IAAI,MAAM,EAAE,CAAC;QACvB,MAAM,GAAG,GAAG,CAAC,CAAC,IAAI,CAAC,aAAa,CAAA;QAChC,2CAA2C;QAC3C,IAAI,EAAE,GAAG,CAAC,CAAA;QACV,IAAI,EAAE,GAAG,UAAU,CAAC,MAAM,CAAA;QAC1B,OAAO,EAAE,GAAG,EAAE,EAAE,CAAC;YACf,MAAM,GAAG,GAAG,CAAC,EAAE,GAAG,EAAE,CAAC,KAAK,CAAC,CAAA;YAC3B,IAAI,UAAU,CAAC,GAAG,CAAE,GAAG,GAAG,EAAE,CAAC;gBAC3B,EAAE,GAAG,GAAG,CAAA;YACV,CAAC;iBAAM,CAAC;gBACN,EAAE,GAAG,GAAG,GAAG,CAAC,CAAA;YACd,CAAC;QACH,CAAC;QACD,IAAI,EAAE,GAAG,UAAU,CAAC,MAAM,EAAE,CAAC;YAC3B,CAAC,CAAC,WAAW,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,WAAW,EAAE,UAAU,CAAC,EAAE,CAAE,CAAC,CAAA;QAC1D,CAAC;IACH,CAAC;AACH,CAAC;AAED,6EAA6E;AAC7E,4EAA4E;AAC5E,eAAe;AACf,MAAM,UAAU,YAAY,CAC1B,KAAiB,EACjB,KAAa,EACb,YAAmC;IAEnC,IAAI,KAAK,GAAG,CAAC,GAAG,KAAK,CAAC,MAAM,EAAE,CAAC;QAC7B,OAAO,SAAS,CAAA;IAClB,CAAC;IACD,MAAM,QAAQ,GAAG,IAAI,QAAQ,CAC3B,KAAK,CAAC,MAAM,EACZ,KAAK,CAAC,UAAU,EAChB,KAAK,CAAC,UAAU,CACjB,CAAA;IACD,MAAM,IAAI,GAAG,QAAQ,CAAC,QAAQ,CAAC,KAAK,EAAE,IAAI,CAAC,CAAA;IAC3C,kEAAkE;IAClE,uDAAuD;IACvD,MAAM,UAAU,GAA2B,MAAM,CAAC,MAAM,CAAC,IAAI,CAAC,CAAA;IAC9D,MAAM,UAAU,GAA0C,EAAE,CAAA;IAC5D,MAAM,OAAO,GAAG,IAAI,WAAW,CAAC,MAAM,CAAC,CAAA;IAEvC,IAAI,CAAC,GAAG,KAAK,GAAG,CAAC,CAAA;IACjB,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,IAAI,EAAE,CAAC,EAAE,EAAE,CAAC;QAC9B,IAAI,CAAC,GAAG,CAAC,GAAG,KAAK,CAAC,MAAM,EAAE,CAAC;YACzB,OAAO,SAAS,CAAA;QAClB,CAAC;QACD,MAAM,KAAK,GAAG,QAAQ,CAAC,QAAQ,CAAC,CAAC,EAAE,IAAI,CAAC,CAAA;QACxC,IAAI,CAAC,GAAG,CAAC,GAAG,KAAK,GAAG,KAAK,CAAC,MAAM,EAAE,CAAC;YACjC,OAAO,SAAS,CAAA;QAClB,CAAC;QACD,MAAM,OAAO,GAAG,YAAY,CAC1B,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,QAAQ,CAAC,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,GAAG,KAAK,GAAG,CAAC,CAAC,CAAC,CACzD,CAAA;QACD,MAAM,IAAI,GAAG,QAAQ,CAAC,QAAQ,CAAC,CAAC,GAAG,KAAK,GAAG,CAAC,EAAE,IAAI,CAAC,CAAA;QACnD,UAAU,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;QACvB,UAAU,CAAC,IAAI,CAAC,EAAE,OAAO,EAAE,MAAM,EAAE,IAAI,EAAE,CAAC,CAAA;QAC1C,CAAC,IAAI,CAAC,GAAG,KAAK,CAAA;IAChB,CAAC;IACD,6EAA6E;IAC7E,OAAO,EAAE,UAAU,EAAE,UAAU,EAAE,GAAG,EAAE,CAAC,EAAE,CAAA;AAC3C,CAAC;AAED,2EAA2E;AAC3E,+EAA+E;AAC/E,4CAA4C;AAC5C;;;;;;;;;;;;;;GAcG;AACH,MAAM,UAAU,gBAAgB,CAC9B,KAAiB,EACjB,MAAc,EACd,KAAa,EACb,OAAkC;IAElC,IAAI,QAAQ,GAAG,OAAO,CAAC,CAAC,CAAC,OAAO,CAAC,aAAa,CAAC,CAAC,CAAC,QAAQ,CAAA;IACzD,IAAI,OAAO,GAAG,OAAO,CAAC,CAAC,CAAC,OAAO,CAAC,YAAY,CAAC,CAAC,CAAC,CAAC,CAAA;IAChD,IAAI,KAAK,GAAG,KAAK,CAAA;IACjB,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,KAAK,EAAE,CAAC,EAAE,EAAE,CAAC;QAC/B,MAAM,CAAC,GAAG,MAAM,GAAG,CAAC,GAAG,CAAC,CAAA;QACxB,MAAM,KAAK,GACT,KAAK,CAAC,CAAC,GAAG,CAAC,CAAE,GAAG,aAAa;YAC7B,KAAK,CAAC,CAAC,GAAG