Need Help Creating a Regular Expression

Worm · 08-16-2010, 03:49 AM

Please forgive my level of competency here, but I have tried and tried and am lost. I am attempting to import my years-old collection of .pdf e-books and need to create a Regular Expression so that Calibre will properly import and map fields. I have about 3,000 e-books in the following format:

Star Wars - [Boba Fett 01] - The Fight to Survive (by Terry Bisson).pdf

The Output I would like in Calibre would be:

Book Title:
The Fight to Survive

Author:
Terry Bisson

Series:
Star Wars - Boba Fett

and then the #1 for the sequence in the series.

Is this possible via a Regular Expression?

Manichean · 08-16-2010, 07:53 AM

Should be possible. Take a look at the preferences, specifically the add/save books- page, which links to a handy reference for regular expressions as well as offering a test input thingy. You should be good from there.

chaley · 08-16-2010, 09:02 AM

I can get close, but not perfect.

Testing using the file name you supplied

Code:

Star Wars - [Boba Fett 01] - The Fight to Survive (by Terry Bisson).pdf

and assuming that *all* the files have this format, even if they do not have series, then the regular expression

Code:

(?P<series>(.+?))(?P<series_index>\d+)\] - (?P<title>.+) \(by (?P<author>.+)\)

almost works. The problem comes from the '[' and ']' characters. The series name should be the first characters up to the 2 digits followed by the ']', with the '[' removed ('Star Wars - Boba Fett'). Unfortunately, there is no way (that I know of) to remove characters within a regular expression group, so you are stuck with generating series names like 'Star Wars - [Boba Fett'.

I see two ways to deal with the extra '['. The first is to rename the files before importing to calibre. I would generate a text file containing all the book names, then edit that file to create a batch/shell script to rename the books. stripping away the leading '['.

If the batch operation isn't something you want to try, then I would run the import, then use the 'manage series' dialog available from the tag browser to manually remove the '[' character from each series. Manually correcting the series this way shouldn't take too long, around 1 to 2 seconds per series.

See the attached screenshot to see the regexp in action.

Worm · 08-16-2010, 12:41 PM

Brilliant and my most sincere thanks! I will try this after work today.

EricLandes · 08-16-2010, 03:38 PM

If you want to do a lot of batch file-renaming before importing into Calibre, look long and hard at the Name It Your Own Way tool.

That tool has saved me hours of time.

chaley · 08-16-2010, 03:57 PM

Quote:

Originally Posted by EricLandes

If you want to do a lot of batch file-renaming before importing into Calibre, look long and hard at the Name It Your Own Way tool.

That tool has saved me hours of time.

Thanks. I like the regular expression matching. Saves me from having to use vim to create batch files.

The tool is at http://www.niyow.com/.

Worm · 08-17-2010, 01:57 AM

I'm on a Mac so Terminal is great so long as I get the commands right. Else, this is a great GUI on OSX: http://manytricks.com/namemangler/.

Those Regular Expressions worked perfectly for my files, by the way. Thank you again!

edbro · 08-17-2010, 10:31 AM

Quote:

Originally Posted by chaley

Thanks. I like the regular expression matching. Saves me from having to use vim to create batch files.

The tool is at http://www.niyow.com/.

How much is it?

I hate software that will not tell you how much it costs unless you install it on your PC. I see no link on the site where you can see the cost or buy it. They don't even tell you it is shareware but I inferred it by the "download free" label.

chaley · 08-17-2010, 10:41 AM

Quote:

Originally Posted by edbro

How much is it?

The news section of the website says that it is freeware.

The license agreement (included in the spoiler) says that same thing.

Spoiler:

plunderydoo · 08-18-2010, 02:20 PM

try this Tool for renaming:
http://www.joejoesoft.com/cms/showpage.php?cid=108

It can do regex and is extremly flexible - and its free.

