RegEx Function: Title Case

phossler · 12-26-2014, 07:33 PM

Trying to use the wonderful new F&R functions to replace ALL CAPS headings

The fragment in the manual works unless there's a in the Hx tags

So if it looks like

<h1> CHAPTER 2 AWAY WE GO</h1>

it seems that the embedded causes it to not find

I cannot figure out how to modify the Find expression to handle that case

What is the modifications needed?

Thanks

kovidgoyal · 12-26-2014, 10:06 PM

(?s)<h\d>(.+?)</h\d>

phossler · 12-27-2014, 11:06 AM

Thanks, but it doesn't seem to find the test text

Is there a setting or something I should change?

jbacelar · 12-27-2014, 01:01 PM

Quote:

Originally Posted by phossler

Thanks, but it doesn't seem to find the test text

Is there a setting or something I should change?

If you've picked the expression of Kovid by the method of copying and pasting, likely that expression in the editor has a space at the end.
And naturally finds nothing.

phossler · 12-27-2014, 01:38 PM

Yep, it was a trailing space

However, the < and > in the gets replaced by named enetities

The end result seems to be they're treated like normal text instead of a tag

So I'm wondering if it's even possible to title case multiple lines

BTW, I do it that way so that the TOC has both lines as an entry:

Chapter 1 Now It All Starts
Chapter 2 And It Keeps On Going

kovidgoyal · 12-27-2014, 10:23 PM

(?s)<h\d>(.+?) (.+?)</h\d>

phossler · 12-28-2014, 11:35 AM

Thanks, but that RE doesn't seem to find anything. I made sure that there was no invisible trailing space in the Find expression ... this time

Quote:

calibre, version 2.14.0
ERROR: Not found: No matches were found for <pre style="font-style:italic">(?s)<h\d>(.+?) (.+?)& lt;/h\d></pre>

Sorry to be a pest, but I just finished manually TitleCasing a book with 40+ two line headings using the [Change Case] option, so I wanted to be ready the next time

Thanks

theducks · 12-28-2014, 12:22 PM

the BR did not include a space and self closing slash

Code:

(?s)<h\d>(.+?)<br\s*/>(.+?)</h\d>

Literals in the pattern must match exactly what you want to match

the \s* handles the unpretty (no space) condition of your original

kovidgoyal · 12-28-2014, 12:44 PM

(?s)<h\d>(.+?)<br\s*/?>(.+?)</h\d>

phossler · 12-28-2014, 01:55 PM

Thanks to both of you for your help, but the Find string is still either finding too much or nothing, as well as returning entities which end up being ignored (i.e. the 's)

The User Manual example

Code:

<[Hh][1-6][^>]*>([^<>]+)</[hH][1-6]>

works fine if there are no < or > between the Hx tags and that's what I started with, i.e. trying to make it handle the case where there might be tags or not. I hadn't even gotten to the point of handling spaces

I'd appreciate any more ideas

kovidgoyal · 12-28-2014, 10:55 PM

I tried the following expression and it works both for heading tags with inside and without br inside.

Code:

<[Hh][1-6][^>]*>([^<>]*)(?:<[^<>]+>){0,1}([^<>]*)</[hH][1-6]>

phossler · 12-29-2014, 09:16 AM

Thanks very much

It will take me a while to understand it, but hopefully I'll learn some new reg ex techniques

BetterRed · 12-29-2014, 01:34 PM

Quote:

Originally Posted by phossler

Thanks very much

It will take me a while to understand it, but hopefully I'll learn some new reg ex techniques

Maybe this can help Regular Expression Analyzer

I've no idea if its any good, but I find it useful to try to unravel obscure RE.

BR

phossler · 12-29-2014, 02:54 PM

@BR -- thanks. Very helpful.

I'm still looking for the RE that says "Do what I want" and not "What I Told You To Do"

phossler · 12-30-2014, 05:07 PM

I'm wondering if I have an issue with the actual TitleCasing RegEx-Function?

For this fragment ...

Code:

<body>

  <h1>TEST1 TEST1 TEST1 TEST1 TEST1 </h1>

  <p>NOW IS THE TIME and this should remain mixed case</p>

  <h1>TEST2 TEST2 TEST2 <br/><br/>TEST3 TEST3 </h1>

  <p>NOW IS THE TIME and this should remain mixed case</p>

  <h1>TEST4 <i>TEST4 TEST4 TEST4</i> TEST4 </h1>

</body>

... my Find expression

Code:

\<([Hh][1-6])\>.*?\</\1\>

... will find step through and highlight just the three H1 open/close pairs and what is between them (#2 in picture), but a Replace doesn't do anything except make the <h into a <H (#1 in picture)

It would seem to me that if the built in Title Case would ignore tags and what's in them, and just act on the 'Find' it should / would work for any embedded tags inside Hx pairs not just 

Am I misunderstanding?

