edfcore 0.4.433 → 0.4.435

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 = "0.4.433";
+export declare const VERSION = "0.4.435";
 //# 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 = '0.4.433';
+export const VERSION = '0.4.435';
 //# sourceMappingURL=constants.js.map

package/docs/CHANGELOG.md

@@ -6,6 +6,39 @@ 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.
 
+## 0.4.435
+
+- **Added** a check that the announce script cuts one release and that a dry run cuts none.
+  `npm run announce` is the last step of every batch and the only command here that writes to
+  GitHub. Everything else — the bump, the commit, the tag, the publish — is either reversible or
+  gated by `npm run check`. A release is neither: it is public the moment it exists, it notifies
+  watchers, and a second one over the same range is not something anyone undoes quietly.
+- Three properties keep that safe and none was checked. Exactly one mutating call, because one
+  release per version is precisely what this script replaced and a per-version loop reintroduced by
+  someone would look like a fix. A dry run that exits rather than falling through, or the flag
+  people use to preview a batch would announce it. And a refusal to announce past a tag with no
+  changelog entry, because the notes *are* the changelog entries — a missing one produces a release
+  whose body skips a version, in the one artefact a reader trusts to be complete.
+- The title is pinned too: `edfcore <first>–<last>` with an en dash, dropping the range for a
+  single version, cut on the newest tag so it points at the code the range ends with. A small thing
+  to get wrong and a permanent one, since release titles are what the release list looks like
+  forever.
+
+## 0.4.434
+
+- **Added** a check that asking a source for nothing costs nothing, in every adapter. A zero-length
+  read is what a caller gets from `end - start` when a window selects no samples, from a range
+  computed off a record count that turned out to be zero, or from a loop whose last iteration has
+  nothing left to take. `assertReadRange` allows it, so every adapter has to decide what to do with
+  it, and each decides separately.
+- The wrong answers are quiet ones. Handing the request to the transport gets a `bytes=100--1`
+  range no server will honour, or a `Blob.slice(100, 100)` and a promise allocated for nothing.
+  Neither is an error a caller could act on, and both are about a request nobody meant to make.
+- All five adapters are checked in one table rather than one at a time, because the contract is
+  about `ByteSource` and an adapter added later inherits it. The offset still has to be real:
+  `offset === byteLength` is legal, since that is where a read of nothing sits at the end of a
+  file, and past it the offset is outside the source whether or not any bytes would have been read.
+
 ## 0.4.433
 
 - **Added** tests for what a conformant header is allowed to say. `validateHeader` raises three

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "edfcore",
-  "version": "0.4.433",
+  "version": "0.4.435",
   "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 = '0.4.433';
+export const VERSION = '0.4.435';