edf2csv 0.6.7 → 0.6.10
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/README.md +3 -2
- package/docs/CHANGELOG.md +32 -0
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -76,8 +76,9 @@ npx edf2csv recording.edf --start 30m --duration 5m
|
|
|
76
76
|
-V, --version Version
|
|
77
77
|
```
|
|
78
78
|
|
|
79
|
-
Exit codes: `0` success, `1` the file couldn't be read or written
|
|
80
|
-
was
|
|
79
|
+
Exit codes: `0` success, `1` the file couldn't be read or written — or `--strict` was given and
|
|
80
|
+
the recording raised a warning, in which case the output was still written — `2` the command
|
|
81
|
+
was wrong. An interrupted run exits 130.
|
|
81
82
|
|
|
82
83
|
## What it doesn't change
|
|
83
84
|
|
package/docs/CHANGELOG.md
CHANGED
|
@@ -3,6 +3,38 @@
|
|
|
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
|
+
## 0.6.10
|
|
7
|
+
|
|
8
|
+
### Fixed: the README's exit codes contradicted its own options list
|
|
9
|
+
|
|
10
|
+
`1` was given as "the file couldn't be read or written", twenty lines under an options list
|
|
11
|
+
that says `--strict` "Exit 1 if the recording raised any warning". A `--strict` run over a
|
|
12
|
+
discontinuous recording writes all four files and exits 1, so the one page a reader meets first
|
|
13
|
+
described that outcome as a failure to read or write a file that was read and written. It also
|
|
14
|
+
skipped 130, which is what Ctrl-C gives and what a wrapper script has to know about.
|
|
15
|
+
|
|
16
|
+
## 0.6.9
|
|
17
|
+
|
|
18
|
+
### Fixed: package-lock.json had been claiming 0.5.51 for fifty-eight releases
|
|
19
|
+
|
|
20
|
+
The lockfile carries the package's version in two places and both had stopped moving. Nothing
|
|
21
|
+
breaks on it — `npm ci` does not read the field, and `npm install` rewrites it — but anyone
|
|
22
|
+
checking out a tag and opening the file is told a version other than the one they checked out.
|
|
23
|
+
|
|
24
|
+
CITATION.cff had exactly this happen, 107 releases of it, and the answer was a test asserting
|
|
25
|
+
it against `package.json`. That test now checks the lockfile in the same breath.
|
|
26
|
+
|
|
27
|
+
## 0.6.8
|
|
28
|
+
|
|
29
|
+
### Fixed: the `--info` table's `OUTPUT` column has three values and the reference listed two
|
|
30
|
+
|
|
31
|
+
`(no samples)` has been printed since 0.4.53, for a channel that declares zero samples per data
|
|
32
|
+
record, and the page enumerating the column said only "the file the channel would land in, or
|
|
33
|
+
`(not selected)` when `--channels` excludes it". A reader meeting the third value had nowhere
|
|
34
|
+
to look it up, and the obvious guess — that it means the same as `(not selected)` — is the one
|
|
35
|
+
0.4.53 exists to contradict: a channel asked for by name and carrying nothing is not a channel
|
|
36
|
+
that went unasked for.
|
|
37
|
+
|
|
6
38
|
## 0.6.7
|
|
7
39
|
|
|
8
40
|
### Fixed: `--info` and `--info --json` listed the same warnings in different orders
|
package/package.json
CHANGED