>>> "NK" == Nikolay Kudryavtsev <nikolay.kudryavtsev@gmail.com> writes:

> Incidentally I have a somewhat related question I've been wondering
> about recently.

> So I have a massive txt ebook that I've been slowly reediting into an
> org file, for the purposes of easier navigation and textual analysis.

> It is also fill-paragraphed. I've been considering the idea of
> removing all superfluous line breaks.

> Now, mechanically it's quite doable - you check for sentence end
> punctuation signs at line endings and if there isn't one you remove
> the line break. There would of course be some false positives...

> This task seems like a somewhat common editing task, so I've been
> wondering if someone has a working code for it already, maybe there
> are some gotchas there, that I haven't though about.

I usually run first  unfill-paragraph-or-region (found in
org-mode-map), which is an interactive compiled Lisp function in
‘next-longline.el’.

And then use either 

    1. Virtual-auto-fill-mode on

    2. Or auto-fill-on, and run fill-paragraph and friends


But of course as you said, false positives. Etc but as a first approach it is quite good.



-- 
Warning: Content may be disturbing to some audiences
I strongly condemn Hamas bestialic terroristic attack on Israel, especially the despicable pogroms.
I strongly condemn Putin's war of aggression against Ukraine.
I support to deliver weapons to Ukraine's military. 
I support the NATO membership of Ukraine.
I support the EU membership of Ukraine. 
https://addons.thunderbird.net/en-US/thunderbird/addon/gmail-conversation-view/