Hi, y’all. Running Linux Mint and I have the puzzle presented above.
From what I gather, I’m using rename (1p) which makes mention of Perl and in the man page it says it will also run as file-rename. I’m not sure if this is the right rename utility for the common argument
s/old_pattern/new_pattern/
but any time I try to run anything (including -n), I just get an angle bracket > and have to ctrl-c out.
I’d also need some details on how the wildcards work, which seems to be lacking in the documentation.
Edit: Instructions unclear. I have a bunch of episodes that are very wordy. I’m moving them onto DVD and truncated on my player the directory will look like:
Star Trek The Next Gene…
Star Trek The Next Gene…
Star Trek The Next Gene…
Star Trek The Next Gene…
Star Trek The Next Gene…
so I want to take (sample episode)
Star Trek The Next Generation Season 1 Episode 1 - Encounter at Far Point
and
-
Replace 'Star Trek The Next Generation Season ’ with ‘S0’
-
Replace 'Episode ’ with ‘E0’ or ‘E’ depending on digits
-
Keep episode title as is.
So it looks like
S01E01 - Encounter at Farpoint.mkv
For anyone looking for a GUI tool I like to use krename.
Tested this with perl-rename and it seems to work fine:
perl-rename -n 's/.* (\d{1,2}).* (\d{1,2})/sprintf "S%02dE%02d",$1,$2/e'*
. represents any single character
‘*’ matches the previous character 1 or more times
'.* ’ matches any series of characters before a space
() encloses something you want to keep
\ is the escape character
d represents any number
{1,2} means the number will be 1 or 2 digits
In the sprintf, %02d is to format each number with 2 digits.
$1 and $2 pass the first and second values enclosed in parentheses from the first argument into the 2 d’s in the “S%02dE%02d” of the second argument.
This link has a better explanation of the /e modifier than I could give you xD
https://www.geeksforgeeks.org/perl/perl-e-modifier-in-regular-expression/
Cool! Thanks!
Is this standard PERL stuff? How I learn what all these things mean? One day I hope to write my own arguments.
No idea lol
I just wanted a rename utility and Arch didn’t have the GUI one I used on Mint so I gradually pieced some of this stuff together. If there’s one source online that lays it all out I haven’t found it yet!
If you’d like me to explain the one here I’ll do my best ;)
Edit: nvm, I edited the original comment to explain everything
I think you want something like this:
$ ls 'Star Trek Next Generation Season 01 Episode 22 - Bob.mkv' 'Star Trek Next Generation Season 1 Episode 1 - Encounter at Farpoint.mkv' $ file-rename 's/.*Season (\d+) Episode (\d+)/sprintf("S%02dE%02d", $1, $2)/ie' *.mkv $ ls 'S01E01 - Encounter at Farpoint.mkv' 'S01E22 - Bob.mkv'If you are getting an angle bracket, it’s probably because you didn’t add
*.mkvat the end.Thanks! I’ll dig into this tonight. The man page could use a bit of work, too, if you’re feeling generous. A lot is assumed to be known.
Edit: Your entry is winning “shortest code” so fa r…
It’s so short because it’s special-purpose rather than general purpose. Perl is a programming language that can do lots and lots of different things. file-rename does one thing, and one thing very well: it renames files using regex
This is the kinda small script I use Claude for. So yeh, it’s LLM generated. Downvote away.
But I am terrible at writing bash scripts!#!/usr/bin/env bash # rename-episodes.sh — run inside the folder, or pass a directory as $1 shopt -s nullglob cd "${1:-.}" || exit 1 for f in *.mkv; do # Match: ... Season N ... Episode N - Title.mkv if [[ $f =~ Season\ ([0-9]+)\ Episode\ ([0-9]+)\ -\ (.+)\.mkv$ ]]; then season="${BASH_REMATCH[1]}" episode="${BASH_REMATCH[2]}" title="${BASH_REMATCH[3]}" # Zero-pad to two digits new=$(printf "S%02dE%02d - %s.mkv" "$season" "$episode" "$title") if [[ "$f" != "$new" ]]; then echo "mv: $f -> $new" # uncomment the next line to actually move the file # mv -n -- "$f" "$new" fi else echo "skip (no match): $f" fi doneI’ve commented out the
mvcommand so you can test/fiddle/play around with it without clobbering your files.Some notes from Claude:
Two practical notes:
If your files aren’t all .mkv, change the glob (*.mkv) and the regex anchor accordingly, or loop over *.{mkv,mp4,avi}.
This assumes the literal words “Season” and “Episode” appear. If your real filenames vary (e.g. “S1”, “1x01”, “Ep 1”), the regex needs adjusting
Edit: I just realised that specifically
file-renameis mentioned, and looks like you are getting appropriate help in other threads.I recommend using a tool designed just for this: https://www.tweaking4all.com/home-theatre/rename-my-tv-series-v2/
Share the rename commands you’ve tried
I’ll look into it. I’m going to be running a media server eventually, so thinking learning the command line arguments would be a good investment.
In this case, I’m moving these onto DVD and the truncated file names will all look the same on the DVD player, so I need to shorten 24*7 files.
Re-phrased the title. It’s a plain language question.
what’s the outcome you’re trying to achieve? you’ve got a folder full of mkv files and you just want to rename them?
why not keep it simple and use a for loop with mv and awk to perform the rename?
editted for clarity. thanks. I have a bad habit of not providing context. underdeveloped theory of mind.
Because I need to preserve the Season and Episode information.
Replace 'Star Trek The Next Generation Season ’ with ‘S0’
Replace 'Episode ’ with ‘E0’ or ‘E’ depending on digits
Keep episode title as is.
To me, your problem feels like the command you’re trying to execute isn’t complete syntax-wise. It may be useful if you’d quote the actual command you’re trying to execute.
This doesn’t answer your question at all, so my apologies up front.
I’m not at all intending to advertise here, and I’m not at all affiliated with them, but in the past, I’ve sidestepped having to write scripts by using FileBot. It’s always just worked for me.
mv “Star Trek Next Generation Season 1 Episode 1 - Encounter at Farpoint.mkv” “S01E01 - Encounter at Far Point.mkv”Or am I missing something about what you’re trying to do?
I think he wants to rename more than one file. Sounds like he wants to rename entire folders using that formula.
Yes, the “etc.” part. Imagine I have 7*24 episodes following this convention. How do I rename them all in one swell foop?
find command with an exec. Basically, find can list files matching a pattern and then run a command for each of them. The exec will probably be a bit gnarly, though, so if you want something a little more palatable, you want a script that grabs the names of the files into a variable, then takes each entry via a for loop (find command and a bash while read might work), stores it in a variable, changes it based on your pattern (sed might work here), stores the changed name in another variable and then mv $former-name $new-name
If the naming of each series is that consistent, you could just use parameters in bash and build the new title out of it and then do the rename, in a loop. It would be a very short shell script.
eta: using printf to format the new title variable will let it handle the number formatting clearnly, like 01 for 1
eta: something like this
for f in *"Season "*; do s_ep="${f#*Season }"; s="${s_ep%% Episode*}"; ep_t="${s_ep#*Episode }"; ep="${ep_t%% - *}"; t="${ep_t#* - }"; printf -v new "S%02dE%02d - %s" "$s" "$ep" "$t"; mv "$f" "$new"; done
mv
mmv







