Server install check; one meaning for spread and filter - #901
Merged
Merged
Conversation
`install` created generated sources (`clock`, recipes) while resolving imports, before checking the program's exports and remaining imports. A program rejected for a taken export name or an unknown import left those sources installed until a later drop. Now every check runs first. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The two backends disagreed on two corners of the language: - A bare `$n` in a projection whose row is not a tuple. The row backend spliced a list's elements (so the tuple's arity varied with the data) and panicked on a scalar; the columnar backend kept either as one field. Both now splice a tuple's fields and keep any other value as one field, which is the only layout a column can hold. - A filter predicate that is not an `Int`. The row backend dropped every row; the columnar backend panicked on an opaque mask error. Both now reject it, naming the shape they got. `Spread` is only produced inside projection tuples, so the row evaluator's list-intro splice goes too. Adds `spread_values` to the corgi/vec gate, and a test that both backends reject a tuple predicate. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Two fixes, one commit each.
Server: check a program before installing its generated sources.
installcreated generated sources (clock, recipes) while resolving imports, before checking the program's exports and remaining imports, so a program rejected for a taken export name or an unknown import left them installed. Every check now runs first. New test:tests/server_install.rs.DDIR: one meaning for spread and filter across backends. The backends disagreed on two corners of the language:
$nwhose row is not a tuple: the row backend spliced a list's elements (arity varying with the data) and panicked on a scalar; corgi kept either as one field. Both now splice a tuple's fields and keep any other value as one field.Int: the row backend dropped every row; corgi panicked on an opaque mask error. Both now reject it, naming the shape.Spreadis only produced inside projection tuples, so the row evaluator's list-intro splice goes too. Addsspread_valuesto the corgi/vec gate and a test that both backends reject a tuple predicate; both new tests fail on master-next.Validation: workspace tests pass; AoC 2023 passes 33/33 on both vec and corgi; clippy warning count unchanged.
🤖 Generated with Claude Code