two column pdf wrong order

[jingangdidi]

Thank you for creating this wonderful tool!
When utilizing extract_text to extract content from a two-column PDF file, the order is incorrectly arranged, with the content on the right column of the first page is followed by that on the right column of the second page). But some two-column PDF file in arxiv no such problem.

The extracted result:

The raw pdf content:

pdf file: 2_column.pdf

[jrmuizel]

Yeah, the library makes no effort to properly assign text to columns yet. The approach I'm thinking of for this would be to gather all the text and positions and then assign columns in a second pass.