Refactor pdf2txt.py into an entry point

[lrq3000]

It is impossible to use pdf2txt.py, the main PDF 2 Text script of the package, from a Python function without calling a shell. This adds additional unnecessary overhead in addition to encoding issues.

This could all be easily avoidable by refactoring pdf2txt.py as an entry point, as described here and here.

[pietermarsman]

This would have to look something like this:

entry_points = {'console_scripts': [
    'pdf2txt=pdfminer.pdf2txt:main',
    ...
]},

https://github.com/pdfminer/pdfminer.six/issues/27#issuecomment-274097265

[pietermarsman]

I realized that this is not as easy as I thought. The console script needs to be in the pdfminer package, but it is not. It is in the tools/ directory.

Mayb ewe should move the content of tools/ to the package as well and just keep some aliases there to not have any breaking changes?

[pietermarsman]

I also realize now that this question already has a perfectly valid solution. To use pdfminer from a Python function (or anywhere) you can use the high-level api:

from pdfminer.high_level import extract_text

text = extract_text("example.pdf")
print(text)

All the features from the command-line are also available throught extract_text and related methods.

[pietermarsman]

The request to add pdf2txt as an entrypoint is also made in issue #724, but it is slighly more on-point so I'm keeping that one open to track that request.