How to remove blank spaces in a word with RegEx?

[RbnJrg]

Hi friends:

I have something like:

<h1 title="My Title">C H A P T E R  1</h1>

and I want:

<h1 title="My Title">CHAPTER 1</h1>

How can I do that with RegEx?

TIA

Rubén

[Vroni]

Depends. It can be impossible or very easy. If its always the word CHAPTER, tats easy, and i guess you woudn't have asked

If all letters are capital you can just search for ([A-Z])_ and reaplce it with \1. The probality that a capital letter is followed by a blank is low, but not 0. So yo have to go through the text one at a time to make a decision.

In any other case i would assume its not possible as Regex in sigil cant be limited to work only inside a specific element (AFAIR Calibre can do that) So you are catching too many wrong positives if your heading is mixed-case.

[KevinH]

Simply add that element to the reflex pattern if you only want to search inside a specific tag element

[RbnJrg]

Originally Posted by Vroni

Depends. It can be impossible or very easy. If it's always the word CHAPTER, that's easy, and i guess you woudn't have asked

Yes, not always the word is CHAPTER.

If all letters are capital you can just search for ([A-Z])_ and reaplce it with \1. The probality that a capital letter is followed by a blank is low, but not 0. So yo have to go through the text one at a time to make a decision.

Not always the letters are in uppercase. Sometimes only the first letter is capitalized.

In any other case i would assume it's not possible as Regex in sigil can't be limited to work only inside a specific element (AFAIR Calibre can do that). So you are catching too many wrong positives if your heading is mixed-case.

I didn't know that Calibre could work with RegEx inside a specific element. I'll do a research to see what can I get there. Thanks for the info.

[RbnJrg]

Originally Posted by KevinH

Simply add that element to the reflex pattern if you only want to search inside a specific tag element

And how do I do that?

[jackie_w]

@RbnJrg,

This is stretching my limited regex lookahead knowledge The closest I could get (tested in calibre) was something like:

Find:\b([a-z])\s+(?=(\w\s+\w))
Replace:\1

with case-sensitivity disabled.

I think you'd need to replace them one-by-one not 'Replace All'.

[Doitsu]

Originally Posted by RbnJrg

I didn't know that Calibre could work with RegEx inside a specific element. I'll do a research to see what can I get there. Thanks for the info.

I rarely use Calibe Editor, but the following should work for you.

• Paste <h1([^>]*)>([^<]+)</h1> in the Find box.
• Select Regex-function from the Mode drop-down box.
• Click Create/Edit, paste the following code in the Code box:
  
  
  

  def replace(match, number, file_name, metadata, dictionaries, data, functions, *args, **kwargs):
      return '<h1' + match.group(1) + '>'  + match.group(2).replace('  ', '§').replace(' ', '').replace('§', ' ') + '</h1>'

  enter a function name, e.g. Ruben, and click OK.

[RbnJrg]

Originally Posted by jackie_w

@RbnJrg,

This is stretching my limited regex lookahead knowledge The closest I could get (tested in calibre) was something like:

Find:\b([a-z])\s+(?=(\w\s+\w))
Replace:\1

with case-sensitivity disabled.

I think you'd need to replace them one-by-one not 'Replace All'.

Thank you very much for your response Jackie. It almost worked When I have, for example:

C H A P T E R  V I I I

after applying your code, I have:

CHAPTERVII I

Where my strings are all in uppercase, I can use your code with \b([A-Z])\s+(?=(\w\s+\w)) and case-sensivity enabled and I can do the work at once. So, if it were possible to solve the issue, your RegEx is very usefull.

[RbnJrg]

Originally Posted by Doitsu

I rarely use Calibe Editor, but the following should work for you.

• Paste <h1([^>]*)>([^<]+)</h1> in the Find box.
• Select Regex-function from the Mode drop-down box.
• Click Create/Edit, paste the following code in the Code box:
  
  
  

  def replace(match, number, file_name, metadata, dictionaries, data, functions, *args, **kwargs):
      return '<h1' + match.group(1) + '>'  + match.group(2).replace('  ', '§').replace(' ', '').replace('§', ' ') + '</h1>'

  enter a function name, e.g. Ruben, and click OK.

Thank you very much Doitsu; your code worked like a charm!

[Notjohn]

Originally Posted by RbnJrg

Thank you very much Doitsu; your code worked like a charm!

Sometimes I log onto the Sigil just to marvel at the things people come up with!

[Vroni]

By the was, are there any other spaces besides blank spaces?

[RbnJrg]

Originally Posted by Vroni

By the way, are there any other spaces besides blank spaces?

No, only blank spaces.

[Vroni]

I would like to have seven red lines, some painted with red ink, some with green ink and some with transparent ink.

"The experts, Youtube"