edf2csv 0.5.92 → 0.5.94

package/CHANGELOG.md

@@ -3,6 +3,53 @@
 Notable changes to edf2csv. Versions follow [semantic versioning](https://semver.org); while the
 major version is 0, a minor bump may contain breaking changes.
 
+## 0.5.94
+
+### Fixed: two conversions with identical output disagreed about whether they were whole
+
+```
+$ edf2csv tenmin.edf --out a            # whole_recording: true
+$ edf2csv tenmin.edf --out b --end 600.3 # whole_recording: false
+$ diff a/signals.csv b/signals.csv       # no output
+```
+
+Byte-identical files, 60,030 rows each, described two ways. `whole_recording` is
+`clampedEnd >= latest`, and `latest` is `recordCount * recordDuration` — 6003 records of 0.1s
+is 600.3000000000001, not the 600.3 that `--info` prints as the recording's length. So naming
+the length exactly makes the conversion partial by a rounding error, on the field a pipeline
+reads to decide whether it has the lot.
+
+The same arithmetic 0.5.91 fixed one field over, and both now go through one `sameInstant`
+comparison so they cannot drift apart: a relative epsilon, well below any real sample interval
+and well above the gap between two routes to the same quantity. A window that really is partial
+still says so.
+
+## 0.5.93
+
+### Fixed: the padding after a TAL was counted as an annotation that could not be read
+
+```
+$ edf2csv space-padded.edf --out out
+warning: 2 annotation entries were unreadable and could not be exported.
+
+$ cat out/annotations.csv
+onset_s,duration_s,description,record_index
+0.5,,Lights off,0
+```
+
+The file holds one annotation and it was exported in full. The two "entries" are the spaces
+filling the rest of each record's annotation slot.
+
+EDF+ pads that slot with `0x00`, which the decoder skips because NUL is also what ends one TAL
+and starts the next. Writers pad with spaces, and a run of spaces after the last TAL is a
+non-empty chunk that does not begin with a sign — so it took the malformed branch. Under
+`--strict` that is a failed run over the whitespace at the end of a slot, on a file that lost
+nothing.
+
+Whitespace-only chunks are padding now — space, tab, CR, LF and NUL. A chunk of anything else
+that does not parse is still counted and still reported, which is the case the warning exists
+for.
+
 ## 0.5.92
 
 ### Fixed: `--info` redirected into a full filesystem wrote nothing and exited 0

package/dist/convert/time-range.js

@@ -133,6 +133,18 @@ export function countSamplesInRange(options) {
     const last = Math.max(0, Math.min(options.samplesPerRecord, upper));
     return Math.max(0, last - first);
 }
+/**
+ * Whether two instants differ only by the arithmetic that produced them.
+ *
+ * A relative epsilon, because the gap between doubles grows with magnitude — the same shape
+ * the long layout uses to decide two sample times are one instant. Well below any real sample
+ * interval, and well above the rounding that two routes to one quantity produce: a recording's
+ * length is `recordCount * recordDuration`, which for 6003 records of 0.1s is not the 600.3 it
+ * prints as.
+ */
+function sameInstant(a, b) {
+    return Math.abs(a - b) <= Math.max(Math.abs(a), Math.abs(b)) * 1e-12;
+}
 /**
  * Turn a requested window into both an exact time span and the record range that
  * contains it. Records are the unit the file can be read in; the exact span is what
@@ -170,9 +182,7 @@ export function resolveRange(options) {
       instant: well below any real interval, and well above the rounding that two different
       routes to the same quantity produce.
     */
-    const atTheEnd = startSeconds >= latest ||
-        Math.abs(startSeconds - latest) <= Math.max(Math.abs(startSeconds), Math.abs(latest)) * 1e-12;
-    if (atTheEnd) {
+    if (startSeconds >= latest || sameInstant(startSeconds, latest)) {
         /*
           Quote what was typed. Reporting the parsed seconds meant `--start 4h` came back as
           "--start 14400s is at or past the end", which reads as a value the user never gave.
@@ -229,7 +239,17 @@ export function resolveRange(options) {
         endSeconds: clampedEnd,
         startRecord,
         endRecord,
-        isWholeRecording: startSeconds <= earliest && clampedEnd >= latest,
+        /*
+          The same rounding, one field over.
+    
+          `latest` is `recordCount * recordDuration`, and 6003 records of 0.1s is
+          600.3000000000001. So `--end 600.3` on a recording of exactly that length wrote every
+          sample it has — byte-identical to a bare conversion — and metadata.json recorded
+          `whole_recording: false` for it, while the bare run recorded true. One conversion, two
+          answers, on the field a pipeline reads to decide whether it has the lot.
+        */
+        isWholeRecording: (startSeconds <= earliest || sameInstant(startSeconds, earliest)) &&
+            (clampedEnd >= latest || sameInstant(clampedEnd, latest)),
         recordingStartSeconds: earliest,
         recordingEndSeconds: latest,
     };

package/dist/convert/time-range.js.map

@@ -1 +1 @@
-{"version":3,"file":"time-range.js","sourceRoot":"","sources":["../../src/convert/time-range.ts"],"names":[],"mappings":"AAAA;;;;;;;GAOG;AAEH,OAAO,EAAE,cAAc,EAAE,MAAM,qBAAqB,CAAC;AAErD,MAAM,OAAO,cAAe,SAAQ,KAAK;IACvC,YAAY,OAAe;QACzB,KAAK,CAAC,OAAO,CAAC,CAAC;QACf,IAAI,CAAC,IAAI,GAAG,gBAAgB,CAAC;IAC/B,CAAC;CACF;AAED,MAAM,YAAY,GAA2B;IAC3C,CAAC,EAAE,IAAI;IACP,EAAE,EAAE,IAAI;IACR,GAAG,EAAE,IAAI;IACT,IAAI,EAAE,IAAI;IACV,KAAK,EAAE,IAAI;IACX,CAAC,EAAE,EAAE;IACL,GAAG,EAAE,EAAE;IACP,IAAI,EAAE,EAAE;IACR,MAAM,EAAE,EAAE;IACV,OAAO,EAAE,EAAE;IACX,CAAC,EAAE,CAAC;IACJ,GAAG,EAAE,CAAC;IACN,IAAI,EAAE,CAAC;IACP,MAAM,EAAE,CAAC;IACT,OAAO,EAAE,CAAC;IACV,EAAE,EAAE,KAAK;CACV,CAAC;AAEF,MAAM,KAAK,GAAG,6CAA6C,CAAC;AAC5D,MAAM,UAAU,GAAG,+BAA+B,CAAC;AAEnD;;;;GAIG;AACH,MAAM,UAAU,aAAa,CAAC,KAAa,EAAE,UAAkB;IAC7D,MAAM,IAAI,GAAG,KAAK,CAAC,IAAI,EAAE,CAAC,WAAW,EAAE,CAAC;IACxC,IAAI,IAAI,KAAK,EAAE,EAAE,CAAC;QAChB,MAAM,IAAI,cAAc,CAAC,GAAG,UAAU,mDAAmD,CAAC,CAAC;IAC7F,CAAC;IAED,6BAA6B;IAC7B,IAAI,kBAAkB,CAAC,IAAI,CAAC,IAAI,CAAC;QAAE,OAAO,YAAY,CAAC,MAAM,CAAC,IAAI,CAAC,EAAE,UAAU,EAAE,KAAK,CAAC,CAAC;IAExF,iCAAiC;IACjC,MAAM,KAAK,GAAG,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IAC/B,IAAI,KAAK,EAAE,CAAC;QACV,MAAM,KAAK,GAAG,KAAK,CAAC,CAAC,CAAC,KAAK,SAAS,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,MAAM,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC;QAC5D,MAAM,OAAO,GAAG,MAAM,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC;QACjC,MAAM,OAAO,GAAG,MAAM,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC;QACjC,IAAI,OAAO,IAAI,EAAE,IAAI,OAAO,IAAI,EAAE,EAAE,CAAC;YACnC,MAAM,IAAI,cAAc,CACtB,GAAG,UAAU,KAAK,KAAK,iDAAiD,CACzE,CAAC;QACJ,CAAC;QACD,OAAO,YAAY,CAAC,KAAK,GAAG,IAAI,GAAG,OAAO,GAAG,EAAE,GAAG,OAAO,EAAE,UAAU,EAAE,KAAK,CAAC,CAAC;IAChF,CAAC;IAED,kCAAkC;IAClC,UAAU,CAAC,SAAS,GAAG,CAAC,CAAC;IACzB,IAAI,KAAK,GAAG,CAAC,CAAC;IACd,IAAI,OAAO,GAAG,CAAC,CAAC;IAChB,IAAI,QAAQ,GAAG,CAAC,CAAC;IACjB,IAAI,KAA6B,CAAC;IAClC,qFAAqF;IACrF,MAAM,IAAI,GAAG,IAAI,GAAG,EAAU,CAAC;IAC/B,OAAO,CAAC,KAAK,GAAG,UAAU,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC,KAAK,IAAI,EAAE,CAAC;QAChD,MAAM,MAAM,GAAG,MAAM,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC;QAChC,MAAM,IAAI,GAAG,KAAK,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;QAC5B,MAAM,KAAK,GAAG,YAAY,CAAC,IAAI,CAAC,CAAC;QACjC,IAAI,KAAK,KAAK,SAAS,EAAE,CAAC;YACxB,MAAM,IAAI,cAAc,CACtB,GAAG,UAAU,KAAK,KAAK,2BAA2B,IAAI,wBAAwB,CAC/E,CAAC;QACJ,CAAC;QAED,iFAAiF;QACjF,qFAAqF;QACrF,qFAAqF;QACrF,IAAI,IAAI,CAAC,GAAG,CAAC,KAAK,CAAC,EAAE,CAAC;YACpB,MAAM,IAAI,cAAc,CACtB,GAAG,UAAU,KAAK,KAAK,+BAA+B;gBACpD,sCAAsC,CACzC,CAAC;QACJ,CAAC;QACD,IAAI,CAAC,GAAG,CAAC,KAAK,CAAC,CAAC;QAEhB,KAAK,IAAI,MAAM,GAAG,KAAK,CAAC;QACxB,OAAO,EAAE,CAAC;QACV,QAAQ,IAAI,KAAK,CAAC,CAAC,CAAC,CAAC,MAAM,CAAC;IAC9B,CAAC;IAED,8DAA8D;IAC9D,IAAI,OAAO,KAAK,CAAC,IAAI,QAAQ,KAAK,IAAI,CAAC,OAAO,CAAC,OAAO,EAAE,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC;QACnE,MAAM,IAAI,cAAc,CACtB,GAAG,UAAU,KAAK,KAAK,gCAAgC;YACrD,6DAA6D,CAChE,CAAC;IACJ,CAAC;IAED,OAAO,YAAY,CAAC,KAAK,EAAE,UAAU,EAAE,KAAK,CAAC,CAAC;AAChD,CAAC;AAED,SAAS,YAAY,CAAC,KAAa,EAAE,UAAkB,EAAE,KAAa;IACpE,IAAI,CAAC,MAAM,CAAC,QAAQ,CAAC,KAAK,CAAC,IAAI,KAAK,GAAG,CAAC,EAAE,CAAC;QACzC,MAAM,IAAI,cAAc,CAAC,GAAG,UAAU,KAAK,KAAK,qCAAqC,CAAC,CAAC;IACzF,CAAC;IACD,OAAO,KAAK,CAAC;AACf,CAAC;AAmBD;;;;;;GAMG;AACH,MAAM,CAAC,MAAM,kBAAkB,GAAG,IAAI,CAAC;AAEvC;;;;;;;;;GASG;AACH,MAAM,UAAU,YAAY,CAAC,IAAY;IACvC,IAAI,CAAC,CAAC,IAAI,GAAG,CAAC,CAAC,IAAI,CAAC,MAAM,CAAC,QAAQ,CAAC,IAAI,CAAC;QAAE,OAAO,kBAAkB,CAAC;IACrE,OAAO,IAAI,CAAC,GAAG,CAAC,kBAAkB,EAAE,CAAC,GAAG,IAAI,GAAG,CAAC,CAAC,CAAC;AACpD,CAAC;AAED,+EAA+E;AAC/E,MAAM,UAAU,mBAAmB,CACjC,IAAY,EACZ,YAAoB,EACpB,UAAkB,EAClB,YAAoB,kBAAkB;IAEtC,OAAO,IAAI,IAAI,YAAY,GAAG,SAAS,IAAI,IAAI,GAAG,UAAU,GAAG,SAAS,CAAC;AAC3E,CAAC;AAED,mFAAmF;AACnF,MAAM,UAAU,mBAAmB,CAAC,OAMnC;IACC,MAAM,KAAK,GAAG,YAAY,CAAC,OAAO,CAAC,IAAI,CAAC,CAAC;IACzC,MAAM,KAAK,GAAG,IAAI,CAAC,IAAI,CACrB,CAAC,OAAO,CAAC,YAAY,GAAG,KAAK,GAAG,OAAO,CAAC,WAAW,CAAC,GAAG,OAAO,CAAC,IAAI,CACpE,CAAC;IACF,MAAM,KAAK,GAAG,IAAI,CAAC,IAAI,CACrB,CAAC,OAAO,CAAC,UAAU,GAAG,KAAK,GAAG,OAAO,CAAC,WAAW,CAAC,GAAG,OAAO,CAAC,IAAI,CAClE,CAAC;IACF,MAAM,KAAK,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,gBAAgB,EAAE,KAAK,CAAC,CAAC,CAAC;IACrE,MAAM,IAAI,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,gBAAgB,EAAE,KAAK,CAAC,CAAC,CAAC;IACpE,OAAO,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,GAAG,KAAK,CAAC,CAAC;AACnC,CAAC;AAED;;;;GAIG;AACH,MAAM,UAAU,YAAY,CAAC,OAe5B;IACC,4EAA4E;IAC5E,gFAAgF;IAChF,6EAA6E;IAC7E,4EAA4E;IAC5E,2EAA2E;IAC3E,MAAM,EAAE,QAAQ,EAAE,MAAM,EAAE,GAAG,IAAI,CAAC,OAAO,CAAC,YAAY,EAAE,OAAO,CAAC,WAAW,EAAE,OAAO,CAAC,cAAc,CAAC,CAAC;IAErG,IAAI,OAAO,CAAC,QAAQ,KAAK,SAAS,IAAI,OAAO,CAAC,GAAG,KAAK,SAAS,EAAE,CAAC;QAChE,MAAM,IAAI,cAAc,CAAC,2CAA2C,CAAC,CAAC;IACxE,CAAC;IAED,MAAM,YAAY,GAAG,OAAO,CAAC,KAAK,IAAI,QAAQ,CAAC;IAE/C,IAAI,UAAkB,CAAC;IACvB,IAAI,OAAO,CAAC,QAAQ,KAAK,SAAS;QAAE,UAAU,GAAG,YAAY,GAAG,OAAO,CAAC,QAAQ,CAAC;SAC5E,IAAI,OAAO,CAAC,GAAG,KAAK,SAAS;QAAE,UAAU,GAAG,OAAO,CAAC,GAAG,CAAC;;QACxD,UAAU,GAAG,MAAM,CAAC;IAEzB;;;;;;;;;;;;;MAaE;IACF,MAAM,QAAQ,GACZ,YAAY,IAAI,MAAM;QACtB,IAAI,CAAC,GAAG,CAAC,YAAY,GAAG,MAAM,CAAC,IAAI,IAAI,CAAC,GAAG,CAAC,IAAI,CAAC,GAAG,CAAC,YAAY,CAAC,EAAE,IAAI,CAAC,GAAG,CAAC,MAAM,CAAC,CAAC,GAAG,KAAK,CAAC;IAChG,IAAI,QAAQ,EAAE,CAAC;QACb;;;;;;;;;UASE;QACF,MAAM,IAAI,cAAc;QACtB;;;;;;;;;;;;UAYE;QACF;;;;;;;UAOE;QACF,QAAQ,KAAK,CAAC;YACZ,CAAC,CAAC,WAAW,MAAM,CAAC,OAAO,CAAC,SAAS,EAAE,YAAY,CAAC,iCAAiC;gBACnF,GAAG,cAAc,CAAC,MAAM,CAAC,aAAa;YACxC,CAAC,CAAC,WAAW,MAAM,CAAC,OAAO,CAAC,SAAS,EAAE,YAAY,CAAC,iCAAiC;gBACnF,GAAG,cAAc,CAAC,MAAM,GAAG,QAAQ,CAAC,8BAA8B;gBAClE,GAAG,aAAa,CAAC,QAAQ,CAAC,OAAO,aAAa,CAAC,MAAM,CAAC,GAAG,CAC9D,CAAC;IACJ,CAAC;IACD,IAAI,UAAU