Download the PDFs from the releases page.
I started in 2015 from @gnijuohz's repo, but now (in 2018) I've re-written pretty much every part of the process.
docopt
requests
& requests_cache
lxml
First, find out a "topic url" for what you want to download. Eg:
https://www.geeksforgeeks.org/tag/samsung/
https://www.geeksforgeeks.org/category/dynamic-programming/
Create a JSON containing links of all posts on that topic
python3.6 list_links.py https://www.geeksforgeeks.org/tag/samsung/
This JSON can now be edited by hand, to remove some links, re-order them etc.
Now fetch the actual posts
python3.6 download_html.py JSON/Samsung.json
Finally, convert the HTML to a PDF using Pandoc
python3.6 html_to_pdf.py HTML/Samsung.html
Things will work only if you're really lucky. This project has taught me how fragile my HTML to PDF pipeline really is. There's just too many things that can go wrong.
List of Topic URLs that have I've fetched. You can download these from the releases page.
Algorithms
https://www.geeksforgeeks.org/category/algorithm/divide-and-conquer/
https://www.geeksforgeeks.org/category/algorithm/combinatorial/
https://www.geeksforgeeks.org/category/algorithm/randomized/
https://www.geeksforgeeks.org/category/algorithm/game-theory/
https://www.geeksforgeeks.org/category/algorithm/pattern-searching/
https://www.geeksforgeeks.org/category/algorithm/branch-and-bound/
Data Strucutres
https://www.geeksforgeeks.org/category/data-structures/hash/
https://www.geeksforgeeks.org/category/data-structures/matrix/
https://www.geeksforgeeks.org/category/data-structures/heap/
https://www.geeksforgeeks.org/category/data-structures/stack/
https://www.geeksforgeeks.org/category/data-structures/queue/
https://www.geeksforgeeks.org/category/data-structures/linked-list/
https://www.geeksforgeeks.org/category/data-structures/c-strings/
https://www.geeksforgeeks.org/category/data-structures/c-arrays/
https://www.geeksforgeeks.org/category/data-structures/binary-search-tree/
Companies
To Compare more of the gfg scrapper