edfcore 0.4.453 → 0.4.455
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/constants.d.ts +1 -1
- package/dist/constants.js +1 -1
- package/docs/CHANGELOG.md +37 -0
- package/package.json +1 -1
- package/src/constants.ts +1 -1
package/dist/constants.d.ts
CHANGED
|
@@ -109,5 +109,5 @@ export declare const SIGNAL_FIELD_BLOCK_OFFSETS: {
|
|
|
109
109
|
readonly reserved: 224;
|
|
110
110
|
};
|
|
111
111
|
/** Published package version. Kept in sync with package.json by a test. */
|
|
112
|
-
export declare const VERSION = "0.4.
|
|
112
|
+
export declare const VERSION = "0.4.455";
|
|
113
113
|
//# sourceMappingURL=constants.d.ts.map
|
package/dist/constants.js
CHANGED
package/docs/CHANGELOG.md
CHANGED
|
@@ -6,6 +6,43 @@ alone does not tell you whether you were affected.
|
|
|
6
6
|
edfcore is pre-1.0. Patch releases have carried behaviour changes where the old behaviour was a
|
|
7
7
|
defect; those are called out below.
|
|
8
8
|
|
|
9
|
+
## 0.4.455
|
|
10
|
+
|
|
11
|
+
- **Fixed** a test that refused what it was named after by accident. `timeline.test.ts` listed
|
|
12
|
+
"the probes are out of order" among the arrays `buildTimelineFromProbes` must refuse, and its
|
|
13
|
+
probes were records 0, 3, 1 — whose LAST entry is record 1 rather than record 3 on a four-record
|
|
14
|
+
file. `assertProbeShape` checks the two ends before it checks the order, so that array was
|
|
15
|
+
refused with "received probes for records 0..1, but the start offset comes from record 0 and the
|
|
16
|
+
span ends at record 3", and the ordering loop underneath had never run once in the suite.
|
|
17
|
+
- The case now runs 0, 2, 1, 3: both ends correct, so nothing before the loop has anything to say
|
|
18
|
+
about it, and rising onsets throughout, so the monotonicity check that runs next cannot claim it
|
|
19
|
+
either. Deleting the loop outright now fails one test instead of none.
|
|
20
|
+
- Every case in that table now asserts the sentence its guard produces, not just `RangeError`.
|
|
21
|
+
Three different refusals share the class, so asserting the class alone proves only that SOME
|
|
22
|
+
guard fired — which is exactly how the case above drifted without anyone noticing.
|
|
23
|
+
- Added the case the loop must NOT refuse: four probes in order, one of them intermediate.
|
|
24
|
+
`assertProbeShape` documents intermediate probes as optional rather than unwelcome, and a loop
|
|
25
|
+
that refused every array of three or more would have passed the disorder case just as well.
|
|
26
|
+
|
|
27
|
+
## 0.4.454
|
|
28
|
+
|
|
29
|
+
- **Added** a test for `validateHeader`'s own `DATE_FIELDS_DISAGREE` check on the ordinary case:
|
|
30
|
+
a `dd.mm.yy` field that resolves to 2051 beside a recording identification `Startdate` that
|
|
31
|
+
spells 1951 out.
|
|
32
|
+
- `checkDates` reports that code from two arms, and only the EDF+ `yy` escape arm had ever been
|
|
33
|
+
run through `validateHeader`. The ordinary arm — the case the code is named for and the one
|
|
34
|
+
every doc page uses as its example — was covered five times over in `resolveStartTime`, which
|
|
35
|
+
is a different function in a different file that happens to emit the same code on the read
|
|
36
|
+
path. Deleting the comparison in `validate.ts` left the suite green.
|
|
37
|
+
- That matters because the two are deliberately independent. `validateHeader` is documented as
|
|
38
|
+
standing on its own so a caller can sweep a header it did not open, and 0.3.81 fixed this exact
|
|
39
|
+
shape of bug in the other direction: the escape arm was missing here, so the parser and the
|
|
40
|
+
conformance sweep disagreed about whether one file had a defect.
|
|
41
|
+
- The assertions are on `specReference`, not on the code. Both functions cite EDF+ additional
|
|
42
|
+
specification 4; only this one cites it as "(local recording identification)" and prints the
|
|
43
|
+
header's date as a resolved day rather than as raw bytes. A test that asserted the code alone
|
|
44
|
+
would have passed on the parser's diagnostic and proved nothing about the function it names.
|
|
45
|
+
|
|
9
46
|
## 0.4.453
|
|
10
47
|
|
|
11
48
|
- **Added** tests for the date comparison behind `DATE_IMPLAUSIBLE`, at every level rather than
|
package/package.json
CHANGED
package/src/constants.ts
CHANGED