,IAAI,YAAY,EAAE,CAAC;QAC/B,uFAAuF;QACvF,sFAAsF;QACtF,qFAAqF;QACrF,MAAM,QAAQ,GACZ,OAAO,CAAC,GAAG,KAAK,SAAS,IAAI,OAAO,CAAC,OAAO,KAAK,SAAS;YACxD,CAAC,CAAC,MAAM,CAAC,OAAO,CAAC,OAAO,EAAE,UAAU,CAAC;YACrC,CAAC,CAAC,aAAa,CAAC,UAAU,CAAC,CAAC;QAChC,MAAM,UAAU,GAAG,MAAM,CAAC,OAAO,CAAC,SAAS,EAAE,YAAY,CAAC,CAAC;QAC3D,MAAM,IAAI,cAAc,CACtB,gCAAgC,QAAQ,qCAAqC,UAAU,GAAG,CAC3F,CAAC;IACJ,CAAC;IAED,MAAM,UAAU,GAAG,IAAI,CAAC,GAAG,CAAC,UAAU,EAAE,MAAM,CAAC,CAAC;IAChD,MAAM,EAAE,WAAW,EAAE,SAAS,EAAE,GAAG,aAAa,CAAC,OAAO,EAAE,YAAY,EAAE,UAAU,CAAC,CAAC;IAEpF,OAAO;QACL,YAAY;QACZ,UAAU,EAAE,UAAU;QACtB,WAAW;QACX,SAAS;QACT,gBAAgB,EAAE,YAAY,IAAI,QAAQ,IAAI,UAAU,IAAI,MAAM;QAClE,qBAAqB,EAAE,QAAQ;QAC/B,mBAAmB,EAAE,MAAM;KAC5B,CAAC;AACJ,CAAC;AAED,SAAS,IAAI,CACX,YAA6C,EAC7C,WAAmB,EACnB,cAAsB;IAEtB,IAAI,CAAC,YAAY,IAAI,YAAY,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;QAC/C,OAAO,EAAE,QAAQ,EAAE,CAAC,EAAE,MAAM,EAAE,WAAW,GAAG,cAAc,EAAE,CAAC;IAC/D,CAAC;IACD,IAAI,QAAQ,GAAG,QAAQ,CAAC;IACxB,IAAI,MAAM,GAAG,CAAC,QAAQ,CAAC;IACvB,KAAK,MAAM,KAAK,IAAI,YAAY,EAAE,CAAC;QACjC,IAAI,KAAK,GAAG,QAAQ;YAAE,QAAQ,GAAG,KAAK,CAAC;QACvC,IAAI,KAAK,GAAG,cAAc,GAAG,MAAM;YAAE,MAAM,GAAG,KAAK,GAAG,cAAc,CAAC;IACvE,CAAC;IACD,IAAI,CAAC,MAAM,CAAC,QAAQ,CAAC,QAAQ,CAAC,IAAI,CAAC,MAAM,CAAC,QAAQ,CAAC,MAAM,CAAC,EAAE,CAAC;QAC3D,OAAO,EAAE,QAAQ,EAAE,CAAC,EAAE,MAAM,EAAE,WAAW,GAAG,cAAc,EAAE,CAAC;IAC/D,CAAC;IACD,OAAO,EAAE,QAAQ,EAAE,MAAM,EAAE,CAAC;AAC9B,CAAC;AAED,sEAAsE;AACtE,SAAS,aAAa,CACpB,OAAwG,EACxG,YAAoB,EACpB,UAAkB;IAElB,MAAM,MAAM,GAAG,OAAO,CAAC,YAAY,CAAC;IACpC,IAAI,CAAC,MAAM,IAAI,MAAM,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;QACnC,OAAO;YACL,WAAW,EAAE,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,CAAC,KAAK,CAAC,YAAY,GAAG,OAAO,CAAC,cAAc,CAAC,CAAC;YAC3E,SAAS,EAAE,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,WAAW,EAAE,IAAI,CAAC,IAAI,CAAC,UAAU,GAAG,OAAO,CAAC,cAAc,CAAC,CAAC;SACzF,CAAC;IACJ,CAAC;IAED,IAAI,KAAK,GAAG,OAAO,CAAC,WAAW,CAAC;IAChC,IAAI,IAAI,GAAG,CAAC,CAAC;IACb,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,MAAM,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACvC,MAAM,KAAK,GAAG,MAAM,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC;QAC7B,IAAI,KAAK,GAAG,OAAO,CAAC,cAAc,GAAG,YAAY,IAAI,KAAK,GAAG,UAAU,EAAE,CAAC;YACxE,IAAI,CAAC,GAAG,KAAK;gBAAE,KAAK,GAAG,CAAC,CAAC;YACzB,IAAI,CAAC,GAAG,CAAC,GAAG,IAAI;gBAAE,IAAI,GAAG,CAAC,GAAG,CAAC,CAAC;QACjC,CAAC;IACH,CAAC;IACD,OAAO,KAAK,GAAG,IAAI,CAAC,CAAC,CAAC,EAAE,WAAW,EAAE,KAAK,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAC,CAAC,EAAE,WAAW,EAAE,CAAC,EAAE,SAAS,EAAE,CAAC,EAAE,CAAC;AACnG,CAAC;AAED;;;;GAIG;AACH,SAAS,MAAM,CAAC,IAAwB,EAAE,OAAe;IACvD,OAAO,IAAI,KAAK,SAAS,CAAC,CAAC,CAAC,aAAa,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,IAAI,IAAI,GAAG,CAAC;AACnE,CAAC;AAED,SAAS,aAAa,CAAC,OAAe;IACpC,OAAO,GAAG,MAAM,CAAC,OAAO,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC;AAC1C,CAAC","sourcesContent":["/**\n * Parsing for the time-range options.\n *\n * Researchers write offsets in whatever form is natural for the recording in front\n * of them: seconds for a short ECG strip, `30m` into a sleep study, `01:23:45` when\n * reading off a clock. All three are accepted; anything ambiguous is rejected with\n * a message that shows the forms that work.\n */\n\nimport { formatDuration } from '../format/number.js';\n\nexport class TimeRangeError extends Error {\n  constructor(message: string) {\n    super(message);\n    this.name = 'TimeRangeError';\n  }\n}\n\nconst UNIT_SECONDS: Record<string, number> = {\n  h: 3600,\n  hr: 3600,\n  hrs: 3600,\n  hour: 3600,\n  hours: 3600,\n  m: 60,\n  min: 60,\n  mins: 60,\n  minute: 60,\n  minutes: 60,\n  s: 1,\n  sec: 1,\n  secs: 1,\n  second: 1,\n  seconds: 1,\n  ms: 0.001,\n};\n\nconst CLOCK = /^(?:(\\d+):)?(\\d{1,2}):(\\d{1,2}(?:\\.\\d+)?)$/u;\nconst UNIT_TOKEN = /(\\d+(?:\\.\\d+)?)\\s*([a-z]+)/giu;\n\n/**\n * Parse a duration or offset into seconds.\n *\n * Accepted: `90`, `90s`, `5m`, `1h30m`, `1h 30m 15s`, `00:30:00`, `30:00`, `250ms`.\n */\nexport function parseTimeSpec(input: string, optionName: string): number {\n  const text = input.trim().toLowerCase();\n  if (text === '') {\n    throw new TimeRangeError(`${optionName} is empty. Try a value like 30s, 5m, or 00:30:00.`);\n  }\n\n  // Bare number means seconds.\n  if (/^\\d+(?:\\.\\d+)?$/u.test(text)) return assertFinite(Number(text), optionName, input);\n\n  // Clock form: hh:mm:ss or mm:ss.\n  const clock = CLOCK.exec(text);\n  if (clock) {\n    const hours = clock[1] === undefined ? 0 : Number(clock[1]);\n    const minutes = Number(clock[2]);\n    const seconds = Number(clock[3]);\n    if (minutes >= 60 || seconds >= 60) {\n      throw new TimeRangeError(\n        `${optionName} \"${input}\" has a minutes or seconds field of 60 or more.`,\n      );\n    }\n    return assertFinite(hours * 3600 + minutes * 60 + seconds, optionName, input);\n  }\n\n  // Unit form: 1h30m, 5 min, 250ms.\n  UNIT_TOKEN.lastIndex = 0;\n  let total = 0;\n  let matched = 0;\n  let consumed = 0;\n  let match: RegExpExecArray | null;\n  // Which units have already been seen, so a repeat can be rejected rather than added.\n  const seen = new Set<number>();\n  while ((match = UNIT_TOKEN.exec(text)) !== null) {\n    const amount = Number(match[1]);\n    const unit = match[2] ?? '';\n    const scale = UNIT_SECONDS[unit];\n    if (scale === undefined) {\n      throw new TimeRangeError(\n        `${optionName} \"${input}\" uses an unknown unit \"${unit}\". Use h, m, s, or ms.`,\n      );\n    }\n\n    // \"1h1h\" is a typo, not a request for two hours. Summing repeated units silently\n    // turned a slip into a plausible window that was quietly the wrong length. Units are\n    // keyed by their scale so the aliases collapse together: \"1h30min20m\" is caught too.\n    if (seen.has(scale)) {\n      throw new TimeRangeError(\n        `${optionName} \"${input}\" gives the same unit twice. ` +\n          `Combine each unit once, as in 1h30m.`,\n      );\n    }\n    seen.add(scale);\n\n    total += amount * scale;\n    matched++;\n    consumed += match[0].length;\n  }\n\n  // Reject partially-understood input like \"5x\" or \"1h banana\".\n  if (matched === 0 || consumed !== text.replace(/\\s+/gu, '').length) {\n    throw new TimeRangeError(\n      `${optionName} \"${input}\" is not a time I understand. ` +\n        `Try 30s, 5m, 1h30m, 00:30:00, or a plain number of seconds.`,\n    );\n  }\n\n  return assertFinite(total, optionName, input);\n}\n\nfunction assertFinite(value: number, optionName: string, input: string): number {\n  if (!Number.isFinite(value) || value < 0) {\n    throw new TimeRangeError(`${optionName} \"${input}\" is not a valid non-negative time.`);\n  }\n  return value;\n}\n\nexport interface ResolvedRange {\n  /** Inclusive start, in seconds from the beginning of the recording. */\n  startSeconds: number;\n  /** Exclusive end, in seconds. */\n  endSeconds: number;\n  /** First data record touching the window. */\n  startRecord: number;\n  /** One past the last data record touching the window. */\n  endRecord: number;\n  /** True when the window covers the whole recording. */\n  isWholeRecording: boolean;\n  /** Earliest record start, including EDF+D timing gaps. */\n  recordingStartSeconds: number;\n  /** End of the latest record, including EDF+D timing gaps. */\n  recordingEndSeconds: number;\n}\n\n/**\n * Slack for comparisons at sample/window boundaries.\n *\n * A nanosecond is far below any real sampling interval — 20 kHz is 50 microseconds — so it\n * absorbs the arithmetic error in `recordStart + sample / rate` without reaching a\n * neighbouring sample.\n */\nexport const BOUNDARY_TOLERANCE = 1e-9;\n\n/**\n * The slack to use for a channel sampled this often.\n *\n * Never as much as half a sample interval, because slack that reaches the next sample stops\n * being slack. A fixed nanosecond was applied whatever the rate, and the format does not\n * oblige the interval to be larger than it: EDF's record duration is an 8-character field\n * that accepts `1e-9`. A recording of two 1 ns records holding ten samples each wrote ten of\n * its twenty rows — the window ends at 2e-9, the comparison asked for `time < 2e-9 - 1e-9`,\n * and the entire second record failed it. Exit 0, no warning, half the samples gone.\n */\nexport function toleranceFor(rate: number): number {\n  if (!(rate > 0) || !Number.isFinite(rate)) return BOUNDARY_TOLERANCE;\n  return Math.min(BOUNDARY_TOLERANCE, 1 / rate / 2);\n}\n\n/** Match the exact half-open boundary rules used while writing signal rows. */\nexport function sampleTimeIsInRange(\n  time: number,\n  startSeconds: number,\n  endSeconds: number,\n  tolerance: number = BOUNDARY_TOLERANCE,\n): boolean {\n  return time >= startSeconds - tolerance && time < endSeconds - tolerance;\n}\n\n/** Count samples from one record that fall inside a half-open requested window. */\nexport function countSamplesInRange(options: {\n  recordStart: number;\n  rate: number;\n  samplesPerRecord: number;\n  startSeconds: number;\n  endSeconds: number;\n}): number {\n  const slack = toleranceFor(options.rate);\n  const lower = Math.ceil(\n    (options.startSeconds - slack - options.recordStart) * options.rate,\n  );\n  const upper = Math.ceil(\n    (options.endSeconds - slack - options.recordStart) * options.rate,\n  );\n  const first = Math.max(0, Math.min(options.samplesPerRecord, lower));\n  const last = Math.max(0, Math.min(options.samplesPerRecord, upper));\n  return Math.max(0, last - first);\n}\n\n/**\n * Turn a requested window into both an exact time span and the record range that\n * contains it. Records are the unit the file can be read in; the exact span is what\n * decides which samples inside those records are actually written.\n */\nexport function resolveRange(options: {\n  start?: number | undefined;\n  /** The `--start` value exactly as typed, quoted back in the past-the-end error. */\n  startText?: string | undefined;\n  duration?: number | undefined;\n  end?: number | undefined;\n  /** The `--end` value exactly as typed, quoted back in the window error. */\n  endText?: string | undefined;\n  recordDuration: number;\n  recordCount: number;\n  /**\n   * True start time of each data record, for discontinuous files. When absent,\n   * records are assumed to sit end to end.\n   */\n  recordStarts?: Float64Array | null | undefined;\n}): ResolvedRange {\n  // For a continuous file the recording spans recordCount * recordDuration. A\n  // discontinuous one does not: a 10-second recording with a 95-second gap in the\n  // middle still ends at 105 seconds. Deriving the span from the records' real\n  // positions is what stops a requested window from being clipped back to the\n  // amount of *data* in the file and silently discarding everything past it.