08-16-2010, 09:02 AM	#3
chaley Grand Sorcerer Posts: 12,525 Karma: 8065948 Join Date: Jan 2010 Location: Notts, England Device: Kobo Libra 2	I can get close, but not perfect. Testing using the file name you supplied Code: Star Wars - [Boba Fett 01] - The Fight to Survive (by Terry Bisson).pdf and assuming that all the files have this format, even if they do not have series, then the regular expression Code: (?P<series>(.+?))(?P<series_index>\d+)\] - (?P<title>.+) \(by (?P<author>.+)\) almost works. The problem comes from the '[' and ']' characters. The series name should be the first characters up to the 2 digits followed by the ']', with the '[' removed ('Star Wars - Boba Fett'). Unfortunately, there is no way (that I know of) to remove characters within a regular expression group, so you are stuck with generating series names like 'Star Wars - [Boba Fett'. I see two ways to deal with the extra '['. The first is to rename the files before importing to calibre. I would generate a text file containing all the book names, then edit that file to create a batch/shell script to rename the books. stripping away the leading '['. If the batch operation isn't something you want to try, then I would run the import, then use the 'manage series' dialog available from the tag browser to manually remove the '[' character from each series. Manually correcting the series this way shouldn't take too long, around 1 to 2 seconds per series. See the attached screenshot to see the regexp in action. Attached Thumbnails

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Regular Expression Help	Azhad	Calibre	86	09-27-2011 03:37 PM
Regular Expression Help	smartmart	Calibre	5	10-17-2010 06:19 AM
Help!! Having trouble with regular expression	Partzz	Calibre	2	09-14-2010 01:32 PM
Regular Expression Help Needed	dloyer4	Calibre	1	07-25-2010 11:37 PM
Help with the regular expression	Dysonco	Calibre	9	03-22-2010 11:45 PM

08-16-2010, 03:49 AM	#1
Worm Junior Member Posts: 3 Karma: 10 Join Date: Aug 2010 Device: Nuvi	Need Help Creating a Regular Expression Please forgive my level of competency here, but I have tried and tried and am lost. I am attempting to import my years-old collection of .pdf e-books and need to create a Regular Expression so that Calibre will properly import and map fields. I have about 3,000 e-books in the following format: Star Wars - [Boba Fett 01] - The Fight to Survive (by Terry Bisson).pdf The Output I would like in Calibre would be: Book Title: The Fight to Survive Author: Terry Bisson Series: Star Wars - Boba Fett and then the #1 for the sequence in the series. Is this possible via a Regular Expression?

08-16-2010, 07:53 AM	#2
Manichean Wizard Posts: 3,130 Karma: 91256 Join Date: Feb 2008 Location: Germany Device: Cybook Gen3	Should be possible. Take a look at the preferences, specifically the add/save books- page, which links to a handy reference for regular expressions as well as offering a test input thingy. You should be good from there.

08-16-2010, 12:41 PM	#4
Worm Junior Member Posts: 3 Karma: 10 Join Date: Aug 2010 Device: Nuvi	Brilliant and my most sincere thanks! I will try this after work today.

08-16-2010, 03:38 PM	#5
EricLandes Connoisseur Posts: 95 Karma: 8282 Join Date: Jan 2010 Device: Kindle PW, Kobo Aura HD, Galaxy Note 10.1	If you want to do a lot of batch file-renaming before importing into Calibre, look long and hard at the Name It Your Own Way tool. That tool has saved me hours of time.

08-17-2010, 01:57 AM	#7
Worm Junior Member Posts: 3 Karma: 10 Join Date: Aug 2010 Device: Nuvi	I'm on a Mac so Terminal is great so long as I get the commands right. Else, this is a great GUI on OSX: http://manytricks.com/namemangler/. Those Regular Expressions worked perfectly for my files, by the way. Thank you again!

08-18-2010, 02:20 PM	#10
plunderydoo Enthusiast Posts: 38 Karma: 412 Join Date: Sep 2009 Device: WinMobile, Hanvon N516 w. OpenInkPot, eLyricon EBX-500, Iphone 3GS	try this Tool for renaming: http://www.joejoesoft.com/cms/showpage.php?cid=108 It can do regex and is extremly flexible - and its free.

Advert

Advert