Skip to content

t27f: deps card bands show the most imported specs first (Closes #10762) - #10767

Merged
gHashTag merged 9 commits into
masterfrom
claude/t27f-deps-rank
Oct 19, 2026
Merged

t27f: deps card bands show the most imported specs first (Closes #10762)#10767
gHashTag merged 9 commits into
masterfrom
claude/t27f-deps-rank

Conversation

@gHashTag

@gHashTag gHashTag commented Oct 19, 2026

Copy link
Copy Markdown
Owner

Closes #10762

What

Each band of the deps card (#10720) now shows its most imported specs first. Before this change, the three rows above the hub base/types were the first three of its 378 importers in edge-list order, which is alphabetical: three account/* specs. The card's own header said these were "not the most important ones". It now shows compiler/lexer, compiler/parser and igla/race/bram_weights.

  • Score: the number of distinct specs that import a spec, counted over the whole edge list. A repeated edge line counts once.
  • Ties keep the edge-list order, so a band where every score is equal is drawn exactly as before.
  • Subtitle: most imported first in each band. The card says how its rows were chosen.
  • Unchanged: the direct counts, the transitive counts and +N more. Only which names are drawn, and their order, change.
  • New check: every fixed text must pass dcard_text_ok. An earlier subtitle had a :, which ffmpeg read as an option separator, so it refused the whole graph while every test passed.

This branch also carries #10747's commits (the #10728 long-name abbreviation) until that PR merges.

Tests (22, all pass)

The five new tests:

  • Ranking: a band shows its most imported specs first. This is checked on both bands, with a hidden spec and +1 more.
  • Ties: equal scores keep the edge-list order.
  • Repeated lines: a repeated edge line, including a \r line, counts one importer.
  • Name check: the hidden-name check follows the ranking, not the edge-list order. This test is a negative control.
  • Fixed texts: every fixed text can be drawn, and a text that ffmpeg would split is refused. The second of these is a negative control.

Evidence (Railway lab, 7853e2e)

  • Mutants: 21 of 21 fail at least one named test.
    • The 12 new mutants:
      • the score ignored;
      • ties repeated, and ties taken last;
      • a repeated line counted;
      • every line treated as seen;
      • out-degree in place of in-degree;
      • edge-list order in the check, and in the draw;
      • a zero first score;
      • a space refused, an empty text accepted, and a comma subtitle.
    • The 9 older mutants still fail.
  • Build: t27c gen and zig build-exe -lc -O ReleaseSafe pass for this branch and for the t27f: deps card draws a long spec name abbreviated by a stated rule, not refused #10747 base.
  • Edge list: 2484 edge lines, every one with 2 fields; 133 lines are repeats; 813 specs.
    • Most imported: base/types 378, math/constants 225, trios-chat/common/helpers 29.
  • Corpus sweep. Each card from the new binary was compared with the base binary's card. The base's old subtitle was normalized first.
    • 124 cards are reordered, 689 are the same, and 0 are lost or refused.
    • 0 cards have a count that differs from the base.
    • 0 drawn rows score higher than the row above them. 1647 rows were checked. 3 rows have abbreviated names, so their scores are unknown.
  • Renders: two cards were rendered at 1080x1350 and passed every pixel check from t27f: deps card -- one spec's importers and imports as a social card #10720, with 0 white pixels outside x 100..980. A deliberately wrong expectation misses on both.
    • base/types, 378 importers;
    • analog/linear, the first reordered card that is not the hub.
  • Hub time: 539 ms.
  • Seals and ratchet: t27c seal --verify reports all hashes MATCH. The deps_card seal was re-saved. t27c suite --ratchet --corpus-only is RATCHET CLEAN.

Not claimed

  • "Most imported" is the only measure. It is not a measure of importance, of recent change, or of size.
  • The ranking is quadratic in the edge list. The hub, with 378 importers in 2484 lines, is the largest band measured.
  • 292 drawn rows have score 0, because no spec imports them. They keep the edge-list order.

Skill

.claude/skills/t27f/SKILL.md gets the rule, the sweep numbers and three lab traps:

  • ffmpeg splits drawtext on : and , inside the quoted text, and only a render catches it;
  • pkill -f '[r]k.sh' still kills an fs-lab command that also names /root/rk.sh;
  • seal --verify does not check the closure, so re-save the seal after a merge that changes an imported spec.

