Book editor help - function for EPUB correction

[soucedz]

Hey guys, i need help (since i don't know anything about Pyhton or RegEx functions) with creating a function that eliminates the unnecessary paragraphs that occur when converting PDFs to EPUB.
i have tried using Find&Replace with a simple expression like:

</p> <p class="calibre2">[a-z]

since correct paragraphs are succeeded with a capital letter, but the problem is that i don't want it to select the matched lower case letter, i tried something like:

</p> <p class="calibre2">?([a-z])

But the matched lower case letter still gets selected.

Thanks in advance.

[theducks]

I have a series of 'cleanups' I use
Note: This is copied from Sigils saved search, ignore leading numbers an the line with Name= (describes what it does) and the escape before the \ (\\) should be a Single\

80\Name=Cleanup/Joins/Join to lower
80\Find="([[:alpha:],]\x201d*)</p>\\s*<p\\b[^>]*>([a-z\x201c])"
80\Replace=\\1 \\2
81\Name=Cleanup/Joins/Join to upper
81\Find="([[:alpha:],]\x201d*)</p>\\s*<p\\b[^>]*>([A-Z\x201c])"
81\Replace=\\1 \\2
87\Name=Cleanup/Joins/Honorifics
87\Find="(Mr|Mrs|Ms|Dr|Prof)\\.</p>\\s+<p class=\"calibre\\d+\">([A-Z])"
87\Replace=\\1. \\2
88\Name=Cleanup/Joins/de BR w/punct
88\Find="([[:punct:]])<br class=\"calibre4\" />\\s+(\"*[A-Za-z\x201c])"
88\Replace="\\1</p><p class=\"calibre4\">\\2"

* adjust the RED
Note: I kept it simple and replace the capture

(Wishlist PI: Import Sigil saved searches)

[Divingduck]

It is maybe better to check your conversion preferences first. Your problem is a very common issue for a wrong conversion setup for PDF.

Reduce the standard line unwrapping factor of 0.45 at PDF input preferences to a value between 0.25 to 0.12

You will find out that this will reduce the most of your problem to a minimum.

[DNSB]

Originally Posted by soucedz

Hey guys, i need help (since i don't know anything about Pyhton or RegEx functions) with creating a function that eliminates the unnecessary paragraphs that occur when converting PDFs to EPUB.
i have tried using Find&Replace with a simple expression like:

</p> <p class="calibre2">[a-z]

since correct paragraphs are succeeded with a capital letter, but the problem is that i don't want it to select the matched lower case letter, i tried something like:

</p> <p class="calibre2">?([a-z])

But the matched lower case letter still gets selected.

Thanks in advance.

I've used a regex similar to your second example with the replacement string being " \1" (a space followed by whatever lower case letter was selected in the search).

[deback]

"Reduce the standard line unwrapping factor of 0.45 at PDF input preferences to a value between 0.25 to 0.12."

Enable Heuristics and change the line unwrap factor to 0.22. This will help to keep paragraphs together, so editing will be minimal.