TOC based on Capitalized Words

buckm56 · 06-02-2011, 12:50 PM

I have a book which isn't well formatted to created a TOC. The only format clue is that all letters of the first couple of words of the chapter are CAPs. Is there any formula that can create a TOC from this.

Thanks

theducks · 06-02-2011, 01:25 PM

Quote:

Originally Posted by buckm56

I have a book which isn't well formatted to created a TOC. The only format clue is that all letters of the first couple of words of the chapter are CAPs. Is there any formula that can create a TOC from this.

Thanks

If you convert to EPUB, then use Sigil's REGEX (match case ticked),
you can quickly step through (just Find-next or Replace)
change the pattern to match those skipped: repeat
BTW: if the Chapter start uses 'small-caps' a case sensitive S&R will not work, but the small-caps class is your best ever trigger

Assuming

each 'chapter' started in a separate segment, I would have to see the <body> tag through to the lines that would make the TOC entry.
If a chapter also starts mid-file,

to see if there is a pattern to trap.

Manichean · 06-03-2011, 02:19 AM

You could do the same thing in the Calibre conversion process, I believe- you can use regex in XPath, so you'd have to formulate an expression that matches capitalized words.

Another thing to try would be to activate the relevant heuristic options.

ldolse · 06-03-2011, 06:46 PM

If I understand the use case he's describing it's not one currently covered by Heuristics. That said, next time I get back to tweaking that it's a potential test case to add, but one that works pretty differently from the existing function as you wouldn't want to wrap those lines in H2 tags.

XPATH with Regex is about the only thing that would work, but all the TOC entries would then be a paragraph long.

theducks · 06-03-2011, 07:13 PM

Quote:

Originally Posted by ldolse

If I understand the use case he's describing it's not one currently covered by Heuristics. That said, next time I get back to tweaking that it's a potential test case to add, but one that works pretty differently from the existing function as you wouldn't want to wrap those lines in H2 tags.

XPATH with Regex is about the only thing that would work, but all the TOC entries would then be a paragraph long.

Good point.
I have done a non-visible header inserted before the back referenced text is rebuilt.

<h2 class="very short" title="\1"> <h2> < The original source preceeding Stuff>\1

ldolse · 06-03-2011, 11:16 PM

Quote:

Originally Posted by theducks

Good point.
I have done a non-visible header inserted before the back referenced text is rebuilt.

<h2 class="very short" title="\1"> <h2> < The original source preceeding Stuff>\1

That's kind of how I was thinking I would do it for Heuristics too, still fiddly because lots of books that do have first paragraphs with that formatting have normal chapter headings anyway. That fact could also help reduce false positives for some books though.

06-02-2011, 12:50 PM	#1
buckm56 Member Posts: 19 Karma: 10 Join Date: Feb 2010 Location: Colorado, US Device: Kindle 3	TOC based on Capitalized Words I have a book which isn't well formatted to created a TOC. The only format clue is that all letters of the first couple of words of the chapter are CAPs. Is there any formula that can create a TOC from this. Thanks

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
ePub TOC to mobi TOC	edmnddntes	Conversion	5	01-24-2011 02:56 AM
Creating TOC for books based on Text Files	crich70	Kindle Formats	4	09-13-2010 12:47 AM
Should ''internet'' be capitalized or lowercase?	taglines	Lounge	18	07-06-2010 04:15 AM
Detect chapter headings with capitalized words	fiendmish	Calibre	6	05-31-2010 10:45 AM
Making a TOC for LRFs? Issues with Calibre + LRF TOC editor not working	Magitek	LRF	0	05-06-2009 01:25 PM

06-03-2011, 02:19 AM	#3
Manichean Wizard Posts: 3,130 Karma: 91256 Join Date: Feb 2008 Location: Germany Device: Cybook Gen3	You could do the same thing in the Calibre conversion process, I believe- you can use regex in XPath, so you'd have to formulate an expression that matches capitalized words. Another thing to try would be to activate the relevant heuristic options.

06-03-2011, 06:46 PM	#4
ldolse Wizard Posts: 1,337 Karma: 123455 Join Date: Apr 2009 Location: Malaysia Device: PRS-650, iPhone	If I understand the use case he's describing it's not one currently covered by Heuristics. That said, next time I get back to tweaking that it's a potential test case to add, but one that works pretty differently from the existing function as you wouldn't want to wrap those lines in H2 tags. XPATH with Regex is about the only thing that would work, but all the TOC entries would then be a paragraph long.

Advert

Advert