xpath exp for just roman unumerals as Chapter headings

[p3aul]

I need an xpath for just roman numerals to put in "Structure Detection" > "Detect Chapters at" part of the conversion dialog. Each Chapter heading is on a line by itself and consists of just a roman numeral.Thanks,
Paul

[DoctorOhh]

Originally Posted by p3aul

I need an xpath for just roman numerals to put in "Structure Detection" > "Detect Chapters at" part of the conversion dialog. Each Chapter heading is on a line by itself and consists of just a roman numeral.Thanks,
Paul

I don't have an xpath for you, but do have a shot in the dark that might work. Try enabling Heuristic Processing and checking the Detect and markup unformatted chapter headings and sub headings.

Good Luck.

[p3aul]

Thanks, but that was the first thing tried. I've never really had much luck with Heuristic Processing. What seems logically simple, HP just ignores.
here is wan example of the first chapter:
I: In which Two Men go forth and One Arrives

What could be more obvious than a Roman Numeral?
This is the latest thing I've tried:
//*[re:test(., '^[I-XXIII]+:$')]

[wallcraft]

Originally Posted by p3aul

here is wan example of the first chapter:
I: In which Two Men go forth and One Arrives

What could be more obvious than a Roman Numeral?
This is the latest thing I've tried:
//*[re:test(., '^[I-XXIII]+:$')]

The "$" matches the end of a line, which isn't right here. Try

//*[re:test(., '^[IXV]+:')]

There is no need to repeat elements in the "[ ]" list and "I-X" means all characters from I to X. If you want a "-" put it at the end of the list.

[theducks]

I would include 'LC' in your set of possible Roman numerals (50,100) for other books that have more that 39 chapters (XL onwards)

[p3aul]

Hi Wallcraft! once again you've come through for me! I like your suggestion also, "theducks"!

My reason for putting in the dash was to include numerals from I(1) to XXIII(23) in case the chapters went that high. I knew they didn't go above XXIII. and all I had to use was a previous xpath statement you gave me: //*[re:test(., '^CHAPTER [I-XXIII]+$')] so I deleted "CHAPTER"

In the case of the xpath above, wallcraft, you only used one set of RN; IXV but the program detected all the chapters in the book which went up to XVIII. Why did it do that? I'm confused. Why wouldn't my "I-XXIII" work?
Thanks,
Paul

[p3aul]

OH, I think I see! Does the "IXV" mean all characters that are I or X or V ? Well I have another xpath to add to my list!

[wallcraft]

Originally Posted by p3aul

In the case of the xpath above, wallcraft, you only used one set of RN; IXV but the program detected all the chapters in the book which went up to XVIII.

The "+" means one or more instances of the preceding character, which (from the []) can be an I or a X or a V. As theducks said, [IXVLC] would cover more cases.

[p3aul]

Thanks all! I've been collecting all these xpath statements in a file I call Calibre tips and tricks. I'mm also learning a little about xpath along the way.
Paul