🤖 Generated with Claude Code

gHashTag and others added 8 commits October 19, 2026 07:15
@gHashTag @claude
t27f: deps card draws a long spec name abbreviated by a stated rule, …
418f63b 
…not refused (Refs #10728)

A shown name past 58 bytes keeps its first segment, `/...`, and the
longest run of whole trailing segments that fits; when none fits, `...`
and its last 55 bytes. Only drawn bytes are checked for quotability, and
the text size follows the drawn length.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@gHashTag @claude
t27f: deps card -- a 58-byte name is checked whole, the test that kil…
4f2e74b 
…ls a `<` bound (Refs #10728)

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@gHashTag @claude
t27f: re-seal deps_card; skill records the long-name rule, its sweep …
6d1ca80 
…and four lab traps (Refs #10728)

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@gHashTag @claude
Merge master into claude/t27f-deps-short-names (Refs #10728)
996a821 
The deps_card seal conflicted because its closure holds claw/t27f_gate.t27, changed on
master by #10745; the seal is re-saved on the lab in the next commit.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@gHashTag @claude
t27f: re-seal deps_card on the merged closure (claw/t27f_gate changed…
9bb4dfe 
… in #10745) (Refs #10728)

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@gHashTag @claude
t27f: deps card bands show the most-imported neighbours first, ties i…
4b334ce 
…n edge-list order (Closes #10762)

Each band ranks its specs by how many distinct specs import them in the
whole edge list; a repeated line counts once, equal scores keep the
edge-list order, and the subtitle states the order. Counts and +N more
are unchanged. The hidden-name check follows the ranking.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@gHashTag @claude
t27f: deps card subtitle without a colon, which ffmpeg split the grap…
7853e2e 
…h on; test every fixed text (Refs #10762)

The render of the hub card failed: drawtext read the ':' in
'each band: most imported first' as an option separator and refused
the graph. The tests compared the graph text, so they passed.
dcard_text_ok now checks every fixed text the card draws.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@gHashTag @claude
t27f: re-seal deps_card for the ranked bands; skill gets the ranking,…
51c3d29 
… its sweep and three lab traps (Refs #10762)

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This was referenced Oct 19, 2026
Merged
Merged
Merged
Merged
Merged
Merged
Merged
@github-actions

github-actions Bot commented Oct 19, 2026

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-10-19 01:55:43 UTC

Summary

Status Count
Total Open PRs 50
PRs with Failing Checks 47
PRs with All Checks Green 3
READY 2
FAILING 47
PENDING 0
NO CHECKS YET 0

These columns do not partition: 2 + 47 + 0 + 0 = 49, and there are 50 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=fad12492937a != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

@gHashTag
gHashTag enabled auto-merge (squash) October 19, 2026 01:59
@github-actions github-actions Bot mentioned this pull request Oct 19, 2026
Merged
@gHashTag @claude
Merge master into claude/t27f-deps-rank; ledger the two edge-list rea…
f5d8ab8 
…ders' exact 2^21-bit buffers (Refs #10762)

The merge keeps this branch's deps_card.t27 and its seal (master holds the
squashed #10747 text, which this branch already contains). Master's t27b
ledger already lists deps_card and why_card, which clears the t27b-native
ratchet for them.

The Verilog widths check failed because why_card.t27 (on master since
#10719) and deps_card.t27 each read the corpus edge list (136,837 bytes)
into a [262144]u8, printed as [2097151:0]. Ledgered with the arithmetic,
as suggest_feed.t27 and token_floor.t27 are.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@github-actions github-actions Bot mentioned this pull request Oct 19, 2026
Merged
@github-actions

github-actions Bot commented Oct 19, 2026

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-10-19 02:13:28 UTC

Summary

Status Count
Total Open PRs 50
PRs with Failing Checks 45
PRs with All Checks Green 5
READY 2
FAILING 45
PENDING 0
NO CHECKS YET 0

These columns do not partition: 2 + 45 + 0 + 0 = 47, and there are 50 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=fad12492937a != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

@github-actions github-actions Bot mentioned this pull request Oct 19, 2026
Merged
@gHashTag
gHashTag merged commit 4d8f1aa into master Oct 19, 2026
33 of 34 checks passed
This was referenced Oct 19, 2026
Closed
Closed
Open
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

t27f: deps card bands show the most-imported neighbours first, not edge-list order

1 participant

@gHashTag