From: Perl 5 Porters <perl5-porters@africa.nicoh.com>
Date: Sun, 17 Mar 1996 09:33:03 +0000 (+0000)
Subject: Move quotewords() explanation into body of routine
X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=commitdiff_plain;h=0dbfbeefeba4b4e7aa312cb453da87aff56a7769;p=p5sagit%2Fp5-mst-13.2.git

Move quotewords() explanation into body of routine
---

diff --git a/lib/Text/ParseWords.pm b/lib/Text/ParseWords.pm
index 97d7beb..8995138 100644
--- a/lib/Text/ParseWords.pm
+++ b/lib/Text/ParseWords.pm
@@ -40,32 +40,6 @@ to be a mis-feature.
 string in $_ and then pulling off words a bit at a time until $_
 is exhausted.
 
-The inner "for" loop builds up each word (or $field) one $snippet
-at a time.  A $snippet is a quoted string, a backslashed character,
-or an unquoted string.  We fall out of the "for" loop when we reach
-the end of $_ or when we hit a delimiter.  Falling out of the "for"
-loop, we push the $field we've been building up onto the list of
-@words we'll be returning, and then loop back and pull another word
-off of $_.
-
-The first two cases inside the "for" loop deal with quoted strings.
-The first case matches a double quoted string, removes it from $_,
-and assigns the double quoted string to $snippet in the body of the
-conditional.  The second case handles single quoted strings.  In
-the third case we've found a quote at the current beginning of $_,
-but it didn't match the quoted string regexps in the first two cases,
-so it must be an unbalanced quote and we croak with an error (which can
-be caught by eval()).
-
-The next case handles backslashed characters, and the next case is the
-exit case on reaching the end of the string or finding a delimiter.
-
-Otherwise, we've found an unquoted thing and we pull of characters one
-at a time until we reach something that could start another $snippet--
-a quote of some sort, a backslash, or the delimiter.  This one character
-at a time behavior was necessary if the delimiter was going to be a
-regexp (love to hear it if you can figure out a better way).
-
 =head1 AUTHORS
 
 Hal Pomeranz (pomeranz@netcom.com), 23 March 1994
@@ -87,6 +61,33 @@ sub shellwords {
 
 
 sub quotewords {
+
+# The inner "for" loop builds up each word (or $field) one $snippet
+# at a time.  A $snippet is a quoted string, a backslashed character,
+# or an unquoted string.  We fall out of the "for" loop when we reach
+# the end of $_ or when we hit a delimiter.  Falling out of the "for"
+# loop, we push the $field we've been building up onto the list of
+# @words we'll be returning, and then loop back and pull another word
+# off of $_.
+#
+# The first two cases inside the "for" loop deal with quoted strings.
+# The first case matches a double quoted string, removes it from $_,
+# and assigns the double quoted string to $snippet in the body of the
+# conditional.  The second case handles single quoted strings.  In
+# the third case we've found a quote at the current beginning of $_,
+# but it didn't match the quoted string regexps in the first two cases,
+# so it must be an unbalanced quote and we croak with an error (which can
+# be caught by eval()).
+#
+# The next case handles backslashed characters, and the next case is the
+# exit case on reaching the end of the string or finding a delimiter.
+#
+# Otherwise, we've found an unquoted thing and we pull of characters one
+# at a time until we reach something that could start another $snippet--
+# a quote of some sort, a backslash, or the delimiter.  This one character
+# at a time behavior was necessary if the delimiter was going to be a
+# regexp (love to hear it if you can figure out a better way).
+
     local($delim, $keep, @lines) = @_;
     local(@words,$snippet,$field,$_);