pdf2info

Pdf2info is a simple table extractor library using tabula.

Installation

Choose one of following:

Local

Using local python3 interpreter:

$ python3 -m pip install -r requirements

Virtualenv

Creating a new virtual environment:

Assuming $ python3 -m pip install virtualenv :

$ python3 -m pip install virtualenv
$ virtualenv venv
$ source venv/bin/activate
(venv) $ python3 -m pip install -r requirements.txt

Conda

Creating a conda environment (assumg you have conda/miniconda installed) :

$ conda create --name venv --file requirements.txt

How to use

Launch extraction

Place all your necessary PDFs in a single dicrectory, then call process_folder.py script:

$ python tables_from_dir.py --dir=path/to/your/dir --out=path/to/out/folder

This will create one csv file per table.

Logging can be read in results.log file. If need to check console live log, add --log-console param:

$ python tables_from_dir.py --dir=path/to/your/dir --out=path/to/out/folder --log-console

If you only need to extract from one pdf, use tables_from_file.py instead of tables_from_dir.py

Recreate tab2know results

Check analysis/tab2know_tests istructions.

Recreate pdf2info results

Check analysis/pdf2info_tests istructions.

Comparison

camelot:

tabula:

linesearch:

tab2know: