I am too lazy to sort my scanned mail by hand, so I automated it. It can be used as a cron job to add ocr information to scanned pdfs in a specific folder and to sort the pdfs according to predefined keywords.
This is just a private project. Please back up your data regularly. I cannot guarantee that some bugs or misusage corrupt or delete your data.
The package needs tesseract-ocr to be installed:
sudo apt-get install tesseract-ocr
To install the mailcategorizer we execute:
git clone git@github.com:sandro-elsweijer/mailcategorizer.git
cd mailcategorizer
pip install .
We can define the directory, where the scan pdfs are located in the config.yml.
We also define, into which folder the files should be sorted.
Lastly, we define the categories for our pdfs. The key value is the name of the folder the pdfs of this category get sorted into. The value is a semicolon-separated list of all keywords, which have to appear in the pdf.
pdfs which match in multiple categories are saved into a dedicated multiple_matches folder. pdfs with no match are saved into the no_matches folder.
When calling the mailcategorizer, we have to provide it with the location of the config file:
mailcategorizer path/to/config.yml
I suggest to automate this process with a cron job, so every time a document is scanned, it gets sorted automatically.