Need to know : how to use this script

[HarshPrajapati2309]

I am trying to use this script but couldnt find the way

[jbarlow83]

Did you follow the instructions in the README?
If that succeeds, ocrmypdf and this plugin will be installed.

[sliedes]

I think the question may be about how to use this with OCRmyPDF; i.e. yes, it's installable, but how to run OCRmyPDF with it without running into NotImplementedError: EasyOCR does not support hOCR output.

And I suspect the answer may be to use the OCRmyPDF option --pdf-renderer sandwich?

[jbarlow83]

Yes, you need to use --pdf-renderer sandwich.

[sliedes]

BTW, I made a PR to EasyOCR that adds hOCR output (though didn't test it with this plugin): JaidedAI/EasyOCR#1275

[jbarlow83]

@sliedes Can you contribute your hOCR change here since EasyOCR doesn't seem to merge many PRs?

[sliedes]

@sliedes Can you contribute your hOCR change here since EasyOCR doesn't seem to merge many PRs?

Certainly! Do you need me to open the PR here, or can you just pick it from the EasyOCR repository?

[jbarlow83]

Please open a PR here - that would be great, thank you!

[sanjay-nit]

Even after using the --pdf-renderer sandwich option, when I copy text from the converted PDF and paste it somewhere, it outputs garbled text.

 €(.|'; 0 #^| ?|*)+ 0 "^| 6&: 9@ ,/.$_-0'; €, {6 +//:'6+. 6$|6( &.+, > &/&< ,//|-"$  60) 0 ,|*)+$ ') +| '|*)+'$ €'"-'*0); && +0/06* +0 #^| €"+.:) '*&6:*, $; *$*$; :€*..)9 *^:*..* €"-3/:' +|&*€)+ 0/|$< 6( *^| +|)*,

[jbarlow83]

Even after using the --pdf-renderer sandwich option, when I copy text from the converted PDF and paste it somewhere, it outputs garbled text.

Open a new issue, provide your input file, command line arguments, and the program you are using to extract text.

Otherwise is it not possible to work out what is going wrong.

[alaminkouser]

Set language

ocrmypdf.ocr(os.path.join(save_path, "input.pdf"), os.path.join(save_path, "output.pdf"), force_ocr=True, pdf_renderer="sandwich", language="eng")