Provide a standalone version

[svenk]

I like your project (have found it on https://news.ycombinator.com/item?id=23157408). It took me quite a bit time to dig throught the code to find the relevant lines in https://github.com/baicunko/scanyourpdf/blob/master/pdfwebsite/views.py

In order to basically use these two central lines (calling convert and gs), which form the algorithm, without all the django overhead, it would be nice to have a standalone version, either as python script (would be also simple to extend) or as shell script (would require to maintain several files, not so useful).

Compare probably with the CLI tools in https://gitlab.com/edouardklein/falsisign

[baicunko]

Someone on HackerNews made a Gist with the commands. I will start implement his script within the source code as well as start developing a stand-alone app for privacy-minded documents which cannot be sent to a third party. I will let update this as soon as HN traffic decreases a bit.

[DanielHabenicht]

From a web perspective view, this could also be done on the client-side: https://github.com/KnicKnic/WASM-ImageMagick

[svenk]

Interesting. That project only coveres ImageMagick. What about GhostScript in WASM?

[DanielHabenicht]

Something like this? https://ghostscript.com/~tor/wasm/readme.html

[apurvmishra99]

Hey, so I was looking into something like this myself and when I I saw this I made into a python CLI app. You guys can check it out here PDFtoScan.

Or you can directly try it out by installing it using
pip install pdf-to-scan

[rwv]

I re-wrote the logic in JavaScript and built a pure frontend site to do the scanning.

lookscanned.io