\n  const { earliest, latest } = span(options.recordStarts, options.recordCount, options.recordDuration);\n\n  if (options.duration !== undefined && options.end !== undefined) {\n    throw new TimeRangeError('Use either --duration or --end, not both.');\n  }\n\n  const startSeconds = options.start ?? earliest;\n\n  let endSeconds: number;\n  if (options.duration !== undefined) endSeconds = startSeconds + options.duration;\n  else if (options.end !== undefined) endSeconds = options.end;\n  else endSeconds = latest;\n\n  /*\n    At the end, allowing for the arithmetic that produced the end.\n\n    `latest` is `recordCount * recordDuration`, and with a fractional duration that is not the\n    number it prints as: 6003 records of 0.1s is 600.3000000000001, not 600.3. So `--start\n    600.3` on a recording `--info` calls \"10m 0.3s\" was accepted by a hair, converted nothing,\n    and exited 0 with a signals.csv holding its header — which is the empty conversion this\n    error exists to prevent, and which the same command on a whole-second recording is refused\n    for.\n\n    A relative epsilon, the same shape the long layout uses to decide two sample times are one\n    instant: well below any real interval, and well above the rounding that two different\n    routes to the same quantity produce.\n  */\n  const atTheEnd =\n    startSeconds >= latest ||\n    Math.abs(startSeconds - latest) <= Math.max(Math.abs(startSeconds), Math.abs(latest)) * 1e-12;\n  if (atTheEnd) {\n    /*\n      Quote what was typed. Reporting the parsed seconds meant `--start 4h` came back as\n      \"--start 14400s is at or past the end\", which reads as a value the user never gave.\n\n      In quotation marks, which the parse errors above have always used and this did not.\n      Without them the value ran into the sentence: `--start \"  5s  \"` printed as\n      `--start   5s   is at or past the end`, where the value appears to be `5s   is` and\n      the surrounding spaces — the actual reason a shell-built argument went wrong — are\n      invisible.\n    */\n    throw new TimeRangeError(\n      /*\n        The recording's length in the same words --info uses for it.\n\n        `formatSeconds` renders a bare number of seconds, so this message and the Duration\n        line disagreed about one file in one session: --info said \"6m 40s\" and the error said\n        \"400s\". On an overnight recording it read \"7950s recording\", leaving the reader to\n        divide by 3600 to find out whether their --start was reasonable — which is the one\n        question this message exists to answer. cli-reference.md has always documented it\n        humanised (\"2h 12m 30s\"), a form no input could produce.\n\n        The typed value keeps `quoted`, since that is the user's own text and should come\n        back exactly as they wrote it.\n      */\n      /*\n        `latest` is where the recording ends on its own clock, which is its length only when\n        it starts at zero. A file timed from its first record's timekeeping TAL need not: one\n        whose records run 1000s to 1003s is three seconds long, and this called it \"this\n        16m 43s recording\" — while --info two lines away said \"Duration 3s\". Where the\n        recording sits is the useful thing to say in that case, and it is the number --start\n        has to be given.\n      */\n      earliest === 0\n        ? `--start ${quoted(options.startText, startSeconds)} is at or past the end of this ` +\n          `${formatDuration(latest)} recording.`\n        : `--start ${quoted(options.startText, startSeconds)} is at or past the end of this ` +\n          `${formatDuration(latest - earliest)} recording, which runs from ` +\n          `${formatSeconds(earliest)} to ${formatSeconds(latest)}.`,\n    );\n  }\n  if (endSeconds <= startSeconds) {\n    // Quote whatever the caller actually gave, for the same reason as the error above. The\n    // end is only echoed when --end was passed: with --duration the end is computed here,\n    // so there is no typed value to quote and the arithmetic result is the honest thing.\n    const endShown =\n      options.end !== undefined && options.endText !== undefined\n        ? quoted(options.endText, endSeconds)\n        : formatSeconds(endSeconds);\n    const startShown = quoted(options.startText, startSeconds);\n    throw new TimeRangeError(\n      `The requested window ends at ${endShown}, which is not after its start at ${startShown}.`,\n    );\n  }\n\n  const clampedEnd = Math.min(endSeconds, latest);\n  const { startRecord, endRecord } = selectRecords(options, startSeconds, clampedEnd);\n\n  return {\n    startSeconds,\n    endSeconds: clampedEnd,\n    startRecord,\n    endRecord,\n    isWholeRecording: startSeconds <= earliest && clampedEnd >= latest,\n    recordingStartSeconds: earliest,\n    recordingEndSeconds: latest,\n  };\n}\n\nfunction span(\n  recordStarts: Float64Array | null | undefined,\n  recordCount: number,\n  recordDuration: number,\n): { earliest: number; latest: number } {\n  if (!recordStarts || recordStarts.length === 0) {\n    return { earliest: 0, latest: recordCount * recordDuration };\n  }\n  let earliest = Infinity;\n  let latest = -Infinity;\n  for (const start of recordStarts) {\n    if (start < earliest) earliest = start;\n    if (start + recordDuration > latest) latest = start + recordDuration;\n  }\n  if (!Number.isFinite(earliest) || !Number.isFinite(latest)) {\n    return { earliest: 0, latest: recordCount * recordDuration };\n  }\n  return { earliest, latest };\n}\n\n/** Every record whose own time span overlaps the requested window. */\nfunction selectRecords(\n  options: { recordDuration: number; recordCount: number; recordStarts?: Float64Array | null | undefined },\n  startSeconds: number,\n  endSeconds: number,\n): { startRecord: number; endRecord: number } {\n  const starts = options.recordStarts;\n  if (!starts || starts.length === 0) {\n    return {\n      startRecord: Math.max(0, Math.floor(startSeconds / options.recordDuration)),\n      endRecord: Math.min(options.recordCount, Math.ceil(endSeconds / options.recordDuration)),\n    };\n  }\n\n  let first = options.recordCount;\n  let last = 0;\n  for (let i = 0; i < starts.length; i++) {\n    const begin = starts[i] ?? 0;\n    if (begin + options.recordDuration > startSeconds && begin < endSeconds) {\n      if (i < first) first = i;\n      if (i + 1 > last) last = i + 1;\n    }\n  }\n  return first < last ? { startRecord: first, endRecord: last } : { startRecord: 0, endRecord: 0 };\n}\n\n/**\n * The value as the caller typed it, in quotation marks, or the parsed seconds if they gave\n * none. The marks show where the value begins and ends, which matters most for the values\n * that went wrong because of what surrounds them.\n */\nfunction quoted(text: string | undefined, seconds: number): string {\n  return text === undefined ? formatSeconds(seconds) : `\"${text}\"`;\n}\n\nfunction formatSeconds(seconds: number): string {\n  return `${Number(seconds.toFixed(3))}s`;\n}\n"]}
+{"version":3,"file":"time-range.js","sourceRoot":"","sources":["../../src/convert/time-range.ts"],"names":[],"mappings":"AAAA;;;;;;;GAOG;AAEH,OAAO,EAAE,cAAc,EAAE,MAAM,qBAAqB,CAAC;AAErD,MAAM,OAAO,cAAe,SAAQ,KAAK;IACvC,YAAY,OAAe;QACzB,KAAK,CAAC,OAAO,CAAC,CAAC;QACf,IAAI,CAAC,IAAI,GAAG,gBAAgB,CAAC;IAC/B,CAAC;CACF;AAED,MAAM,YAAY,GAA2B;IAC3C,CAAC,EAAE,IAAI;IACP,EAAE,EAAE,IAAI;IACR,GAAG,EAAE,IAAI;IACT,IAAI,EAAE,IAAI;IACV,KAAK,EAAE,IAAI;IACX,CAAC,EAAE,EAAE;IACL,GAAG,EAAE,EAAE;IACP,IAAI,EAAE,EAAE;IACR,MAAM,EAAE,EAAE;IACV,OAAO,EAAE,EAAE;IACX,CAAC,EAAE,CAAC;IACJ,GAAG,EAAE,CAAC;IACN,IAAI,EAAE,CAAC;IACP,MAAM,EAAE,CAAC;IACT,OAAO,EAAE,CAAC;IACV,EAAE,EAAE,KAAK;CACV,CAAC;AAEF,MAAM,KAAK,GAAG,6CAA6C,CAAC;AAC5D,MAAM,UAAU,GAAG,+BAA+B,CAAC;AAEnD;;;;GAIG;AACH,MAAM,UAAU,aAAa,CAAC,KAAa,EAAE,UAAkB;IAC7D,MAAM,IAAI,GAAG,KAAK,CAAC,IAAI,EAAE,CAAC,WAAW,EAAE,CAAC;IACxC,IAAI,IAAI,KAAK,EAAE,EAAE,CAAC;QAChB,MAAM,IAAI,cAAc,CAAC,GAAG,UAAU,mDAAmD,CAAC,CAAC;IAC7F,CAAC;IAED,6BAA6B;IAC7B,IAAI,kBAAkB,CAAC,IAAI,CAAC,IAAI,CAAC;QAAE,OAAO,YAAY,CAAC,MAAM,CAAC,IAAI,CAAC,EAAE,UAAU,EAAE,KAAK,CAAC,CAAC;IAExF,iCAAiC;IACjC,MAAM,KAAK,GAAG,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IAC/B,IAAI,KAAK,EAAE,CAAC;QACV,MAAM,KAAK,GAAG,KAAK,CAAC,CAAC,CAAC,KAAK,SAAS,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,MAAM,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC;QAC5D,MAAM,OAAO,GAAG,MAAM,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC;QACjC,MAAM,OAAO,GAAG,MAAM,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC;QACjC,IAAI,OAAO,IAAI,EAAE,IAAI,OAAO,IAAI,EAAE,EAAE,CAAC;YACnC,MAAM,IAAI,cAAc,CACtB,GAAG,UAAU,KAAK,KAAK,iDAAiD,CACzE,CAAC;QACJ,CAAC;QACD,OAAO,YAAY,CAAC,KAAK,GAAG,IAAI,GAAG,OAAO,GAAG,EAAE,GAAG,OAAO,EAAE,UAAU,EAAE,KAAK,CAAC,CAAC;IAChF,CAAC;IAED,kCAAkC;IAClC,UAAU,CAAC,SAAS,GAAG,CAAC,CAAC;IACzB,IAAI,KAAK,GAAG,CAAC,CAAC;IACd,IAAI,OAAO,GAAG,CAAC,CAAC;IAChB,IAAI,QAAQ,GAAG,CAAC,CAAC;IACjB,IAAI,KAA6B,CAAC;IAClC,qFAAqF;IACrF,MAAM,IAAI,GAAG,IAAI,GAAG,EAAU,CAAC;IAC/B,OAAO,CAAC,KAAK,GAAG,UAAU,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC,KAAK,IAAI,EAAE,CAAC;QAChD,MAAM,MAAM,GAAG,MAAM,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC;QAChC,MAAM,IAAI,GAAG,KAAK,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;QAC5B,MAAM,KAAK,GAAG,YAAY,CAAC,IAAI,CAAC,CAAC;QACjC,IAAI,KAAK,KAAK,SAAS,EAAE,CAAC;YACxB,MAAM,IAAI,cAAc,CACtB,GAAG,UAAU,KAAK,KAAK,2BAA2B,IAAI,wBAAwB,CAC/E,CAAC;QACJ,CAAC;QAED,iFAAiF;QACjF,qFAAqF;QACrF,qFAAqF;QACrF,IAAI,IAAI,CAAC,GAAG,CAAC,KAAK,CAAC,EAAE,CAAC;YACpB,MAAM,IAAI,cAAc,CACtB,GAAG,UAAU,KAAK,KAAK,+BAA+B;gBACpD,sCAAsC,CACzC,CAAC;QACJ,CAAC;QACD,IAAI,CAAC,GAAG,CAAC,KAAK,CAAC,CAAC;QAEhB,KAAK,IAAI,MAAM,GAAG,KAAK,CAAC;QACxB,OAAO,EAAE,CAAC;QACV,QAAQ,IAAI,KAAK,CAAC,CAAC,CAAC,CAAC,MAAM,CAAC;IAC9B,CAAC;IAED,8DAA8D;IAC9D,IAAI,OAAO,KAAK,CAAC,IAAI,QAAQ,KAAK,IAAI,CAAC,OAAO,CAAC,OAAO,EAAE,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC;QACnE,MAAM,IAAI,cAAc,CACtB,GAAG,UAAU,KAAK,KAAK,gCAAgC;YACrD,6DAA6D,CAChE,CAAC;IACJ,CAAC;IAED,OAAO,YAAY,CAAC,KAAK,EAAE,UAAU,EAAE,KAAK,CAAC,CAAC;AAChD,CAAC;AAED,SAAS,YAAY,CAAC,KAAa,EAAE,UAAkB,EAAE,KAAa;IACpE,IAAI,CAAC,MAAM,CAAC,QAAQ,CAAC,KAAK,CAAC,IAAI,KAAK,GAAG,CAAC,EAAE,CAAC;QACzC,MAAM,IAAI,cAAc,CAAC,GAAG,UAAU,KAAK,KAAK,qCAAqC,CAAC,CAAC;IACzF,CAAC;IACD,OAAO,KAAK,CAAC;AACf,CAAC;AAmBD;;;;;;GAMG;AACH,MAAM,CAAC,MAAM,kBAAkB,GAAG,IAAI,CAAC;AAEvC;;;;;;;;;GASG;AACH,MAAM,UAAU,YAAY,CAAC,IAAY;IACvC,IAAI,CAAC,CAAC,IAAI,GAAG,CAAC,CAAC,IAAI,CAAC,MAAM,CAAC,QAAQ,CAAC,IAAI,CAAC;QAAE,OAAO,kBAAkB,CAAC;IACrE,OAAO,IAAI,CAAC,GAAG,CAAC,kBAAkB,EAAE,CAAC,GAAG,IAAI,GAAG,CAAC,CAAC,CAAC;AACpD,CAAC;AAED,+EAA+E;AAC/E,MAAM,UAAU,mBAAmB,CACjC,IAAY,EACZ,YAAoB,EACpB,UAAkB,EAClB,YAAoB,kBAAkB;IAEtC,OAAO,IAAI,IAAI,YAAY,GAAG,SAAS,IAAI,IAAI,GAAG,UAAU,GAAG,SAAS,CAAC;AAC3E,CAAC;AAED,mFAAmF;AACnF,MAAM,UAAU,mBAAmB,CAAC,OAMnC;IACC,MAAM,KAAK,GAAG,YAAY,CAAC,OAAO,CAAC,IAAI,CAAC,CAAC;IACzC,MAAM,KAAK,GAAG,IAAI,CAAC,IAAI,CACrB,CAAC,OAAO,CAAC,YAAY,GAAG,KAAK,GAAG,OAAO,CAAC,WAAW,CAAC,GAAG,OAAO,CAAC,IAAI,CACpE,CAAC;IACF,MAAM,KAAK,GAAG,IAAI,CAAC,IAAI,CACrB,CAAC,OAAO,CAAC,UAAU,GAAG,KAAK,GAAG,OAAO,CAAC,WAAW,CAAC,GAAG,OAAO,CAAC,IAAI,CAClE,CAAC;IACF,MAAM,KAAK,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,gBAAgB,EAAE,KAAK,CAAC,CAAC,CAAC;IACrE,MAAM,IAAI,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,gBAAgB,EAAE,KAAK,CAAC,CAAC,CAAC;IACpE,OAAO,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,GAAG,KAAK,CAAC,CAAC;AACnC,CAAC;AAED;;;;;;;;GAQG;AACH,SAAS,WAAW,CAAC,CAAS,EAAE,CAAS;IACvC,OAAO,IAAI,CAAC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,IAAI,IAAI,CAAC,GAAG,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,GAAG,KAAK,CAAC;AACvE,CAAC;AAED;;;;GAIG;AACH,MAAM,UAAU,YAAY,CAAC,OAe5B;IACC,4EAA4E;IAC5E,gFAAgF;IAChF,6EAA6E;IAC7E,4EAA4E;IAC5E,2EAA2E;IAC3E,MAAM,EAAE,QAAQ,EAAE,MAAM,EAAE,GAAG,IAAI,CAAC,OAAO,CAAC,YAAY,EAAE,OAAO,CAAC,WAAW,EAAE,OAAO,CAAC,cAAc,CAAC,CAAC;IAErG,IAAI,OAAO,CAAC,QAAQ,KAAK,SAAS,IAAI,OAAO,CAAC,GAAG,KAAK,SAAS,EAAE,CAAC;QAChE,MAAM,IAAI,cAAc,CAAC,2CAA2C,CAAC,CAAC;IACxE,CAAC;IAED,MAAM,YAAY,GAAG,OAAO,CAAC,KAAK,IAAI,QAAQ,CAAC;IAE/C,IAAI,UAAkB,CAAC;IACvB,IAAI,OAAO,CAAC,QAAQ,KAAK,SAAS;QAAE,UAAU,GAAG,YAAY,GAAG,OAAO,CAAC,QAAQ,CAAC;SAC5E,IAAI,OAAO,CAAC,GAAG,KAAK,SAAS;QAAE,UAAU,GAAG,OAAO,CAAC,GAAG,CAAC;;QACxD,UAAU,GAAG,MAAM,CAAC;IAEzB;;;;;;;;;;;;;MAaE;IACF,IAAI,YAAY,IAAI,MAAM,IAAI,WAAW,CAAC,YAAY,EAAE,MAAM,CAAC,EAAE,CAAC;QAChE;;;;;;;;;UASE;QACF,MAAM,IAAI,cAAc;QACtB;;;;;;;;;;;;UAYE;QACF;;;;;;;UAOE;QACF,QAAQ,KAAK,CAAC;YACZ,CAAC,CAAC,WAAW,MAAM,CAAC,OAAO,CAAC,SAAS,EAAE,YAAY,CAAC,iCAAiC;gBACnF,GAAG,cAAc,CAAC,MAAM,CAAC,aAAa;YACxC,CAAC,CAAC,WAAW,MAAM,CAAC,OAAO,CAAC,SAAS,EAAE,YAAY,CAAC,iCAAiC;gBACnF,GAAG,cAAc,CAAC,MAAM,GAAG,QAAQ,CAAC,8BAA8B;gBAClE,GAAG,aAAa,CAAC,QAAQ,CAAC,OAAO,aAAa,CAAC,MAAM,CAAC,GAAG,CAC9D,CAAC;IACJ,CAAC;IACD,IAAI,UAAU,IAAI,YAAY,EAAE,CAAC;QAC/B,uFAAuF;QACvF,sFAAsF;QACtF,qFAAqF;QACrF,MAAM,QAAQ,GACZ,OAAO,CAAC,GAAG,KAAK,SAAS,IAAI,OAAO,CAAC,OAAO,KAAK,SAAS;YACxD,CAAC,CAAC,MAAM,CAAC,OAAO,CAAC,OAAO,EAAE,UAAU,CAAC;YACrC,CAAC,CAAC,aAAa,CAAC,UAAU,CAAC,CAAC;QAChC,MAAM,UAAU,GAAG,MAAM,CAAC,OAAO,CAAC,SAAS,EAAE,YAAY,CAAC,CAAC;QAC3D,MAAM,IAAI,cAAc,CACtB,gCAAgC,QAAQ,qCAAqC,UAAU,GAAG,CAC3F,CAAC;IACJ,CAAC;IAED,MAAM,UAAU,GAAG,IAAI,CAAC,GAAG,CAAC,UAAU,EAAE,MAAM,CAAC,CAAC;IAChD,MAAM,EAAE,WAAW,EAAE,SAAS,EAAE,GAAG,aAAa,CAAC,OAAO,EAAE,YAAY,EAAE,UAAU,CAAC,CAAC;IAEpF,OAAO;QACL,YAAY;QACZ,UAAU,EAAE,UAAU;QACtB,WAAW;QACX,SAAS;QACT;;;;;;;;UAQE;QACF,gBAAgB,EACd,CAAC,YAAY,IAAI,QAAQ,IAAI,WAAW,CAAC,YAAY,EAAE,QAAQ,CAAC,CAAC;YACjE,CAAC,UAAU,IAAI,MAAM,IAAI,WAAW,CAAC,UAAU,EAAE,MAAM,CAAC,CAAC;QAC3D,qBAAqB,EAAE,QAAQ;QAC/B,mBAAmB,EAAE,MAAM;KAC5B,CAAC;AACJ,CAAC;AAED,SAAS,IAAI,CACX,YAA6C,EAC7C,WAAmB,EACnB,cAAsB;IAEtB,IAAI,CAAC,YAAY,IAAI,YAAY,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;QAC/C,OAAO,EAAE,QAAQ,EAAE,CAAC,EAAE,MAAM,EAAE,WAAW,GAAG,cAAc,EAAE,CAAC;IAC/D,CAAC;IACD,IAAI,QAAQ,GAAG,QAAQ,CAAC;IACxB,IAAI,MAAM,GAAG,CAAC,QAAQ,CAAC;IACvB,KAAK,MAAM,KAAK,IAAI,YAAY,EAAE,CAAC;QACjC,IAAI,KAAK,GAAG,QAAQ;YAAE,QAAQ,GAAG,KAAK,CAAC;QACvC,IAAI,KAAK,GAAG,cAAc,GAAG,MAAM;YAAE,MAAM,GAAG,KAAK,GAAG,cAAc,CAAC;IACvE,CAAC;IACD,IAAI,CAAC,MAAM,CAAC,QAAQ,CAAC,QAAQ,CAAC,IAAI,CAAC,MAAM,CAAC,QAAQ,CAAC,MAAM,CAAC,EAAE,CAAC;QAC3D,OAAO,EAAE,QAAQ,EAAE,CAAC,EAAE,MAAM,EAAE,WAAW,GAAG,cAAc,EAAE,CAAC;IAC/D,CAAC;IACD,OAAO,EAAE,QAAQ,EAAE,MAAM,EAAE,CAAC;AAC9B,CAAC;AAED,sEAAsE;AACtE,SAAS,aAAa,CACpB,OAAwG,EACxG,YAAoB,EACpB,UAAkB;IAElB,MAAM,MAAM,GAAG,OAAO,CAAC,YAAY,CAAC;IACpC,IAAI,CAAC,MAAM,IAAI,MAAM,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;QACnC,OAAO;YACL,WAAW,EAAE,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,CAAC,KAAK,CAAC,YAAY,GAAG,OAAO,CAAC,cAAc,CAAC,CAAC;YAC3E,SAAS,EAAE,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,WAAW,EAAE,IAAI,CAAC,IAAI,CAAC,UAAU,GAAG,OAAO,CAAC,cAAc,CAAC,CAAC;SACzF,CAAC;IACJ,CAAC;IAED,IAAI,KAAK,GAAG,OAAO,CAAC,WAAW,CAAC;IAChC,IAAI,IAAI,GAAG,CAAC,CAAC;IACb,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,MAAM,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACvC,MAAM,KAAK,GAAG,MAAM,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC;QAC7B,IAAI,KAAK,GAAG,OAAO,CAAC,cAAc,GAAG,YAAY,IAAI,KAAK,GAAG,UAAU,EAAE,CAAC;YACxE,IAAI,CAAC,GAAG,KAAK;gBAAE,KAAK,GAAG,CAAC,CAAC;YACzB,IAAI,CAAC,GAAG,CAAC,GAAG,IAAI;gBAAE,IAAI,GAAG,CAAC,GAAG,CAAC,CAAC;QACjC,CAAC;IACH,CAAC;IACD,OAAO,KAAK,GAAG,IAAI,CAAC,CAAC,CAAC,EAAE,WAAW,EAAE,KAAK,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAC,CAAC,EAAE,WAAW,EAAE,CAAC,EAAE,SAAS,EAAE,CAAC,EAAE,CAAC;AACnG,CAAC;AAED;;;;GAIG;AACH,SAAS,MAAM,CAAC,IAAwB,EAAE,OAAe;IACvD,OAAO,IAAI,KAAK,SAAS,CAAC,CAAC,CAAC,aAAa,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,IAAI,IAAI,GAAG,CAAC;AACnE,CAAC;AAED,SAAS,aAAa,CAAC,OAAe;IACpC,OAAO,GAAG,MAAM,CAAC,OAAO,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC;AAC1C,CAAC","sourcesContent":["/**\n * Parsing for the time-range options.\n *\n * Researchers write offsets in whatever form is natural for the recording in front\n * of them: seconds for a short ECG strip, `30m` into a sleep study, `01:23:45` when\n * reading off a clock. All three are accepted; anything ambiguous is rejected with\n * a message that shows the forms that work.\n */\n\nimport { formatDuration } from '../format/number.js';\n\nexport class TimeRangeError extends Error {\n  constructor(message: string) {\n    super(message);\n    this.name = 'TimeRangeError';\n  }\n}\n\nconst UNIT_SECONDS: Record<string, number> = {\n  h: 3600,\n  hr: 3600,\n  hrs: 3600,\n  hour: 3600,\n  hours: 3600,\n  m: 60,\n  min: 60,\n  mins: 60,\n  minute: 60,\n  minutes: 60,\n  s: 1,\n  sec: 1,\n  secs: 1,\n  second: 1,\n  seconds: 1,\n  ms: 0.001,\n};\n\nconst CLOCK = /^(?:(\\d+):)?(\\d{1,2}):(\\d{1,2}(?:\\.\\d+)?)$/u;\nconst UNIT_TOKEN = /(\\d+(?:\\.\\d+)?)\\s*([a-z]+)/giu;\n\n/**\n * Parse a duration or offset into seconds.\n *\n * Accepted: `90`, `90s`, `5m`, `1h30m`, `1h 30m 15s`, `00:30:00`, `30:00`, `250ms`.\n */\nexport function parseTimeSpec(input: string, optionName: string): number {\n  const text = input.trim().toLowerCase();\n  if (text === '') {\n    throw new TimeRangeError(`${optionName} is empty. Try a value like 30s, 5m, or 00:30:00.`);\n  }\n\n  // Bare number means seconds.\n  if (/^\\d+(?:\\.\\d+)?$/u.test(text)) return assertFinite(Number(text), optionName, input);\n\n  // Clock form: hh:mm:ss or mm:ss.\n  const clock = CLOCK.exec(text);\n  if (clock) {\n    const hours = clock[1] === undefined ? 0 : Number(clock[1]);\n    const minutes = Number(clock[2]);\n    const seconds = Number(clock[3]);\n    if (minutes >= 60 || seconds >= 60) {\n      throw new TimeRangeError(\n        `${optionName} \"${input}\" has a minutes or seconds field of 60 or more.`,\n      );\n    }\n    return assertFinite(hours * 3600 + minutes * 60 + seconds, optionName, input);\n  }\n\n  // Unit form: 1h30m, 5 min, 250ms.\n  UNIT_TOKEN.lastIndex = 0;\n  let total = 0;\n  let matched = 0;\n  let consumed = 0;\n  let match: RegExpExecArray | null;\n  // Which units have already been seen, so a repeat can be rejected rather than added.\n  const seen = new Set<number>();\n  while ((match = UNIT_TOKEN.exec(text)) !== null) {\n    const amount = Number(match[1]);\n    const unit = match[2] ?? '';\n    const scale = UNIT_SECONDS[unit];\n    if (scale === undefined) {\n      throw new TimeRangeError(\n        `${optionName} \"${input}\" uses an unknown unit \"${unit}\". Use h, m, s, or ms.`,\n      );\n    }\n\n    // \"1h1h\" is a typo, not a request for two hours. Summing repeated units silently\n    // turned a slip into a plausible window that was quietly the wrong length. Units are\n    // keyed by their scale so the aliases collapse together: \"1h30min20m\" is caught too.\n    if (seen.has(scale)) {\n      throw new TimeRangeError(\n        `${optionName} \"${input}\" gives the same unit twice. ` +\n          `Combine each unit once, as in 1h30m.`,\n      );\n    }\n    seen.add(scale);\n\n    total += amount * scale;\n    matched++;\n    consumed += match[0].length;\n  }\n\n  // Reject partially-understood input like \"5x\" or \"1h banana\".\n  if (matched === 0 || consumed !== text.replace(/\\s+/gu, '').length) {\n    throw new TimeRangeError(\n      `${optionName} \"${input}\" is not a time I understand. ` +\n        `Try 30s, 5m, 1h30m, 00:30:00, or a plain number of seconds.`,\n    );\n  }\n\n  return assertFinite(total, optionName, input);\n}\n\nfunction assertFinite(value: number, optionName: string, input: string): number {\n  if (!Number.isFinite(value) || value < 0) {\n    throw new TimeRangeError(`${optionName} \"${input}\" is not a valid non-negative time.`);\n  }\n  return value;\n}\n\nexport interface ResolvedRange {\n  /** Inclusive start, in seconds from the beginning of the recording. */\n  startSeconds: number;\n  /** Exclusive end, in seconds. */\n  endSeconds: number;\n  /** First data record touching the window. */\n  startRecord: number;\n  /** One past the last data record touching the window. */\n  endRecord: number;\n  /** True when the window covers the whole recording. */\n  isWholeRecording: boolean;\n  /** Earliest record start, including EDF+D timing gaps. */\n  recordingStartSeconds: number;\n  /** End of the latest record, including EDF+D timing gaps. */\n  recordingEndSeconds: number;\n}\n\n/**\n * Slack for comparisons at sample/window boundaries.\n *\n * A nanosecond is far below any real sampling interval — 20 kHz is 50 microseconds — so it\n * absorbs the arithmetic error in `recordStart + sample / rate` without reaching a\n * neighbouring sample.\n */\nexport const BOUNDARY_TOLERANCE = 1e-9;\n\n/**\n * The slack to use for a channel sampled this often.\n *\n * Never as much as half a sample interval, because slack that reaches the next sample stops\n * being slack. A fixed nanosecond was applied whatever the rate, and the format does not\n * oblige the interval to be larger than it: EDF's record duration is an 8-character field\n * that accepts `1e-9`. A recording of two 1 ns records holding ten samples each wrote ten of\n * its twenty rows — the window ends at 2e-9, the comparison asked for `time < 2e-9 - 1e-9`,\n * and the entire second record failed it. Exit 0, no warning, half the samples gone.\n */\nexport function toleranceFor(rate: number): number {\n  if (!