underscore regex delimiters #9853
Comments
From chip@seas.upenn.eduCreated by chip@seas.upenn.eduSubstitutions don't work if the delimiter is the underscore character. $ perl -p -e 's#/32##;' Perl Info |
From @rgs2009/8/26 chip@seas.upenn.edu (via RT) <perlbug-followup@perl.org>:
Hmm but _ is a word character. What if I have a function named |
|
The RT System itself - Status changed from 'new' to 'open' |
From zefram@fysh.orgchip@seas.upenn.edu (via RT) wrote:
The underscore is not perceived as a delimiter there, but as part of $ perl -MO=Deparse -e 's#/32##;' Other clues are available if you turn on warnings: $ perl -wce 's_/32__;' And if the substitution had contained other text then it would have $ perl -wce 's_\\__;' I believe the documentation is at fault. perlop(1) says: Any non-alphanumeric, non-whitespace delimiter may replace the Underscore is not alphanumeric or whitespace, but is evidently -zefram |
From @AbigailOn Thu, Aug 27, 2009 at 09:33:13AM +0100, Zefram wrote:
But even then, the documentation isn't correct. You *can* use word characters s _/32__ and s s/32ss are fine. The problem here lies in the tokenization part: the first token of In fact, this issue isn't any different from saying you cannot use C<< _ >> C<< func_ >> which isn't parsed as C<< func (_) >> either. Abigail |
From @demerphq2009/8/27 Abigail <abigail@abigail.be>:
The doocumentation should read "non-perl-word, non whitespace delimiter" We use "alphhanumeric" fairly regularly when we speak of
Right. This is not a bug. Cheers, -- |
From @nwc10On Thu, Aug 27, 2009 at 02:03:46PM +0200, demerphq wrote:
in the implementation. I think we agree that the documentation needs to be improved. Nicholas Clark |
From @davidnicol
Inline Patchdiff --git a/pod/perlop.pod b/pod/perlop.pod
index 0170202..26d0f1e 100644
--- a/pod/perlop.pod
+++ b/pod/perlop.pod
@@ -1167,11 +1167,13 @@ process modifiers are available:
c Do not reset search position on a failed match when /g is in effect.
If "/" is the delimiter then the initial C<m> is optional. With the C<m>
-you can use any pair of non-alphanumeric, non-whitespace characters
+you can use any pair of non-whitespace characters
as delimiters. This is particularly useful for matching path names
that contain "/", to avoid LTS (leaning toothpick syndrome). If "?" is
the delimiter, then the match-only-once rule of C<?PATTERN?> applies.
If "'" is the delimiter, no interpolation is performed on the PATTERN.
+When using a character valid in an identifier, whitespace is required
+after the C<m>.
PATTERN may contain variables, which will be interpolated (and the
pattern recompiled) every time the pattern search is evaluated, except
@@ -1381,13 +1383,13 @@ specific options:
e Evaluate the right side as an expression.
ee Evaluate the right side as a string then eval the result
-Any non-alphanumeric, non-whitespace delimiter may replace the
-slashes. If single quotes are used, no interpretation is done on the
-replacement string (the C</e> modifier overrides this, however). Unlike
-Perl 4, Perl 5 treats backticks as normal delimiters; the replacement
-text is not evaluated as a command. If the
-PATTERN is delimited by bracketing quotes, the REPLACEMENT has its own
-pair of quotes, which may or may not be bracketing quotes, e.g.,
+Any non-whitespace delimiter may replace the slashes. Add space after
+the C<s> when using a character allowed in identifiers. If single quotes
+are used, no interpretation is done on the replacement string (the C</e>
+modifier overrides this, however). Unlike Perl 4, Perl 5 treats backticks
+as normal delimiters; the replacement text is not evaluated as a command.
+If the PATTERN is delimited by bracketing quotes, the REPLACEMENT has
+its own pair of quotes, which may or may not be bracketing quotes, e.g.,
C<s(foo)(bar)> or C<< s<foo>/bar/ >>. A C</e> will cause the
replacement portion to be treated as a full-fledged Perl expression
and evaluated right then and there. It is, however, syntax checked at
diff --git a/t/op/subst.t b/t/op/subst.t
index 06c04e8..fe03599 100755
--- a/t/op/subst.t
+++ b/t/op/subst.t
@@ -7,7 +7,7 @@ BEGIN {
}
require './test.pl';
-plan( tests => 139 );
+plan( tests => 140 );
$x = 'foo';
$_ = "x";
@@ -263,6 +263,9 @@ eval 's{foo} # this is a comment, not a delimiter
{bar};';
ok( ! @?, 'parsing of split subst with comment' );
+$snum = eval '$_="exactly"; s sxsys;m 3(yactl)3;$1' #68804
+ok( $snum eq 'yactl', 'alpha delimiters are allowed' );
+
$_="baacbaa";
$snum = tr/a/b/s;
ok( $_ eq "bbcbb" && $snum == 4,
-- "Yes to health care, no to wealth care!" |
From @rgsThanks, applied to bleadperl. 2009/8/30 David Nicol <davidnicol@gmail.com>:
|
|
@rgs - Status changed from 'open' to 'resolved' |
From @davidnicolOn Sun, Aug 30, 2009 at 7:42 AM, Rafael
this is more thorough, and the test more fun: Inline Patchdiff --git a/pod/perlop.pod b/pod/perlop.pod
index adf0718..31489e1 100644
--- a/pod/perlop.pod
+++ b/pod/perlop.pod
@@ -962,6 +962,9 @@ from the next line. This allows you to write:
s {foo} # Replace foo
{bar} # with bar.
+You can use alphanumerics for quote delimiters too, in which case
+the whitespace is required.
+
The following escape sequences are available in constructs that interpolate
and in transliterations.
X<\t> X<\n> X<\r> X<\f> X<\b> X<\a> X<\e> X<\x> X<\0> X<\c> X<\N>
diff --git a/t/op/subst.t b/t/op/subst.t
index 30af8a2..f1b16ee 100644
--- a/t/op/subst.t
+++ b/t/op/subst.t
@@ -263,8 +263,8 @@ eval 's{foo} # this is a comment, not a delimiter
{bar};';
ok( ! @?, 'parsing of split subst with comment' );
-$snum = eval '$_="exactly"; s sxsys;m 3(yactl)3;$1';
-is( $snum, 'yactl', 'alpha delimiters are allowed' );
+{ my $match = eval '$_= q laurel;y subcostalis;s paperclip;m elite;$&';
+is( $match, 'lit', 'alpha delimiters are allowed' ); }
$_="baacbaa";
$snum = tr/a/b/s; |
Migrated from rt.perl.org#68804 (status was 'resolved')
Searchable as RT68804$
The text was updated successfully, but these errors were encountered: