edfcore 0.4.504 → 0.4.506

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 = "0.4.504";
+export declare const VERSION = "0.4.506";
 //# 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 = '0.4.504';
+export const VERSION = '0.4.506';
 //# sourceMappingURL=constants.js.map

package/docs/CHANGELOG.md

@@ -6,6 +6,37 @@ 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.
 
+## 0.4.506
+
+- **Added** the four-row cost table on `validation.md`, and the claim it exists to support:
+  "conformance costs one traversal rather than two", which is the reason `validateRecording` takes
+  an `index` at all.
+- `budget-invariance.test.ts` checks that the budget cannot change what the sweep REPORTS. Nothing
+  checked what it READS, and every row of that table is a cost a caller plans around — a plain EDF
+  is pure header arithmetic, an EDF+ without an index has to traverse, a complete index means it
+  does not, and `scanSamples` means it does regardless.
+- The sentences under the table are checked with it, including the one most likely to be quietly
+  wrong: a probed index — which is exactly what `openEdf` hands you, and therefore what a caller
+  passes by mistake — is ignored rather than refused, and the sweep reads the file itself. So is
+  an index covering a different number of records, which has the right shape and the wrong length.
+
+## 0.4.505
+
+- **Added** the values `design-decisions.md` prints beside three of its decisions. `AGENTS.md`
+  sends every contributor to that page before proposing an architectural change, so it is the one
+  document read specifically to be argued with — and `design-absences.test.ts` checks the absences
+  it claims, not the demonstrations that justify them.
+- A decision whose example no longer produces what it prints is worse than an undocumented one: a
+  reader weighing the trade-off is reading the example to decide whether the cost is what the page
+  says it is.
+- The three chosen are each the evidence for a REJECTION rather than an illustration of an accepted
+  design — the start time as fields, which is the argument against returning a `Date`; the onset as
+  exact ticks beside its lossy seconds, which is the argument against `parseFloat`; and
+  `clampToDigitalRange` as a separate pure function, which is the argument against clamping on read
+  the way the reference implementation does.
+- The last one is asserted from both sides: the clamp produces the array the page prints, and a
+  read of the same samples does not clamp them.
+
 ## 0.4.504
 
 - **Added** the values `api-primitives.md` prints beside each primitive. That page is the reference

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "edfcore",
-  "version": "0.4.504",
+  "version": "0.4.506",
   "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 = '0.4.504';
+export const VERSION = '0.4.506';