Calling Expression experts

[JCSullivan]

I'm trying to get the expression to ignore anything before the author's name and, perhaps anything after the Title but this is as far as I got.

I'm reading the tutorial http://manual.calibre-ebook.com/regexp.html but not getting far.

Thanks

[speakingtohe]

If you have a lot of books like that you could use search and replace with
\d\d-\d\d-
\d\d\d\d


Probably an expression on import would work but I am wimpy in that area.
Helen

[Perkin]

As long as the filesname follow your example, here's a quick solution.

\d+-\d+-(?P<author>.+) - (?P<title>.+) \d+

If you want the date as published field you can alter it to

\d+-\d+-(?P<author>.+) - (?P<title>.+) (?P<published>\d+)

[JCSullivan]

Originally Posted by speakingtohe

If you have a lot of books like that you could use search and replace with
\d\d-\d\d-
\d\d\d\d


Probably an expression on import would work but I am wimpy in that area.
Helen

Thanks Helen

Originally Posted by Perkin

As long as the filesname follow your example, here's a quick solution.

\d+-\d+-(?P<author>.+) - (?P<title>.+) \d+

If you want the date as published field you can alter it to

\d+-\d+-(?P<author>.+) - (?P<title>.+) (?P<published>\d+)

Thanks Perkin, I'm going to try this out later.

Your "d+" expression now explains better what the tutorial says about "\d is equivalent to [0-9]" and also helps me understand the expression better.

[theducks]

Decoding the above:
anything not inside a ( ) pair is not captured and is discarded.
The term does need to match what is there for this to work (at all )

[JCSullivan]

Thank you - that helps even more.
It's great to learn step by step using examples (tutorials are excellent but sometimes, this is better)
I'm very graphical and all your replies are really appreciated.

[JCSullivan]

Using a test file and an expression with the above information and everything worked well. (image 1)

But if I drag-n-drop it doesn't work (image 2)

Am I doing something wrong.

Thanks

[Perkin]

Remove the [] from the expression, or change to ().

They mean that you want to use inclusive (or exclusive) characters, or a range of characters
e.g. [a-o] means any lower case letter from a to o (a,b,c,d,e,f,g,h,i,j,k,l,m,n,o)
[a-oxz] means any lower case letter from a to o or x or z

[JCSullivan]

Thank you Perkin.

Again, it worked in test mode but not in drag-n-drop as can be seen.
Could it be because I'm using d-n-d????

Thanks

[Perkin]

Just use .+ to match all the garbage after the ' - ' seperator

(?P<title>.+) - .+

[JCSullivan]

Thank you again.

This does NOT work for me.

Could (or has) anyone actually tried it out by using drag-n-drop?

Is it a "bug" (I hate saying that, sorry) or is it a limitation and therefore not available using drag-n-drop.

Obviously it is reading the metadata from the mp3 file because the author's name is NOT part of the file name.

[kiwidude]

Drag and drop is no different to using add. It will use your regular expression for a text file. However I notice you still had "Get metadata from file contents" ticked in a screenshot above, so if you drag/dropped an epub or mobi for instance it is going to use title/author metadata from inside the file, NOT evaluate your regex. If relying on filenames/regex is how you want to do your imports, then untick that other option.

Apologies if I missed it but I haven't seen you post the actual filename of the file you are drag/dropping. I see lots of screenshots of regexes and the result, but not what the actual name of the file is. If your "plan" for the regex was based on a flawed assumption then everyone's suggestions to tweak it aren't going to do any good...

[JCSullivan]

Thanks KiwiDude

That worked great - I knew I was doing something wrong.
It was the actual file name with the exception that it was an ".mp3" rather than a ".txt" Testing would not allow me to use ".mp3" as an extension.

Thank you everyone else with your help and patience - I learnt a lot from your expression examples.