renaming files question

Angharad has asked for the wisdom of the Perl Monks concerning the following question:

Hi there
I have a perl script that creates a particular file. However, depending on the eventual users of the script, sometimes the filename might look like this

1bn9.en.cnt
[download]

And other times like this

1bn9.pd.cnt
[download]

I want to circum-navigate this problem by renaming the file as follows

1bn9.cnt
[download]

before processing it further
I've tried a simple mv command

mv 1bn9.*.cnt 1bn9.cnt
[download]

But that complains about their being multiple files and therefore the destination having to be a directory, which is not what I want at all.
Can anyone suggest a solution ... be it Perl or a Unix command? It would be much appreciated.

Comment on renaming files question Select or Download Code

Replies are listed 'Best First'.
Re: renaming files question by Happy-the-monk (Canon) on Jan 06, 2006 at 13:25 UTC
If you don't care if there is more than one file with such a "middle extension" and want to run the risk of losing the others, this simple trick does it: `perl -wle 'for ( <1bn9..cnt> ) { qx(mv $_ 1bn9.cnt) }'` On the safe side, run it only once: `perl -wle '$_ = <1bn9..cnt>; qx(mv $_ 1bn9.cnt) if $_'` Cheers, Sören	[reply] [d/l] [select]
Re: renaming files question by tirwhan (Abbot) on Jan 06, 2006 at 13:19 UTC
So I guess you only want to handle one of these files, regardless of how many there are? This should do the trick: `my $a_file = (glob("1bn9..cnt")) or die "No file found"; rename $a_file,"1bn9.cnt" or die "Couldn't move file";` [download] There are ten types of people: those that understand binary and those that don't.*	[reply] [d/l]
Re: renaming files question by DungeonKeeper (Novice) on Jan 06, 2006 at 13:56 UTC
It will only complain about multiple files if there are multiple files, in which case you don't want to collapse them into a single file before processing, so perhaps this is what you want: `for my $original ( glob '/path/lbn9*.cnt' ) { rename $original, '/path/lbn9.cnt'; # file processing here... }` [download]	[reply] [d/l]
Re: renaming files question by sh1tn (Priest) on Jan 06, 2006 at 13:46 UTC
`perl -e 'map{ /(1bn.)\..+\.(cnt)/ and rename $_, "$1.$2" }glob "1bn*"'`	[reply] [d/l]
Re: renaming files question by kulls (Hermit) on Jan 06, 2006 at 14:37 UTC
try with the 'cut' command like `echo 1bn9.pd.cnt \| cut -d. f1` [download] to get the first column and `echo 1bn9.pd.cnt \| cut -d. f3` [download] to get the last column and merge it of both. -kulls	[reply] [d/l] [select]
Re^2: renaming files question by Tanktalus (Canon) on Jan 06, 2006 at 16:44 UTC
If you're going to do this in shell, then you can avoid a bit of work: `$ echo lbn9.pd.cnt \| cut -d. -f1,3 lbn9.cnt` [download] However, since I think you're advocating writing the above in perl (with the backticks), it seems kind of redundant: `join '.', (split '.', $name)[0,2]` [download] which avoids all the overhead of forking and execing multiple (2-3) subprocesses. Personally, I find the backticks (or even system in general) to be vastly overused in perl, mostly by people pretending it's the shell rather than using it as perl.	[reply] [d/l] [select]
Re: renaming files question by vek (Prior) on Jan 06, 2006 at 19:13 UTC
Will multiple users be running the original script concurrently? If not, you could modify it to just append to `lbn9.cnt` and you wouldn't have to rename anything. -- vek --	[reply] [d/l]


"be consistent"
	PerlMonks