Re: Can this be a 1 line or simpler regex?
by Abigail-II (Bishop) on Apr 05, 2003 at 07:30 UTC
|
I think the above is fine, but if you insist on less lines:
%& = ("\t" => '\t', "\n" => '\n', "\r" => '\r', "\f" => '\f');
s/[\t\n\r\f]/$&{$&}/g;
I wouldn't call it simpler though.
Abigail | [reply] [Watch: Dir/Any] [d/l] |
Re: Can this be a 1 line or simpler regex?
by BrowserUk (Patriarch) on Apr 05, 2003 at 08:38 UTC
|
Yes, though you might want to lay this out over two lines.
s{([\t\n\r\f])}{ local $_= '\\' . $1; tr[\t\n\r\f][tnrf]; $_ }eg
Examine what is said, not who speaks.
1) When a distinguished but elderly scientist states that something is possible, he is almost certainly right. When he states that something is impossible, he is very probably wrong.
2) The only way of discovering the limits of the possible is to venture a little way past them into the impossible
3) Any sufficiently advanced technology is indistinguishable from magic.
Arthur C. Clarke.
| [reply] [Watch: Dir/Any] [d/l] |
|
Beautiful. Though, without knowing exactly why, that layout bugs me. I'd rather do
s{([\t\n\r\f])}{ local $_= $1; tr[\t\n\r\f][tnrf]; "\\$_" }eg;
Shrug. :)
Makeshifts last the longest. | [reply] [Watch: Dir/Any] [d/l] |
|
| [reply] [Watch: Dir/Any] |
|
Ew. That that sprinkles eval blocks all over. That's also suboptimal. The original poster had the best version already. I fixed my own code to be nearly similar except that it doesn't benefit from perl's ability to compile a static regex only once. Your code is a static regex but you're imposing eval {} at each match point. An entirely different route might be to do some code generation so that the common parts are written only once, eval blocks aren't sprinked all over the place and the world is good again. The only issue is that now its not pretty anymore.
# Keep $escape somewhere or give it a proper name.
my $escape = eval "sub {\n".join("\n",map " \$_[0] =~ s/\\$_/\\\\$_
+/g;", qw/t r n f/)."\n};";
$escape->( $text );
| [reply] [Watch: Dir/Any] [d/l] |
|
|
Re: Can this be a 1 line or simpler regex?
by diotalevi (Canon) on Apr 05, 2003 at 07:35 UTC
|
(fixed the code) I suppose a single regex could be devised for this but it'd be suboptimal compared to what you already have. A short loop would also do this and again, I prefer your original. $text =~ s/\\$r/\\\\$r/g for qw/t n r f/;
for(qw't n r f') {
my $r = "\\$_";
$text =~ s/$r/$r/g;
}
| [reply] [Watch: Dir/Any] [d/l] [select] |
|
| [reply] [Watch: Dir/Any] |
Re: Can this be a 1 line or simpler regex?
by IndyZ (Friar) on Apr 05, 2003 at 07:28 UTC
|
This works:
s/([\t\n\r\f])/%m=("\t","\\t","\n","\\n","\r","\\r","\f","\\f");$m{$1}/ge;
but it just feels dirty.
-- IndyZ | [reply] [Watch: Dir/Any] [d/l] |
Re: Can this be a 1 line or simpler regex?
by Aristotle (Chancellor) on Apr 06, 2003 at 00:02 UTC
|
$_ = "one\ttwo\nthree four five\r";
s/(\t)|(\n)|(\r)|(\f)/
($1 && '\t' || '')
. ($2 && '\n' || '')
. ($3 && '\r' || '')
. ($4 && '\f' || '')
/ge;
Update:
$_ = "one\ttwo\nthree four five\r";
s/(\t)|(\n)|(\r)|(\f)/
$1 ? '\t' :
$2 ? '\n' :
$3 ? '\r' :
$4 ? '\f' :
''
/ge;
Makeshifts last the longest. | [reply] [Watch: Dir/Any] [d/l] [select] |
Re: Can this be a 1 line or simpler regex?
by Enlil (Parson) on Apr 06, 2003 at 03:13 UTC
|
Here is one more way to do it: 1 while(s!(\n)|(\r)|(\t)|(\f)!('\n','\r','\t','\f')[$#--1]!e);
-enlil
| [reply] [Watch: Dir/Any] [d/l] |
Re: Can this be a 1 line or simpler regex?
by I0 (Priest) on Apr 06, 2003 at 05:54 UTC
|
s/([\t\n\r\f])/${{"\t"=>"\\t","\n"=>"\\n","\r"=>"\\r","\f"=>"\\f"}}{$1}/g; | [reply] [Watch: Dir/Any] [d/l] |
A reply falls below the community's threshold of quality. You may see it by logging in. |