Using Regex to replace line breaks

[btired]

I am editing and cleaning up a bunch of Calibre epub conversions and they're messy and full of trash code. Regex has been so helpful to find stuff that regular find and replace can't do efficiently.

I'm still pretty new to using Regex and I was wondering if it can help me find the following: Carriage returns / new lines that are not new lines that end with a tag; then replace them with a blank 'space'.

Here's and example of uncorrected text:

<p class="p2">‘Professor!’
It
was
Vesuvius.
She sounded frightened. ‘Professor!’</p>
<p class="p2">Sara
looked to Robert.</p>

<p class="p1"><br/>‘What is it?’ Sara asked. Robert put his arm around her,
but she barely seemed to notice.</p>

...and ideally, I'd like the corrected file to look like this:

<p class="p2">‘Professor!’ It was Vesuvius. She sounded frightened. ‘Professor!’</p>
<p class="p2">Sara looked to Robert.</p>

<p class="p1"><br/>‘What is it?’ Sara asked. Robert put his arm around her, but she barely seemed to notice.</p>

Is this possible to do with regex? I've tried to figure it out but can't quite get it.

[KevinH]

Start with a copy of your epub and try running Sigil's Mend and Prettify tool.

It may do what you want. If not, regex can.

[theducks]

That is NOT a Calibre conversion (or artifact).

<p><br />...
Gimmie a proper Scene break or first paragraph by using a Top Margin.
I don't even try for a 1 click repair. After EACH success, I save in case a later fix goes off the rails

Once that is done,
I fix the lowercase lowercase ones. (This is the only one I use Replace all after a few tests. I replace the Class= with the actual one used in the book

The rest I step thru Replace-Next or Find (Skips that one). These are fairly few and take only a few minutes.
Then I fix Honorifics (many have a period)
Next is Initials <I can't remember if I do the Before or After Honorifics)

Then I fix Uppercase (eg Names)

Upper-Upper is mostly for acronyms

[btired]

Originally Posted by KevinH

Start with a copy of your epub and try running Sigil's Mend and Prettify tool.

It may do what you want. If not, regex can.

KevinH, thank you so much! That solved my problem and did exactly what I needed it to do - I didn't know that tool existed.

Originally Posted by theducks

That is NOT a Calibre conversion (or artifact).

<p><br />...
Gimmie a proper Scene break or first paragraph by using a Top Margin.

The text I was giving as an example already had the code altered to how I wanted it, it was just the multiple unnecessary carriage returns / new lines that I wanted to fix. I should've been clearer in my explanation

Thanks for those regex's - they'll come in handy in my future editing!

[Tex2002ans]

Hey, btired, welcome to MobileRead!!!

Originally Posted by btired

I'm still pretty new to using Regex and I was wondering if it can help me find the following: Carriage returns / new lines that are not new lines that end with a tag; then replace them with a blank 'space'.

Like KevinH said, the easiest way is to:

• Tools > Reformat HTML > Mend and Prettify All HTML Files

- - -

After that, if you wanted to merge "broken" paragraphs together, then see my answers in:

• 2022: "False paragraph breaks & RegEx"

That would help combine things like:

<p>This is an example</p>
<p>of lines that accidentally</p>
<p>didn't merge into a single paragraph.</p>

and give you:

<p>This is an example of lines that accidentally didn't merge into a single paragraph.</p>