edfcore 0.4.489 → 0.4.491
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 +31 -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 = "0.4.491";
|
|
113
113
|
//# sourceMappingURL=constants.d.ts.map
|
package/dist/constants.js
CHANGED
package/docs/CHANGELOG.md
CHANGED
|
@@ -6,6 +6,37 @@ 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
|
+
## 0.4.491
|
|
10
|
+
|
|
11
|
+
- **Added** the `maxMaterializeBytes` boundary for `toPhysical` and `clampToDigitalRange` — the
|
|
12
|
+
fifth module that compares a requirement against that budget, and the last one still untested at
|
|
13
|
+
the edge.
|
|
14
|
+
- Finding them one at a time is the thing that kept going wrong. 0.4.464 covered three, found by
|
|
15
|
+
grepping for `resolveMaterializeBudget`; 0.4.482 found a fourth in `validate.ts` whose local
|
|
16
|
+
variable the grep walked past; this is the fifth. So the file now enumerates the refusal sites
|
|
17
|
+
out of `src/` and fails on a sixth until it is listed — which is the only way this stops being a
|
|
18
|
+
rule that most of the code follows.
|
|
19
|
+
- Both array widths are checked here rather than one, because they are different numbers reached
|
|
20
|
+
through the same comparison: eight bytes a sample for the float64 result, four for the clamped
|
|
21
|
+
int32. A boundary correct for one and wrong for the other would look tested.
|
|
22
|
+
|
|
23
|
+
## 0.4.490
|
|
24
|
+
|
|
25
|
+
- **Added** a check that every published export is reached by the suite a fresh clone actually
|
|
26
|
+
runs. `tests/corpus/` skips when the corpus is absent — 59 MB of other people's recordings, not
|
|
27
|
+
redistributed — so those tests do not run on a clone, in CI unless fetched, or for a contributor
|
|
28
|
+
who has not asked for them.
|
|
29
|
+
- An export whose only appearance in the suite is inside that directory is an export nothing
|
|
30
|
+
verifies on an ordinary run: it can be renamed, dropped from a barrel or left broken, and the run
|
|
31
|
+
a contributor does stays green. Not hypothetical here — `whole-api.test.ts` calls essentially the
|
|
32
|
+
whole barrel and is one of the files that skips.
|
|
33
|
+
- All 78 pass today. The value is the tripwire: adding an export and testing it only against a real
|
|
34
|
+
file now fails, in the run that would otherwise say nothing.
|
|
35
|
+
- `tests/scratch/` is excluded alongside the corpus for the reason 0.4.479 established — it is
|
|
36
|
+
gitignored, so anything in it satisfies a check on the machine that wrote it and nowhere else.
|
|
37
|
+
- Deliberately coarse, and the file says so: an import is a floor, not proof the export is called.
|
|
38
|
+
The floor is the part that can be checked structurally, and it was not being checked at all.
|
|
39
|
+
|
|
9
40
|
## 0.4.489
|
|
10
41
|
|
|
11
42
|
- **Added** a property test for `summarizeDiagnostics`. Everything that prints a count of a file's
|
package/package.json
CHANGED
package/src/constants.ts
CHANGED