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

fellow monks i think i just found a little mistake i am currently doing the exercise #1 in the tutorials under string matching & regular expressions the original code looks like this 
1: #!/usr/bin/perl
2: #this programs prompts a user for a filename and then
3: #replaces all tabs in that file with 3 spaces.
4: print "What file do you want to \"detabify?\" ";
5: $file=<>;
6: chomp $file;
7: $/=undef;             #changes the input record separator so it'll 
+read the entire file in at once
8: open FILE, "<$file";  
9: $text=<FILE>;         #reads the entire file in
10: close FILE;
11: $text=~s/\t/   /g;    #replaces tabs
12: open FILE, ">$file";
13: print $text;          #write changes to file
14: close FILE;

[download]
shouldnt line 12 open $text instead of $file? snowrider

Replies are listed 'Best First'.
Re: is this correct
by kschwab (Vicar) on Feb 02, 2001 at 22:28 UTC
    It is broken, but not at line 12.

    Line 13 should read: 

    print FILE $text;

    [download]
    As it stands now, it reads in the file, prints the untabbified text to STDOUT, and truncates the supplied user-specified file.
[reply]
[d/l]
      And assuming you actually run the program, you'll see that problem firsthand and quickly fix that line so it acts correctly.

      (Note: this is simply a friendly suggestion that some attempt be made to understand the program/problem when you post.)

      ALL HAIL BRAK!!!

[reply]
        PsychoSpunk this is my second day tring to program in perl since the program wasnt running correctly i did try to fix the problem but since i dont have any experience in perl thats alittle hard without the help of you guys(fellow monks). thanks for your help snowrider
[reply]
      You could also do this... 
      select(FILE);
print $text;
print "blah blah blah\n";

      [download]
      And so on... all prints would go to FILE

      Update:
      Snowrider - This is assuming you have multiple files open and are printing to them as you go... and as I know the basic tasks you will have to perform as you come up to speed with Perl... this might be useful to you.

      i.e.,

      open DRILL_FILE, ">$drill_file";
open ROUT_FILE, ">$rout_file";
select(DRILL_FILE);
print "stuff...\n";
select(ROUT_FILE);
print "other stuff...\n";

      [download]

      Prost,
      Moe

[reply]
[d/l]
[select]
Re: is this correct
by davorg (Chancellor) on Feb 02, 2001 at 22:10 UTC

    No, it looks right to me. $file contains the name of the file that you are processing and $text contains the contents of that file.</p. --
    <http://www.dave.org.uk>

    "Perl makes the fun jobs fun
    and the boring jobs bearable" - me

[reply]
Re: is this correct
by Gloom (Monk) on Feb 03, 2001 at 00:10 UTC
    What's append if the file take 150mb of disk space ?

    Maybe you may use something like this :

    # perl -e 'while(<>){s/\t/   /g;print}' < file > file.tmp;\
> mv file.tmp file;

    [download]

    Update : Thank's arhuman for this comment. This option is really usefull.

    Hope this helps

[reply]
[d/l]
      In case like this (a while loop around one or more files) the perl -ne is usually a good candidate...

      Something like :
      perl -ne 's/\t/   /g;print' file > file.tmp;
mv file.tmp file;

      [download]
[reply]
[d/l]
[select]
        To abstract it a little more

        perl -i.back -pe 's/\t/    /g' file

        The -i means copy original file to file.back and put the output of program in file. The -p is like -n but does while(<>){YOUR_CODE; print;} instead.

[reply]
[d/l]
[select]

Back to Seekers of Perl Wisdom