Repository navigation
Count every cut section on the episode page - #26
Merged
Merged
Conversation
The Ready card, timeline header and What gets cut card counted only the repeats the detector found. They now count what the timeline lists as cut: repeats not kept, hand cuts, missing-from-transcript cuts not kept, and phrase cuts, each counted on its own as the timeline shows them. Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_01U2ixfGbcMQxvMnRGjVYzda
dben
pushed a commit
that referenced
this pull request
Oct 25, 2026
With main's fix (#26) merged, the timeline and the Ready card say 7 sections for the hero episode. The captions shot now pauses part way into a line, so the line being said is highlighted. Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_015LKxkNWZNzhFwjch7F2M6A
dben
added a commit
that referenced
this pull request
Oct 25, 2026
* README as a getting-started guide, with new screenshots - Retake the screenshots from made-up podcasts run through the current app: the timeline (intro, ad breaks, a section cut by hand, outro), the home page, the episode list with a selection, and new ones for the Filters, Settings and Private feed tabs, captions and the phone layout. - Rewrite the README around installing and using pod-block: adding a podcast, downloading, checking the cuts, listening, filters, show settings and phrases, captions, and sharing. - Move block types, how repeats are found, speed and the database to a new HOW-IT-WORKS.md, and point CLI.md and WEBAPP.md at it. - Tone pass over CLI.md, WEBAPP.md and the deployment and whisper READMEs: rewrap uneven lines, untangle the Retention bullets, and correct how long captions outlast clean-up. Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_015LKxkNWZNzhFwjch7F2M6A * Screenshot scripts under docs, and a timeline with every kind of cut - docs/screenshots/make/ retakes the README screenshots in one command (run.sh): made-up feeds, a stand-in transcriber, seeding the app, and the Playwright shots, about six minutes end to end. - The hero show now has a segment theme (a plain Repeat), an ad read its feed's transcript leaves out (Not in transcript), and an outro cut at "This was a production of Podblock Studios" (Outro (phrase)). - Captions and Settings shots move to that show, so they show speaker letters, the transcript check and the outro phrase. Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_015LKxkNWZNzhFwjch7F2M6A * WEBAPP.md with contents, features before design, and a checked future list - Add a contents list, and put the features first: each one (captions, phrases, the transcript check, chapters, the episode page, show settings, the Jobs, Downloads and Timings pages) gets its own section instead of a long bullet under Processing, with labelled bullets. - Move Decisions, Stack, How it fits together, Code layout, Data model, Routes and the UI rules under Design. - Future features, checked against the code and grouped by area; note what's already handled for bumpers and where captions' words are used. - Fixes found on the way: 21 migrations, not 15; original_sha in the data model; GET /episodes/{id}/transcript and POST /downloads/reprocess in Routes; HOW-IT-WORKS.md's breaks use the show's longest ad break (8 minutes by default, not 10) and handle false-start stings. Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_015LKxkNWZNzhFwjch7F2M6A * Retake the screenshots with every cut counted With main's fix (#26) merged, the timeline and the Ready card say 7 sections for the hero episode. The captions shot now pauses part way into a line, so the line being said is highlighted. Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_015LKxkNWZNzhFwjch7F2M6A --------- Co-authored-by: Claude <[email protected]>
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
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.
The section counts on the episode page (Ready card, timeline header, What gets cut card) counted only
report.sections. They now count what the timeline lists as cut.processing.cut_count()counts fromprocessing.timeline()'s output, so the count and the timeline come from the same data: repeats except kept ones, hand cuts, phrase cuts, and missing-from-transcript cuts that aren't kept. Overlapping cuts are counted separately, the same way the timeline lists them.cut_countto the template, so the Ready and What gets cut cards get it even when there's no timeline.tests/test_podcast_pages.py: an episode with 4 found sections, a kept repeat, a hand cut over a break, a missing cut, a kept missing stretch and an outro phrase cut shows 7 in all three places. It fails onmain.uv run pytest -m "not slow"(190 passed) anduv run ruff check .pass.Generated by Claude Code