[Chicago-talk] Q on '-n' and wildcards
walter at torres.ws
Tue Sep 16 14:03:56 CDT 2003
A new question, about wildcards.
I have a script I wrote sometime ago that accepted and processed wildcard
@files = glob ( "*.txt" );
That worked just fine.
Now I'm trying to use a new technique (for me) the -n -s (as suggested by
Mark and Andy).
I have a test script that accepts filenames from the command line, pulls out
the '-o' (mine) filename for the results to be placed.
> test3.pl demo_1.log demo_2.log demo_3.log -0 results.log
It opens each file and runs down the contents, spitting them out to the
result.log just fine.
Next step is to handle wild cards...
> test3.pl demo_*.log -0 results.log
Anyone have any pointers on how to "upgrade" my script to handle wild cards?
Also, the current incarnation of my script places all results into a single
I have a desire to place the results of each file in their own log file, in
a different directory.
> test3.pl demo_*.log -0 /new/path
This should place the results of our munging into a file of the same name as
the original in the given directory.
I know how to handle coding when the file is over...
# built-in loop begins here ================================
# This technique uses Perl built-in methods to
# handle multiple files and iterate through each
# file in turn. Makes life a bit easier!
# We will need to munge each row in turn, later
# This just spits out the file name and row number
print RESULT "$ARGV: $.\n";
# Place end data into log file
# Name of file and number of records processed
&to_log("FILE: $ARGV $. records") if (eof);
# built-in loop ends here ==================================
So, I guess I' asking, is there a START_OF_FILE sort of thing?
I need to open a new RESULT file each time Perl opens an a new input file.
Any pointers, doc URLS, etc that I can read up on this?
I've searched and read till me eyes hurt. I've found nothing, but then
again, I figure I'm not asking the right questions.
Thanks for your help.
More information about the Chicago-talk