Sigil find and replace page numbers

[montalex]

I have a question about removing page numbers in Sigil. I found an epub of "The Sot Weed Factor", which is only available free in an OCRed scanned version. This means that headers have been integrated into the text, so that every page has "The Sot-Weed Factor [xxx]" (where xxx is the page number. It was easy to use "find and replace to get rid of "The Sot-Weed Factor", but the numbers are proving intractable (there are 860 of them).

This is what it looks like:

<p>[ 315 ]</p>

I've tried ever possible version of:

<p>[ \d+ ]</p> or <p>[ (\d+) ]</p>

but nothing works! It seems like the brackets are a problem ( [ ] ). Do you know any solution?

[mrmikel]

The brackets are part of regex. To make them a literal character use \ in front of them.
<p>\[\d+\]</p>

Or <p>\[(.+?)\]</p> to remove any thing with brackets around it.

That is my guess. I am not a regex guru.

[montalex]

Originally Posted by mrmikel

The brackets are part of regex. To make them a literal character use \ in front of them.
<p>\[\d+\]</p>

Or <p>\[(.+?)\]</p> to remove any thing with brackets around it.

That is my guess. I am not a regex guru.

To me, you are a regex guru! <p>\[(.+?)\]</p> worked. Phew! That felt good, like trimming a hedge...

Thanks!

[mrmikel]

You have to be careful of unintended consequences when you have regex selected without thinking and a period is in the search box. It will select all sorts of stuff other than intended. Thank goodness for the undo button!

It too needs the \ to find it in regex.

[montalex]

Originally Posted by mrmikel

You have to be careful of unintended consequences when you have regex selected without thinking and a period is in the search box. It will select all sorts of stuff other than intended. Thank goodness for the undo button!

It too needs the \ to find it in regex.

Indeed! One of the scripts I tried found and replaced half of the book! Yours only removed page numbers between brackets.

Thanks!

[911jason]

I'm also trying to remove page numbers inserted by ABC Amber LIT. The code looks like this:

<p class="calibre3">
<b class="calibre1">Page  1</b></p>
<p class="calibre3">
</p>

How can I remove all instances of PAGE ### through the file?

[Turtle91]

In Sigil, regex search and replace:

search:<p class="calibre3">\s*<b class="calibre1">Page (\d+)</b></p>
replace: (nothing)

Try it on a few first before committing to the replace all...and save before you do!

I would also replace all the <p class="calibre3"> with something that is descriptive. eg <p class="noindent"> or, if it is a basic paragraph type, replace it with just <p>

[theducks]

I suspect you may need

search:<p class="calibre3">\s*<b class="calibre1">Page\s+(\d+)</b></p>

if your example is true (has multiple spaces, depends on # of digits)

[Turtle91]

yup yup! Thanks!