12-27-2014, 11:06 AM	#3
phossler Wizard Posts: 1,071 Karma: 412718 Join Date: Jan 2009 Location: Valley Forge, PA, USA Device: Kindle Paperwhite	Thanks, but it doesn't seem to find the test text Is there a setting or something I should change? Attached Thumbnails

12-27-2014, 01:38 PM	#5
phossler Wizard Posts: 1,071 Karma: 412718 Join Date: Jan 2009 Location: Valley Forge, PA, USA Device: Kindle Paperwhite	Yep, it was a trailing space However, the < and > in the <br/> gets replaced by named enetities The end result seems to be they're treated like normal text instead of a tag So I'm wondering if it's even possible to title case multiple lines BTW, I do it that way so that the TOC has both lines as an entry: Chapter 1 Now It All Starts Chapter 2 And It Keeps On Going Attached Thumbnails

12-27-2014, 10:23 PM	#6
kovidgoyal creator of calibre Posts: 43,857 Karma: 22666666 Join Date: Oct 2006 Location: Mumbai, India Device: Various	(?s)<h\d>(.+?)<br>(.+?)</h\d>

12-28-2014, 12:22 PM	#8
theducks Well trained by Cats Posts: 29,800 Karma: 54830978 Join Date: Aug 2009 Location: The Central Coast of California Device: Kobo Libra2,Kobo Aura2v1, K4NT(Fixed: New Bat.), Galaxy Tab A	the BR did not include a space and self closing slash Code: (?s)<h\d>(.+?)<br\s/>(.+?)</h\d> Literals in the pattern must match exactly* what you want to match the \s* handles the unpretty (no space) condition of your original

12-28-2014, 01:55 PM	#10
phossler Wizard Posts: 1,071 Karma: 412718 Join Date: Jan 2009 Location: Valley Forge, PA, USA Device: Kindle Paperwhite	Thanks to both of you for your help, but the Find string is still either finding too much or nothing, as well as returning entities which end up being ignored (i.e. the <br/> 's) The User Manual example Code: <[Hh][1-6][^>]*>([^<>]+)</[hH][1-6]> works fine if there are no < or > between the Hx tags and that's what I started with, i.e. trying to make it handle the case where there might be <br/> tags or not. I hadn't even gotten to the point of handling spaces I'd appreciate any more ideas Attached Thumbnails

12-26-2014, 07:33 PM	#1
phossler Wizard Posts: 1,071 Karma: 412718 Join Date: Jan 2009 Location: Valley Forge, PA, USA Device: Kindle Paperwhite	RegEx Function: Title Case Trying to use the wonderful new F&R functions to replace ALL CAPS headings The fragment in the manual works unless there's a <br/> in the Hx tags So if it looks like <h1> CHAPTER 2 <br/>AWAY WE GO</h1> it seems that the embedded <br/> causes it to not find I cannot figure out how to modify the Find expression to handle that case What is the modifications needed? Thanks Attached Thumbnails Last edited by phossler; 12-26-2014 at 09:28 PM.

12-26-2014, 10:06 PM	#2
kovidgoyal creator of calibre Posts: 43,857 Karma: 22666666 Join Date: Oct 2006 Location: Mumbai, India Device: Various	(?s)<h\d>(.+?)</h\d>

12-28-2014, 12:44 PM	#9
kovidgoyal creator of calibre Posts: 43,857 Karma: 22666666 Join Date: Oct 2006 Location: Mumbai, India Device: Various	(?s)<h\d>(.+?)<br\s*/?>(.+?)</h\d>

12-28-2014, 10:55 PM	#11
kovidgoyal creator of calibre Posts: 43,857 Karma: 22666666 Join Date: Oct 2006 Location: Mumbai, India Device: Various	I tried the following expression and it works both for heading tags with <br> inside and without br inside. Code: <[Hh][1-6][^>]>([^<>])(?:<[^<>]+>){0,1}([^<>]*)</[hH][1-6]>

12-29-2014, 09:16 AM	#12
phossler Wizard Posts: 1,071 Karma: 412718 Join Date: Jan 2009 Location: Valley Forge, PA, USA Device: Kindle Paperwhite	Thanks very much It will take me a while to understand it, but hopefully I'll learn some new reg ex techniques

12-29-2014, 02:54 PM	#14
phossler Wizard Posts: 1,071 Karma: 412718 Join Date: Jan 2009 Location: Valley Forge, PA, USA Device: Kindle Paperwhite	@BR -- thanks. Very helpful. I'm still looking for the RE that says "Do what I want" and not "What I Told You To Do"

12-30-2014, 05:07 PM	#15
phossler Wizard Posts: 1,071 Karma: 412718 Join Date: Jan 2009 Location: Valley Forge, PA, USA Device: Kindle Paperwhite	I'm wondering if I have an issue with the actual TitleCasing RegEx-Function? For this fragment ... Code: <body> <h1>TEST1 TEST1 TEST1 TEST1 TEST1 </h1> <p>NOW IS THE TIME and this should remain mixed case</p> <h1>TEST2 TEST2 TEST2 <br/><br/>TEST3 TEST3 </h1> <p>NOW IS THE TIME and this should remain mixed case</p> <h1>TEST4 <i>TEST4 TEST4 TEST4</i> TEST4 </h1> </body> ... my Find expression Code: \<([Hh][1-6])\>.?\</\1\> ... will find step through and highlight just the three H1 open/close pairs and what is between them (#2 in picture), but a Replace doesn't do anything except make the <h into a <H (#1 in picture) It would seem to me that if the built in Title Case would ignore tags and what's in them, and just act on the 'Find' it should / would work for any embedded tags inside Hx pairs not just <br/> Am I misunderstanding? Attached Thumbnails Last edited by phossler; 12-30-2014 at 05:10 PM. Reason: Can't type*

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
regex-function convert roman numerals	weberr	Editor	11	09-22-2021 05:15 PM
A regex function to number a mathematical ebook	dmonasse	Editor	3	12-23-2014 02:54 AM
Regex Function - Split unknown word	Paulie_D	Editor	19	12-07-2014 05:12 AM
Regex for Title Case or Sentence case?	Turtle91	Sigil	3	01-19-2013 01:36 PM
Dutch title case function	fvdham	Library Management	8	10-11-2012 10:09 PM

Advert

Advert