From: mike@bill.iac.net <mike@bill.iac.net>
Date: Sat, 27 Jun 1998 03:49:13 +0000 (+0000)
Subject: [ PATCH 5.004 68 ] Text::ParseWords, ^W fixed, version 3.1
X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=commitdiff_plain;h=b174585de5ccc9973ba572393b2b34e1a6a5b749;p=p5sagit%2Fp5-mst-13.2.git

[ PATCH 5.004 68 ] Text::ParseWords, ^W fixed, version 3.1

	Message-ID: <19980627034913.A32220@bill.minivend.com>

p4raw-id: //depot/perl@1249
---

diff --git a/lib/Text/ParseWords.pm b/lib/Text/ParseWords.pm
index d3a89f0..95f0e9b 100644
--- a/lib/Text/ParseWords.pm
+++ b/lib/Text/ParseWords.pm
@@ -1,7 +1,7 @@
 package Text::ParseWords;
 
-use vars qw($VERSION @ISA @EXPORT);
-$VERSION = "3.0";
+use vars qw($VERSION @ISA @EXPORT $PERL_SINGLE_QUOTE);
+$VERSION = "3.1";
 
 require 5.000;
 
@@ -48,22 +48,28 @@ sub nested_quotewords {
 
 
 sub parse_line {
+	# We will be testing undef strings
+	local($^W) = 0;
+
     my($delimiter, $keep, $line) = @_;
     my($quote, $quoted, $unquoted, $delim, $word, @pieces);
 
     while (length($line)) {
-	($quote, $quoted, $unquoted, $delim) =
+
+	($quote, $quoted, undef, $unquoted, $delim, undef) =
 	    $line =~ m/^(["'])                 # a $quote
-                        ((?:\\.|[^\1\\])*?)    # and $quoted text
-                        \1                     # followed by the same quote
+                        ((?:\\.|(?!\1)[^\\])*)    # and $quoted text
+                        \1 		       # followed by the same quote
+                        ([\000-\377]*)	       # and the rest
 		       |                       # --OR--
                        ^((?:\\.|[^\\"'])*?)    # an $unquoted text
-                        (\Z(?!\n)|$delimiter|(?!^)(?=["']))  
+		      (\Z(?!\n)|$delimiter|(?!^)(?=["']))  
                                                # plus EOL, delimiter, or quote
-                      /x;                      # extended layout
+                      ([\000-\377]*)	       # the rest
+		      /x;		       # extended layout
+	return() unless( $quote || length($unquoted) || length($delim));
 
-        return() unless(length($&));
-        $line = $';
+	$line = $+;
 
         if ($keep) {
 	    $quoted = "$quote$quoted$quote";
@@ -71,6 +77,7 @@ sub parse_line {
         else {
 	    $unquoted =~ s/\\(.)/$1/g;
 	    $quoted =~ s/\\(.)/$1/g if ($quote eq '"');
+	    $quoted =~ s/\\([\\'])/$1/g if ( $PERL_SINGLE_QUOTE && $quote eq "'");
 	}
         $word .= ($quote) ? $quoted : $unquoted;
  
diff --git a/t/lib/parsewords.t b/t/lib/parsewords.t
index 21ed0d3..9079179 100755
--- a/t/lib/parsewords.t
+++ b/t/lib/parsewords.t
@@ -7,7 +7,7 @@ BEGIN {
 
 use Text::ParseWords;
 
-print "1..15\n";
+print "1..17\n";
 
 @words = shellwords(qq(foo "bar quiz" zoo));
 print "not " if $words[0] ne 'foo';
@@ -17,11 +17,16 @@ print "ok 2\n";
 print "not " if $words[2] ne 'zoo';
 print "ok 3\n";
 
+# Gonna get some undefined things back
+local($^W) = 0;
+
 # Test quotewords() with other parameters and null last field
 @words = quotewords(':+', 1, 'foo:::"bar:foo":zoo zoo:');
 print "not " unless join(";", @words) eq qq(foo;"bar:foo";zoo zoo;);
 print "ok 4\n";
 
+$^W = 1;
+
 # Test $keep eq 'delimiters' and last field zero
 @words = quotewords('\s+', 'delimiters', '4 3 2 1 0');
 print "not " unless join(";", @words) eq qq(4; ;3; ;2; ;1; ;0);
@@ -66,6 +71,9 @@ print "ok 11\n";
 print "not " if (@words);
 print "ok 12\n";
 
+# Gonna get some more undefined things back
+$^W = 0;
+
 @words = nested_quotewords('s+', 0, $string);
 print "not " if (@words);
 print "ok 13\n";
@@ -79,3 +87,17 @@ print "ok 14\n";
 $result = join('|', parse_line(':', 0, ':"0":'));
 print "not " unless ($result eq '|0|');
 print "ok 15\n";
+
+# Test for \001 in quoted string
+$result = join('|', parse_line(':', 0, ':"' . "\001" . '":'));
+print "not " unless ($result eq "|\1|");
+print "ok 16\n";
+
+$^W = 1;
+
+# Now test perlish single quote behavior
+$Text::ParseWords::PERL_SINGLE_QUOTE = 1;
+$string = 'aaaa"bbbbb" cc\ cc \\\\\"dddd\' eee\\\\\"\\\'ffff\' gg';
+$result = join('|', parse_line('\s+', 0, $string));
+print "not " unless $result eq 'aaaabbbbb|cc cc|\"dddd eee\\\\"\'ffff|gg';
+print "ok 17\n";