edfcore 0.4.377 → 0.4.379
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 +30 -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 = "0.4.379";
|
|
113
113
|
//# sourceMappingURL=constants.d.ts.map
|
package/dist/constants.js
CHANGED
package/docs/CHANGELOG.md
CHANGED
|
@@ -6,6 +6,36 @@ 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
|
+
## 0.4.379
|
|
10
|
+
|
|
11
|
+
- **Added** an execution of the "Oddities that bite implementers" section of `edf-format.md`, which
|
|
12
|
+
is that page's payload: seven paragraphs each naming a thing about EDF that produces a wrong
|
|
13
|
+
answer rather than an error. The page is written for someone about to write their own parser —
|
|
14
|
+
which the comparison page says is what people overwhelmingly do — so these are the claims most
|
|
15
|
+
likely to be acted on by a reader who never installs the package.
|
|
16
|
+
- Each is checked against the library, because that is the only way to check them and because each
|
|
17
|
+
is a place edfcore could regress into the naive behaviour being warned about. A date eighty years
|
|
18
|
+
wrong, a record count silently short, a rate of `Infinity`, a first header byte mangled by the
|
|
19
|
+
decoder that read it, and a redundant header size believed over the computed one — every
|
|
20
|
+
paragraph describes a bug that looks like working software.
|
|
21
|
+
|
|
22
|
+
## 0.4.378
|
|
23
|
+
|
|
24
|
+
- **Added** a check that the browser inspector uploads nothing, which is the most consequential
|
|
25
|
+
claim on the site. `demo.astro` says a browser can read these files "so a researcher never has to
|
|
26
|
+
hand a patient recording to a server", and somebody will drop a real clinical recording on that
|
|
27
|
+
page on the strength of that sentence.
|
|
28
|
+
- It is a claim about an absence, and nothing notices an absence breaking. A copy button that
|
|
29
|
+
reported usage, an error handler that posted a stack trace with a filename in it, an analytics
|
|
30
|
+
snippet in the shared layout — each is a normal thing to add to a website, and each would make
|
|
31
|
+
the sentence false without changing anything a visitor can see. So every source the site ships is
|
|
32
|
+
scanned for a way to send bytes: `fetch`, `XMLHttpRequest`, `sendBeacon`, `WebSocket`,
|
|
33
|
+
`EventSource`, a form, a remote dynamic import.
|
|
34
|
+
- `navigator.clipboard` is allowed by name rather than by permitting `navigator` generally, and the
|
|
35
|
+
pattern is asserted to still reject `navigator.sendBeacon` — the primitive built for exactly the
|
|
36
|
+
thing this forbids. The check covers the layout too, since the promise is about the page a
|
|
37
|
+
visitor is on and the layout wraps it.
|
|
38
|
+
|
|
9
39
|
## 0.4.377
|
|
10
40
|
|
|
11
41
|
- **Added** a check that the exports badge and the API-surface table on the README show the same
|
package/package.json
CHANGED
package/src/constants.ts
CHANGED