edfcore 0.4.347 → 0.4.349

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 = "0.4.347";
+export declare const VERSION = "0.4.349";
 //# 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 = '0.4.347';
+export const VERSION = '0.4.349';
 //# 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.
 
+## 0.4.349
+
+- **Added** an execution of the cost table on `reading-signals.md`: one call for three channels
+  against three calls for one, at one read and 15,380 bytes versus three and 46,140. The file is
+  rebuilt from the sentence above the table, and both rows are measured through a recording source.
+- The table is an argument about how to write a loop, and it is the argument most likely to be
+  ignored, because three calls return the same answers as one and nothing in the result says the
+  caller paid triple. The check also constrains the direction that would look like an improvement:
+  narrowing a multi-signal read to per-signal ranges decodes fewer bytes and turns one request into
+  three. The row that must not move is the first — one read, and the whole record.
+- The three calls are additionally asserted to read the identical byte range each time, which is
+  what makes the extra two waste rather than work.
+
+## 0.4.348
+
+- **Added** a check for what `index.locate` costs, against the two read counts `discontinuous.md`
+  prints — three for the first call, zero for a second one nearby.
+- The zero is the one worth holding. A UI that calls `locate` on every pointer move is either free
+  or a request per frame, and nothing at the call site says which; a memo dropped in a refactor
+  changes no result at all, so no other test would notice. The three is worth holding for its
+  parenthesis rather than its size: records 0 and 5 are already known because `openEdf` probed
+  them, so the search starts from what the open already paid for.
+- Also pinned: that the search reads one record at a time rather than widening to a range, which
+  would still answer correctly and cost far more over HTTP; and the file the page draws, whose span
+  of 16 and coverage of 6 are computed independently and differ by the ten seconds no record
+  covers.
+
 ## 0.4.347
 
 - **Added** an execution of the `bigint` ticks decision on `design-decisions.md`. It is defended

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "edfcore",
-  "version": "0.4.347",
+  "version": "0.4.349",
   "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 = '0.4.347';
+export const VERSION = '0.4.349';