Adding books - need help with regular expressions

[tweebee]

Hello,

I'm new to calibre and I am looking to add my books into the library by parsing the filenames. They are in this format:

ISBN_Title_[Publisher].ext

I would like to parse the ISBN, Title, and Publisher into their appropriate fields while ignoring the underscores and brackets. I have read the manual and searched this forum and I still can't find any info on how to parse the underscores.

Thanks for your help!

Twee

[Starson17]

Originally Posted by tweebee

IThey are in this format:

ISBN_Title_[Publisher].ext

I would like to parse the ISBN, Title, and Publisher into their appropriate fields while ignoring the underscores and brackets.

Try this:

(?P<isbn>.*?)[_ ](?P<title>.*)[_ ]\[(?P<publisher>.*)\]

Underscores are treated specially and are converted to spaces, so they tend to disappear automatically. I made the regex a bit less dense so you can see what it's doing.

[tweebee]

That works beautifully, thank you so much! I had the brackets parsed right, just didn't know how to do the underscores.

Thank you again!

[Starson17]

Originally Posted by tweebee

That works beautifully, thank you so much! I had the brackets parsed right, just didn't know how to do the underscores.

Thank you again!

You're welcome.
If you understand the regex, you'll note that you can get rid of these: "[_ ]" and replace with a simple space. I just put that in to be bit more clear. Also, note the non-greedy selector in the ISBN.

[tweebee]

Originally Posted by Starson17

You're welcome.
If you understand the regex, you'll note that you can get rid of these: "[_ ]" and replace with a simple space. I just put that in to be bit more clear. Also, note the non-greedy selector in the ISBN.

Oh I didn't even think of simply replacing those with the spaces! I was so hung up on putting something there to parse those underscores!

Thanks for that reminder. I'll know better next time.

[Daigomi]

I didn't want to create a new thread, so I hope no-one minds me asking my question here. I've got a very similar situation to tweebee except my format is:

ISBN.Publisher.Title.Date.ext

The problem is that every space is replaced by a full stop, so in practice it looks something like:

0154879871.Bigbook.Publishers.Being.Truly.Happy.20 07.November.pdf

I understand that there is no way for the script to know when the publisher name ends and the book title begins, but I have my books sorted by publisher so I could change the script for each publisher. Basically, I am looking for a way for a way to sort the books as follows:

{isbn}.{Publisher}.{Still Publisher}.{Title}.{Ignore last two}

If someone could help me with a template for this, I'm sure I could figure out how to adapt it to other similar formats.

[Starson17]

Originally Posted by Daigomi

I am looking for a way for a way to sort the books as follows:

{isbn}.{Publisher}.{Still Publisher}.{Title}.{Ignore last two}.pdf

If someone could help me with a template for this, I'm sure I could figure out how to adapt it to other similar formats.

Try this:

(?P<isbn>.*?)\.(?P<publisher>.*?\..*?)\.(?P<title>.*)\..*?\..*?

It parses this:
12345X.Publisher1.Publisher2.Title1.Title2.Title3. Ignore1.Ignore2.pdf

When requesting these, you really should give several examples of the actual file names, not just what you consider to be the pattern. It's important to see the spaces, hyphens, periods,, brackets, etc. in the names. For example, it's not clear if your actual files have the name Title, or {Title} in them. I assumed it didn't have the curly brackets, but if it did, other options would have been available.

[theducks]

Originally Posted by Daigomi;

0154879871.Bigbook.Publishers.Being.Truly.Happy.20 07.November.pdf

I understand that there is no way for the script to know when the publisher name ends and the book title begins, but I have my books sorted by publisher so I could change the script for each publisher.

slightly Side note:
An ISBN contains the Publisher (usually for the parent company) NUMBER

Language digit-Publisher-Book_Number-check_digit
unfortunately you (and many others) have stripped the hyphen
book_number+Publisher = 10 digits
small presses get long publisher numbers and short book numbers
a huge company get lots of book numbers.

[Daigomi]

Originally Posted by Starson17

Try this:

(?P<isbn>.*?)\.(?P<publisher>.*?\..*?)\.(?P<title>.*)\..*?\..*?

It parses this:
12345X.Publisher1.Publisher2.Title1.Title2.Title3. Ignore1.Ignore2.pdf

When requesting these, you really should give several examples of the actual file names, not just what you consider to be the pattern. It's important to see the spaces, hyphens, periods,, brackets, etc. in the names. For example, it's not clear if your actual files have the name Title, or {Title} in them. I assumed it didn't have the curly brackets, but if it did, other options would have been available.

Sorry, I gave a general example but I'll give a few actual examples as well:

0192840975.Oxford.University.Press.USA.Global.Warm ing.A.Very.Short.Introduction.Jan.2005.pdf
041528919X.Routledge.Roman.Berytus.Beirut.in.Late. Antiquity.Apr.2004.pdf
069100899X.Princeton.University.Press.Racism.A.Sho rt.History.May.2002.pdf

It seems to work almost perfectly thanks. The one slight annoyance is that the title and publisher has full stops in it (title: Racism.A.Short.History) but I can probably fix that with search and replace.

Thank you very much!

[Starson17]

Originally Posted by Daigomi

It seems to work almost perfectly thanks. The one slight annoyance is that the title and publisher has full stops in it (title: Racism.A.Short.History)

The input regex can't "change" any part of the filename during the Add Book operation. It can only select a part of the filename character string and insert that substring into the desired title, author or publisher field. Thus, there's no way to change the periods/dots/stops into spaces during import. As you have planned, you have to do that with Search and Replace in a second operation.

[Daigomi]

Originally Posted by Starson17

The input regex can't "change" any part of the filename during the Add Book operation. It can only select a part of the filename character string and insert that substring into the desired title, author or publisher field. Thus, there's no way to change the periods/dots/stops into spaces during import. As you have planned, you have to do that with Search and Replace in a second operation.

Ahh, I didn't know that. Thanks!