RegEx or RE Function to apply [Change Case] Capitialize?

phossler · 05-01-2016, 10:03 AM

Is there a RegEx or RegEx Function that will apply [Change Case] Capitialize to the upper case words at the beginning of a paragraph, which is usually in the first paragraph of each chapter?

It's purely personal opinion but in my Kindle I find the 'paper book' typographic style disconcerting and if there were a easy way to change it'd save a lot of manual work.

I'd like to change the first to the second . It's not 100% since there could be proper names or acronyms in the UPPER CASE TEXT, but I can clean them up manually

Code:

<body>

<p>ALLCAPS ALLCAPS ALLCAPS lower case  lower case  lower case  lower case  lower case </p>

<p>Allcaps allcaps allcaps lower case  lower case  lower case  lower case  lower case </p>

</body>

Thanks

eschwartz · 05-01-2016, 11:44 AM

There are builtin functions to replace captured text as capitalized/lowercased/titlecased...

phossler · 05-01-2016, 01:24 PM

Quote:

Originally Posted by eschwartz

There are builtin functions to replace captured text as capitalized/lowercased/titlecased...

Yes -- Select + Right Click - [Change Case] Capitialize is what I've been doing

I was looking for a less manual way

theducks · 05-01-2016, 01:33 PM

Quote:

Originally Posted by phossler

Yes -- Select + Right Click - [Change Case] Capitialize is what I've been doing

I was looking for a less manual way

You have run into a difference in REGEX engines

Sigil uses PCRE. That engine can do case (\L, \U ) from S&R

phossler · 05-01-2016, 02:13 PM

Quote:

Originally Posted by theducks

You have run into a difference in REGEX engines

Sigil uses PCRE. That engine can do case (\L, \U ) from S&R

Sorry, but I'm not sure I'm following that

Calibre has the manual change case, so are you saying that with the Calibre RE engine, there's no way to change cases?

If so, then I guess I'll have to keep doing it manually, chapter by chapter

theducks · 05-01-2016, 02:29 PM

Quote:

Originally Posted by phossler

Sorry, but I'm not sure I'm following that

Calibre has the manual change case, so are you saying that with the Calibre RE engine, there's no way to change cases?

If so, then I guess I'll have to keep doing it manually, chapter by chapter

There is a Function Mode for the S&R in the editor (use the Help in the Editor, to locate) (I am currently on a system that does not have the editor)

Or use Sigil

if you don't write in Python

kovidgoyal · 05-01-2016, 03:17 PM

http://manual.calibre-ebook.com/func...n-the-document

phossler · 05-01-2016, 04:47 PM

@kovidgoyal

Thanks, I use that all the time to fix things like

Code:

<h1>CHAPTER 1 - THE BEGINNING</h1>

I was really looking to see if there is any automatic way to change paragraph leading upper case words to lower case

Code:

<body>

(before)

<p>ALLCAPS ALLCAPS ALLCAPS lower case  lower case  lower case  lower case  lower case </p>

(after)

<p>Allcaps allcaps allcaps lower case  lower case  lower case  lower case  lower case </p>

</body>

I did try using the Capitialize text function with a Find of <p[^>]*>.+?

That works good for the ALLCAPS text I was using for a test, but it also gets applied to every paragraph.

I was thinking there might be some clever (i.e. beyond me) RegEx way to take the first whole word after a and make it TitleCase, and then the (say) next 4 words and make them lowercase. Then stepping through with Find or Replace&Find would just be a lot easier

eschwartz · 05-01-2016, 04:57 PM

I still don't understand the problem.

According to your example in post #1, capture the first three words and apply the default, builtin regex function.

Although using both the word allcaps and caps letters to indicate your intent, then arguing with yourself, isn't helping.

phossler · 05-01-2016, 05:33 PM

Yes, but if there's 100 chapters in the book that all have the upper case leadin for the first paragraph, then I'd have to select the text for each chapter manually.

I was only wondering if there was a way to find a group of upper case words at the beginning of a paragraph and do all 100 at once

Quote:

Although using both the word allcaps and caps letters to indicate your intent, then arguing with yourself, isn't helping.

I'll try to avoid arguing with myself

BetterRed · 05-01-2016, 06:27 PM

@phossler - as I understand it, you want a One-Click something that marches through a book detecting all the 'first paragraphs' (some book's format the paragraph following a scene break as a 'first paragraph') that have leading Upper Case words and then change the leading Upper Case words to Sentence Case words.

I've yet to find a Case changer that deals with proper names correctly. So 'SMILEY FLEW TO VIENNA; CARSTAIR'S was in the arrivals hall to meet him...' inevitably becomes 'Smiley flew to vienna; carstair's was in the arrivals hall to meet him...'

Some errors may be picked up by a Spellchecker. But not if 'SMILEY FLEW TO DODGE CITY; CAPABILITY BROWN was in the arrivals hall to meet him...'

I suspect a goodly number of 'first paragraphs' in novels, biographies, and histories contain a proper name in the first sentence. So even after a One-Click button waved its magic wand, you'd have to eyeball each 'first paragraph' to sort the wand's curses from it's blessings.

