A would-be simple substitution...

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

Replies are listed 'Best First'.
Re: A would-be simple substitution... by Roger (Parson) on Mar 02, 2004 at 07:33 UTC
Your $time2 is only replacing the bits that "matched", the parts unmatched are not touched. And also in your other attempt: `$time2 =~ s/^(\d{2})(\d{2})(\d{2})\.(\d{2}).*?$/$1:$2:$3.$4/; # ^ # add anchor point` [download]	[reply] [d/l]
Re: A would-be simple substitution... by Tomte (Priest) on Mar 02, 2004 at 07:40 UTC
Shouldn't the substitution of $time2 work? IE -- see two digits after the '.' and then leave it at that? No ;-), because you are substituting the matching part, and the trailing digits do not match, they are therefore left alone. I even tried `$time2 =~ s/^(\d{2})(\d{2})(\d{2})\.(\d{2}).+?/$1:$2:$3.$4/;` [download] but got the same (bad) results. I didn't: `#!/usr/bin/perl # # use strict; use warnings; my $time2 = "123456.745768"; $time2 =~ s/^(\d{2})(\d{2})(\d{2})\.(\d{2}).+?/$1:$2:$3.$4/; print "$time2\n"; __END__ 12:34:56.74768` [download] Notice the missing five? `.+?` is non greedy and matches only what it has to, in this case one single character. take away the `?`, and it does, what you want; I'd change the `.` (dot) to `\d` and the `+` to a `` though, resulting in `s/^(\d{2})(\d{2})(\d{2})\.(\d{2})\d/$1:$2:$3.$4/;` [download] regards, tomte Hlade's Law: If you have a difficult task, give it to a lazy person -- they will find an easier way to do it.	[reply] [d/l] [select]
Re: Re: A would-be simple substitution... by GaijinPunch (Pilgrim) on Mar 02, 2004 at 08:53 UTC
See -- told you it was a long day. :) A cold AND jet lag are a losing combination. I can't tell if I'm about to pass out b/c I'm sick, or my body is just out of whack. Thanks guys.	[reply]
Re: A would-be simple substitution... by Aristotle (Chancellor) on Mar 02, 2004 at 10:24 UTC
Semantically you are composing a new value for the variable; why not write the code that way as well? `$time = "$1:$2:$3.$4" if $time =~ /^(\d{2})(\d{2})(\d{2})\.(\d{2})/;` [download] This also makes it easier to yell if the data didn't conform to expectations: `if($time =~ /^(\d{2})(\d{2})(\d{2})\.(\d{2})/) { $time = "$1:$2:$3.$4"; } else { die "Malformed time: $time\n"; }` [download] Makeshifts last the longest.	[reply] [d/l] [select]
Re: A would-be simple substitution... by gmpassos (Priest) on Mar 02, 2004 at 20:28 UTC
Remember that ".+?" is not doing what you think! If you want to indicate 0 or more matches, the regexp operator is "". The combination of +? and ? tells that it will match until find the next occurrence. Soo, we use this combination for things like that: `## to look until find > /<.?>/ ## if we use just ., the next occurrence will never be ## found, since .* already get everything. ## So +? is just the same thing of .*, but is for 1 or ## more occurrences.` [download] Cheers. Graciliano M. P. "Creativity is the expression of the liberty".	[reply] [d/l]
Re: A would-be simple substitution... by Anonymous Monk on Mar 03, 2004 at 08:28 UTC
This works `#!/usr/bin/perl # # use strict; use warnings; my $time = "123456.745768"; my $time2 = "123456.745768"; $time =~ s/^(\d{2})(\d{2})(\d{2})\.(\d{2})\d{4}/$1:$2:$3.$4/; $time2 =~ s/^(\d{2})(\d{2})(\d{2})\.(\d{2})\d*$/$1:$2:$3.$4/; print "$time\n"; print "$time2\n";` [download]	[reply] [d/l]