Commit graph

5 commits

Author SHA1 Message Date
LUIS NOVO
12bd2655a4 docs: setup 2025-10-20 15:25:28 -03:00
Luis Novo
b5666c4d68
Fix/increase fix: increase API client timeouts for transformation operations timeouts (#170)
* fix: increase API client timeouts for transformation operations

- Increase frontend timeout from 30s to 300s (5 minutes)
- Increase Streamlit API client timeout from 30s to 300s
- Add API_CLIENT_TIMEOUT environment variable for configurability
- Add ESPERANTO_LLM_TIMEOUT environment variable documentation
- Update .env.example with comprehensive timeout documentation

Fixes #131 - API timeout errors during transformation generation
Transformations now have sufficient time to complete on slower
hardware (Ollama, LM Studio) without frontend timeout errors.

Users can now configure timeouts for both the API client layer
(API_CLIENT_TIMEOUT) and the LLM provider layer (ESPERANTO_LLM_TIMEOUT)
to accommodate their specific hardware and network conditions.

* docs: add timeout configuration documentation

- Add comprehensive timeout troubleshooting section to common-issues.md
- Add FAQ entry about timeout errors during transformations
- Document API_CLIENT_TIMEOUT and ESPERANTO_LLM_TIMEOUT usage
- Provide specific timeout recommendations for different hardware/network scenarios
- Link to GitHub issue #131 for reference

* chore: bump

* refactor: improve timeout configuration with validation and consistency

Based on PR review feedback, this commit addresses several improvements:

**Timeout Validation:**
- Add validation to ensure timeout values are between 30s and 3600s
- Invalid values fall back to default 300s with warning logs
- Handles edge cases (negative, zero, invalid strings)

**Fix Hard-coded Timeouts:**
- Replace all hard-coded timeout values in api/client.py
- ask_simple: 300s → self.timeout
- execute_transformation: 120s → self.timeout
- embed_content: 120s → self.timeout
- create_source: 300s → self.timeout
- rebuild_embeddings: Uses smart logic (2x timeout, max 3600s)

**Improved Documentation:**
- Add clarifying comments about ms vs seconds (frontend vs backend)
- Document that frontend uses 300000ms = backend 300s
- Add inline documentation for rebuild_embeddings timeout logic

**Development Dependencies:**
- Add pytest>=8.0.0 to dev dependencies for future test coverage

This makes timeout configuration more robust, consistent, and user-friendly
while maintaining backward compatibility.
2025-10-19 11:37:24 -03:00
LUIS NOVO
9670e3553d remove libmagic references (deprecated) 2025-10-19 09:00:40 -03:00
Luis Novo
b7e656a319
Version 1 (#160)
New front-end
Launch Chat API
Manage Sources
Enable re-embedding of all contents
Sources can be added without a notebook now
Improved settings
Enable model selector on all chats
Background processing for better experience
Dark mode
Improved Notes

Improved Docs: 
- Remove all Streamlit references from documentation
- Update deployment guides with React frontend setup
- Fix Docker environment variables format (SURREAL_URL, SURREAL_PASSWORD)
- Update docker image tag from :latest to :v1-latest
- Change navigation references (Settings → Models to just Models)
- Update development setup to include frontend npm commands
- Add MIGRATION.md guide for users upgrading from Streamlit
- Update quick-start guide with correct environment variables
- Add port 5055 documentation for API access
- Update project structure to reflect frontend/ directory
- Remove outdated source-chat documentation files
2025-10-18 12:46:22 -03:00
LUIS NOVO
b20c62df47 docs: new docs 2025-07-17 12:38:40 -03:00