edf2csv 0.4.75 → 0.4.77

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.
Files changed (2) hide show
  1. package/CHANGELOG.md +36 -0
  2. package/package.json +1 -1
package/CHANGELOG.md CHANGED
@@ -3,6 +3,42 @@
3
3
  Notable changes to edf2csv. Versions follow [semantic versioning](https://semver.org); while the
4
4
  major version is 0, a minor bump may contain breaking changes.
5
5
 
6
+ ## 0.4.77
7
+
8
+ ### Fixed: the landing page showed an annotations.csv the recording could not produce
9
+
10
+ The page says every terminal block and CSV sample on it is real output from one synthetic
11
+ sleep recording, and the recipe for that recording lived in a comment. Which made the claim
12
+ unfalsifiable in practice: checking it meant rebuilding a 19 MB file by hand from prose. The
13
+ page drifted twice anyway — a row figure in 0.4.67, a `UTC` suffix in 0.4.68 — and this
14
+ found a third. Its `annotations.csv` sample shows `Sleep stage W` at onset 0. The recording
15
+ had an annotation channel with no events in it at all, so the real file is a header row and
16
+ nothing else.
17
+
18
+ The recipe is a module now, `test/fixtures/sleep-study.mjs`, and the recording carries the
19
+ sleep staging a sleep study carries. A test rebuilds it, runs `--info`, and compares the
20
+ output against the page character for character; then converts two seconds of it and checks
21
+ each CSV sample really is how that file starts. Samples that are abridged say so with an
22
+ ellipsis and are left alone — which the `metadata.json` sample did not, while showing
23
+ `"version": "0.1.0"`. It is now visibly an extract.
24
+
25
+ ## 0.4.76
26
+
27
+ ### Added: the API reference's examples are run, not just displayed
28
+
29
+ Every `js` block on the page is now extracted, pointed at a real recording, and executed.
30
+ Only paths and the package specifier are rewritten; the code is otherwise exactly what the
31
+ page shows, and each block runs against the first fixture that carries every channel it
32
+ names. A rename, a moved option, a return shape that changed — the kind of drift that leaves
33
+ a page looking perfectly fine — now fails a test instead.
34
+
35
+ The `parseTimeSpec` example is checked differently, because running it proves nothing: it
36
+ asserts its results in comments rather than printing them. Those four claimed values are
37
+ parsed out of the page and compared against what the function returns.
38
+
39
+ Both were confirmed capable of failing before being kept — one by renaming an import on the
40
+ page, the other by changing a claimed `0.25` to `0.5`.
41
+
6
42
  ## 0.4.75
7
43
 
8
44
  ### Fixed: the buffer-reuse warning described something the obvious test disproves
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "edf2csv",
3
- "version": "0.4.75",
3
+ "version": "0.4.77",
4
4
  "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.",
5
5
  "keywords": [
6
6
  "edf",