-
Notifications
You must be signed in to change notification settings - Fork 1
Troubleshooting
Run Developer: Reload Window, then reopen the PDF. Confirm the installed extension is named Inleaf Reader.
Run packaging commands from the repository, not from your reading folder:
cd /path/to/inleaf-reader
npm run compileIf packaging fails, the later installation command will also fail because no VSIX was created.
The PDF probably lacks a text layer. Perform OCR first and open the resulting PDF. A scanned page image alone cannot provide selectable text.
OCR text may differ from the visible page. Open the annotation editor and correct Original text. Header, footer, page-number, and figure exclusion is heuristic and can vary between PDFs.
Run Inleaf Reader: Diagnose Translation Setup.
- ECDICT requires no Python and should work offline for English words.
- Argos requires a compatible Python executable and installed language model.
- LibreTranslate requires a reachable endpoint.
- DeepSeek requires a configured API key and network access.
- Keep VS Code and Inleaf Reader updated.
- Test whether the problem occurs only with a very large scanned PDF.
- Close duplicate reader tabs for the same large document.
- Report repeatable cases with the PDF page count, approximate file size, operating system, VS Code version, and steps to reproduce.
Look for .inleaf-reader beside the PDF and check the matching
*.annotations.json and .bak files. If the PDF was moved, reopen it once so
the recovery index can locate missing sidecars. Recovery never overwrites an
existing destination file.
If the issue persists, open a GitHub issue.