edfcore 0.4.434 → 0.4.436
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 = "0.4.436";
|
|
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
|
+
## 0.4.436
|
|
10
|
+
|
|
11
|
+
- **Added** tests that a real `AbortSignal` reaches `fetch` and a bare `{ aborted }` shim never
|
|
12
|
+
does. `FetchLike` deliberately does not name `signal` — naming it would pull the DOM
|
|
13
|
+
`AbortSignal` into the published types by parameter contravariance, the exact dependency the
|
|
14
|
+
structural shims exist to avoid — so the signal is handed over at runtime, on one line.
|
|
15
|
+
- Both halves matter and they fail differently. Attaching a shim is loud: the platform `fetch`
|
|
16
|
+
throws a `TypeError` on an init whose `signal` is not an `AbortSignal`, so every request from a
|
|
17
|
+
caller using the published `AbortSignalLike` type would fail at once. Not attaching a real one is
|
|
18
|
+
silent, and that is the half nothing had exercised — no test had ever given `httpSource` a
|
|
19
|
+
genuine `AbortSignal`.
|
|
20
|
+
- Without the attach, `throwIfSignalAborted` still rejects the caller's promise at the next poll,
|
|
21
|
+
so an abort looks like it worked, while the request runs to completion. On a range covering a few
|
|
22
|
+
hundred megabytes that is the difference between cancelling a transfer and paying for it, and the
|
|
23
|
+
only visible symptom is a bill, or a phone that stays warm after the user navigated away.
|
|
24
|
+
- The caller's own signal object has to be the one attached, not a copy, or aborting theirs aborts
|
|
25
|
+
nothing. A source-level signal and a per-read one are both checked, since the page documents the
|
|
26
|
+
first as the default for every request and the second as winning over it.
|
|
27
|
+
|
|
28
|
+
## 0.4.435
|
|
29
|
+
|
|
30
|
+
- **Added** a check that the announce script cuts one release and that a dry run cuts none.
|
|
31
|
+
`npm run announce` is the last step of every batch and the only command here that writes to
|
|
32
|
+
GitHub. Everything else — the bump, the commit, the tag, the publish — is either reversible or
|
|
33
|
+
gated by `npm run check`. A release is neither: it is public the moment it exists, it notifies
|
|
34
|
+
watchers, and a second one over the same range is not something anyone undoes quietly.
|
|
35
|
+
- Three properties keep that safe and none was checked. Exactly one mutating call, because one
|
|
36
|
+
release per version is precisely what this script replaced and a per-version loop reintroduced by
|
|
37
|
+
someone would look like a fix. A dry run that exits rather than falling through, or the flag
|
|
38
|
+
people use to preview a batch would announce it. And a refusal to announce past a tag with no
|
|
39
|
+
changelog entry, because the notes *are* the changelog entries — a missing one produces a release
|
|
40
|
+
whose body skips a version, in the one artefact a reader trusts to be complete.
|
|
41
|
+
- The title is pinned too: `edfcore <first>–<last>` with an en dash, dropping the range for a
|
|
42
|
+
single version, cut on the newest tag so it points at the code the range ends with. A small thing
|
|
43
|
+
to get wrong and a permanent one, since release titles are what the release list looks like
|
|
44
|
+
forever.
|
|
45
|
+
|
|
9
46
|
## 0.4.434
|
|
10
47
|
|
|
11
48
|
- **Added** a check that asking a source for nothing costs nothing, in every adapter. A zero-length
|
package/package.json
CHANGED
package/src/constants.ts
CHANGED