Regular expression help

[krendk]

Being a non programmer and after having searched I have not been able to compile a regular expression structure that covers the following file format

<Title> by <Authors> at <Publisher> - <ISBN-10> - <ISBN-13>

Help would nice

Thanks in advance

Kren

[thrawn_aj]

How about this?

(?P<title>.+) by (?P<author>.+) at (?P<publisher>.+) - (?P<ISBN-10>.+) - (?P<ISBN-13>.+)\.pdf

Note1: I added the \.pdf as an example. Be sure to add the file extension you have or the search will fail even if the rest of the regex is correct (maybe that was your problem?).

Note2: I suspect that this will fail dramatically if <title> contains "by" as a separate word.

Note3: I don't recall the actual labels for the 2 isbn fields. My (possibly obsolete) knowledge of this tells me that only 1 isbn field is supported. Dunno how far custom metadata has come. Assuming that isbn-13 is not usable, this is what I would do:

(?P<title>.+) by (?P<author>.+) at (?P<publisher>.+) - (?P<isbn>.+) - ([0-9]+)\.pdf

(if you want to save the ISBN-10 and throw away the ISBN-13. Make appropriate changes if you want it the other way round).

If custom metadata fields are supported (I just don't know), the following might work:

(?P<title>.+) by (?P<author>.+) at (?P<publisher>.+) - (?P<isbn>.+) - (?P<isbn13>.+)\.pdf

(if you want to save the ISBN-10 to the normal Calibre ISBN field and the ISBN-13 to a custom field called "isbn13" - it has to exist of course).

It doesn't help that I'm at work and don't have Calibre in front of me to check this. Good luck though

[Manichean]

Originally Posted by thrawn_aj

Note1: I added the \.pdf as an example. Be sure to add the file extension you have or the search will fail even if the rest of the regex is correct (maybe that was your problem?).

Note 1 to note 1: I don't think you need to add any extensions to the import regexes. From what I recall, Calibre automatically strips the extension from the filename, thus the regex never even seees the extension.

[krendk]

Gentlemen,

thanks for your input - it works except for the publisher - why I don't know

Kren

[thrawn_aj]

Originally Posted by Manichean

Note 1 to note 1: I don't think you need to add any extensions to the import regexes. From what I recall, Calibre automatically strips the extension from the filename, thus the regex never even seees the extension.

You're absolutely right - as I said, I didn't have Calibre to test it with.
The official manual explicitly warned about this so I figured that's what was tripping up OP's initial trials. Unfortunately, after reading it again, I see that I misread the last line - it was merely talking about the test filename, not the real one .