cleader - CLI READER

Store web articles as plain text. No more ‘saved as’ HTML (aka cURL:d) or
‘printed as PDF’, just extracted content in markdown, without all annoying markup.

FEATURES

Uses STDOUT, easy to process further:

python cleader.py http://iamnearlythere.com/what_i_want_from_an_api/ | wc -c

Save to file:

python cleader.py --save http://iamnearlythere.com/what_i_want_from_an_api/

creates the file ‘what_i_want_from_an_interface.md’

Extract only the article’s HTML:

python cleader.py http://iamnearlythere.com/what_i_want_from_an_api/ | markdown_py > what_i_want_from_an_api.html

This example requires ‘pip install markdown’ and is almost what html2text
does, but, again, cutting away all the cruft that’s not the main article’s content.

Very create-your-own-library friendly, subjective example coming up:

echo 'function save() { python cleader.py "$1" --save=~/favorites }' >> ~/.bashrc && . ~/bashrc

After that setup, this CLI call:

save http://iamnearlythere.com/what_i_want_from_an_api/

would create ~/favorites/what_i_want_from_an_interface.md for you
to keep around forever and ever.

… which could easily be turned to a nice UNIX:y archive system:

• listing the latest ‘favorites’: ls -t1 ~/favorites | head -n 5
• weighted search for ‘css’: grep -ic css ~/favorites/* | head -n 5
• showing extracts: head -n 10 ~/favorites/* | less

DEPENDENCIES

• pip install -r requirements.txt
• A (free) Readability API key, from http://www.readability.com/developers/api/parser

QUESTIONS

Read or post issues at https://github.com/chelmertz/cleader

Email [email protected]