(rate > 0) || !Number.isFinite(rate)) return BOUNDARY_TOLERANCE;\n  return Math.min(BOUNDARY_TOLERANCE, 1 / rate / 2);\n}\n\n/** Match the exact half-open boundary rules used while writing signal rows. */\nexport function sampleTimeIsInRange(\n  time: number,\n  startSeconds: number,\n  endSeconds: number,\n  tolerance: number = BOUNDARY_TOLERANCE,\n): boolean {\n  return time >= startSeconds - tolerance && time < endSeconds - tolerance;\n}\n\n/** Count samples from one record that fall inside a half-open requested window. */\nexport function countSamplesInRange(options: {\n  recordStart: number;\n  rate: number;\n  samplesPerRecord: number;\n  startSeconds: number;\n  endSeconds: number;\n}): number {\n  const slack = toleranceFor(options.rate);\n  const lower = Math.ceil(\n    (options.startSeconds - slack - options.recordStart) * options.rate,\n  );\n  const upper = Math.ceil(\n    (options.endSeconds - slack - options.recordStart) * options.rate,\n  );\n  const first = Math.max(0, Math.min(options.samplesPerRecord, lower));\n  const last = Math.max(0, Math.min(options.samplesPerRecord, upper));\n  return Math.max(0, last - first);\n}\n\n/**\n * Whether two instants differ only by the arithmetic that produced them.\n *\n * A relative epsilon, because the gap between doubles grows with magnitude — the same shape\n * the long layout uses to decide two sample times are one instant. Well below any real sample\n * interval, and well above the rounding that two routes to one quantity produce: a recording's\n * length is `recordCount * recordDuration`, which for 6003 records of 0.1s is not the 600.3 it\n * prints as.\n */\nfunction sameInstant(a: number, b: number): boolean {\n  return Math.abs(a - b) <= Math.max(Math.abs(a), Math.abs(b)) * 1e-12;\n}\n\n/**\n * Turn a requested window into both an exact time span and the record range that\n * contains it. Records are the unit the file can be read in; the exact span is what\n * decides which samples inside those records are actually written.\n */\nexport function resolveRange(options: {\n  start?: number | undefined;\n  /** The `--start` value exactly as typed, quoted back in the past-the-end error. */\n  startText?: string | undefined;\n  duration?: number | undefined;\n  end?: number | undefined;\n  /** The `--end` value exactly as typed, quoted back in the window error. */\n  endText?: string | undefined;\n  recordDuration: number;\n  recordCount: number;\n  /**\n   * True start time of each data record, for discontinuous files. When absent,\n   * records are assumed to sit end to end.\n   */\n  recordStarts?: Float64Array | null | undefined;\n}): ResolvedRange {\n  // For a continuous file the recording spans recordCount * recordDuration. A\n  // discontinuous one does not: a 10-second recording with a 95-second gap in the\n  // middle still ends at 105 seconds. Deriving the span from the records' real\n  // positions is what stops a requested window from being clipped back to the\n  // amount of *data* in the file and silently discarding everything past it.\n  const { earliest, latest } = span(options.recordStarts, options.recordCount, options.recordDuration);\n\n  if (options.duration !== undefined && options.end !== undefined) {\n    throw new TimeRangeError('Use either --duration or --end, not both.');\n  }\n\n  const startSeconds = options.start ?? earliest;\n\n  let endSeconds: number;\n  if (options.duration !== undefined) endSeconds = startSeconds + options.duration;\n  else if (options.end !== undefined) endSeconds = options.end;\n  else endSeconds = latest;\n\n  /*\n    At the end, allowing for the arithmetic that produced the end.\n\n    `latest` is `recordCount * recordDuration`, and with a fractional duration that is not the\n    number it prints as: 6003 records of 0.1s is 600.3000000000001, not 600.3. So `--start\n    600.3` on a recording `--info` calls \"10m 0.3s\" was accepted by a hair, converted nothing,\n    and exited 0 with a signals.csv holding its header — which is the empty conversion this\n    error exists to prevent, and which the same command on a whole-second recording is refused\n    for.\n\n    A relative epsilon, the same shape the long layout uses to decide two sample times are one\n    instant: well below any real interval, and well above the rounding that two different\n    routes to the same quantity produce.\n  */\n  if (startSeconds >= latest || sameInstant(startSeconds, latest)) {\n    /*\n      Quote what was typed. Reporting the parsed seconds meant `--start 4h` came back as\n      \"--start 14400s is at or past the end\", which reads as a value the user never gave.\n\n      In quotation marks, which the parse errors above have always used and this did not.\n      Without them the value ran into the sentence: `--start \"  5s  \"` printed as\n      `--start   5s   is at or past the end`, where the value appears to be `5s   is` and\n      the surrounding spaces — the actual reason a shell-built argument went wrong — are\n      invisible.\n    */\n    throw new TimeRangeError(\n      /*\n        The recording's length in the same words --info uses for it.\n\n        `formatSeconds` renders a bare number of seconds, so this message and the Duration\n        line disagreed about one file in one session: --info said \"6m 40s\" and the error said\n        \"400s\". On an overnight recording it read \"7950s recording\", leaving the reader to\n        divide by 3600 to find out whether their --start was reasonable — which is the one\n        question this message exists to answer. cli-reference.md has always documented it\n        humanised (\"2h 12m 30s\"), a form no input could produce.\n\n        The typed value keeps `quoted`, since that is the user's own text and should come\n        back exactly as they wrote it.\n      */\n      /*\n        `latest` is where the recording ends on its own clock, which is its length only when\n        it starts at zero. A file timed from its first record's timekeeping TAL need not: one\n        whose records run 1000s to 1003s is three seconds long, and this called it \"this\n        16m 43s recording\" — while --info two lines away said \"Duration 3s\". Where the\n        recording sits is the useful thing to say in that case, and it is the number --start\n        has to be given.\n      */\n      earliest === 0\n        ? `--start ${quoted(options.startText, startSeconds)} is at or past the end of this ` +\n          `${formatDuration(latest)} recording.`\n        : `--start ${quoted(options.startText, startSeconds)} is at or past the end of this ` +\n          `${formatDuration(latest - earliest)} recording, which runs from ` +\n          `${formatSeconds(earliest)} to ${formatSeconds(latest)}.`,\n    );\n  }\n  if (endSeconds <= startSeconds) {\n    // Quote whatever the caller actually gave, for the same reason as the error above. The\n    // end is only echoed when --end was passed: with --duration the end is computed here,\n    // so there is no typed value to quote and the arithmetic result is the honest thing.\n    const endShown =\n      options.end !== undefined && options.endText !== undefined\n        ? quoted(options.endText, endSeconds)\n        : formatSeconds(endSeconds);\n    const startShown = quoted(options.startText, startSeconds);\n    throw new TimeRangeError(\n      `The requested window ends at ${endShown}, which is not after its start at ${startShown}.`,\n    );\n  }\n\n  const clampedEnd = Math.min(endSeconds, latest);\n  const { startRecord, endRecord } = selectRecords(options, startSeconds, clampedEnd);\n\n  return {\n    startSeconds,\n    endSeconds: clampedEnd,\n    startRecord,\n    endRecord,\n    /*\n      The same rounding, one field over.\n\n      `latest` is `recordCount * recordDuration`, and 6003 records of 0.1s is\n      600.3000000000001. So `--end 600.3` on a recording of exactly that length wrote every\n      sample it has — byte-identical to a bare conversion — and metadata.json recorded\n      `whole_recording: false` for it, while the bare run recorded true. One conversion, two\n      answers, on the field a pipeline reads to decide whether it has the lot.\n    */\n    isWholeRecording:\n      (startSeconds <= earliest || sameInstant(startSeconds, earliest)) &&\n      (clampedEnd >= latest || sameInstant(clampedEnd, latest)),\n    recordingStartSeconds: earliest,\n    recordingEndSeconds: latest,\n  };\n}\n\nfunction span(\n  recordStarts: Float64Array | null | undefined,\n  recordCount: number,\n  recordDuration: number,\n): { earliest: number; latest: number } {\n  if (!recordStarts || recordStarts.length === 0) {\n    return { earliest: 0, latest: recordCount * recordDuration };\n  }\n  let earliest = Infinity;\n  let latest = -Infinity;\n  for (const start of recordStarts) {\n    if (start < earliest) earliest = start;\n    if (start + recordDuration > latest) latest = start + recordDuration;\n  }\n  if (!Number.isFinite(earliest) || !Number.isFinite(latest)) {\n    return { earliest: 0, latest: recordCount * recordDuration };\n  }\n  return { earliest, latest };\n}\n\n/** Every record whose own time span overlaps the requested window. */\nfunction selectRecords(\n  options: { recordDuration: number; recordCount: number; recordStarts?: Float64Array | null | undefined },\n  startSeconds: number,\n  endSeconds: number,\n): { startRecord: number; endRecord: number } {\n  const starts = options.recordStarts;\n  if (!starts || starts.length === 0) {\n    return {\n      startRecord: Math.max(0, Math.floor(startSeconds / options.recordDuration)),\n      endRecord: Math.min(options.recordCount, Math.ceil(endSeconds / options.recordDuration)),\n    };\n  }\n\n  let first = options.recordCount;\n  let last = 0;\n  for (let i = 0; i < starts.length; i++) {\n    const begin = starts[i] ?? 0;\n    if (begin + options.recordDuration > startSeconds && begin < endSeconds) {\n      if (i < first) first = i;\n      if (i + 1 > last) last = i + 1;\n    }\n  }\n  return first < last ? { startRecord: first, endRecord: last } : { startRecord: 0, endRecord: 0 };\n}\n\n/**\n * The value as the caller typed it, in quotation marks, or the parsed seconds if they gave\n * none. The marks show where the value begins and ends, which matters most for the values\n * that went wrong because of what surrounds them.\n */\nfunction quoted(text: string | undefined, seconds: number): string {\n  return text === undefined ? formatSeconds(seconds) : `\"${text}\"`;\n}\n\nfunction formatSeconds(seconds: number): string {\n  return `${Number(seconds.toFixed(3))}s`;\n}\n"]}

package/dist/edf/annotations.js

@@ -39,6 +39,23 @@ export function decodeRecordAnnotations(bytes, recordIndex, carriesTimekeeping =
             continue;
         if (i > start) {
             const chunk = bytes.subarray(start, i);
+            /*
+              Padding is not a lost annotation.
