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

Hello Monks ,

Pls. guide me which way to go with respect to scenario given below.

CASE:

PERL script executes set of commands with status of each command logged in a file; output from each command varies in width and also their category

Some samples category is given here

Project status :

DB update :

How do I format these categories / lines to look orderly (as given below)?

Project Status :

DB Update :

Any pointers would be of great help?

/RamThen/

Replies are listed 'Best First'.
Re: Pointers to formatting output
by BrowserUk (Patriarch) on Jan 09, 2004 at 13:42 UTC

    Reading between the lines of your post --literally. I had to look at the source of the html to work out what you were after:) -- I think you want one of these two: 

    printf "%20s: \n", $_ for 'this', 'that', 'and this other bit';

                this:
                that:
  and this other bit:

printf "%-20s: \n", $_ for 'this', 'that', 'and this other bit';

this                :
that                :
and this other bit  :

    [download]
    Examine what is said, not who speaks.
    "Efficiency is intelligent laziness." -David Dunham
    "Think for yourself!" - Abigail
    Timing (and a little luck) are everything!
[reply]
[d/l]

      thanks for pointers

      I followed suggestion given by BrowserUk and was able to log action as per my requirements

      Now when I mail this log using Net::SMTP module I do not get contents in same format as they are in log. I declared content-type to text in script

      From log :

      - Project : test_ch

      - Product : test

      Same log while sent as mail :

      - Project : test_ch

      - Product : test

      Does any body know reason for this behaviour ?

      /RamThen/

[reply]
Re: Pointers to formatting output
by Roger (Parson) on Jan 09, 2004 at 13:39 UTC
    I see that you are looking for a way to 'capitalize' the first letter of every word? And also align the colon ':'? (You have to be more specific with your question next time, otherwise it becomes a game of 'spotting the difference' again. ;-)
    my @text = (
   "Project status :",
   "DB update :",
   "Some column :",
);

my $max_line_width = 0;

# capitalize the first letter of every word, and record
# the maximum width of the lines seen.

my @formatted_text =
   map { s/(\w+)/\u\L$1/g;
         $max_line_width = length if $max_line_width < length;
         $_ } @text;

# insert spaces before the colon ':' to pad the string to 
# the maximum line width recorded

my @formatted_text =
   map { substr($_, -1, 0) = ' ' x ($max_line_width - length);
         $_ } @formatted_text;

print "$_\n" for @formatted_text;

    [download]
    And the output -
    Project Status :
Db Update      :
Some Column    :

    [download]
[reply]
[d/l]
[select]
Re: Pointers to formatting output
by Hena (Friar) on Jan 09, 2004 at 13:25 UTC
    For simple formatting (columns same size, numbers with 0 in front etc.) use printf (sprintf for internal handling of scalars). This is probably what you want. Next stage might be usage of print <<"END". That will allow scalars to be used ('END' wont).

    Most advanced way is then 'formats'. These allow much more control in what is printed where and how (man perlform).
[reply]
Re: Pointers to formatting output
by Hofmator (Curate) on Jan 09, 2004 at 13:25 UTC
    I'm not sure what you are trying to achieve, but have a look at printf or here at the monastery Using (s)printf().

    -- Hofmator

[reply]
Re: Pointers to formatting output
by calin (Deacon) on Jan 09, 2004 at 13:47 UTC
    Update: It looks that the OP doesn't want grouping by category ; he wants prettyfication of the lines themselves (see BrowserUk's comment above). The rest of my post below is moot.

    (hope I understand your question correctly)

    If a regexp like m/^(.+:)\s*$/ matches a category headline, and you have a logfile like: 

    Project status:
project created

DB update:
database initialized

Project status:
world domintation ahead

DB update:
database low on memory

DB update:
database borks ; all data lost

Project status:
shattered dreams

Project status:
starting all over again

    [download]

    then the following program will give you a starting point: 

    #!/usr/bin/perl

my %cats;
my $cat = "_UNDEFINED:";

open my $fh, '<', 'categories.txt' or die;

while (<$fh>) {
  next if m/^\s*$/;  #skip blank lines
  if (m/^(.+:)\s*$/) {
    $cat = $1;
    next;
  }
  $cats{$cat} .= ">> $_";
}

for (sort keys %cats) {
  print "$_\n";
  print $cats{$_};
}

close $fh;

    [download]

    The output is: 

    DB update:
>> database initialized
>> database low on memory
>> database borks ; all data lost
Project status:
>> project created
>> world domintation ahead
>> shattered dreams
>> starting all over again

    [download]
[reply]
[d/l]
[select]

Back to Seekers of Perl Wisdom