summaryrefslogtreecommitdiff
path: root/lib
diff options
context:
space:
mode:
authorusa <usa@b2dd03c8-39d4-4d8f-98ff-823fe69b080e>2014-02-14 04:24:27 +0000
committerusa <usa@b2dd03c8-39d4-4d8f-98ff-823fe69b080e>2014-02-14 04:24:27 +0000
commita20b957bafba111f95d238b66013037fe30baadb (patch)
tree532eefae940d070287ba6bc82706904dd6aea36b /lib
parenta540b10c96c187005c5037bcc942262234c43b45 (diff)
merge revision(s) 40848: [Backport #8425]
* lib/webrick/htmlutils.rb (WEBrick::HTMLUtils#escape): replace HTML meta chars even in non-ascii string. [Bug #8425] [ruby-core:55052] * lib/webrick/httputils.rb (WEBrick::HTTPUtils#{_escape,_unescape}): fix %-escape encodings. [Bug #8425] [ruby-core:55052] git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_9_3@44934 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Diffstat (limited to 'lib')
-rw-r--r--lib/webrick/htmlutils.rb5
-rw-r--r--lib/webrick/httputils.rb14
2 files changed, 15 insertions, 4 deletions
diff --git a/lib/webrick/htmlutils.rb b/lib/webrick/htmlutils.rb
index ed901f1ce2..90994f18b8 100644
--- a/lib/webrick/htmlutils.rb
+++ b/lib/webrick/htmlutils.rb
@@ -15,12 +15,13 @@ module WEBrick
# Escapes &, ", > and < in +string+
def escape(string)
- str = string ? string.dup : ""
+ return "" unless string
+ str = string.dup.force_encoding('binary')
str.gsub!(/&/n, '&amp;')
str.gsub!(/\"/n, '&quot;')
str.gsub!(/>/n, '&gt;')
str.gsub!(/</n, '&lt;')
- str
+ str.force_encoding(string.encoding)
end
module_function :escape
diff --git a/lib/webrick/httputils.rb b/lib/webrick/httputils.rb
index f029dacb56..d99573762c 100644
--- a/lib/webrick/httputils.rb
+++ b/lib/webrick/httputils.rb
@@ -350,8 +350,18 @@ module WEBrick
def _make_regex(str) /([#{Regexp.escape(str)}])/n end
def _make_regex!(str) /([^#{Regexp.escape(str)}])/n end
- def _escape(str, regex) str.gsub(regex){ "%%%02X" % $1.ord } end
- def _unescape(str, regex) str.gsub(regex){ $1.hex.chr } end
+ def _escape(str, regex)
+ str = str.dup.force_encoding('binary')
+ str.gsub!(regex) {"%%%02X" % $1.ord}
+ # %-escaped string should contain US-ASCII only
+ str.force_encoding(Encoding::US_ASCII)
+ end
+ def _unescape(str, regex)
+ str = str.dup.force_encoding('binary')
+ str.gsub!(regex) {$1.hex.chr}
+ # encoding of %-unescaped string is unknown
+ str
+ end
UNESCAPED = _make_regex(control+space+delims+unwise+nonascii)
UNESCAPED_FORM = _make_regex(reserved+control+delims+unwise+nonascii)