+      
+              The spec pads the slot with NUL, which the loop above already skips because it is what
+              separates one TAL from the next. Writers pad with spaces instead, and a run of spaces
+              after the last TAL is a non-empty chunk — so a file holding one perfectly readable
+              event, exported in full, was told "2 annotation entries were unreadable and could not
+              be exported", one per record. Nothing was lost. Under --strict that is a failed run
+              over the whitespace at the end of a slot.
+      
+              Only whitespace. A chunk of anything else that does not parse is a real loss and is
+              still counted, which is the case this warning exists for.
+            */
+            if (chunk.every(isPaddingByte)) {
+                start = i + 1;
+                continue;
+            }
             const parsed = parseTal(chunk, recordIndex);
             // The timekeeping TAL is the one in first POSITION, whether or not it decodes.
             // Clearing this flag only on a successful parse meant that an unreadable first TAL
@@ -94,6 +111,10 @@ export function decodeRecordAnnotations(bytes, recordIndex, carriesTimekeeping =
         negativeDurations,
     };
 }
+/** Space, tab, CR, LF or NUL — what a writer fills the rest of the slot with. */
+function isPaddingByte(byte) {
+    return byte === 0x20 || byte === 0x09 || byte === 0x0d || byte === 0x0a || byte === 0x00;
+}
 function parseTal(chunk, recordIndex) {
     // The onset must be explicitly signed; anything else is not a TAL.
     const first = chunk[0];

package/dist/edf/annotations.js.map

@@ -1 +1 @@
-{"version":3,"file":"annotations.js","sourceRoot":"","sources":["../../src/edf/annotations.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;GAaG;AAEH,OAAO,EAAE,UAAU,EAAE,MAAM,YAAY,CAAC;AAExC,MAAM,QAAQ,GAAG,IAAI,CAAC,CAAC,yDAAyD;AAChF,MAAM,YAAY,GAAG,IAAI,CAAC,CAAC,gCAAgC;AAC3D,MAAM,OAAO,GAAG,IAAI,CAAC;AAErB,MAAM,aAAa,GAAG,MAAM,CAAC,YAAY,CAAC,QAAQ,CAAC,CAAC;AACpD,MAAM,iBAAiB,GAAG,MAAM,CAAC,YAAY,CAAC,YAAY,CAAC,CAAC;AAuD5D;;;;;;GAMG;AACH,MAAM,UAAU,uBAAuB,CACrC,KAAiB,EACjB,WAAmB,EACnB,kBAAkB,GAAG,IAAI;IAEzB,MAAM,WAAW,GAAiB,EAAE,CAAC;IACrC,IAAI,WAAW,GAAkB,IAAI,CAAC;IACtC,IAAI,UAAU,GAAG,IAAI,CAAC;IACtB,IAAI,SAAS,GAAG,CAAC,CAAC;IAClB,IAAI,oBAAoB,GAAG,CAAC,CAAC;IAC7B,IAAI,mBAAmB,GAAG,CAAC,CAAC;IAC5B,IAAI,iBAAiB,GAAG,CAAC,CAAC;IAE1B,IAAI,KAAK,GAAG,CAAC,CAAC;IACd,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,IAAI,KAAK,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACvC,IAAI,CAAC,KAAK,KAAK,CAAC,MAAM,IAAI,KAAK,CAAC,CAAC,CAAC,KAAK,OAAO;YAAE,SAAS;QAEzD,IAAI,CAAC,GAAG,KAAK,EAAE,CAAC;YACd,MAAM,KAAK,GAAG,KAAK,CAAC,QAAQ,CAAC,KAAK,EAAE,CAAC,CAAC,CAAC;YACvC,MAAM,MAAM,GAAG,QAAQ,CAAC,KAAK,EAAE,WAAW,CAAC,CAAC;YAE5C,+EAA+E;YAC/E,mFAAmF;YACnF,+EAA+E;YAC/E,iFAAiF;YACjF,+EAA+E;YAC/E,uEAAuE;YACvE;;;;;;;;;cASE;YACF,MAAM,aAAa,GAAG,UAAU,IAAI,kBAAkB,CAAC;YACvD,UAAU,GAAG,KAAK,CAAC;YAEnB,IAAI,MAAM,EAAE,CAAC;gBACX,IAAI,aAAa;oBAAE,WAAW,GAAG,MAAM,CAAC,KAAK,CAAC;gBAC9C,KAAK,MAAM,UAAU,IAAI,MAAM,CAAC,WAAW;oBAAE,WAAW,CAAC,IAAI,CAAC,UAAU,CAAC,CAAC;gBAC1E,mBAAmB,IAAI,MAAM,CAAC,mBAAmB,CAAC;gBAClD,iBAAiB,IAAI,MAAM,CAAC,iBAAiB,CAAC;YAChD,CAAC;iBAAM,CAAC;gBACN;;;;;;;;;kBASE;gBACF,IAAI,aAAa;oBAAE,oBAAoB,EAAE,CAAC;;oBACrC,SAAS,EAAE,CAAC;YACnB,CAAC;QACH,CAAC;QACD,KAAK,GAAG,CAAC,GAAG,CAAC,CAAC;IAChB,CAAC;IAED,OAAO;QACL,WAAW;QACX,WAAW;QACX,SAAS;QACT,oBAAoB;QACpB,mBAAmB;QACnB,iBAAiB;KAClB,CAAC;AACJ,CAAC;AAWD,SAAS,QAAQ,CAAC,KAAiB,EAAE,WAAmB;IACtD,mEAAmE;IACnE,MAAM,KAAK,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC;IACvB,IAAI,KAAK,KAAK,IAAI,CAAC,OAAO,IAAI,KAAK,KAAK,IAAI,CAAC,OAAO;QAAE,OAAO,IAAI,CAAC;IAElE,MAAM,IAAI,GAAG,UAAU,CAAC,KAAK,CAAC,CAAC;IAC/B,MAAM,KAAK,GAAG,IAAI,CAAC,KAAK,CAAC,aAAa,CAAC,CAAC;IACxC,MAAM,IAAI,GAAG,KAAK,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;IAE5B,IAAI,SAAS,GAAG,IAAI,CAAC;IACrB,IAAI,YAAY,GAAkB,IAAI,CAAC;IACvC,MAAM,WAAW,GAAG,IAAI,CAAC,OAAO,CAAC,iBAAiB,CAAC,CAAC;IACpD,IAAI,WAAW,IAAI,CAAC,EAAE,CAAC;QACrB,SAAS,GAAG,IAAI,CAAC,KAAK,CAAC,CAAC,EAAE,WAAW,CAAC,CAAC;QACvC,YAAY,GAAG,IAAI,CAAC,KAAK,CAAC,WAAW,GAAG,CAAC,CAAC,CAAC;IAC7C,CAAC;IAED,MAAM,KAAK,GAAG,MAAM,CAAC,SAAS,CAAC,CAAC;IAChC,IAAI,CAAC,MAAM,CAAC,QAAQ,CAAC,KAAK,CAAC;QAAE,OAAO,IAAI,CAAC;IAEzC;;;;;;;;;;MAUE;IACF,IAAI,QAAQ,GAAkB,IAAI,CAAC;IACnC,IAAI,kBAAkB,GAAG,KAAK,CAAC;IAC/B,IAAI,YAAY,KAAK,IAAI,IAAI,YAAY,KAAK,EAAE,EAAE,CAAC;QACjD,MAAM,CAAC,GAAG,MAAM,CAAC,YAAY,CAAC,CAAC;QAC/B,IAAI,MAAM,CAAC,QAAQ,CAAC,CAAC,CAAC;YAAE,QAAQ,GAAG,CAAC,CAAC;;YAChC,kBAAkB,GAAG,IAAI,CAAC;IACjC,CAAC;IAED;;;;;;;;;MASE;IACF,MAAM,gBAAgB,GAAG,QAAQ,KAAK,IAAI,IAAI,QAAQ,GAAG,CAAC,CAAC;IAE3D,MAAM,WAAW,GAAiB,EAAE,CAAC;IACrC,KAAK,MAAM,GAAG,IAAI,KAAK,CAAC,KAAK,CAAC,CAAC,CAAC,EAAE,CAAC;QACjC,gFAAgF;QAChF,IAAI,GAAG,KAAK,EAAE;YAAE,SAAS;QACzB,WAAW,CAAC,IAAI,CAAC;YACf,KAAK;YACL,QAAQ;YACR,IAAI,EAAE,GAAG;YACT,WAAW;YACX,GAAG,CAAC,kBAAkB,CAAC,CAAC,CAAC,EAAE,kBAAkB,EAAE,IAAI,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;SAC5D,CAAC,CAAC;IACL,CAAC;IAED,yFAAyF;IACzF,+CAA+C;IAC/C,OAAO;QACL,KAAK;QACL,WAAW;QACX,mBAAmB,EAAE,kBAAkB,CAAC,CAAC,CAAC,WAAW,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC;QAChE,iBAAiB,EAAE,gBAAgB,CAAC,CAAC,CAAC,WAAW,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC;KAC7D,CAAC;AACJ,CAAC;AAED,OAAO,EAAE,QAAQ,EAAE,YAAY,EAAE,OAAO,EAAE,CAAC","sourcesContent":["/**\n * EDF+ annotation (TAL) decoding.\n *\n * The annotations channel stores UTF-8 text in place of samples. Its bytes are a\n * run of Time-stamped Annotation Lists, each terminated by a NUL, with the rest\n * of the channel NUL-padded:\n *\n *   +<onset>[<0x15><duration>]<0x14><text><0x14>...<0x00>\n *\n * The first TAL of every data record must carry that record's start time and no\n * text; that is how an EDF+D file states where each record actually sits in time.\n *\n *   +1.25<0x15>0.5<0x14>Seizure onset<0x14><0x00>\n */\n\nimport { decodeUtf8 } from './bytes.js';\n\nconst SEP_TEXT = 0x14; // separates onset/duration from text, and text from text\nconst SEP_DURATION = 0x15; // separates onset from duration\nconst TAL_END = 0x00;\n\nconst TEXT_SEP_CHAR = String.fromCharCode(SEP_TEXT);\nconst DURATION_SEP_CHAR = String.fromCharCode(SEP_DURATION);\n\nexport interface Annotation {\n  /** Seconds from the start of the recording. */\n  onset: number;\n  /**\n   * Seconds, or null when the TAL stated no duration that could be read.\n   *\n   * Null covers two cases the file distinguishes and this field does not: a TAL that omitted\n   * the duration, and a TAL that stated one which is not a number. They are told apart by\n   * `unreadableDurations`, which is what raises the warning; the value itself has nowhere\n   * honest to put \"the file said `abc`\".\n   */\n  duration: number | null;\n  text: string;\n  /** Index of the data record this annotation was stored in. */\n  recordIndex: number;\n  /**\n   * True when the file stated a duration that could not be read.\n   *\n   * `duration` is null either way, which is the ambiguity the counts beside it exist to\n   * flag — and those counts were of the whole file while `annotations.csv` is filtered to\n   * the requested window. A conversion of one second of a recording warned that \"1\n   * annotation states a duration that is not a number, so its duration_s cell is empty\"\n   * about an event two seconds outside it, and failed `--strict` for it. Carrying the fact\n   * on the event lets the count be taken where the window has already been applied.\n   */\n  durationUnreadable?: boolean;\n}\n\nexport interface DecodedRecordAnnotations {\n  /** Record start time in seconds, from the leading timekeeping TAL. */\n  recordStart: number | null;\n  annotations: Annotation[];\n  /** Non-empty chunks that were not valid TALs, so the caller can report them. */\n  malformed: number;\n  /** Unreadable TALs in first position, which carry a record's start time, not an event. */\n  malformedTimekeeping: number;\n  /**\n   * Events kept whose stated duration could not be read.\n   *\n   * Counted apart again, for the same reason the two above are: the entry was exported and\n   * nothing about it is missing except the one field, so calling it an entry that \"could not\n   * be exported\" describes a loss that did not happen and hides the one that did.\n   */\n  unreadableDurations: number;\n  /**\n   * Events kept whose stated duration is a readable number below zero.\n   *\n   * Separate from the count above because the value survives: it is written to the CSV as\n   * the file gave it, and what is wrong with it is arithmetic rather than parsing.\n   */\n  negativeDurations: number;\n}\n\n/**\n * Decode one data record's annotation bytes.\n *\n * Malformed TALs are skipped rather than thrown, because a single bad annotation\n * should not cost the user an entire conversion. The count of skipped chunks is\n * returned so the caller can tell the user rather than losing them in silence.\n */\nexport function decodeRecordAnnotations(\n  bytes: Uint8Array,\n  recordIndex: number,\n  carriesTimekeeping = true,\n): DecodedRecordAnnotations {\n  const annotations: Annotation[] = [];\n  let recordStart: number | null = null;\n  let isFirstTal = true;\n  let malformed = 0;\n  let malformedTimekeeping = 0;\n  let unreadableDurations = 0;\n  let negativeDurations = 0;\n\n  let start = 0;\n  for (let i = 0; i <= bytes.length; i++) {\n    if (i !== bytes.length && bytes[i] !== TAL_END) continue;\n\n    if (i > start) {\n      const chunk = bytes.subarray(start, i);\n      const parsed = parseTal(chunk, recordIndex);\n\n      // The timekeeping TAL is the one in first POSITION, whether or not it decodes.\n      // Clearing this flag only on a successful parse meant that an unreadable first TAL\n      // promoted the next ordinary annotation to timekeeping, and its onset silently\n      // became the record's start time — shifting every sample in that record. Leaving\n      // recordStart null instead is what the caller already handles, with a fallback\n      // timestamp and an ANNOTATION_DECODE_FAILED warning naming the record.\n      /*\n        Only one annotation channel carries a record's start time.