Can't seem to convert anything properly

[Orion Invictus]

I'm trying to convert a couple scientific articles into the AZW3 format for my Kindle Scribe (first generation, or whatever they're calling it). I've been using the Scribe as a study tool, but reading in plain PDF format doesn't allow me to use all its features. These two articles are a sort of test to see if Calibre can meet my needs, seeing as how Amazon's own conversion tool can't.

One of the articles always fails with an "index out of bounds" error when trying to parse headers (maybe it doesn't match the regex?), while the other is seemingly defining different margins for each line.

The problem with the second article's conversion is thus:

Basically, it
does exactly what you're read-
ing here, where you can't exactly tell when a li-
ne is going to
end because there's no consistency
between them.

Can someone help me figure out what I'm doing wrong? If Calibre just isn't made for scientific articles (usually written using LaTeX, if it matters), is there another software that is?

[PeterT]

You might like to follow the advice from this post --> https://www.mobileread.com/forums/sh...d.php?t=186697

Sent from my Pixel 7a using Tapatalk

[Orion Invictus]

Originally Posted by PeterT

You might like to follow the advice from this post --> https://www.mobileread.com/forums/sh...d.php?t=186697

Sent from my Pixel 7a using Tapatalk

I missed an even more relevant sticky thread, about how PDF is just about the worst format to convert from, as well as the fact that Calibre doesn't handle two columns very well. Since scientific articles are in PDF and almost always two columns, this was a pointless effort on my part.

Thanks for the help, a mod can probably close/delete this, I doubt it's gonna help anyone.

[retiredbiker]

Originally Posted by Orion Invictus

I missed an even more relevant sticky thread, about how PDF is just about the worst format to convert from, as well as the fact that Calibre doesn't handle two columns very well. Since scientific articles are in PDF and almost always two columns, this was a pointless effort on my part.

Thanks for the help, a mod can probably close/delete this, I doubt it's gonna help anyone.

I do a lot of converting old magazines, usually two columns, to epubs. Also with centered text boxes, advertisements, "continued on page nn", and whatnot. There are no magic shortcuts. OCR a column at a time, put the text into Writer, proofread and correct, and then finally you can make an epub.

[Orion Invictus]

Originally Posted by retiredbiker

I do a lot of converting old magazines, usually two columns, to epubs. Also with centered text boxes, advertisements, "continued on page nn", and whatnot. There are no magic shortcuts. OCR a column at a time, put the text into Writer, proofread and correct, and then finally you can make an epub.

It's not a scan, though. It already has the text and images actually embedded (if that's the right word) in the PDF. OCR would just be copy/paste. And I don't have time to correct the dozens of articles I'd need to read (especially because, ideally, I'd only be reading them once while summarizing them).

I just wanted to be able to use the Scribe's features that can only be used in e-books, like writing notes by hand in the middle of the text (as opposed to on top of the text or with the keyboard) and going through pages with a single touch (small thing, but useful). Sadly, it seems like that's just not happening.

Thank you both for the help, though. I'll make do with what I've got.