This script allows OCR'ing documents and moving them according to information in the document.

I created it to split a large PDF and move the seperated pages according to the information in the PDFs.

It looks for things like "Herr Vorname mittlerer Name Nachname" and moves them to done/Herr Vorname mittlerer Name Nachname.pdf. Also works with Frau Soundso Nachname.

If a file by the name already exists, "-0" is automatically appended. If this already exists, "-1" will be appended and so on.

Files whose names cannot be determined will be moved to the manual folder.

You need to install Tesseract Version 5.0.0-alpha (this can be done automatically with https://github.com/NormanTUD/UsefulFreeHomeServer), you also need GhostScript, ImageMagick and pdftk.

Create a folder called todo and put one or many PDFs in that folder. Then run

perl parse.pl --todo=/path/to/TODO --tmp=/path/to/TMP --manual=/path/to/MANUAL --done=/path/to/DONE --crop=4000x240+260+2200

If no paths are specified, it looks in the same folder where the script itself is in.

The --crop parameter can be left empty, but this is not recommended. It crops the image of the PDF before OCR, so that

• OCR is much faster

• OCR recognition is much more reliable

The crop parameter gets passed directly to ImageMagick when set. It is specified this way:

${WIDTH}x${HEIGHT}+{YSTART}+{XSTART}

and should be set according to your document.

Assuming you try to get German-like names, you can use the check.pl in whatever folder you are in. It will check for names that the script may have gotten wrongly. Use it with perl check.pl.

You can also use the missing.pl to get documents that are missing.

Both of these will not yet work with --todo, --tmp and --done and assume that you did not use any special paths.