\n\n        This flagged the first TAL of *every* annotation channel as timekeeping. In a second\n        channel the first TAL is an ordinary event — so when one failed to parse, the event\n        was dropped and counted as a lost timekeeping entry, which produced the warning\n        \"3 data records carry a timekeeping annotation that could not be read\" followed by\n        \"No event was lost\". Three events had been lost, and the timekeeping in that file was\n        perfectly readable. Both sentences false, about the same three records.\n      */\n      const isTimekeeping = isFirstTal && carriesTimekeeping;\n      isFirstTal = false;\n\n      if (parsed) {\n        if (isTimekeeping) recordStart = parsed.onset;\n        for (const annotation of parsed.annotations) annotations.push(annotation);\n        unreadableDurations += parsed.unreadableDurations;\n        negativeDurations += parsed.negativeDurations;\n      } else {\n        /*\n          Counted apart from the events, because losing one is a different loss.\n\n          A timekeeping TAL is never exported — it says where the record sits, not what\n          happened — so counting it among the entries that \"could not be exported\" both\n          overstated what was lost from annotations.csv and said nothing about the thing that\n          actually went missing, which is a record's position in time. A file with one\n          unreadable timekeeping TAL and three perfectly good events reported \"1 annotation\n          entry was unreadable and could not be exported\" while exporting all three.\n        */\n        if (isTimekeeping) malformedTimekeeping++;\n        else malformed++;\n      }\n    }\n    start = i + 1;\n  }\n\n  return {\n    recordStart,\n    annotations,\n    malformed,\n    malformedTimekeeping,\n    unreadableDurations,\n    negativeDurations,\n  };\n}\n\ninterface ParsedTal {\n  onset: number;\n  annotations: Annotation[];\n  /** How many of those annotations carry a duration the file stated and this could not read. */\n  unreadableDurations: number;\n  /** How many carry a duration that read as a number below zero. */\n  negativeDurations: number;\n}\n\nfunction parseTal(chunk: Uint8Array, recordIndex: number): ParsedTal | null {\n  // The onset must be explicitly signed; anything else is not a TAL.\n  const first = chunk[0];\n  if (first !== 0x2b /* + */ && first !== 0x2d /* - */) return null;\n\n  const text = decodeUtf8(chunk);\n  const parts = text.split(TEXT_SEP_CHAR);\n  const head = parts[0] ?? '';\n\n  let onsetText = head;\n  let durationText: string | null = null;\n  const durationSep = head.indexOf(DURATION_SEP_CHAR);\n  if (durationSep >= 0) {\n    onsetText = head.slice(0, durationSep);\n    durationText = head.slice(durationSep + 1);\n  }\n\n  const onset = Number(onsetText);\n  if (!Number.isFinite(onset)) return null;\n\n  /*\n    A duration the file stated and this could not read is not the same as no duration.\n\n    Both came out as `null` and so as an empty `duration_s` cell, which the documentation\n    defines as meaning the file gave no duration — so an event whose duration was written as\n    `abc` was exported as an event with no duration, indistinguishable from one beside it\n    that genuinely had none, and nothing anywhere said a field had been dropped. The onset is\n    already held to this standard: one that is not a number costs the whole TAL and is\n    reported. A duration is one field of an otherwise readable event, so the event is kept —\n    but it is counted, and the run says so.\n  */\n  let duration: number | null = null;\n  let durationUnreadable = false;\n  if (durationText !== null && durationText !== '') {\n    const d = Number(durationText);\n    if (Number.isFinite(d)) duration = d;\n    else durationUnreadable = true;\n  }\n\n  /*\n    A duration is a length of time, and a length below zero is not one.\n\n    The value is kept and written as the file gave it — inventing a zero, or dropping it to\n    an empty cell, would put a number in annotations.csv that no writer wrote, which is the\n    one thing this tool does not do. But it is reported, because everything downstream\n    quietly does the wrong thing with it: the recipe this documentation gives for the samples\n    an event covers is `onset_s + duration_s`, which for a duration of -3 ends three seconds\n    before the event starts and selects nothing at all, with no error anywhere.\n  */\n  const durationNegative = duration !== null && duration < 0;\n\n  const annotations: Annotation[] = [];\n  for (const raw of parts.slice(1)) {\n    // A trailing separator yields an empty segment; a timekeeping TAL is all empty.\n    if (raw === '') continue;\n    annotations.push({\n      onset,\n      duration,\n      text: raw,\n      recordIndex,\n      ...(durationUnreadable ? { durationUnreadable: true } : {}),\n    });\n  }\n\n  // Per event rather than per TAL: one TAL may carry several texts, and each becomes a row\n  // of annotations.csv with the same cell in it.\n  return {\n    onset,\n    annotations,\n    unreadableDurations: durationUnreadable ? annotations.length : 0,\n    negativeDurations: durationNegative ? annotations.length : 0,\n  };\n}\n\nexport { SEP_TEXT, SEP_DURATION, TAL_END };\n"]}
+{"version":3,"file":"annotations.js","sourceRoot":"","sources":["../../src/edf/annotations.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;GAaG;AAEH,OAAO,EAAE,UAAU,EAAE,MAAM,YAAY,CAAC;AAExC,MAAM,QAAQ,GAAG,IAAI,CAAC,CAAC,yDAAyD;AAChF,MAAM,YAAY,GAAG,IAAI,CAAC,CAAC,gCAAgC;AAC3D,MAAM,OAAO,GAAG,IAAI,CAAC;AAErB,MAAM,aAAa,GAAG,MAAM,CAAC,YAAY,CAAC,QAAQ,CAAC,CAAC;AACpD,MAAM,iBAAiB,GAAG,MAAM,CAAC,YAAY,CAAC,YAAY,CAAC,CAAC;AAuD5D;;;;;;GAMG;AACH,MAAM,UAAU,uBAAuB,CACrC,KAAiB,EACjB,WAAmB,EACnB,kBAAkB,GAAG,IAAI;IAEzB,MAAM,WAAW,GAAiB,EAAE,CAAC;IACrC,IAAI,WAAW,GAAkB,IAAI,CAAC;IACtC,IAAI,UAAU,GAAG,IAAI,CAAC;IACtB,IAAI,SAAS,GAAG,CAAC,CAAC;IAClB,IAAI,oBAAoB,GAAG,CAAC,CAAC;IAC7B,IAAI,mBAAmB,GAAG,CAAC,CAAC;IAC5B,IAAI,iBAAiB,GAAG,CAAC,CAAC;IAE1B,IAAI,KAAK,GAAG,CAAC,CAAC;IACd,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,IAAI,KAAK,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACvC,IAAI,CAAC,KAAK,KAAK,CAAC,MAAM,IAAI,KAAK,CAAC,CAAC,CAAC,KAAK,OAAO;YAAE,SAAS;QAEzD,IAAI,CAAC,GAAG,KAAK,EAAE,CAAC;YACd,MAAM,KAAK,GAAG,KAAK,CAAC,QAAQ,CAAC,KAAK,EAAE,CAAC,CAAC,CAAC;YACvC;;;;;;;;;;;;cAYE;YACF,IAAI,KAAK,CAAC,KAAK,CAAC,aAAa,CAAC,EAAE,CAAC;gBAC/B,KAAK,GAAG,CAAC,GAAG,CAAC,CAAC;gBACd,SAAS;YACX,CAAC;YACD,MAAM,MAAM,GAAG,QAAQ,CAAC,KAAK,EAAE,WAAW,CAAC,CAAC;YAE5C,+EAA+E;YAC/E,mFAAmF;YACnF,+EAA+E;YAC/E,iFAAiF;YACjF,+EAA+E;YAC/E,uEAAuE;YACvE;;;;;;;;;cASE;YACF,MAAM,aAAa,GAAG,UAAU,IAAI,kBAAkB,CAAC;YACvD,UAAU,GAAG,KAAK,CAAC;YAEnB,IAAI,MAAM,EAAE,CAAC;gBACX,IAAI,aAAa;oBAAE,WAAW,GAAG,MAAM,CAAC,KAAK,CAAC;gBAC9C,KAAK,MAAM,UAAU,IAAI,MAAM,CAAC,WAAW;oBAAE,WAAW,CAAC,IAAI,CAAC,UAAU,CAAC,CAAC;gBAC1E,mBAAmB,IAAI,MAAM,CAAC,mBAAmB,CAAC;gBAClD,iBAAiB,IAAI,MAAM,CAAC,iBAAiB,CAAC;YAChD,CAAC;iBAAM,CAAC;gBACN;;;;;;;;;kBASE;gBACF,IAAI,aAAa;oBAAE,oBAAoB,EAAE,CAAC;;oBACrC,SAAS,EAAE,CAAC;YACnB,CAAC;QACH,CAAC;QACD,KAAK,GAAG,CAAC,GAAG,CAAC,CAAC;IAChB,CAAC;IAED,OAAO;QACL,WAAW;QACX,WAAW;QACX,SAAS;QACT,oBAAoB;QACpB,mBAAmB;QACnB,iBAAiB;KAClB,CAAC;AACJ,CAAC;AAWD,iFAAiF;AACjF,SAAS,aAAa,CAAC,IAAY;IACjC,OAAO,IAAI,KAAK,IAAI,IAAI,IAAI,KAAK,IAAI,IAAI,IAAI,KAAK,IAAI,IAAI,IAAI,KAAK,IAAI,IAAI,IAAI,KAAK,IAAI,CAAC;AAC3F,CAAC;AAED,SAAS,QAAQ,CAAC,KAAiB,EAAE,WAAmB;IACtD,mEAAmE;IACnE,MAAM,KAAK,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC;IACvB,IAAI,KAAK,KAAK,IAAI,CAAC,OAAO,IAAI,KAAK,KAAK,IAAI,CAAC,OAAO;QAAE,OAAO,IAAI,CAAC;IAElE,MAAM,IAAI,GAAG,UAAU,CAAC,KAAK,CAAC,CAAC;IAC/B,MAAM,KAAK,GAAG,IAAI,CAAC,KAAK,CAAC,aAAa,CAAC,CAAC;IACxC,MAAM,IAAI,GAAG,KAAK,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;IAE5B,IAAI,SAAS,GAAG,IAAI,CAAC;IACrB,IAAI,YAAY,GAAkB,IAAI,CAAC;IACvC,MAAM,WAAW,GAAG,IAAI,CAAC,OAAO,CAAC,iBAAiB,CAAC,CAAC;IACpD,IAAI,WAAW,IAAI,CAAC,EAAE,CAAC;QACrB,SAAS,GAAG,IAAI,CAAC,KAAK,CAAC,CAAC,EAAE,WAAW,CAAC,CAAC;QACvC,YAAY,GAAG,IAAI,CAAC,KAAK,CAAC,WAAW,GAAG,CAAC,CAAC,CAAC;IAC7C,CAAC;IAED,MAAM,KAAK,GAAG,MAAM,CAAC,SAAS,CAAC,CAAC;IAChC,IAAI,CAAC,MAAM,CAAC,QAAQ,CAAC,KAAK,CAAC;QAAE,OAAO,IAAI,CAAC;IAEzC;;;;;;;;;;MAUE;IACF,IAAI,QAAQ,GAAkB,IAAI,CAAC;IACnC,IAAI,kBAAkB,GAAG,KAAK,CAAC;IAC/B,IAAI,YAAY,KAAK,IAAI,IAAI,YAAY,KAAK,EAAE,EAAE,CAAC;QACjD,MAAM,CAAC,GAAG,MAAM,CAAC,YAAY,CAAC,CAAC;QAC/B,IAAI,MAAM,CAAC,QAAQ,CAAC,CAAC,CAAC;YAAE,QAAQ,GAAG,CAAC,CAAC;;YAChC,kBAAkB,GAAG,IAAI,CAAC;IACjC,CAAC;IAED;;;;;;;;;MASE;IACF,MAAM,gBAAgB,GAAG,QAAQ,KAAK,IAAI,IAAI,QAAQ,GAAG,CAAC,CAAC;IAE3D,MAAM,WAAW,GAAiB,EAAE,CAAC;IACrC,KAAK,MAAM,GAAG,IAAI,KAAK,CAAC,KAAK,CAAC,CAAC,CAAC,EAAE,CAAC;QACjC,gFAAgF;QAChF,IAAI,GAAG,KAAK,EAAE;YAAE,SAAS;QACzB,WAAW,CAAC,IAAI,CAAC;YACf,KAAK;YACL,QAAQ;YACR,IAAI,EAAE,GAAG;YACT,WAAW;YACX,GAAG,CAAC,kBAAkB,CAAC,CAAC,CAAC,EAAE,kBAAkB,EAAE,IAAI,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;SAC5D,CAAC,CAAC;IACL,CAAC;IAED,yFAAyF;IACzF,+CAA+C;IAC/C,OAAO;QACL,KAAK;QACL,WAAW;QACX,mBAAmB,EAAE,kBAAkB,CAAC,CAAC,CAAC,WAAW,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC;QAChE,iBAAiB,EAAE,gBAAgB,CAAC,CAAC,CAAC,WAAW,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC;KAC7D,CAAC;AACJ,CAAC;AAED,OAAO,EAAE,QAAQ,EAAE,YAAY,EAAE,OAAO,EAAE,CAAC","sourcesContent":["/**\n * EDF+ annotation (TAL) decoding.\n *\n * The annotations channel stores UTF-8 text in place of samples. Its bytes are a\n * run of Time-stamped Annotation Lists, each terminated by a NUL, with the rest\n * of the channel NUL-padded:\n *\n *   +<onset>[<0x15><duration>]<0x14><text><0x14>...<0x00>\n *\n * The first TAL of every data record must carry that record's start time and no\n * text; that is how an EDF+D file states where each record actually sits in time.\n *\n *   +1.25<0x15>0.5<0x14>Seizure onset<0x14><0x00>\n */\n\nimport { decodeUtf8 } from './bytes.js';\n\nconst SEP_TEXT = 0x14; // separates onset/duration from text, and text from text\nconst SEP_DURATION = 0x15; // separates onset from duration\nconst TAL_END = 0x00;\n\nconst TEXT_SEP_CHAR = String.fromCharCode(SEP_TEXT);\nconst DURATION_SEP_CHAR = String.fromCharCode(SEP_DURATION);\n\nexport interface Annotation {\n  /** Seconds from the start of the recording. */\n  onset: number;\n  /**\n   * Seconds, or null when the TAL stated no duration that could be read.