3)Embed the command line arguments into the top of your script.pl, perhaps as you are sending it across the wire to perl. This would require some pre-processing of the script and some assumptions of where you can embed the code - if I were to do it this way, I would probably have a token of some sort in the script that I would replace with my parameters --> The cmd line argumetns are also dynamic in nature..but i dont understand this entire statment correctely. can you give an example ?
In your script you could have something like:
At this point, you would use something locally to replace '%%REPLACEDTOKEN%%' with your parameters, and pipe that to a perl call on the remote machine. sed -e .... < script.pl | ssh user@remote perl# Earlier stuff in the script... # Fill in the arguments if running remotely @ARGV ||= qw(%%REPLACEDTOKEN%%); # ... and back it out if @ARGV should really be empty @ARGV = () if $ARGV[0] eq "\%\%REPLACEDTOKEN\%\%"; # Rest of your script
I still have concerns that this is the most robust solution to your problem.
Do your scripts use any temporary files? If so, I would seriously reconsider your answer to number 2. If you can write temp files, you can also write a temporary perl script.
--MidLifeXis
In reply to Re^3: run perl script with cmd line in shell
by MidLifeXis
in thread run perl script with cmd line in shell
by Anonymous Monk
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |