On Dec 30, 2003, at 8:45 AM, Kevin Old wrote: [..]
$ko =~ s/\r/\\page/g; # change returns to \page
$ko =~ s/\f//g; # remove formfeeds
$ko =~ s/\t/\\tab/g; # change tabs to \tab
$ko =~ s/\\\\/\\/g; # escape backslashes
$ko =~ s/{/\\{/g; # escape left curly
$ko =~ s/}/\\}/g; # escape right curly
$ko =~ s/\n/\n\\par /g; # replace newlines with \n\par
$ko =~ s/\\par\s+\n\\par\s+\n\\par\s+\n+/\\par \n/migs; # cosolidate
"blank" lines
$ko =~ s/\\f0\\fs22\n\\page/\\f0\\fs22\n\\par /migs; # remove first page
break
This is not so much for laziness, but is more for learning what/how to consolidate.
Sometimes 'consolidation' does not save the nation.
but the lines that amuse me are
$ko =~ s/\n/\n\\par /g; # replace newlines with \n\par $ko =~ s/\\par\s+\n\\par\s+\n\\par\s+\n+/\\par \n/migs; # cosolidate
first we put some stuff in and then decide to rip it out...
What if we tried to solve the 'blank line' problem first?
There is also that minor buggaboo about
"\r\n" - our friends <CR><LF>that first line in the package would take say
Joe\r\n\r\nBob
and convert it into
Joe\page\n\page\nBob
and then we hit the \\par packer and would get to
Joe\page\n\par\page\n\parBob
and now we do not see our 'empty line'
So I would always start with the
take out what we want to remove
then dither with what is left.Say something along the line of
$ko =~ s/\r\n\s*\r\n/\r\n/mig ;
that would take us from our initial state to
Joe\r\nBob
and perchance a bit more sanity in the process.
Logically the next line would do the remove of things like the \f and then we start worrying about the rest of the substitutions.
I presume you have already peeked
perldoc perlretut
as well as picked up the book on mastering regular expressions.
ciao drieux
---
-- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] <http://learn.perl.org/> <http://learn.perl.org/first-response>
