docs(pages): document local-model setup, tool-calling requirement, and LLM timeouts#11
docs(pages): document local-model setup, tool-calling requirement, and LLM timeouts#11chethanuk wants to merge 3 commits into
Conversation
|
Caution The consumer version of Gemini Code Assist on GitHub has been sunset. All code review activity has officially ceased. |
|
Warning Review limit reached
Next review available in: 48 minutes Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available. How can I continue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews. How do review limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please refer docs for additional details. Review details⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (7)
📝 WalkthroughWalkthroughDocumentation in English, Japanese, and Chinese adds Ollama custom-provider setup, HTTP timeout configuration, and troubleshooting guidance for models without native tool-calling support. ChangesOllama documentation
Estimated code review effort: 1 (Trivial) | ~5 minutes Suggested reviewers: 🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 3
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@pages/src/content/docs/en/configuration.md`:
- Around line 108-113: Update the Ollama timeout examples in
pages/src/content/docs/en/configuration.md (108-113),
pages/src/content/docs/ja/configuration.md (106-112), and
pages/src/content/docs/zh/configuration.md (100-105) consistently: either add
valid non-empty model and api_key fields to each complete provider object, or
explicitly state that only timeout_sec should be merged into an existing
provider configuration.
In `@pages/src/content/docs/en/faq.md`:
- Around line 58-61: Update the diagnostic log fences in
pages/src/content/docs/en/faq.md lines 58-61, pages/src/content/docs/ja/faq.md
lines 58-61, and pages/src/content/docs/zh/faq.md lines 53-56 to use the text
language identifier on each opening fence.
In `@pages/src/content/docs/ja/configuration.md`:
- Around line 88-89: Update the “No tool calls parsed” links to use stable
fragments matching the generated headings or explicit stable IDs. Apply this to
pages/src/content/docs/ja/configuration.md lines 88-89 and
pages/src/content/docs/zh/configuration.md lines 84-84; also correct the
self-link fragments in pages/src/content/docs/ja/faq.md lines 199-201 and
pages/src/content/docs/zh/faq.md lines 185-186.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: fdf436bb-dd3a-43b7-ae57-f9ce2dfbae61
📒 Files selected for processing (6)
pages/src/content/docs/en/configuration.mdpages/src/content/docs/en/faq.mdpages/src/content/docs/ja/configuration.mdpages/src/content/docs/ja/faq.mdpages/src/content/docs/zh/configuration.mdpages/src/content/docs/zh/faq.md
| ``` | ||
| [ocr] No tool calls parsed for src/foo.go, retrying... | ||
| [ocr] Max tool requests reached for src/foo.go. | ||
| ``` |
There was a problem hiding this comment.
📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick win
Add text language identifiers to all diagnostic log fences.
pages/src/content/docs/en/faq.md#L58-L61: change the opening fence to```text.pages/src/content/docs/ja/faq.md#L58-L61: change the opening fence to```text.pages/src/content/docs/zh/faq.md#L53-L56: change the opening fence to```text.
🧰 Tools
🪛 markdownlint-cli2 (0.23.0)
[warning] 58-58: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
📍 Affects 3 files
pages/src/content/docs/en/faq.md#L58-L61(this comment)pages/src/content/docs/ja/faq.md#L58-L61pages/src/content/docs/zh/faq.md#L53-L56
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@pages/src/content/docs/en/faq.md` around lines 58 - 61, Update the diagnostic
log fences in pages/src/content/docs/en/faq.md lines 58-61,
pages/src/content/docs/ja/faq.md lines 58-61, and
pages/src/content/docs/zh/faq.md lines 53-56 to use the text language identifier
on each opening fence.
Source: Linters/SAST tools
| ["No tool calls parsed"（ローカルモデル / Ollama）](../faq/#no-tool-calls-parsed-ollama)を | ||
| 参照してください。 |
There was a problem hiding this comment.
🎯 Functional Correctness | 🟡 Minor | ⚡ Quick win
Use stable anchors for localized “No tool calls parsed” links. The current fragments do not consistently match headings containing localized model labels.
pages/src/content/docs/ja/configuration.md#L88-L89: replace the fragment with the exact generated slug or an explicit stable ID.pages/src/content/docs/zh/configuration.md#L84-L84: replace the fragment with the exact generated slug or an explicit stable ID.pages/src/content/docs/ja/faq.md#L199-L201: fix the self-link fragment.pages/src/content/docs/zh/faq.md#L185-L186: fix the self-link fragment.