\n   *\n   * Null covers two cases the file distinguishes and this field does not: a TAL that omitted\n   * the duration, and a TAL that stated one which is not a number. They are told apart by\n   * `unreadableDurations`, which is what raises the warning; the value itself has nowhere\n   * honest to put \"the file said `abc`\".\n   */\n  duration: number | null;\n  text: string;\n  /** Index of the data record this annotation was stored in. */\n  recordIndex: number;\n  /**\n   * True when the file stated a duration that could not be read.\n   *\n   * `duration` is null either way, which is the ambiguity the counts beside it exist to\n   * flag — and those counts were of the whole file while `annotations.csv` is filtered to\n   * the requested window. A conversion of one second of a recording warned that \"1\n   * annotation states a duration that is not a number, so its duration_s cell is empty\"\n   * about an event two seconds outside it, and failed `--strict` for it. Carrying the fact\n   * on the event lets the count be taken where the window has already been applied.\n   */\n  durationUnreadable?: boolean;\n}\n\nexport interface DecodedRecordAnnotations {\n  /** Record start time in seconds, from the leading timekeeping TAL. */\n  recordStart: number | null;\n  annotations: Annotation[];\n  /** Non-empty chunks that were not valid TALs, so the caller can report them. */\n  malformed: number;\n  /** Unreadable TALs in first position, which carry a record's start time, not an event. */\n  malformedTimekeeping: number;\n  /**\n   * Events kept whose stated duration could not be read.\n   *\n   * Counted apart again, for the same reason the two above are: the entry was exported and\n   * nothing about it is missing except the one field, so calling it an entry that \"could not\n   * be exported\" describes a loss that did not happen and hides the one that did.\n   */\n  unreadableDurations: number;\n  /**\n   * Events kept whose stated duration is a readable number below zero.\n   *\n   * Separate from the count above because the value survives: it is written to the CSV as\n   * the file gave it, and what is wrong with it is arithmetic rather than parsing.\n   */\n  negativeDurations: number;\n}\n\n/**\n * Decode one data record's annotation bytes.\n *\n * Malformed TALs are skipped rather than thrown, because a single bad annotation\n * should not cost the user an entire conversion. The count of skipped chunks is\n * returned so the caller can tell the user rather than losing them in silence.\n */\nexport function decodeRecordAnnotations(\n  bytes: Uint8Array,\n  recordIndex: number,\n  carriesTimekeeping = true,\n): DecodedRecordAnnotations {\n  const annotations: Annotation[] = [];\n  let recordStart: number | null = null;\n  let isFirstTal = true;\n  let malformed = 0;\n  let malformedTimekeeping = 0;\n  let unreadableDurations = 0;\n  let negativeDurations = 0;\n\n  let start = 0;\n  for (let i = 0; i <= bytes.length; i++) {\n    if (i !== bytes.length && bytes[i] !== TAL_END) continue;\n\n    if (i > start) {\n      const chunk = bytes.subarray(start, i);\n      /*\n        Padding is not a lost annotation.\n\n        The spec pads the slot with NUL, which the loop above already skips because it is what\n        separates one TAL from the next. Writers pad with spaces instead, and a run of spaces\n        after the last TAL is a non-empty chunk — so a file holding one perfectly readable\n        event, exported in full, was told \"2 annotation entries were unreadable and could not\n        be exported\", one per record. Nothing was lost. Under --strict that is a failed run\n        over the whitespace at the end of a slot.\n\n        Only whitespace. A chunk of anything else that does not parse is a real loss and is\n        still counted, which is the case this warning exists for.\n      */\n      if (chunk.every(isPaddingByte)) {\n        start = i + 1;\n        continue;\n      }\n      const parsed = parseTal(chunk, recordIndex);\n\n      // The timekeeping TAL is the one in first POSITION, whether or not it decodes.\n      // Clearing this flag only on a successful parse meant that an unreadable first TAL\n      // promoted the next ordinary annotation to timekeeping, and its onset silently\n      // became the record's start time — shifting every sample in that record. Leaving\n      // recordStart null instead is what the caller already handles, with a fallback\n      // timestamp and an ANNOTATION_DECODE_FAILED warning naming the record.\n      /*\n        Only one annotation channel carries a record's start time.\n\n        This flagged the first TAL of *every* annotation channel as timekeeping. In a second\n        channel the first TAL is an ordinary event — so when one failed to parse, the event\n        was dropped and counted as a lost timekeeping entry, which produced the warning\n        \"3 data records carry a timekeeping annotation that could not be read\" followed by\n        \"No event was lost\". Three events had been lost, and the timekeeping in that file was\n        perfectly readable. Both sentences false, about the same three records.\n      */\n      const isTimekeeping = isFirstTal && carriesTimekeeping;\n      isFirstTal = false;\n\n      if (parsed) {\n        if (isTimekeeping) recordStart = parsed.onset;\n        for (const annotation of parsed.annotations) annotations.push(annotation);\n        unreadableDurations += parsed.unreadableDurations;\n        negativeDurations += parsed.negativeDurations;\n      } else {\n        /*\n          Counted apart from the events, because losing one is a different loss.\n\n          A timekeeping TAL is never exported — it says where the record sits, not what\n          happened — so counting it among the entries that \"could not be exported\" both\n          overstated what was lost from annotations.csv and said nothing about the thing that\n          actually went missing, which is a record's position in time. A file with one\n          unreadable timekeeping TAL and three perfectly good events reported \"1 annotation\n          entry was unreadable and could not be exported\" while exporting all three.\n        */\n        if (isTimekeeping) malformedTimekeeping++;\n        else malformed++;\n      }\n    }\n    start = i + 1;\n  }\n\n  return {\n    recordStart,\n    annotations,\n    malformed,\n    malformedTimekeeping,\n    unreadableDurations,\n    negativeDurations,\n  };\n}\n\ninterface ParsedTal {\n  onset: number;\n  annotations: Annotation[];\n  /** How many of those annotations carry a duration the file stated and this could not read. */\n  unreadableDurations: number;\n  /** How many carry a duration that read as a number below zero. */\n  negativeDurations: number;\n}\n\n/** Space, tab, CR, LF or NUL — what a writer fills the rest of the slot with. */\nfunction isPaddingByte(byte: number): boolean {\n  return byte === 0x20 || byte === 0x09 || byte === 0x0d || byte === 0x0a || byte === 0x00;\n}\n\nfunction parseTal(chunk: Uint8Array, recordIndex: number): ParsedTal | null {\n  // The onset must be explicitly signed; anything else is not a TAL.\n  const first = chunk[0];\n  if (first !== 0x2b /* + */ && first !== 0x2d /* - */) return null;\n\n  const text = decodeUtf8(chunk);\n  const parts = text.split(TEXT_SEP_CHAR);\n  const head = parts[0] ?? '';\n\n  let onsetText = head;\n  let durationText: string | null = null;\n  const durationSep = head.indexOf(DURATION_SEP_CHAR);\n  if (durationSep >= 0) {\n    onsetText = head.slice(0, durationSep);\n    durationText = head.slice(durationSep + 1);\n  }\n\n  const onset = Number(onsetText);\n  if (!Number.isFinite(onset)) return null;\n\n  /*\n    A duration the file stated and this could not read is not the same as no duration.\n\n    Both came out as `null` and so as an empty `duration_s` cell, which the documentation\n    defines as meaning the file gave no duration — so an event whose duration was written as\n    `abc` was exported as an event with no duration, indistinguishable from one beside it\n    that genuinely had none, and nothing anywhere said a field had been dropped. The onset is\n    already held to this standard: one that is not a number costs the whole TAL and is\n    reported. A duration is one field of an otherwise readable event, so the event is kept —\n    but it is counted, and the run says so.\n  */\n  let duration: number | null = null;\n  let durationUnreadable = false;\n  if (durationText !== null && durationText !== '') {\n    const d = Number(durationText);\n    if (Number.isFinite(d)) duration = d;\n    else durationUnreadable = true;\n  }\n\n  /*\n    A duration is a length of time, and a length below zero is not one.\n\n    The value is kept and written as the file gave it — inventing a zero, or dropping it to\n    an empty cell, would put a number in annotations.csv that no writer wrote, which is the\n    one thing this tool does not do. But it is reported, because everything downstream\n    quietly does the wrong thing with it: the recipe this documentation gives for the samples\n    an event covers is `onset_s + duration_s`, which for a duration of -3 ends three seconds\n    before the event starts and selects nothing at all, with no error anywhere.\n  */\n  const durationNegative = duration !== null && duration < 0;\n\n  const annotations: Annotation[] = [];\n  for (const raw of parts.slice(1)) {\n    // A trailing separator yields an empty segment; a timekeeping TAL is all empty.\n    if (raw === '') continue;\n    annotations.push({\n      onset,\n      duration,\n      text: raw,\n      recordIndex,\n      ...(durationUnreadable ? { durationUnreadable: true } : {}),\n    });\n  }\n\n  // Per event rather than per TAL: one TAL may carry several texts, and each becomes a row\n  // of annotations.csv with the same cell in it.\n  return {\n    onset,\n    annotations,\n    unreadableDurations: durationUnreadable ? annotations.length : 0,\n    negativeDurations: durationNegative ? annotations.length : 0,\n  };\n}\n\nexport { SEP_TEXT, SEP_DURATION, TAL_END };\n"]}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "edf2csv",
-  "version": "0.5.92",
+  "version": "0.5.94",
   "description": "Convert EDF, EDF+ and BDF biosignal recordings (European Data Format) to CSV from the command line. Local, streaming, and never resamples or alters units.",
   "keywords": [
     "edf",