edfcore 0.4.260 → 0.4.262

package/dist/constants.d.ts

@@ -109,5 +109,5 @@ export declare const SIGNAL_FIELD_BLOCK_OFFSETS: {
     readonly reserved: 224;
 };
 /** Published package version. Kept in sync with package.json by a test. */
-export declare const VERSION = "0.4.260";
+export declare const VERSION = "0.4.262";
 //# sourceMappingURL=constants.d.ts.map

package/dist/constants.js

@@ -79,5 +79,5 @@ export const SIGNAL_FIELD_BLOCK_OFFSETS = {
     reserved: 224,
 };
 /** Published package version. Kept in sync with package.json by a test. */
-export const VERSION = '0.4.260';
+export const VERSION = '0.4.262';
 //# sourceMappingURL=constants.js.map

package/docs/CHANGELOG.md

@@ -6,6 +6,33 @@ alone does not tell you whether you were affected.
 edfcore is pre-1.0. Patch releases have carried behaviour changes where the old behaviour was a
 defect; those are called out below.
 
+## 0.4.262
+
+- **Fixed** the worked example on `annotations.md` — read the sample under each sleep-stage event
+  — which was the other complete program that failed to compile on nothing but an unnarrowed
+  index. It ended `toPhysical(signal, chunk.signals[0].digital)`, and `chunk.signals[0]` is
+  `T | undefined` under `noUncheckedIndexedAccess` even though the call asked for exactly one
+  signal. It narrows with a `continue` now, which is the shape the loop around it already uses
+  twice, and the numbered comment says the thing worth knowing: asking for one signal does not
+  tell the compiler you got one.
+
+  That closes both of the complete-but-unsound examples the 102-fence sweep in 0.4.261 turned up.
+  The rest of the site's failures are fragments referencing a `recording` or a `header` declared
+  in an earlier block on the same page, which is what a reference page is for.
+
+## 0.4.261
+
+- **Fixed** the opening example of `reading-signals.md`, which did not compile. It is the first
+  complete program on the page a reader lands on from "how do I read a signal", and it had the
+  same defect the README quick start had one release ago: `const [chunk] = await readWindow(...)`
+  followed by `chunk.signals[0].digital`, which under `noUncheckedIndexedAccess` is `TS18048` and
+  `TS2532`. Found by extracting every fenced example on the site that imports from `edfcore` and
+  compiling all 102 of them; two were complete programs failing on nothing but this, and this was
+  one. It also now says why the guard is there, because the reason is the same fact the page
+  teaches: a window inside an EDF+D gap really does select nothing.
+- Compiled it in `documented-examples.test-d.ts` alongside the other four, with the same
+  narrowing check the README quick start got.
+
 ## 0.4.260
 
 - **Fixed** the README's quick start, which did not compile. It is the first code most people run

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "edfcore",
-  "version": "0.4.260",
+  "version": "0.4.262",
   "description": "Modern, typed, zero-dependency reader for EDF, EDF+, BDF and BDF+ biosignal files. Works in browsers and Node with true random access.",
   "keywords": [
     "edf",

package/src/constants.ts

@@ -93,4 +93,4 @@ export const SIGNAL_FIELD_BLOCK_OFFSETS = {
 } as const;
 
 /** Published package version. Kept in sync with package.json by a test. */
-export const VERSION = '0.4.260';
+export const VERSION = '0.4.262';