singularity/singularity-forge
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1

fix: robust prose slice header parsing — handle H1-H4, bold, dots, no-separator variants (#1248)

Browse source 
The prose fallback parser only matched H2 (## S01:) headers with
colon/dash separators. LLMs produce many variants that silently
produced 0 slices, permanently blocking auto-mode.

Expanded the regex from:
  /^##\s+(?:Slice\s+)?(S\d+)[:\s—–-]+\s*(.+)/gm
to:
  /^#{1,4}\s+\*{0,2}(?:Slice\s+)?(S\d+)\*{0,2}[:\s.—–-]*\s*(.+)/gm

Now handles:
  - H1 through H4 headers (# ## ### ####)
  - Bold-wrapped: **S01: Title**, **S01**: Title
  - Dot separator: S01. Title
  - Space-only separator: S01 Title (no punctuation)
  - Non-zero-padded IDs: S1, S01, S001
  - No-space after colon: S01:Title
  - All previous separators: colon, hyphen, em dash, en dash

Also strips trailing bold markers from titles and skips matches
with empty titles.

Fixes #1243
This commit is contained in:
Tom Boucher 2026-03-18 17:02:51 -04:00 • committed by GitHub
parent de1256d352
commit 7e2ca68161
1 changed files with 22 additions and 7 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

29
src/resources/extensions/gsd/roadmap-slices.ts
View file

@ -96,24 +96,39 @@ export function parseRoadmapSlices(content: string): RoadmapSliceEntry[] {
/**
* Fallback parser for prose-style roadmaps where the LLM wrote
* `## Slice S01: Title` headers instead of the machine-readable
* `## Slices` checklist. Extracts slice IDs and titles so auto-mode
* can at least identify slices and plan them.
* slice headers instead of the machine-readable `## Slices` checklist.
* Extracts slice IDs and titles so auto-mode can at least identify
* slices and plan them.
*
* Also handles `## S01: Title` and `## S01 — Title` variants.
* Handles these LLM-generated variants:
* ## S01: Title (H2, colon separator)
* ### S01: Title (H3)
* #### S01: Title (H4)
* ## Slice S01: Title (with "Slice" prefix)
* ## S01 — Title (em dash)
* ## S01 – Title (en dash)
* ## S01 - Title (hyphen)
* ## S01. Title (dot separator)
* ## S01 Title (space only, no separator)
* ## **S01: Title** (bold-wrapped)
* ## **S01**: Title (bold ID only)
* ## S1: Title (non-zero-padded ID)
*/
function parseProseSliceHeaders(content: string): RoadmapSliceEntry[] {
const slices: RoadmapSliceEntry[] = [];
const headerPattern = /^##\s+(?:Slice\s+)?(S\d+)[:\s—–-]+\s*(.+)/gm;
// Match H1–H4 headers containing S<digits> with optional "Slice" prefix and bold markers.
// Separator after the ID is flexible: colon, dash, em/en dash, dot, or just whitespace.
const headerPattern = /^#{1,4}\s+\*{0,2}(?:Slice\s+)?(S\d+)\*{0,2}[:\s.—–-]*\s*(.+)/gm;
let match: RegExpExecArray | null;
while ((match = headerPattern.exec(content)) !== null) {
const id = match[1]!;
const title = match[2]!.trim();
let title = match[2]!.trim().replace(/\*{1,2}$/g, "").trim(); // strip trailing bold markers
if (!title) continue; // skip if we only matched the ID with no title
// Try to extract depends from prose: "Depends on: S01" or "**Depends on:** S01, S02"
const afterHeader = content.slice(match.index + match[0].length);
const nextHeader = afterHeader.search(/^##\s/m);
const nextHeader = afterHeader.search(/^#{1,4}\s/m);
const section = nextHeader !== -1 ? afterHeader.slice(0, nextHeader) : afterHeader.slice(0, 500);
const depsMatch = section.match(/\*{0,2}Depends\s+on:?\*{0,2}\s*(.+)/i);