,CAAC,CAAE,GAAG,WAAW;YAC3B,KAAK,CAAC,CAAC,GAAG,CAAC,CAAE,GAAG,SAAS;YACzB,KAAK,CAAC,CAAC,GAAG,CAAC,CAAE,GAAG,OAAO;YACvB,KAAK,CAAC,CAAC,GAAG,CAAC,CAAE,GAAG,KAAK;YACrB,KAAK,CAAC,CAAC,GAAG,CAAC,CAAE,CAAA;QACf,MAAM,IAAI,GAAG,CAAC,KAAK,CAAC,CAAC,GAAG,CAAC,CAAE,IAAI,CAAC,CAAC,GAAG,KAAK,CAAC,CAAC,CAAE,CAAA;QAC7C,IACE,CAAC,KAAK,KAAK,CAAC,IAAI,IAAI,KAAK,CAAC,CAAC;YAC3B,CAAC,KAAK,GAAG,QAAQ,IAAI,CAAC,KAAK,KAAK,QAAQ,IAAI,IAAI,GAAG,OAAO,CAAC,CAAC,EAC5D,CAAC;YACD,QAAQ,GAAG,KAAK,CAAA;YAChB,OAAO,GAAG,IAAI,CAAA;YACd,KAAK,GAAG,IAAI,CAAA;QACd,CAAC;IACH,CAAC;IACD,OAAO,KAAK,CAAC,CAAC,CAAC,IAAI,aAAa,CAAC,QAAQ,EAAE,OAAO,CAAC,CAAC,CAAC,CAAC,OAAO,CAAA;AAC/D,CAAC;AAED,4FAA4F;AAC5F,MAAM,UAAU,cAAc,CAC5B,UAAsB,EACtB,eAAwC,CAAC,CAAC,EAAE,CAAC,CAAC;IAE9C,MAAM,OAAO,GAAG,IAAI,WAAW,EAAE,CAAA;IACjC,IAAI,SAAS,GAAG,CAAC,CAAA;IACjB,IAAI,aAAa,GAAG,CAAC,CAAA;IACrB,MAAM,WAAW,GAAa,EAAE,CAAA;IAChC,MAAM,WAAW,GAA2B,MAAM,CAAC,MAAM,CAAC,IAAI,CAAC,CAAA;IAC/D,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,UAAU,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QAC3C,IAAI,CAAC,UAAU,CAAC,CAAC,CAAC,EAAE,CAAC;YACnB,IAAI,aAAa,GAAG,CAAC,EAAE,CAAC;gBACtB,MAAM,OAAO,GAAG,YAAY,CAC1B,OAAO,CAAC,MAAM,CAAC,UAAU,CAAC,QAAQ,CAAC,aAAa,EAAE,CAAC,CAAC,CAAC,CACtD,CAAA;gBACD,WAAW,CAAC,SAAS,CAAC,GAAG,OAAO,CAAA;gBAChC,WAAW,CAAC,OAAO,CAAC,GAAG,SAAS,CAAA;YAClC,CAAC;YACD,aAAa,GAAG,CAAC,GAAG,CAAC,CAAA;YACrB,SAAS,EAAE,CAAA;QACb,CAAC;IACH,CAAC;IACD,OAAO,EAAE,WAAW,EAAE,WAAW,EAAE,CAAA;AACrC,CAAC;AAED,MAAM,UAAU,gBAAgB,CAAC,IAAkB;IACjD,IAAI,WAAW,GAAG,CAAC,CAAA;IACnB,KAAK,MAAM,KAAK,IAAI,IAAI,EAAE,CAAC;QACzB,WAAW,IAAI,KAAK,CAAC,MAAM,CAAA;IAC7B,CAAC;IACD,MAAM,WAAW,GAAG,IAAI,UAAU,CAAC,WAAW,CAAC,CAAA;IAC/C,IAAI,MAAM,GAAG,CAAC,CAAA;IACd,KAAK,MAAM,KAAK,IAAI,IAAI,EAAE,CAAC;QACzB,WAAW,CAAC,GAAG,CAAC,KAAK,EAAE,MAAM,CAAC,CAAA;QAC9B,MAAM,IAAI,KAAK,CAAC,MAAM,CAAA;IACxB,CAAC;IACD,OAAO,WAAW,CAAA;AACpB,CAAC;AAQD,0EAA0E;AAC1E,4EAA4E;AAC5E,6EAA6E;AAC7E,iDAAiD;AACjD,MAAM,UAAU,aAAa,CAC3B,OAAY,EACZ,KAAa,EACb,GAAW,EACX,GAAW,EACX,MAAW,EAAE;IAEb,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,OAAO,CAAC,MAAM,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC;QAC/C,MAAM,CAAC,GAAG,OAAO,CAAC,CAAC,CAAE,CAAA;QACrB,IAAI,CAAC,CAAC,MAAM,KAAK,KAAK,EAAE,CAAC;YACvB,IAAI,CAAC,CAAC,KAAK,IAAI,GAAG,EAAE,CAAC;gBACnB,MAAK;YACP,CAAC;iBAAM,IAAI,CAAC,CAAC,GAAG,IAAI,GAAG,EAAE,CAAC;gBACxB,GAAG,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;YACb,CAAC;QACH,CAAC;aAAM,IAAI,CAAC,CAAC,MAAM,GAAG,KAAK,EAAE,CAAC;YAC5B,MAAK;QACP,CAAC;IACH,CAAC;IACD,OAAO,GAAG,CAAA;AACZ,CAAC"}
|
package/package.json
CHANGED
package/src/util.ts
CHANGED
|
