edfcore 0.2.51 → 0.2.53

package/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.
 
+## 0.2.53
+
+- **Added** a check that `validateRecording`'s sample scan sees what pyEDFlib sees. The observed
+  digital minimum and maximum are the only numbers in a validation report derived from every sample
+  rather than from the header, so they are the ones worth checking against another reader — and a
+  sampled window cannot check them, because the extremes of a 22-hour recording are very unlikely
+  to fall inside the 256 samples the goldens record. The whole-signal extremes now come from
+  pyEDFlib, across all five corpus files.
+- Also pins `outOfDigitalRangeCount` against those extremes. It is a claim about the DECLARATION
+  rather than about the samples — a non-zero count means the header's digital range is wrong, and
+  edfcore never clamps — so recomputing whether any sample could be outside the declared bounds
+  from pyEDFlib's own observations is the independent form of that check.
+
+## 0.2.52
+
+- **Added** streaming equivalence on the 22-hour recording. The documented claim is that a streamed
+  chunk and a read chunk are the same object in every respect; on a 40-record fixture a chunking
+  mistake often cancels out, and over 2,650 records it cannot. 7,950,000 samples are concatenated
+  from 42 streamed chunks and compared element by element with a single `readWindow`, and no chunk
+  is allowed to hold more than its own records.
+- Also pins that streaming yields its first chunk after reading under one percent of a 48 MB file.
+  Bounded memory is half the claim; not having to read the whole file before yielding anything is
+  the other half, and a byte counter is the evidence.
+- The comparison is a loop rather than `toEqual`. Deep equality over two 7.95-million-element typed
+  arrays took 45 seconds and, on failure, printed a diff nobody could read; the loop takes under a
+  second and names the first differing sample, which is the only part anyone would look at.
+
 ## 0.2.51
 
 - **Added** the tests only a real, large recording can support, against the 22-hour sleep-edfx

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 = "0.2.51";
+export declare const VERSION = "0.2.53";
 //# 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 = '0.2.51';
+export const VERSION = '0.2.53';
 //# sourceMappingURL=constants.js.map

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "edfcore",
-  "version": "0.2.51",
+  "version": "0.2.53",
   "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 = '0.2.51';
+export const VERSION = '0.2.53';