3
0
mirror of https://github.com/larvit/adf-codec.git synced 2026-11-18 22:25:44 +01:00

Answer product-owner round 2: 1 🔴, 1 🟡 (0 misread); each plain spelling reads back one value and the rest ride the carry

This commit is contained in:
2026-11-16 23:32:02 +01:00
committed by Lillem4n
parent 355714343d
commit 54ec1cb61f
2 changed files with 16 additions and 16 deletions
+3 -3
View File
@@ -1501,9 +1501,9 @@ Valid while Jira's editor writes these defaults.
Real Jira documents hold attributes that wiki markup can't spell: `order` 1 on every ordered list,
`accessLevel` `""` on every mention, `isNumberColumnEnabled` `false` and `layout` `default` on
every table, `layout` `center` on `mediaSingle`, `language` `""` on a code block. A plain spelling
stands for exactly those values and the reader writes them, so a Jira-authored document writes as
wiki markup, not as JSON. Any other value, `localId` and a mention's `text` among them, rides the
carry: the lossless writer keeps each attribute (`docs/decisions.md` §Ids stay site-local), and the
stands for exactly that one value and the reader writes it, so a Jira-authored document writes as
wiki markup, not as JSON. Any other value, an absent attribute, `localId` and a mention's `text`
among them, rides the carry, so the plain spelling reads back deep-equal: the lossless writer keeps each attribute (`docs/decisions.md` §Ids stay site-local), and the
lossy writer drops it. `[~name]` reads as text, not as a mention with `id` `name`, since Jira Cloud
names a user by account id and the writer would otherwise write a different account back.
The maintainer is asked whether the lossless writer may drop a mention's `text` and any `localId`,