Editing text file

[freekngeek]

Hi, All

I have text file which looks like this

*|DSPF 1.0
*
*|DESIGN "TOP_1b_1a"
*|DATE "Tue Jun 18 11:42:12 2013"
*|VENDOR "Mentor Graphics Corp."
*|PROGRAM "Calibre xRC v2012.2_36.25"
*|DIVIDER /
*|DELIMITER :
* Nominal Temperature: 25C
* Circuit Temperature: 25C
*
.subckt TOP_1b_1a  PC_M1_M2_50_500_50_500_50_500_50_500_50_bot
+ PC_M1_M2_50_100_50_100_50_100_50_100_50_bot
+ PC_M1_M2_500_250_500_250_500_250_500_250_500_ctr
* PC_M1_M2_500_250_500_250_500_250_500_250_500_low
+ PC_M1_M2_75_500_75_500_75_500_75_500_75_bot
+ PC_M1_M2_75_100_75_100_75_100_75_100_75_bot
* PC_M1_M2_150_50_150_50_150_50_150_50_150_high
+ PC_M1_M2_100_500_100_500_100_500_100_500_100_bot
+ PC_M1_M2_100_100_100_100_100_100_100_100_100_bot
rM6_M7_JA_50_100_50_100_50_100_50_100_50_ctr/0 R435:pos
+ M6_M7_JA_50_100_50_100_50_100_50_100_50_ctr 238.22
rM6_M7_JA_50_100_50_100_50_100_50_100_50_ctr/1 R434:neg
+ M6_M7_JA_50_100_50_100_50_100_50_100_50_ctr 238.22
*
*|NET M6_M7_JA_50_100_50_100_50_100_50_100_50_high 0.0
*|I (R435:neg R435 neg B 0.0 10 3151)
R0 R0:pos R0:neg lvsres R=0.001 $X=9.975 $Y=10.999
R1 R1:pos R1:neg lvsres R=0.001 $X=9.975 $Y=122
R2 R2:pos R2:neg lvsres R=0.001 $X=9.975 $Y=243.999
R3 R3:pos R3:neg lvsres R=0.001 $X=9.975 $Y=355
R4 R4:pos R4:neg lvsres R=0.001 $X=159.963 $Y=10.999
R5 R5:pos R5:neg lvsres R=0.001 $X=159.963 $Y=122

What I want is to edit my text file using perl script. I want all the lines to be in one line which start with '+' sign, and leave rest of lines as it is. So, whenever it encounters '+' sign, script should print all the lines in one single line and then removes the '+' sign.

I would appreciate any kind of help or ideas to build this script. Thank you

[kcott]

G'day freekngeek,

I think this is probably the algorithm you're after but see notes at the end.

#!/usr/bin/env perl

use strict;
use warnings;

my $in_plus_lines = 0;

while (<DATA>) {
    chomp;

    if (/^\+(.*)$/) {
        print "\n" unless $in_plus_lines;
        $in_plus_lines = 1;
        print $1;
    }
    else {
        $in_plus_lines = 0;
        print "\n" if $. > 1;
        print "$_";
    }
}

print "\n";

__DATA__
* asterisk_line_1
*
.dot_line
+ plus_line_1
+ plus_line_2
* asterisk_line_3
+ plus_line_3
+ plus_line_4
* asterisk_line_4
+ plus_line_5
+ plus_line_6
plain_text_line_1
+ plus_line_7
plain_text_line_2
+ plus_line_8
*
*asterisk_line_6
*asterisk_line_7
plain_text_line_3

Output:

$ pm_cat_plus_lines.pl
* asterisk_line_1
*
.dot_line
 plus_line_1 plus_line_2
* asterisk_line_3
 plus_line_3 plus_line_4
* asterisk_line_4
 plus_line_5 plus_line_6
plain_text_line_1
 plus_line_7
plain_text_line_2
 plus_line_8
*
*asterisk_line_6
*asterisk_line_7
plain_text_line_3

Notes:

• Your sample input data could be substantially reduced: quite a few lines at the beginning and end could be removed; the length of lines could also be reduced as they can become hard to read when concatenated (due to wrapping).
• Expected output would have been useful. I don't believe your prosaic description of the wanted output is correct; e.g. "I want all the lines to be in one line which start with '+' sign" doesn't look right — I think you want groups of lines starting with '+', not all lines.
• The differences in data in consecutive lines is hard to discern, e.g. it takes a fair amount of effort to tell "PC_M1_M2_75_500_75_500_75_500_75_500_75_bot" and "PC_M1_M2_75_100_75_100_75_100_75_100_75_bot" apart.
• According to "and then removes the '+' sign", you end up with a leading space when the "+" lines are concatenated. Is that what you wanted? (Expected output would have made that obvious.)

As you can see, I've used completely different input; hopefully, the output is easier to read. Given the points above, my algorithm may be incorrect: there may be enough for you to correct it for your needs; if not, please address those points before asking follow-up questions.

[hdb]

or like this:

my $data;
{
  $/=undef;
  $data = <DATA>;
}
$data =~ s/\n\+//g;
print $data;

[CountZero]

One small correction to make it even better:

local $/=undef;

[hbm]

Open the file and a new file. Read the original file line by line. If the line starts with '+', remove the plus and the newline and accumulate it. If the line doesn't start with '+', print any accumulated lines and the current line to the new file.

[choroba]

No need to accumulate. You can print every line, having possibly removed the newline and the + if needed.