Repository navigation
Conversation
This branch has not been deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Make Yesterday/daily reports and day digests honor the configured note language on macOS and Windows/Linux. For example, Russian notes with an English interface now request Russian prose on every generation pass, including truncation retries and post-model session additions.
Daily evidence has no single transcript, so the interface language supplies the fallback for Match Transcript. JSON keys, status values, source IDs, parser-required headings and quoted evidence retain their canonical form. Saved reports and original evidence are not rewritten.
Validation
testDreamRequestsKeepRussianAfterOutputLimitAndParseTheReplyandtestDigestLanguageReachesEveryPassAndRetryAndSeparatesCachedLanguages; restoring it passes both. Tests inspect all passes and retries and exercise parsed responses/cache separation.runtime::tests::healthy_cuda_helper_is_reclaimed_before_a_failed_probefails withGPU is busy; an isolated copy of clean masterdd49d90reproduces the exact same failure. This is not presented as a fully green suite.cargo test -p lokalbot-ai --lib --locked --jobs 3 --no-fail-fast. The app build also exited 0 (1m 08s). SHA verification covered all 872 source files for the combined corporate-call commit25803918plus this language commit27d4e581(six non-overlapping Rust files).cargo fmt --all -- --checkandgit diff --checkpassed.Still required before Ready:
GitHub Actions currently reports
action_requiredfor all five workflows: maintainer authorization is required to run workflows from this fork. No upstream CI pass is claimed.Changed test expectations
None. Three existing fake TextEngine implementations now recognize the original system-prompt prefix rather than requiring exact equality with the unadorned base prompt. The old routing silently selected the wrong fake response once the language directive was appended; all existing expected report contents/assertions are retained.
Linked issues
None.
Risk / rollout notes
Draft: not ready for merge. The change touches scheduled and explicit daily generation, digest caching and model-free fallbacks on both implementations. Canonical parser/schema tokens remain unchanged. A model can still ignore an instruction; the real-model smoke test is outstanding.
Custom note languages and languages without a UI catalog use the requested language for model-generated prose and the interface catalog for deterministic fallback labels. Proper names, quoted evidence and stored actions remain verbatim. Existing English reports require explicit regeneration. Rollback is code-only; no saved report or library migration is performed.