I always found the capitalisation of leading words in print irritating, especially as it's often arbitrary with no regard as to the author's phrasing. To me they're a printer's affectation.

Getting rid of them in e-books. whilst tempting, usually requires more effort than I'm willing to expend. Unless of course the book is in need of serious tin-bashing, in which case selecting the text, changing the case, and correcting proper names may be worth the effort.

BR

phossler · 05-01-2016, 07:19 PM

@BR -- agree 100%

Was not even thinking of limited to just first paragraphs. I was going for any tag

I was thinking that ...

if the first word after a is all caps, then it would be capitialized

if the second, third, and fourth (maybe more) words after a were all caps, then they would be lower cased

Even stepping through with [Find] or [Replace, Find Next] to catch:

'SMILEY FLEW TO DODGE CITY; CAPABILITY BROWN was in the arrivals hall to meet him...'

to be fixed manually would be a lot faster then manually going to each file, highlighting the 'offending text,' and right clicking to change the case, etc.

'Smiley flew to Dodge City; Capaility Brown was in the arrivals hall to meet him...'

I tried 2 RegExs to do it in two passes, but never got even the first one to work.

I think theducks' comment about differences in RegEx engines means that there is no way to automate the process

BetterRed · 05-01-2016, 09:04 PM

The only instances I've seen of upper casing the first few words of a paragraph have been in the first paragraph of a chapter or scene - hence 'first paragraph'.

I doubt there's a regex engine that will differentiate between - 'LOS VEGAS IS CITY ONE SHOULD DODGE' and 'I HAD A GREAT TIME IN DODGE CITY.'

I just opened first three 'real' books in my Test library and looked at words 2-6 in the first paragraph of the first chapter. A factual history had no proper nouns, a Regency novel had 'HYDE PARK', and an international banking exposé had 'LONG-TERM CAPITAL MANAGEMENT'. Of those six words the only one that would even be flagged as a spelling error is 'hyde', the other five ('park', 'long', 'term', 'capital', and 'management') are valid in lower case - yet they would all be patently wrong in the context they are used.

Whilst ever there are no multi-lingual, deep-intelligence enabled regex engines which are capable of further deep-learning, us humans will always have something to do

BR

phossler · 05-01-2016, 09:39 PM

Quote:

Whilst ever there are no multi-lingual, deep-intelligence enabled regex engines which are capable of further deep-learning, us humans will always have something to do

Roger that

kovidgoyal · 05-01-2016, 10:56 PM

Quote:

Originally Posted by phossler

Yes, but if there's 100 chapters in the book that all have the upper case leadin for the first paragraph, then I'd have to select the text for each chapter manually.

No you wont. Simply run the S&R on "All text files" it will end up title casing all paragraphs in your book. But that is no problem since you want all paragraphs to have the same casing behavior.

05-01-2016, 04:47 PM	#8
phossler Wizard Posts: 1,085 Karma: 412718 Join Date: Jan 2009 Location: Valley Forge, PA, USA Device: Kindle Paperwhite	@kovidgoyal Thanks, I use that all the time to fix things like Code: <h1>CHAPTER 1 - THE BEGINNING</h1> I was really looking to see if there is any automatic way to change paragraph leading upper case words to lower case Code: <body> (before) <p>ALLCAPS ALLCAPS ALLCAPS lower case lower case lower case lower case lower case </p> (after) <p>Allcaps allcaps allcaps lower case lower case lower case lower case lower case </p> </body> I did try using the Capitialize text function with a Find of <p[^>]*>.+?</p> That works good for the ALLCAPS text I was using for a test, but it also gets applied to every paragraph. I was thinking there might be some clever (i.e. beyond me) RegEx way to take the first whole word after a <p> and make it TitleCase, and then the (say) next 4 words and make them lowercase. Then stepping through with Find or Replace&Find would just be a lot easier

05-01-2016, 07:19 PM	#12
phossler Wizard Posts: 1,085 Karma: 412718 Join Date: Jan 2009 Location: Valley Forge, PA, USA Device: Kindle Paperwhite	@BR -- agree 100% Was not even thinking of limited to just first paragraphs. I was going for any <p> tag I was thinking that ... if the first word after a <p> is all caps, then it would be capitialized if the second, third, and fourth (maybe more) words after a <p> were all caps, then they would be lower cased Even stepping through with [Find] or [Replace, Find Next] to catch: 'SMILEY FLEW TO DODGE CITY; CAPABILITY BROWN was in the arrivals hall to meet him...' to be fixed manually would be a lot faster then manually going to each file, highlighting the 'offending text,' and right clicking to change the case, etc. 'Smiley flew to Dodge City; Capaility Brown was in the arrivals hall to meet him...' I tried 2 RegExs to do it in two passes, but never got even the first one to work. I think theducks' comment about differences in RegEx engines means that there is no way to automate the process

