rename and regex

[rebeltaz]

How can you batch rename files while retaining certain portions and replacing others? For example:

original filename:
prefix.Y01Z18.suffix.ext

renamed to:
y01z18.ext

All files have this same format but: the suffix is not always the same text; the prefix is not always the same; the numbers after Y and Z change from file to file.

What I am looking for is something along the lines of

rename 's/.+YxxZxx.+\.ext/yxxzxx\.ext/' *.ext

where xx leaves that digit intact. I know about a program but that is overkill for what I want and it actually adds more than I want to the filenames.

Thanks...

[steeldriver]

You can do that by grouping the bits you want to keep between parentheses, and then substituting those groups using $1, $2, etc. in the replacement string e.g.

$ rename -nv 's/.*[.]S([0-9]+)E([0-9]+)[.].*[.]mp4/s$1e$2.mp4/' *.mp4
prefix.S01E18.suffix.mp4 renamed as s01e18.mp4

[rebeltaz]

Thank you SO much! That is exactly what I was looking for!

[rebeltaz]

One thing.. what does the [] around the dots do? Thanks.

[steeldriver]

In general [ ] defines a range of characters for example later we have [0-9] meaning any single digit, but in the case of [.] it's just a way to escape the regex meaning of '.' (= any single character) so that it matches a literal period instead - an alternative way to do the same thing is with a backslash escape ie \. but sometimes [.] is easier to read

[HiImTye]

'[.]' is the same as '\.' - it means match one of the characters inside (or one not inside) the square brackets, ie [xyzX] will match any of x, y, x, or X, while [^xyzX] will match any that is NOT x, y, z, or X. as usual you can use *+? etc to specify repetition.

anyway you do this because '.' means 'any character'

edit: a better way would be

rename -n 's|.*([a-zA-Z][0-9][0-9][a-zA-Z][0-9][0-9]\.[a-zA-Z0-9]*)|\L$1|' fileOrFiles

which will match any of

blahblah.S11E11.ext
blahy32Z32.extension
really long filename.with.extra.stuff.A12B33.extension

and turn them into

s11e11.ext
y32z32.extension
a12b33.extension

[rebeltaz]

Oh, OK.. that was what confused me. I am used to seeing it as \.

Thanks again!

[rebeltaz]

edit: a better way would be

rename -n 's|.*([a-zA-Z][0-9][0-9][a-zA-Z][0-9][0-9]\.[a-zA-Z0-9])|\L$1|' fileOrFiles

which will match any of

blahblah.S11E11.ext
blahy32Z32.extension
really long filename.with.extra.stuff.A12B33.extension

and turn them into

s11e11.ext
y32z32.extension
a12b33.extension

OK... I guess '|' is taking the place of '/'. It looks like that just extracts the a12b34 pattern and carries it to the other side along with the extension? I like that. What does the 'L' in 'L$1' do?

[HiImTye]

\L converts everything that follows to lowercase, while \l converts the next letter to lowercase. I'd share a good info page but I need to set up Firefox Sync on my tablet again

and yeah, I like pipes, it prevents picket fencing

edit: I edited my previous post so thst it grabs more than one letter extensions /braindead

[rebeltaz]

\L converts everything that follows to lowercase, while \l converts the next letter to lowercase. I'd share a good info page but I need to set up Firefox Sync on my tablet again

and yeah, I like pipes, it prevents picket fencing

edit: I edited my previous post so thst it grabs more than one letter extensions /braindead

I just now got around to trying this and it is grabbing everything after the first dot after the A12B33.

e.g.
really.long.filename.with.extra.stuff.A12B33.and.m ore.stuff.extension

becomes

a12b33.and.more.stuff.extension

instead of

a12b33.extension


I have been trying to figure out on my own how to match the characters after the last dot, but nothing is working..