I have a doc with 2 tables. I converted it to docx (with Word) and then trying to use pandoc to convert it to md.

I got some strange error, and when looking at the native I found this:

pandoc -t native others/Mirrors-Lens.docx -o others/Mirrors-Lens.native is kind of normal
pandoc -t markdown others/Mirrors-Lens.docx | pandoc -f markdown -t native -o others/Mirrors-Lens-round.native is hugely different from the above

One of the main difference is that both tables (originally multi-columned) collapsed to 1 column only in the later round-trip. (Another annoyance to me is I actually tried to use my pantable2csv filter to capture the table from the docx directly into csv but resulted in error.)

For the 2nd table, I’m guessing the existence of <, > causes problems. For the 1st I have no clue since the native looks fine to me.

But since I do not own the copyright of the file, I am not going to share this publicly (isn’t really important but just in case…). If anyone is interested in help solving the puzzle / debug pandoc, please give me your github account and I can open a private repository and invite you. Thanks!