3
0
mirror of https://github.com/larvit/adf-codec.git synced 2026-11-17 13:50:25 +01:00
Commit Graph

209 Commits

Author SHA1 Message Date
lilleman f32945350a Answer stability round 1: 0 🔴, 2 🟡; seed the withheld-mark carries once per line, pin the single highlight around a link 2026-10-14 21:10:41 +02:00
lilleman c9f9276df9 Answer architecture round 1: 1 🔴, 2 🟡; read a node's taken mark types once, let the highlight ask takesMark, correct the README's code-span bullet 2026-10-14 21:10:41 +02:00
lilleman 278ebbbad5 Give code text no mark but code, link and annotation, in every reader 2026-10-14 21:10:41 +02:00
lilleman 6c389b9c83 Answer prose round 1: 0 🔴, 8 🟡; say which marks a node refuses, and name the fix in the directive mark refusal 2026-10-14 15:08:46 +02:00
lilleman 1ece44c882 Answer stability round 1: 0 🔴, 1 🟡; pin the directive mark guard to the mark's own type 2026-10-14 15:08:46 +02:00
lilleman 6842f386ae Answer architecture round 1: 2 🔴, 4 🟡; refuse a link around only nodes that take none, mark the change breaking, file code-text marks 2026-10-14 15:08:46 +02:00
lilleman 1ae7ba6992 Give an inline node no mark Atlassian's schema withholds from it, in every reader 2026-10-14 15:08:46 +02:00
lilleman 0789e87c31 Name the marker a list takes after a list of its kind, from the list right before it 2026-10-14 13:18:14 +02:00
lilleman 39a7f16b95 Answer prose round 1: 0 🔴, 6 🟡; list unsupported-node-shape causes as sub-items with their remedies 2026-10-14 13:18:14 +02:00
lilleman d69ffac722 Answer stability round 1: 1 🔴, 2 🟡; spell list hints from the node and the marker alternation, pin both pair-fault sites 2026-10-14 13:18:14 +02:00
lilleman 3391154518 Name the CommonMark spelling a refused directive takes, split the unsupported-node-shape row by cause, and pass readAttributes its pair fault 2026-10-14 13:18:14 +02:00
lilleman aefcd4b9e3 Answer prose round 1: 0 🔴, 6 🟡; the changelog names what breaks, the fallback reads as untyped mark, attribute or value, the spec's value list and marks paragraph split, messages name the JSON array to write 2026-10-14 10:48:12 +02:00
lilleman 168ca1cea2 Answer stability round 1: 0 🔴, 2 🟡; a mark's malformed attrs name their cause, the docs name a value of the wrong type as a fallback 2026-10-14 10:48:12 +02:00
lilleman c8ea357fa7 Answer architecture round 1: 2 🔴, 5 🟡; nesting counts the JSON fallback alone, one mark attrs read and spell, one mark table, a malformed mark is unsupported-node-shape, the old colwidth names the comma list, README and notes name the attrs fallback 2026-10-14 10:48:12 +02:00
lilleman e882b5de7a Spell colwidth as a comma list and a block's marks by name 2026-10-14 10:48:12 +02:00
lilleman 5f1c64b371 Answer prose round 1: 0 🔴, 7 🟡; the guarantee splits per reader, the spec's portable sentence follows the named errors, item 112 names rows holding no pipe 2026-10-14 01:17:39 +02:00
lilleman 8a5e7dfa45 Answer stability round 1: 1 🔴, 2 🟡; a line indented four columns stays code under a table, tests pin the claim checks, the error row names a ragged body row 2026-10-14 01:17:39 +02:00
lilleman 2cc6ea3f57 Answer architecture round 1: 0 🔴, 5 🟡; comments and docs name the narrowing, item 112 files GFM's ragged rows, a paragraph's lines carry their positions 2026-10-14 01:17:39 +02:00
lilleman ec9492635a Read GFM's pipe table without leading pipes and with alignment colons in portableMarkdownToAdf 2026-10-14 01:17:39 +02:00
lilleman 6152808194 Answer prose round 1: 0 🔴, 2 🟡; the slot's rule joins the node-tables bullet, the remedies offer the space alike, the message says lines 2026-10-14 00:26:19 +02:00
lilleman 5cd07cafec Answer architecture round 1: 0 🔴, 2 🟡; the slot's line ending is a value the slot does not take, and its message offers the space again 2026-10-14 00:26:19 +02:00
lilleman 06df26727a Fold unspellable-whitespace into unsupported-node-shape 2026-10-14 00:26:19 +02:00
lilleman 667360f6e6 Answer prose round 1: 0 🔴, 9 🟡; the retry paragraph, the guarantees, the changelog and the code-list entry each say one thing once, the engine-limit message names what it covers, the persona line excepts engine-limit by the maintainer's call, and the error row's split is filed as item 111 2026-10-13 20:10:29 +02:00
lilleman a0fdfe377e Answer architecture round 2: 1 🔴, 1 🟡; the CommonMark ask answers yes or no, the memo's callers name themselves, the bound's policy reads at its constant, Firefox's bare out-of-memory string is engine-limit, and the floor entry no longer points at item 110 2026-10-13 20:10:29 +02:00
lilleman 3be5265a61 Give the reader's length-refusal test a timeout past Bun's default 2026-10-13 20:10:29 +02:00
lilleman 03b4ab2e44 Spell a line in memory linear in its length, so a paragraph up to V8's longest string converts within Node's default heap 2026-10-13 20:10:29 +02:00
