Is there a better way.

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

Replies are listed 'Best First'.
Re: Is there a better way. by Roy Johnson (Monsignor) on Feb 23, 2009 at 22:42 UTC
`@uploads = grep /$pattern/, @files;` ? Caution: Contents may have been coded under pressure.	[reply] [d/l]
Re^2: Is there a better way. by Anonymous Monk on Feb 23, 2009 at 22:49 UTC
Thanks I will give that a try.	[reply]
Re^2: Is there a better way. by afresh1 (Hermit) on Feb 25, 2009 at 21:34 UTC
The way I normally do this is similar, but skips using the temporary `@files` array. `opendir my $dh, $dir or die "Unable to opendir [$dir]: $!"; my @uploads = grep /$pattern/, readdir ($dh); closedir $dh;` [download] Sometimes I will add a `map { "$dir/$_" }` between `=` and `grep`, but it depends on the use. l8rZ, -- andrew	[reply] [d/l] [select]
Re: Is there a better way. by ig (Vicar) on Feb 23, 2009 at 23:42 UTC
Maybe: `chdir(Directory) or die "Directory: $!"; @uploads = glob("$caseNumber");` [download] or `my @uploads = map { $_ =~ m!./(.)! } glob("Directory/$caseNumber") +;` [download] If your directory or $caseNumber have spaces in them then you will want bsd_glob, which you can get by: `use File::Glob qw(:glob);` [download]	[reply] [d/l] [select]
Re: Is there a better way. by Illuminatus (Curate) on Feb 23, 2009 at 22:57 UTC
Do you really want the code to run faster, or are you just looking for a more elegant solution? my $caseNumber = $ARGV[0]; my $dir="directory"; my @files=`ls $dir/$caseNumber`; # will leave newlines [download] 6 of one...	[reply] [d/l]
Re^2: Is there a better way. by jwkrahn (Abbot) on Feb 23, 2009 at 23:44 UTC
Instead of running an external program to get a list of file names just use Perl's built-in glob function: `my @files = glob "$dir/$caseNumber";` [download]	[reply] [d/l]
Re^3: Is there a better way. by Lawliet (Curate) on Feb 24, 2009 at 01:47 UTC
Without digressing too much, why is it better to use Perl's built-in functions? And you didn't even know bears could type.	[reply]
Re^4: Is there a better way. by graff (Chancellor) on Feb 24, 2009 at 03:28 UTC
Re^4: Is there a better way. by jwkrahn (Abbot) on Feb 24, 2009 at 03:21 UTC
Re^4: Is there a better way. by k0d3Monkee (Initiate) on Feb 25, 2009 at 00:08 UTC
Re: Is there a better way. by ysth (Canon) on Feb 24, 2009 at 03:18 UTC
`my $caseNumber = $ARGV[0]; my $pattern = "(.?)$caseNumber(.?)";` [download] The ending (.*?) is pointless there. It's saying "must match ..., followed by zero or more characters". Since there will always be zero or more characters, just leave it off. And unless case numbers are a fixed length and distinguishable from the surrounding parts of the filename, that's not a good thing to do, since a case number "13" will match a file for case number "113", "130", etc. -- Online Fortune Cookie Search Office Space merchandise	[reply] [d/l]
Re: Is there a better way. by tokpela (Chaplain) on Feb 24, 2009 at 09:41 UTC
I don't see the solution of pulling into an initial @files array scalable. I would iterate over the directory handle - what if you want to search a directory that has thousands of files (ie lots of memory)? `my $caseNumber = $ARGV[0]; my $pattern = "(.?)$caseNumber(.?)"; my $count = 0; my @uploads; ##Open logfiles directory opendir DIR, "Directory" or die "Unable to open Directory, please try +again"; while (my $filename = readdir(DIR)) { if ($filename =~ /$pattern/) { my $filepath = "$directory/$filename"; # make full filepath push(@uploads, $filepath); } } closedir DIR; $count = scalar(@uploads);` [download] Also, I added the directory to the filename since I imagine that you will probably want to access the files at some point from the array.	[reply] [d/l]