https://github.com/robertknight/pdf-text-checker
A small web app to check whether a PDF has selectable text
https://github.com/robertknight/pdf-text-checker
Last synced: 2 days ago
JSON representation
A small web app to check whether a PDF has selectable text
- Host: GitHub
- URL: https://github.com/robertknight/pdf-text-checker
- Owner: robertknight
- License: bsd-3-clause
- Created: 2020-10-03T10:31:52.000Z (over 4 years ago)
- Default Branch: main
- Last Pushed: 2020-10-04T09:35:16.000Z (over 4 years ago)
- Last Synced: 2025-01-09T15:43:09.327Z (4 months ago)
- Language: JavaScript
- Homepage:
- Size: 10.7 KB
- Stars: 0
- Watchers: 3
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# PDF text checker
This is a small web application that retrieves a PDF from a URL and
checks whether it has selectable text that can be extracted by
[PDF.js](https://mozilla.github.io/pdf.js/)
A PDF that has selectable text is suitable for use with tools like [Hypothesis](https://hypothes.is).
PDFs without selectable text will require pre-processing with an
[OCR](https://en.wikipedia.org/wiki/Optical_character_recognition) tool.
**[➡️ View live demo](https://robertknight.github.io/pdf-text-checker/check-pdf-text.html)**
## Usage
Start a local HTTP server with:
```
python3 -m http.server 5002
```
Then browse to http://localhost:5002/check-pdf-text.html. Enter the URL
of a PDF to fetch and check and click "Check PDF".
## Notes
This tool relies on a proxy server to bypass cross-origin restrictions on fetching
URLs. It currently uses https://cors-anywhere.herokuapp.com for this purpose.
Checking whether a PDF's text layer is usable for the purposes of copying and
annotating text is not a binary classification. A PDF may have selectable text
that is garbled or otherwise problematic. This tool uses very basic heuristics
and the result does not guarantee that the PDF will have "usable" text.