edfcore 0.6.8 → 0.6.9
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 +24 -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 = "0.6.
|
|
112
|
+
export declare const VERSION = "0.6.9";
|
|
113
113
|
//# sourceMappingURL=constants.d.ts.map
|
package/dist/constants.js
CHANGED
package/docs/CHANGELOG.md
CHANGED
|
@@ -6,6 +6,30 @@ 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
|
+
## 0.6.9
|
|
10
|
+
|
|
11
|
+
- **Added** `corpus.yml`, so the corpus suite is run by something other than a developer
|
|
12
|
+
remembering to. `tests/corpus/` is the only place this library is checked against bytes it did
|
|
13
|
+
not write itself — EDFlib's generator, a PhysioNet polysomnogram, a CHB-MIT seizure recording,
|
|
14
|
+
and the parity goldens generated from pyEDFlib and MNE — and every test in it skips when
|
|
15
|
+
`tests/corpus/files/` is empty. `ci.yml` never fetched and `scripts/release.mjs` runs
|
|
16
|
+
`npm run check`, which does not either, so those tests executed only on a machine where someone
|
|
17
|
+
had happened to run `corpus:fetch`.
|
|
18
|
+
- 0.4.219 has the bug report in it. It records ungating one check "because CI never fetches, so a
|
|
19
|
+
gated version would have been the one check there that runs nowhere" — a workaround for one
|
|
20
|
+
check, with the other hundred-odd left where they were.
|
|
21
|
+
- Weekly and cached on the manifest hash, not per-push: the files come from third-party servers
|
|
22
|
+
that owe this project nothing, a run that hits the cache downloads nothing, and a corpus host
|
|
23
|
+
being down is not a broken library. It gates no merge and no release. `fetch-corpus.mjs` checks
|
|
24
|
+
every file against its `sha256` either way, so a restored cache is verified rather than trusted.
|
|
25
|
+
- `the-corpus-runs-somewhere.test.ts` checks that a workflow fetches, that the same workflow runs
|
|
26
|
+
the suite afterwards rather than only fetching, that no workflow fetching the corpus runs on a
|
|
27
|
+
push or a pull request, and that `corpus:fetch` is still unreachable from `npm run check` — which
|
|
28
|
+
is the premise the whole arrangement rests on. It also checks the offline default from the other
|
|
29
|
+
side: every corpus file that opens something in the download directory asks whether it is there
|
|
30
|
+
first, so a clone with no corpus skips rather than fails.
|
|
31
|
+
- No library change.
|
|
32
|
+
|
|
9
33
|
## 0.6.8
|
|
10
34
|
|
|
11
35
|
- **Fixed** `AGENTS.md` quoting the corpus download as ~59 MB. It is ~102 MB, which is what
|
package/package.json
CHANGED
package/src/constants.ts
CHANGED