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
RegEx Function: Title Case	phossler	Editor	29	07-04-2020 10:52 AM
Regex Function about «» and “”	senhal	Editor	8	04-06-2016 02:12 AM
Regex Function - Split unknown word	Paulie_D	Editor	19	12-07-2014 05:12 AM
Change Case with Regex Problem	nqk	Editor	4	07-25-2014 10:38 PM
Regex for Title Case or Sentence case?	Turtle91	Sigil	3	01-19-2013 01:36 PM

05-01-2016, 10:03 AM	#1
phossler Wizard Posts: 1,085 Karma: 412718 Join Date: Jan 2009 Location: Valley Forge, PA, USA Device: Kindle Paperwhite	RegEx or RE Function to apply [Change Case] Capitialize? Is there a RegEx or RegEx Function that will apply [Change Case] Capitialize to the upper case words at the beginning of a paragraph, which is usually in the first paragraph of each chapter? It's purely personal opinion but in my Kindle I find the 'paper book' typographic style disconcerting and if there were a easy way to change it'd save a lot of manual work. I'd like to change the first <p> to the second <p>. It's not 100% since there could be proper names or acronyms in the UPPER CASE TEXT, but I can clean them up manually Code: <body> <p>ALLCAPS ALLCAPS ALLCAPS lower case lower case lower case lower case lower case </p> <p>Allcaps allcaps allcaps lower case lower case lower case lower case lower case </p> </body> Thanks Last edited by phossler; 05-01-2016 at 11:15 AM. Reason: spelling

05-01-2016, 11:44 AM	#2
eschwartz Ex-Helpdesk Junkie Posts: 19,421 Karma: 85397180 Join Date: Nov 2012 Location: The Beaten Path, USA, Roundworld, This Side of Infinity Device: Kindle Touch fw5.3.7 (Wifi only)	There are builtin functions to replace captured text as capitalized/lowercased/titlecased...

05-01-2016, 03:17 PM	#7
kovidgoyal creator of calibre Posts: 44,149 Karma: 22670164 Join Date: Oct 2006 Location: Mumbai, India Device: Various	http://manual.calibre-ebook.com/func...n-the-document

05-01-2016, 04:57 PM	#9
eschwartz Ex-Helpdesk Junkie Posts: 19,421 Karma: 85397180 Join Date: Nov 2012 Location: The Beaten Path, USA, Roundworld, This Side of Infinity Device: Kindle Touch fw5.3.7 (Wifi only)	I still don't understand the problem. According to your example in post #1, capture the first three words and apply the default, builtin regex function. Although using both the word allcaps and caps letters to indicate your intent, then arguing with yourself, isn't helping.

05-01-2016, 06:27 PM	#11
BetterRed null operator (he/him) Posts: 20,771 Karma: 27405072 Join Date: Mar 2012 Location: Sydney Australia Device: none	@phossler - as I understand it, you want a One-Click something that marches through a book detecting all the 'first paragraphs' (some book's format the paragraph following a scene break as a 'first paragraph') that have leading Upper Case words and then change the leading Upper Case words to Sentence Case words. I've yet to find a Case changer that deals with proper names correctly. So 'SMILEY FLEW TO VIENNA; CARSTAIR'S was in the arrivals hall to meet him...' inevitably becomes 'Smiley flew to vienna; carstair's was in the arrivals hall to meet him...' Some errors may be picked up by a Spellchecker. But not if 'SMILEY FLEW TO DODGE CITY; CAPABILITY BROWN was in the arrivals hall to meet him...' I suspect a goodly number of 'first paragraphs' in novels, biographies, and histories contain a proper name in the first sentence. So even after a One-Click button waved its magic wand, you'd have to eyeball each 'first paragraph' to sort the wand's curses from it's blessings. I always found the capitalisation of leading words in print irritating, especially as it's often arbitrary with no regard as to the author's phrasing. To me they're a printer's affectation. Getting rid of them in e-books. whilst tempting, usually requires more effort than I'm willing to expend. Unless of course the book is in need of serious tin-bashing, in which case selecting the text, changing the case, and correcting proper names may be worth the effort. BR

05-01-2016, 09:04 PM	#13
BetterRed null operator (he/him) Posts: 20,771 Karma: 27405072 Join Date: Mar 2012 Location: Sydney Australia Device: none	The only instances I've seen of upper casing the first few words of a paragraph have been in the first paragraph of a chapter or scene - hence 'first paragraph'. I doubt there's a regex engine that will differentiate between - 'LOS VEGAS IS CITY ONE SHOULD DODGE' and 'I HAD A GREAT TIME IN DODGE CITY.' I just opened first three 'real' books in my Test library and looked at words 2-6 in the first paragraph of the first chapter. A factual history had no proper nouns, a Regency novel had 'HYDE PARK', and an international banking exposé had 'LONG-TERM CAPITAL MANAGEMENT'. Of those six words the only one that would even be flagged as a spelling error is 'hyde', the other five ('park', 'long', 'term', 'capital', and 'management') are valid in lower case - yet they would all be patently wrong in the context they are used. Whilst ever there are no multi-lingual, deep-intelligence enabled regex engines which are capable of further deep-learning, us humans will always have something to do BR

Advert

Advert