If like me you were wondering if MS actually provided their own parsers for their Office file formats… they did not.
It seems to just be a bunch of random pyxyz 3rd-party support libraries all mashed together.
What do you mean by parser? Office docs are just zipped XML files. They are trivial to parse. The hard part is all the quirks the document renderers have, which makes it impossible to perfectly match the output. But markdown can’t handle any complex formatting anyway
Maybe the people that wrote their parser have left the company? Typical big software corp problem.
I mean, the parser would still be there even if the people left the company, right? The source code remains.
It might also be somewhere, but nobody knows where.
Reading through the source code, it’s more of a repackaging of other open source libraries, probably for its AI effort.
