Skip to content

Render review questions through in2lambda.draft.export.render - #20

Merged
peterbjohnson merged 5 commits into
mainfrom
wb/t25
Sep 21, 2026
Merged

peterbjohnson merged 5 commits into
mainfrom
wb/t25

Conversation

@peterbjohnson

Copy link
Copy Markdown
Member

package.render looks for in2lambda.draft.render and raises RenderUnavailable when it is absent, so every review prints in2lambda render is not there yet and names questions by source lines. The pinned in2lambda has the function at in2lambda.draft.export.render(draft, output_dir) -> list[Path]: it writes one PDF per question, named question_000_Question_1.pdf and so on, compiled as the PDF generator compiles them, and needs pandoc and xelatex.

Call it. In package.py, render(draft, out_dir) calls in2lambda.draft.export.render, maps each written PDF back to its question key (q1, q2, ... in the order the set holds them, which is the order Set.problems() gives), and returns that mapping; delete RenderUnavailable and the not there yet message. Where xelatex is not installed, in2lambda's own error is reported as the render stage line and the review carries on with not rendered, as today. Update tests/test_pipeline.py's review tests to assert that a review lists each question's PDF path, with the compile skipped where needs_compiler says so, and remove docs/how-it-works.md's and README.md's not there yet passages, describing the render line as 2 questions to OUT/render. Done when in2lambda-agent run tests/fixtures/... --review sample on a machine with xelatex prints PDF paths in the listing, and the docs test passes.


Workbench ticket t25.

@peterbjohnson
peterbjohnson merged commit 48b1e99 into main Sep 21, 2026
1 check passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant