Regex examples

[ReaderRabbit]

Sorry, this is such an obvious question and is probably answered somewhere but I didn't find it.

What would be the best way to find and eliminate page numbers such as:

He glanced 190</p>

<p class="calibre1">up at the big clock

[Doitsu]

Originally Posted by ReaderRabbit

What would be the best way to find and eliminate page numbers such as:

He glanced 190</p>

<p class="calibre1">up at the big clock

Assuming that each page number is preceded by a space, the following quick & dirty regex should work:

\d+</p>\s+<p class=".*?">

(Replace with nothing.)

[mzmm]

Originally Posted by Ripplinger

I couldn't get that to work at all and was about to give up and then realized you didn't use the curly smart quotes. Once I changed it to smart quotes, it would work somewhat, but it will also pick up any sentence or paragraph that doesn't immediately start with a quote. So it would pick up paragraphs like this:

Pamela shuddered. “We’ve been making ourselves polite to a murderess.”

And there's usually far too many of those types of sentences to want to read through over 500 of them to find the beginning quote buried further in.

try this? it'll probably still miss some (like if the closing quote butts up against a </span> instead of a </p> for example) so you'd probably want to scan the text afterwards but it might save you some copy/pasting.

find: (<p[^>]*>)(?:\s+)?([^“]+?”)(?:\s+)?(</p>)

replace: \1“\2\3

[ReaderRabbit]

Originally Posted by Doitsu

Assuming that each page number is preceded by a space, the following quick & dirty regex should work:

\d+</p>\s+<p class=".*?">

(Replace with nothing.)

Thanks so much! Works perfectly.

What about page numbers like this: <p class="calibre1">200</p>

I used to be able to find them with the 'Wildcard' search and replace. I am using version version 0.6.2 of Sigil. Where has that feature gone?

I ♥ brainiacs

[Doitsu]

Originally Posted by ReaderRabbit

What about page numbers like this: <p class="calibre1">200</p>

Use: <p class=".*?">\d+</p>

[Ripplinger]

Originally Posted by mzmm

try this? it'll probably still miss some (like if the closing quote butts up against a </span> instead of a </p> for example) so you'd probably want to scan the text afterwards but it might save you some copy/pasting.

find: (<p[^>]*>)(?:\s+)?([^“]+?”)(?:\s+)?(</p>)

replace: \1“\2\3

That's getting closer and it will pick up more, but it won't work on a sentence where the end quote isn't directly before the </p>, so it wouldn't pick up this sentence if the beginning quote was missing:

“But how would you deal with these miseries out of the past?” I asked.

Even if there were 2 regex strings to run to be able to pick up both sentence structures would still be much better than just hoping you find them all in proofing.

[ReaderRabbit]

Originally Posted by Doitsu

Use: <p class=".*?">\d+</p>

This is the same command, right? I did not see a difference but it works.


Has anyone compiled a list of basic commands, such as finding book numbers? I checked: 'Useful RegEx commandos for ebook corrections' suggested earlier in this thread but did not find anything I could use.

Thanks again

[ReaderRabbit]

Is there a command to fix this type of error?

<p class="calibre1">“That makes her sound like an ungrateful victim,”</p>

<p class="calibre1">Peggy said.</p>

[theducks]

Originally Posted by Doitsu

Use: <p class=".*?">\d+</p>

Just watch out if your chapters are also a digits only and NOT coded to a h#. fix those first, then clean

[theducks]

Originally Posted by ReaderRabbit

Is there a command to fix this type of error?

<p class="calibre1">“That makes her sound like an ungrateful victim,”</p>

<p class="calibre1">Peggy said.</p>

Use the Saved search: Join Paragraphs (included by default )

[ReaderRabbit]

Originally Posted by theducks

Use the Saved search: Join Paragraphs (included by default )

Join Paragraphs did not work on this example.

[theducks]

Originally Posted by ReaderRabbit

Join Paragraphs did not work on this example.

Got it (The why): Comma Quote not quote comma (I think that nees to be fixed, then it would work

[mzmm]

Originally Posted by Ripplinger

That's getting closer and it will pick up more, but it won't work on a sentence where the end quote isn't directly before the </p>, so it wouldn't pick up this sentence if the beginning quote was missing:

“But how would you deal with these miseries out of the past?” I asked.

Even if there were 2 regex strings to run to be able to pick up both sentence structures would still be much better than just hoping you find them all in proofing.

this should work.

(<p[^>]*>)(?:\s+)?([^“”]+?”)(.*?)(?:\s+)?(</p>)

\1“\2\3\4

[Ripplinger]

Originally Posted by mzmm

this should work.

(<p[^>]*>)(?:\s+)?([^“”]+?”)(.*?)(?:\s+)?(</p>)

\1“\2\3\4

That worked like a charm! It caught all instances no matter where the ending quote was located.

Thanks!

[ReaderRabbit]

Originally Posted by mzmm

this should work.

(<p[^>]*>)(?:\s+)?([^“”]+?”)(.*?)(?:\s+)?(</p>)

\1“\2\3\4

Thanks so much! Don't know how you all figure this stuff out. It's way beyond me but I like my books to be as correct as possible.

I ♥ brainiacs