This guide provides solutions to common issues you might encounter when installing or using OARC-Crawlers.
- Virtual Environment Issues
- Installation Problems
- Dependency Conflicts
- Runtime Errors
- Platform-Specific Issues
- CLI & API Issues
- MCP & VS Code Integration
- Getting Help
A clean Python environment is essential. Use uv for best results.
# List installed packages
pip list
# Example output (minimal, clean venv)
# oarc-crawlers 0.1.2
# pip 25.0.1
# setuptools 78.1.0
# wheel 0.45.1
# Install uv if missing
pip install uv
# Deactivate any conda/venv
conda deactivate || true
.venv\Scripts\deactivate || true
# Remove all packages except pip/setuptools/wheel
pip freeze | Select-String -Pattern "^(?!pip)" | ForEach-Object { pip uninstall -y $_.ToString().Trim() }
# Re-activate and check
.venv\Scripts\activate
pip listIf you see unexpected packages or conflicts, check your Python path:
which python # Unix/MacOS
where python # Windows
# Should point to your venv, not system Python.
# If not, deactivate/reactivate or recreate the venv.- Check your internet connection.
- Upgrade pip and uv:
pip install --upgrade pip uv
- Try both
uv pip install oarc-crawlersandpip install oarc-crawlers.
- Windows: Run PowerShell as administrator.
- Unix/MacOS: Use
sudoif needed.
- OARC-Crawlers requires Python 3.10 or 3.11.
- Check your version:
python --version
- If not 3.10/3.11, create a new venv:
uv venv --python 3.11
- Always use a clean venv.
- If you see dependency errors, try:
uv pip install oarc-crawlers==<latest_version>
- Avoid mixing system and venv packages.
- Update oarc-crawlers (
uv pip install -U oarc-crawlers). - Check the video URL and regional restrictions.
- Use
--verbosefor more error details. - If you see "Video unavailable" or "No streams found", try a different video or check for age/content restrictions.
- Check your internet connection.
- Some sites block bots; try setting a user-agent:
headers = {'User-Agent': 'Mozilla/5.0'} crawler = WebCrawler(data_dir="./data", headers=headers)
- For SSL errors, update your Python and CA certificates.
- Use authenticated requests if possible.
- Wait and retry if rate-limited.
- Cache results to avoid repeated requests.
- If you get "Paper not found" or "API error", check the arXiv ID or query.
- For "Too many requests", wait and retry.
- Path length limits: keep data directories short.
- PowerShell execution policy: run
Set-ExecutionPolicy RemoteSignedif needed. - File locking: close files before re-running commands.
- SSL errors: update Python and run
Install Certificates.commandif needed. - XCode tools may be required for some dependencies.
- Install build essentials (
sudo apt install build-essential). - Use
uvto avoid system package conflicts.
- Ensure your venv is activated.
- On Windows, check that
%USERPROFILE%\AppData\Roaming\Python\Scriptsis in your PATH. - Try
python -m oarc_crawlersas a fallback.
- Use
oarc-crawlers --helpandoarc-crawlers [command] --helpfor options. - For argument errors, check the Cheat Sheet or CLI.md.
- Use correct imports:
from oarc_crawlers import YTCrawler, GHCrawler, ArxivCrawler, DDGCrawler, WebCrawler
- All async methods must be awaited inside an
async deffunction. - Use
asyncio.run(...)to run top-level async code.
- If you see "Parquet file not found" or "cannot read file", check the path and permissions.
- Use
oarc-crawlers data view <file>to inspect Parquet files.
- Ensure port 3000 is free or specify
--port. - Use
oarc-crawlers mcp run --verbosefor debug output.
- Check your MCP server config in VS Code settings:
{ "mcp.servers": [ { "name": "OARC Crawlers", "port": 3000, "transport": "ws" } ] } - Make sure the MCP server is running before starting VS Code.
- Errors are returned as structured JSON with
error.codeanderror.message. - See VSCodeMCP.md for error code meanings.
- Use
--verbosefor detailed logs. - Check docs/CHEATSHEET.md and docs/Examples.md for usage.
- For API reference: see docs/API.md.
- For unresolved issues:
- Search GitHub Issues
- Open a new issue with error details, OS, Python version, and steps to reproduce.