why don't you just change the case of the file name before you start? then you could have your script always look for the lower case file name?
eval 'exec perl $0 ${1+"$@"}'
if 0;
# don't modify below here
#-------------------------
$VERSION='2.0';
# chcase 2.0
# Changes case of filenames
# http://www.blemished.net/chcase.html
# supermike@blemished.net
#
use Getopt::Std;
use Cwd;
use File::Basename;
sub HELP_MESSAGE {
print<<EOT;
chcase $VERSION
USAGE:
chcase [-terdlouCcqn] [-x '<perl exp>'] FILE...
-t : Test mode (don't actually rename any files)
-e : Print examples
-r : Rename recursively
-d : Also rename directories
-l : Rename & follow symbolic links (default is not to)
-o : Overwrite if file exists
-u : Change to uppercase (default is lower)
-C : Capitalize each word
-c : Capitalize first character only
-q : Quiet mode (no output)
-n : No escape characters (for bold/inverse output)
-x '<perl exp>' : Perl expression to operate on filename
like s/// or tr/// (you need the quotes)
case of filename not changed when this option used
EOT
exit;
}
sub EXAMPLES() {
print<<EOT;
chcase $VERSION
EXAMPLES:
\$ chcase My.DOC *.JPG /tmp/FileName.txt
=> files are changed to lowercase
\$ chcase -rdt .
=> show what would happen if we renamed all files and sub dirs
recursively in pwd to lowercase; remove t to do it for real
\$ chcase -rdoc /some/dir ./images/ *png
=> files and dirs recursively capitalized,
overwrite any existing already capitalized file
\$ chcase -x 's/projectx/PlanB/; s/ /_/g' *.doc
=> renames *.doc files substituting "projectx" with "PlanB",
and replacing all spaces with underscores
\$ find ./music/ -name "*mp3" | xargs chcase -x 'tr/a-zA-Z0-9.-_/_/cs'
=> find mp3 files under music/ dir, transliterate (and squash)
non-alphanumeric characters except for . or - to _
EOT
exit;
}
sub msg {
return if $opt_q; my($ol,$new,$err)=@_;
$out=' 'x($lvl-1).$s.$ol; $out.=" => $new" if $new;
$out.=" : $i$err" if $err; print "$out$n\n";
}
sub chcase {
$_=$old=shift;
$dn=dirname($_).'/'; $dn="" if ($dn eq './');
$_=basename($_);
if ($opt_x) { eval $opt_x; } else { $_=$opt_u ? uc : lc; }
$_=$dn.$_;
return($old) if (($old eq $_) or -d or (-f and !$opt_o));
if ($opt_t or rename($old, $_)) { msg($old,$_); return($opt_t ? $old : $_); }
else { msg($old,$_,$!); return($old); }
}
sub proc {
shift;
if (-l and !$opt_l) { return; }
if (-f) { chcase($_); }
elsif (-d) {
s/\/+$//;
if ($opt_d) { $_=chcase($_); }
if ($opt_r) {
my $cwd=cwd();
if (chdir($_)) { msg("$b$_/"); } else { msg("$b$_/",0,$!); return; }
$lvl++; $s='|__';
opendir(D,'.');
foreach (readdir(D)) { proc($_) unless (($_ eq '.') or ($_ eq '..')); }
closedir(D);
chdir($cwd);
$s='' if (--$lvl == 0);
}
}
}
sub init() {
$Getopt::Std::STANDARD_HELP_VERSION=true;
getopts( 'erdouqtnlCcx:' ) or HELP_MESSAGE();
EXAMPLES() if $opt_e;
unless ($opt_n) {$b="\033[01m"; $i="\033[03m"; $n="\033[0m"; }
$opt_x='s/([^_-\s]+)/\u\L$1/g' if $opt_C;
$opt_x='$_=ucfirst(lc)' if $opt_c;
}
init();
while ($_=shift @ARGV) { proc($_); }
-Charles
I have a file I am accessing that constantly changes from all lowercase to all UPPERCASE (this is caused by a wierd bug in some proprietary database we use). My script use to copy this file with no problems until we upgraded our database. So I thought the solution would be easy with an "if" statement:
if
[ -e partmast.dbf ]
then
dbFile="partmast.dbf"
else
dbFile="PARTMAST.DBF"
fi
The strange thing is that "-e" which means "file exists" ignores case. So no matter if the file is really lowercase or UPPERCASE, the result is always dbFile="partmast.dbf"
I thought UNIX/Linux always adhered to proper case. Anyway, if anyone can help me, that would be great. Thank you in advance!