@@ -98,7 +98,50 @@ export function optimizeChunks(chunks: Chunk[], lowest?: OffsetCoords) {
|
|
|
98
98
|
}
|
|
99
99
|
}
|
|
100
100
|
|
|
101
|
-
return mergedChunks
|
|
101
|
+
return makeDisjoint(mergedChunks)
|
|
102
|
+
}
|
|
103
|
+
|
|
104
|
+
/**
|
|
105
|
+
* Trim any chunk that starts before its predecessor ends, so the spans a query
|
|
106
|
+
* reads never overlap.
|
|
107
|
+
*
|
|
108
|
+
* Merging alone does not guarantee this. Two chunks from different bins can
|
|
109
|
+
* overlap inside a single BGZF block while the merge above still declines to
|
|
110
|
+
* join them, because the combined span would exceed its 5MB cap — on
|
|
111
|
+
* test/data/out.bam that leaves `3804:0-4977599:16404` next to
|
|
112
|
+
* `4977599:5843-9719917:27612`, whose 10561 shared bytes of block 4977599 get
|
|
113
|
+
* fetched, decompressed and decoded by both. Every record in the overlap is
|
|
114
|
+
* then returned TWICE from getRecordsForRange (5 of out.bam's 6551), which a
|
|
115
|
+
* consumer sees as a duplicated read: rendered twice, counted twice in
|
|
116
|
+
* coverage, and colliding on any id derived from fileOffset.
|
|
117
|
+
*
|
|
118
|
+
* Safe because a BAI chunk's `maxv` is the virtual offset just past its last
|
|
119
|
+
* record — a record boundary — so no record begins inside the trimmed span and
|
|
120
|
+
* the union of the chunks is unchanged. Verified by record counts: the
|
|
121
|
+
* duplicates disappear and the number of DISTINCT records is identical.
|
|
122
|
+
*/
|
|
123
|
+
function makeDisjoint(chunks: Chunk[]) {
|
|
124
|
+
const out: Chunk[] = [chunks[0]!]
|
|
125
|
+
for (let i = 1; i < chunks.length; i++) {
|
|
126
|
+
const chunk = chunks[i]!
|
|
127
|
+
const prevMax = out[out.length - 1]!.maxv
|
|
128
|
+
const cmp =
|
|
129
|
+
chunk.minv.blockPosition - prevMax.blockPosition ||
|
|
130
|
+
chunk.minv.dataPosition - prevMax.dataPosition
|
|
131
|
+
if (cmp >= 0) {
|
|
132
|
+
out.push(chunk)
|
|
133
|
+
continue
|
|
134
|
+
}
|
|
135
|
+
// starts inside the previous chunk: begin where that one ended, and drop it
|
|
136
|
+
// entirely if that leaves nothing
|
|
137
|
+
const stillHasData =
|
|
138
|
+
chunk.maxv.blockPosition - prevMax.blockPosition ||
|
|
139
|
+
chunk.maxv.dataPosition - prevMax.dataPosition
|
|
140
|
+
if (stillHasData > 0) {
|
|
141
|
+
out.push(new Chunk(prevMax, chunk.maxv, chunk.bin, chunk.endPosition))
|
|
142
|
+
}
|
|
143
|
+
}
|
|
144
|
+
return out
|
|
102
145
|
}
|
|
103
146
|
|
|
104
147
|
// The pseudo-bin's mapped-record count, a little-endian uint64. Read as two
|