Repository navigation
spec(t27f): first real reel -- the t27 basics course as a 27-cell map (Closes #9996) - #9997
Merged
Merged
Conversation
…Closes #9996) specs/t27f/reel_t27_basics.t27 draws one real topic with the reel camera and the deck look: the 27 lessons of gHashTag/trinity's t27 basics course, in order, one reel row per three course modules. A cell's status is earned from two measurements (lesson page live on t27.ai, lesson spec passes on the lab), and the slide text of all five demo scenes is checked against deck.t27 for every preset. Runs under both t27c and t27b with the same verdict on every test. t27b refused the first draft for comparing a str's u64 .len with an i32; lengths are cast now. Refs gHashTag/trinity#1677 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
gHashTag
enabled auto-merge (squash)
October 17, 2026 15:41
Contributor
This was referenced Oct 17, 2026
Merged
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Closes #9996
Refs gHashTag/trinity#1677
What
specs/t27f/reel_t27_basics.t27: the first real T27F reel. One topic, the t27 basics course (gHashTag/trinityapps/website/specs/course/t27-basics.t27atee51235), drawn with the reel camera (reel.t27) and the deck look (deck.t27).lenses_are_whole_modules).a_status_is_earned_not_chosen): a cell is published only when its lesson page is live AND its lesson spec passes; verified when only the spec passes; draft otherwise. Measured 2026-10-17: all 27 pages att27.ai/learn/<id>/return 200 with the right title (a made-up id returns 404), and all 27 lesson specs pass on the lab. 23 of those 27 specs report vacuous passes because their asserts are on constants; a planted wrong assert in lesson 11 turns it red, so the checks are real.deck.t27on every preset (every_slide_fits_every_preset). One headline line per slide: the stage strip holds only one on wide and square (text_fits(2, 0, STAGE, wide)is false), and the viewer wraps titles greedily, so a headline that never wraps is the only one drawn the same everywhere.the_camera_is_the_demo_camera):t27f_viewdraws the reel.t27 demo choreography, so lens 0 and focus cell 10 (lesson 11) are pinned.Two compilers: t27c and t27b
The new spec,
reel.t27anddeck.t27were run under t27c (Zig backend) and under t27b (t27's own AArch64 backend,t27b test --check, qemu-aarch64 on the Railway lab).Same verdict on every test. t27b also refused the first draft: it compared a str's u64
.lenwith an i32, which t27c accepted. Every length is now cast to i32 before the compare.The t27b binary used was built 2026-10-10 from master. The t27c binary was built from 90fc25b.
Negative controls (both compilers)
a_status_is_earned_not_chosenevery_slide_fits_every_presetevery_slide_fits_every_presetRestoring the file returns both compilers to 6/6.
Stills
t27f_viewdrew all 15 stills (5 scenes x wide/square/tall) from this text and refused none. On the split slides every row is drawn whole on one line, and the glass card is at full opacity.One finding: a viewer process started before t27-mcp
07e2c7c(STILL_CARD_ALPHA, gHashTag/t27-mcp#4) still draws split stills with the card at opacity 0. Restart the MCP server after updating it.Not claimed
No GIF or MP4 is rendered or posted; these are stills for the chat. CI:
cli-triandt27b-native-ratchetare red on master itself (cli-tri: #9634, fix #9639), so they are not signals for this diff.🤖 Generated with Claude Code