metadata interfering with title

[commentator8]

Hi all,

I have just started taming my ebook collection, and after adding many many books to calibre, i am now going over all the books and manually fixing those with no metadata. Generally its a speller/missing apostrophe causing it but sometimes it can be more stubborn.

First thing: When i get it to fetch metadata (often after giving it a isbn) it will change the title to include the series in brackets or following a colon. Is there any way to either prevent this or go through all entries and then remove the extra bit?

I find the easiest way to get all data on a book is to google (or amazon/isbndb) the isbn and then let calibre run with that to get all the info on the book. To emphasise - this is for books that calibre wouldn't do natively based on bad title/author/something.

Thanks!

[DoctorOhh]

Originally Posted by commentator8

First thing: When i get it to fetch metadata (often after giving it a isbn) it will change the title to include the series in brackets or following a colon. Is there any way to either prevent this or go through all entries and then remove the extra bit?

Go to Preferences - Behavior and Uncheck Overwrite author and title, see attached.

[commentator8]

Thanks, did the trick.

Any ideas on the other two things (fixing already changed titles, isbn lookup from title, not opposite)?

[DoctorOhh]

Originally Posted by commentator8

fixing already changed titles

Now I'm confused. You mean besides manually editing the title in the edit metadata window or double clicking in the library list and changing the Title?

Originally Posted by commentator8

isbn lookup from title, not opposite)?

I always use download metadata to get the isbn, rarely the other way around. If the ISBN you have doesn't retrieve the correct title, remove any ISBN in the edit metadata and fetch metadata based on title and author. Then select the returned metadata entry that suits you or your book. See attached.

[commentator8]

Re the first point, i had already mangled quite a few names by doing a mass ctrl-d on my books. So i was wondering if there is the ability to fix this by playing with the raw data, or by some regex option to play with the fields.

Guess not. Although if you know where the database is stored (if it is in an accessible form) i may try and play with it like that.

As to point two, i have quite a few books that wont find metadata, and i hadn't really paid attention to the fact that they only now pick up the one book to update to because of the dodgy isbn that the first update gave them. So yes - if i delete it it inevitable gives me more options, and while a slow painful process, it gets there.

Thanks for the help.

[DoctorOhh]

Originally Posted by commentator8

Re the first point, i had already mangled quite a few names by doing a mass ctrl-d on my books. So i was wondering if there is the ability to fix this by playing with the raw data, or by some regex option to play with the fields.

If you select multiple books then go to the bulk edit metadata window there is a tab that allows for regex search and replace (use at your own risk). I'm not sure how this would help you but it is available and I thought I would point it out.

[commentator8]

That looks promising. I was thinking of removing anything post colon or in brackets...

I'll let you know if it works.

[commentator8]

I've found that these will work:
\(.+?\) - anything in brackets
(:+.+) - anything after :

Is there anyway to combine them (and make them safer - to any regex guys out there)?

[Manichean]

There's an introduction into regular expressions that might help you.

[commentator8]

Originally Posted by Manichean

There's an introduction into regular expressions that might help you.

Great job there, its a world better than most of the stuff i found on the net (although there is still room for the regex testing grounds).

Thanks

[Manichean]

Originally Posted by commentator8

Great job there, its a world better than most of the stuff i found on the net (although there is still room for the regex testing grounds).

Thanks. But what do you mean by the parenthesised (sp?) remark?

[commentator8]

I was referring to sites like
http://rejex.heroku.com/
http://gskinner.com/RegExr/
which let you play around and see the results.

[Manichean]

I see. I didn't include such sites on purpose, as Calibre has excellent testing facilities in all places that regexpes are used.