3
0
mirror of https://github.com/larvit/adf-codec.git synced 2026-11-12 19:58:28 +01:00

Answer prose round 1: item 78 names what it records and that it breaks

This commit is contained in:
2026-10-05 08:45:45 +02:00
committed by Lillem4n
parent 328ec88f52
commit a4ff89df07
+6 -5
View File
@@ -73,11 +73,12 @@ Lands after items 6, 59, 60, 61 and 62. The CommonMark spec suite also runs agai
Today `null` and `undefined` throw a `TypeError` reading `length`, and `42` reads as an empty
document. A JavaScript caller, or a TypeScript one past an untyped boundary, can pass any of these.
No existing code reads true of a non-string input (`docs/decisions.md` §The code list), so the
maintainer chose a new one, `not-a-string`, at `{ line: 1, offset: 0 }` with an empty path
(2026-10-05). The chunk adds it to `ConvertErrorCode`, gives it a row in the README's parsing table,
and lands the rule in `docs/decisions.md` §Which code a cause takes. Found while checking the other
ADF converters for item 71, 2026-10-05.
No existing code reads true of a non-string input (`docs/decisions.md` §The code list). The
maintainer chose a new one on 2026-10-05: `not-a-string`, with position `{ line: 1, offset: 0 }`
and an empty path. The chunk adds it to `ConvertErrorCode` and the README's parsing table, and
records in `docs/decisions.md` §Which code a cause takes that a value that is not a string is
`not-a-string`. Breaking (§The code list): the changelog entry says so. Found while checking the
other ADF converters for item 71.
### 45. Replace `isAdfDocument` with a reader returning `Result<AdfDocument>`.