Skip to content

fix: parse single-line fenced LLM replies in instrument classifier - #47

Merged
Anai-Guo merged 1 commit into
mainfrom
chore/auto-maint-2026-10-18
Oct 19, 2026
Merged

Anai-Guo merged 1 commit into
mainfrom
chore/auto-maint-2026-10-18

Conversation

@Anai-Guo

Copy link
Copy Markdown
Owner

自动维护改动

类型:维护(小 bug 修复)
改动:discovery/classifier.classify_with_llm 剥离 markdown 围栏时 text.index("\n") 对单行围栏回复(如 ```{...}```)抛 ValueError,改为 find() 并回退为去掉三个反引号(与 analyzer / skill generator / paper-pilot client 的既有修复一致);tests/test_classifier.py 新增 4 个参数化回归用例(多行 / 单行 / 裸围栏 / 无围栏)。
测试:通过(修复前单行用例失败、修复后全过;本地全量 517 passed)
参考:无

本 PR 由每日维护任务生成,请人工 review 后再合并。

🤖 Generated with Claude Code

classify_with_llm stripped the opening markdown fence with str.index("\n"),
which raises ValueError when the model returns a single-line fenced reply
such as ```{...}```. Use find() and fall back to dropping the three
backticks, matching the fix already applied in the analyzer, skill
generator and paper-pilot client. Adds parametrized regression tests.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
@Anai-Guo
Anai-Guo merged commit e021792 into main Oct 19, 2026
3 checks passed
@Anai-Guo
Anai-Guo deleted the chore/auto-maint-2026-10-18 branch October 19, 2026 10:29
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