📍 Affects 4 files
pages/src/content/docs/ja/configuration.md#L88-L89(this comment)pages/src/content/docs/zh/configuration.md#L84-L84pages/src/content/docs/ja/faq.md#L199-L201pages/src/content/docs/zh/faq.md#L185-L186
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@pages/src/content/docs/ja/configuration.md` around lines 88 - 89, Update the
“No tool calls parsed” links to use stable fragments matching the generated
headings or explicit stable IDs. Apply this to
pages/src/content/docs/ja/configuration.md lines 88-89 and
pages/src/content/docs/zh/configuration.md lines 84-84; also correct the
self-link fragments in pages/src/content/docs/ja/faq.md lines 199-201 and
pages/src/content/docs/zh/faq.md lines 185-186.
Source: Linters/SAST tools
Two changes per locale (en/zh/ja), addressing alibaba#234: - New "No tool calls parsed" entry under Configuration & startup: the symptom loop, the durable rule that the model must support native tool calling (deepseek-r1 narrates calls in content and can never work; qwen3 works), the Ollama tools-tag search link, and the maintainer's curl snippet to verify a model emits structured tool_calls without OCR in the loop. - The existing "Max tool requests reached" entry (where users actually land) gains a 4th cause bullet cross-linking the new entry. Anchors follow generateHeadingId (pages/src/utils/headingId.ts), the site's actual slugger, and were verified against the rendered DOM in all three locales. Code blocks are byte-identical across locales per i18n convention; heading counts stay in parity.
Two additions per locale (en/zh/ja), addressing alibaba#234: - Custom providers: a copy-paste Ollama example (127.0.0.1:11434/v1, protocol openai) with the note that custom providers require a non-empty api_key placeholder (resolver has no env fallback for them) and a pointer to the FAQ tool-calling rule. - New Timeouts subsection: providers.<name>.timeout_sec / llm.timeout_sec / OCR_LLM_TIMEOUT, the 300s default, and the caveat that timeout_sec is not supported by 'ocr config set' (config_cmd has no timeout handling) so config.json must be edited directly. The ja Timeouts heading is タイムアウト（Timeouts） so the site slugger (which strips katakana) still yields a linkable #timeouts anchor. Code blocks byte-identical across locales; heading parity kept.
marked percent-encodes non-ASCII hrefs (#超时 renders as #%E8%B6%85%E6%97%B6), but heading ids are raw text from generateHeadingId, so handleContentClick's getElementById never matched for CJK anchors: same-page clicks silently no-oped and cross-page anchor scrolls exhausted their retries at the top of the page. This affected every pre-existing zh in-page anchor (e.g. faq 复用已有的环境变量) as well as the zh links added for alibaba#234. Decode the fragment (with a malformed-input guard) at both lookup sites.
1a673f9 to
9172a34
Compare
1 participant
Fork-side review PR for upstream alibaba#234 (local-model docs). Two commits so either half can drop:
en/zh/ja faq.md): new"No tool calls parsed" (local models / Ollama)entry + 4th cause bullet in the existing "Max tool requests reached" entry.en/zh/ja configuration.md): Ollama custom-provider example + new Timeouts subsection (timeout_secknobs,OCR_LLM_TIMEOUT, 300s default,ocr config setcaveat).Verification (branch-local)
npm run typecheck— cleannpm run build— compiles; 2 warnings, byte-parity with main's 2 pre-existing bundle-size warnings (zero new)generateHeadingId(strips katakana, single-hyphen runs), not github-slugger; links written to match it###heading count 27/27/27 (faq), 8/8/8 (configuration)https://ollama.com/search?c=tools→ 200git show --stat: commit A touches only 3 faq.md, commit B only 3 configuration.mdSide-finding (pre-existing, not touched): several existing in-page links use github-slugger anchors (e.g.
#filtering--rules) that don't match the site renderer's ids — broken on the docs site today.Summary by CodeRabbit