This is a fork of the pdfminer tool, with a specific focus on extracting semantic XML out of OCR-ed PDF.
It extracts pdf content page by page, and also identifies words and lines using distinct tags.
python lc_setup.py install
You can also install it within a virtualenv.
python lc_pdf2txt.py
Provides various options, of interest to us are XML specific options which have been added:
-B make_brief
Which disables character level font glyphs if that is too verbose for you.
-t xml
Outputs XML
lc_pdf2txt.py -B -t xml -o test.xml ./akn_mu_act_1923-10-13_act_14-1923_eng_main.pdf
will convert akn_mu_act_1923-10-13_act_14-1923_eng_main.pdf to test.xml.
We typically don't need character level font-glyphs