CM Analyst wrote:
I am struggling to modify an existing Perl script that was originally written for a Unix version of
ClearCase (SCM tool).


The script is supposed provide "keyword expansion"
functionality for files that are checked into the SCM
system.

My task is to make the same script run on Windows and
I am not having too much luck.

The script compiles without errors but it is NOT
expanding and writing the keyword values to the file
when it is checked back into Clearcase.


Can anyone point me in right direction...

TIA.

Amad.


############## use strict; use warnings;

Very good start. :-)


use Env;

You are not using this module anywhere so why is it here?


##EJM begin added to fix warnings/errors
my $log_expand = 0;
my $fulldate = 0;
my $basename = 0;
my $newident = "\$Id: ...\$";
my $user = "amad";

You do not use $user anywhere, are you sure you need it?


##EJM End

my $copyright = "";

#if (-f "NO_EXPAND")
#{
#    print "*** NOTICE *** - Not expanding
keywords...\n";
#exit 0;
#}

system("clearprompt proceed -mask proceed -prompt
\"Test to see if the script is firing...\"");

$log_expand=1;

#if(-f "NO_LOG_EXPAND" eq "No")

-f returns true or false, it will never be equal to the string "No".


#{
    $log_expand=0;
#    print "*** NOTICE *** - Not expanding
\$Log\$keyword\n";
#}
my $c = "c:\\Program
Files\\Rational\\ClearCase\\bin\\cleartool.exe";

You do not use $c anywhere, are you sure you need it?


# create nice alias for ClearCase environment vars

my $opkind = $ENV{'CLEARCASE_OP_KIND'};
my $idstr =  $ENV{'CLEARCASE_ID_STR'};
my $eltype = $ENV{'CLEARCASE_ELTYPE'};
my $file = $ENV{'CLEARCASE_XPN'};
my $comment = $ENV{'CLEARCASE_COMMENT'};
my $no_log = $ENV{'NO_LOG'};

You do not use $comment or $no_log anywhere, are you sure you need them?


#print "$file = $ENV{'CLEARCASE_XPN'}";
#print "$idstr =  $ENV{'CLEARCASE_ID_STR'}";

# only process keywords for text_file
#
if ($eltype ne "text_file")
{
exit 0;
}

# make a standard date string
#
chop ($fulldate=`date '+%d-%h-%Y %H:%M'`);

You do not use $fulldate anywhere, are you sure you need it?


chop ($copyright=`date '+%Y'`);
$copyright = "\$Copyright: (C) ".$copyright." My
Company, Inc. \$";

There is no need to call an external program to get the date and AFAIK those won't work on Windows anyway:


use POSIX 'strftime';

my $fulldate  = strftime '%d-%h-%Y %H:%M', localtime;
my $copyright = strftime '$Copyright: (C) %Y My Company, Inc. $', localtime;


# remove all but the basename of $file
#

($basename = $file) = [EMAIL PROTECTED]/@@;

use File::Basename;

my $basename = basename( $file );


# create an Id expansion based on the ClearCase
operation type
#
if ($opkind eq "checkin")
{
    $newident = "\$Id: [EMAIL PROTECTED]@$idstr  \$";

exit 0;
}

#elsif ($opkind eq "checkout")
#{
#    $newident = "\$Id: CHECKEDOUT [EMAIL PROTECTED]@$idstr
  \$";
#}

my $found_id = 0;
my $found_log = 0;

my $outfile = "$file.$$";
if (!open (IN, $file)) {

exit 0;
}
open (OUT, ">$outfile") || die "Cannot create out
file";

# read in the source file, expand keywords and write
the result
# back out
#

while (<IN>)
{
# expand the Id keyword
       if ($found_id == 0 && /\$Id.*\$/)
        {
            s/\$Id.*\$/$newident/;
            $found_id = 1;
        }

There is no need to use the same regular expression twice:

       if ($found_id == 0 && s/\$Id.*\$/$newident/)
        {
            $found_id = 1;
        }


        print OUT  $_;
}
# expand the Copyright keyword
#
if (/\$Copyright.*\$/) {
        s/\$Copyright.*\$/$copyright/;
}

The while loop has ended so the only thing in $_ is the last line from $file and even if you changed that it is not printed out anywhere. And again, there is no need to test the match before doing the substitution, just do the substitution.



# for a checkin operation we look to expand
the Log keyword
#
if ($log_expand == 1) {
        if (/\$Log\$/) {
                if ($opkind eq "checkin") {
                }
        }
}

This block of code is not doing anything?


# remember all the stuff from the beginning of the
line to the Log keyword
# so that when we create a multiline expansion we can
prefix each of our
# lines with the same junk, presumably some language's
comment characters

(my $prefix) = /(^.*)\$Log\$/;

You do not use $prefix anywhere, are you sure you need it?


# close everthing up

close (IN);
close (OUT);

rename ($file, "$file.bak") || die "Cannot rename
$file to $file.bak";
rename ($outfile, $file) || die "Cannot rename files";
unlink ("$file.bak") || die "Cannot remove $file.bak";

if ($found_id == 0)
{
   print "*** NOTICE - $file does not contain a \$Id\$
keyword\n";
}

if ($log_expand == 1 && $found_log == 0)

{
   print "*** NOTICE - $file does not contain a
\$Log\$ keyword\n";

}

# tell ClearCase to continue

exit 0;


John
--
use Perl;
program
fulfillment

--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
<http://learn.perl.org/> <http://learn.perl.org/first-response>




Reply via email to