Cleaning up with Search & Replace

[zakman]

So I just imported a list of my books from Goodreads into Calibre and need to do some cleanup. All of the imported books that are part of a series are now titled like so (apologies for extra ASCII):

Title---------------------------------------------------------Series
=======================================
Book Name (The Blah Saga, #7)-------------------------(empty)

If possible, I'd like to change the metadata so it's more in line with the rest of my library:

Title------------------Series
===============================
Book Name----------The Blah Saga [7]

Is this possible? If so, how? I'm pretty new to maniupulating the data like this in Calibre, so the more details, the better. Thanks!

[chaley]

Moderator Notice
Moved from Template Questions to its own thread

[chaley]

Originally Posted by zakman

So I just imported a list of my books from Goodreads into Calibre and need to do some cleanup. All of the imported books that are part of a series are now titled like so (apologies for extra ASCII):

Title---------------------------------------------------------Series
=======================================
Book Name (The Blah Saga, #7)-------------------------(empty)

If possible, I'd like to change the metadata so it's more in line with the rest of my library:

Title------------------Series
===============================
Book Name----------The Blah Saga [7]

Is this possible? If so, how? I'm pretty new to maniupulating the data like this in Calibre, so the more details, the better. Thanks!

You can do this with calibre's bulk metadata search and replace. It requires 3 search/replace operations.

The steps:

1. *Back up your library*. Things can go very wrong and you might need to get back to ground 0.
2. Assuming you have more than one book, press E to open bulk metadata edit.
3. Click on the Search and replace tab.
4. If needed, change the Search mode to "Regular Expression.
5. Do the first search and replace to copy the series number to calibre's series_index column.
   
   Press the Apply button.
6. Do the second S/R to copy the series name to calibre's series column.
   
   Press the Apply button.
7. Finally, remove the extraneous information from the title.
   
   Press the OK button to apply the S/R and close the dialog.

The regular expression

^(.*) \((.*), #(\d*)\)

says Starting at the beginning of the line (the title) find all characters up to " (", capturing them in the first capture group. Next the characters up to ", #" (the series name) are captured in group 2. Finally, the digits up to the closing ')' are captured in group 3.

Note that the images show the regular expression

^(.*).*\((.*), #(\d*)\)

The one above replaces the ".*\(" with " \(", which is cleaner. I didn't bother to remake the images.

[zakman]

Thanks. After working through this a couple of times (and cleaning up between each one), I am now getting a "could not convert string to float" error when trying to run the first S/R. This happens even on a smaller selection of books that I know for sure should match up with the regular expression. Time to restore my backup, I guess?

EDIT: Working with it some more, it appears to only match on books where the series name is a single word. For example:



If the screenshot didn't come through:

The title "The Android's Dream (The Android's Dream #1)" does not appear to be recognized by the expression and is unchanged from the "Text Test" to the "Test Result" columns.

The title "Animal Farm (FarmyAnimal, #1)" IS recognized and "Test Result" shows a "1".

Continuing through the rest of the S/R's looks like they are working as planned. Seems to be just the multi-word series giving me problems.

[chaley]

Originally Posted by zakman

Thanks. After working through this a couple of times (and cleaning up between each one), I am now getting a "could not convert string to float" error when trying to run the first S/R. This happens even on a smaller selection of books that I know for sure should match up with the regular expression. Time to restore my backup, I guess?

That means that at least one of the books has something odd where the series number should be, where by 'odd' I mean there isn't a number there. Check each book to be sure that what follows the '#' character is a number.

[dunhill]

I have the name of the series and the number before the title, how would the regular expression be to clean it?
example:
Series Name #015 - Book Title
I want only the title to remain in its field