lilleman 7bc69355c3 Answer prose round 1: the refusal messages name what passed the bound, the decisions read one idea a sentence, and the changelog gives the readers' fix its own bullet 2026-10-13 01:30:12 +02:00
lilleman 7caa8237f0 Answer architecture round 6: each block is held to the bound as it is built, the reader's ask takes the engine guard, and the chunk ships under a worse comprehension ruling by the maintainer's call, with item 110 to fill the memo in one place 2026-10-13 01:30:12 +02:00
lilleman a0bb09b80d Answer architecture round 4: the forecast and the indent leave, a writer refuses a document whose text alone passes the bound, and the joins are the one bound left 2026-10-13 01:30:12 +02:00
lilleman ae1a47f1bc Answer architecture round 3: the forecast never overcounts, the carry's JSON and the portable code block join through joinedWithin, and the list's choice of form stands in its own function 2026-10-13 01:30:12 +02:00
lilleman 3414c8690a Answer architecture round 2: every builder joins its pieces through one bounded join, the readers take the engine guard, every refusal stands at the document's path, and a portable list's refusal names its own place 2026-10-13 01:30:12 +02:00
lilleman e3aa790ec8 Refuse markdown past V8's longest string on every engine, before building it 2026-10-13 01:30:12 +02:00
lilleman 62d80a69b7 Answer prose round 1: the README says which refusal differs between engines, items 106 and 107 read one idea a sentence, and two test titles and a comment read on one pass 2026-10-11 21:05:07 +02:00
lilleman a9331b4125 Return a result where a conversion's text would pass the engine's longest string 2026-10-11 21:05:07 +02:00
lilleman 580bc862f2 Write an external image a link marks as [![alt](url)](href) in both writers 2026-10-11 13:31:18 +02:00
lilleman cc40fb95f2 Answer prose round 1: the README drops the parse-path sentence and the comments say what stands in and why 2026-10-11 12:13:51 +02:00
lilleman a38b61347a Report a refusal's path where the node stands once the task list moves it 2026-10-11 12:13:51 +02:00
lilleman cea9137495 Answer prose round 1: the comments and test name say what ends a list, the README says because, the changelog leads with the rename and gives the definition's trigger, and items 56 and 98 state their claims plainly 2026-10-10 13:20:05 +02:00
lilleman fcc6640f6e Answer architecture round 2: a list definitions end joins a set of ended lists, which a named endList fills and openedList reads 2026-10-10 13:20:05 +02:00
lilleman 7d00b11b2c Answer architecture round 1: closeLeaf records the block definitions end beside the blocks it pushes, openedList's comment names the exception, the README drops its counts, the changelog puts other changes last, and item 98 files the blank-line re-scan 2026-10-10 13:20:05 +02:00
lilleman fa7f4ef987 Read two lists a link reference definition parts as two lists, and put the changelog's breaking entries first 2026-10-10 13:20:05 +02:00
lilleman 3741b9e5b3 Answer prose round 1: the comments say why, items 95, 96, 91 and 34 read in one pass, the floor's decision drops the rule CLAUDE.md holds, and item 97 files the changelog's order 2026-10-09 19:17:20 +02:00
lilleman 3e9c56d399 Answer architecture round 3: the comprehension floor's decision names item 95, the directive read takes its index from the claimed prefix, and only the claimed prefix leaves directive-syntax.ts 2026-10-09 19:17:20 +02:00
lilleman af9b0d082c Answer architecture round 2: the escape walk owns its escapes and names every token kind, a directive read takes the claimed prefix, the reader computes one end, imports sort, and item 96 covers unclosed backtick runs 2026-10-09 19:17:20 +02:00
lilleman 652ae60179 Answer architecture round 1: a token's width is true for every kind that carries one, the escape walk threads one record, the == step and the directive precondition say why, and item 96 files the inline HTML scan 2026-10-09 19:17:20 +02:00
lilleman 59ead3e8e9 Escape a line against the inline reader's own tokens, in time linear in a run 2026-10-09 19:17:20 +02:00
lilleman 167bf1ff9b Answer prose round 1: the decision says right after and keeps only its reason, the README counts the link-definition exception, and the comments and spec say which list alternates 2026-10-09 15:39:13 +02:00
lilleman 559ad4ec3c Answer architecture round 1: a * marker before a first line of * alone takes a backslash, the marker choice is a decision, the spec names which lists alternate, and the stale listBreak test goes 2026-10-09 15:39:13 +02:00
lilleman 40580e6948 Read a list whose bullet or ordered delimiter changes as two lists, and alternate the marker between adjacent lists in place of !adf:listBreak 2026-10-09 15:39:13 +02:00
lilleman 180840532d Answer architecture round 2: itemParts takes one run of task lists over the item, the properties share blockTaskItemChildren, and readBlock's bullet case names who reads a task list first 2026-10-09 14:37:06 +02:00