Need a regex for importing books

[DoctorOhh]

Originally Posted by theducks

No reasonable way to determine which words belong together with (spaces) and which are new fields.

There is one method, but I don't have the skills to write the simple regex required. Essentially it would be something like (this doesn't work, but I'm sure someone can correct it)

(?P<isbn>.+) ? (?P<title>[^_]+)

the important part is the isbn, don't really care what garbage is in the title field because we will write over it when fetching metadata. Once the isbn field is set, download metadata and have overwrite Author and Title and most everything checked in the metadata configuration window.

[theducks]

Originally Posted by dwanthny

the important part is the isbn, don't really care what garbage is in the title field because we will write over it when fetching metadata. Once the isbn field is set download metadata and have overwrite Author and Title and most everything checked in the metadata configuration window.

I did not think of that (mostly because I turned those Off because I got tired of it 'fixing' my normalized names.
I use Robert A. Heinlein, Not Robert Anson Heinlein or Robert Heinlein)
I use the (base)Book title:The Lark and the Wren
not the (?) title: Bardic Voices Book 1:The Lark and the Wren

[DoctorOhh]

Originally Posted by theducks

I did not think of that (mostly because I turned those Off because I got tired of it 'fixing' my normalized names.

I never have title or author checked when fetching metadata either. Now hopefully someone can correct the regex.

[theducks]

Originally Posted by dwanthny

I never have title or author checked when fetching metadata either. Now hopefully someone can correct the regex.

won't a escaped Period work between those fields?

(?P<isbn>.+)\.(?P<title>[^_]+)

[DoctorOhh]

Originally Posted by theducks

won't a escaped Period work between those fields?

(?P<isbn>.+)\.(?P<title>[^_]+)

When I tested it with this
01505798756X.Silly Press.The.Strange.Professional.Title.Jul.1985.epub

It put 1985 in the title and the rest as isbn. Sort of want it to break on the first period, not the last. You can test it under adding books preferences.

[sdspieg]

Yes, but all I want is the isbn number - the rest I can get through a bulk download of covers and metadata. And the isbn always is prior to the first dot. Is there no way to just extract that and then copy it to the isbn-field?

Thanks,

-Stephan

[theducks]

Originally Posted by dwanthny

When I tested it with this
01505798756X.Silly Press.The.Strange.Professional.Title.Jul.1985.epub

It put 1985 in the title and the rest as isbn. Sort of want it to break on the first period, not the last. You can test it under adding books preferences.

(?P<isbn>[0-9X?)(?P<title>[^_]+)

works for the ISBN part

[PeterT]

Did you mean

Originally Posted by theducks

(?P<isbn>[0-9X]?)(?P<title>[^_]+)

[Serpentine]

Surely something simple like :

(?mi)^(?P<isbn>[\d\-x]{9,17})

Would work better? It includes some leeway for -'s which are often used to space isbn's.

And for fun, here's an over-complex regex to pull out the potentially useful stuff from that formatting :

(?P<isbn>[\d\-xX]{9,17})\.(?:(?P<publisher>(?:[\w]+\s[\w]*)+)\.)?(?P<title>.+?)(?P<datefudge>(?:Jan|Feb|Mar|Apr|May|Jun|Jul|Aug|Sep|Oct|Nov|Dec)\w*\.\d{2,4})?(?:\.[^\.]+[\n\r$])

[sdspieg]

I appreciate all of your assistance, but can somebody please give me some more info on where exactly I have to enter what (see my original posting?)

Because when I enter the regex from theducks in the 'search' field it turns
0313306419.Greenwood.Press.Rudolfo.A.Anaya.A.Criti cal.Companion.Oct.1999
into .Grnwood.rss.Rudofo.A.Anaya.A.Crca.Companon.Oc. (and then puts that in the isbn)...

The simple one from Serpentine does just get rid of the isbn
.Greenwood.Press.Rudolfo.A.Anaya.A.Critical.Compan ion.Oct.1999
but then again puts that in the isbn field, instead of the isbn itself

I'm sure I'm doing sthg wrong here, but would really appreciate your help.

Thanks!

-Stephan

[Serpentine]

I was assuming you were using the ISBN import option on a list of files, and were structuring the text yourself using the regex. It's most likely the easiest way to fetch the metadata and such automatically.

Would you like a step by step of doing that + regex?

[sdspieg]

Isbn-import? You mean the extract-isbn plugin (which is still very much hit-'n-miss)?

So I guess my answer is - yes please, a step-by-step sounds great!

-Stephan

[Serpentine]

Nah, there's a nice easy way to do it without the plugin - I've never use the plugin.
I was going to do it manually, but using a script is far easier.

Anyway, assuming you have Python installed here's the script:
Right click me -> save as

1. Save that to the folder which contains all your books - Don't worry if your books are in subfolders, it will crawl all the subfolders too.
2. If your python is using the default settings, and you have .py files associated: Just double click it, this should run it, a file called "bookisbnlist.txt" will be created in the folder.
3. Open the text file.
4. Swap to Calibre. Add Books (in the top menu bar) -> Add from ISBN
5. Copy and paste in the content of that text file.
6. Add them.

If you don't have Python, or don't have the files associated - figure it out
The python is written in a rather readable manner - so if you want to try other patterns, you should be able to edit it easily enough.

[Serpentine]

Originally Posted by salines

"Fielding, Joy-ebbes - Tanz, Püppchen, tanz.pdf"

(?P<author>.+?)\s+-\s+(?:(?P<series>.+?)\s*(?:(?P<series_index>\d+(?:\.\d*)?))?\s+-\s+)?(?P<title>.+)

Try that, it's a bit more robust and seems to fit your other cases.

Originally Posted by salines

Should I switch the used regex if I add book for series and none series?

It should deal with books no series now, I had placed something in the wrong conditional

[DoctorOhh]

Originally Posted by sdspieg

Isbn-import? You mean the extract-isbn plugin (which is still very much hit-'n-miss)?

So I guess my answer is - yes please, a step-by-step sounds great!

-Stephan

Go to Preferences - Adding Books and try any of the REGEXs above that people say will work in the Regular Expression box. Apply the settings then add the book. You can test the expression first by placing the name of a test file (including the extension of .epub or .mobi or whatever) name area and click the test button.

If all goes well just add books like normal and the isbn field will be filled in, then download the metadata and have it fill in the fields.