Permalink
Browse files

Implement a workaround for a bug in ruby-1.9.3p0.

The bug is that an error would be raised while attempting to convert a
template from one encoding to another.

Please see http://redmine.ruby-lang.org/issues/5564 for more details.

The workaround is to load all conversions into memory ahead of time,
and will only happen if the ruby version is *exactly* 1.9.3p0. The
hope is obviously that the underlying problem will be resolved in
the next patchlevel release of 1.9.3.
  • Loading branch information...
1 parent 7776055 commit fc988115f75bdca6aa95454208b49125a5041117 @jonleighton jonleighton committed Nov 6, 2011
View
@@ -62,6 +62,19 @@
persistent between requests so if you need to manipulate the environment
for your test you need to do it before the cookie jar is created.
+## Rails 3.1.2 (unreleased) ##
+
+* Implement a workaround for a bug in ruby-1.9.3p0 where an error would be raised
+ while attempting to convert a template from one encoding to another.
+
+ Please see http://redmine.ruby-lang.org/issues/5564 for details of the bug.
+
+ The workaround is to load all conversions into memory ahead of time, and will
+ only happen if the ruby version is *exactly* 1.9.3p0. The hope is obviously that
+ the underlying problem will be resolved in the next patchlevel release of
+ 1.9.3.
+
+ *Jon Leighton*
## Rails 3.1.1 (unreleased) ##
Binary file not shown.
@@ -3,6 +3,32 @@
require 'active_support/core_ext/object/try'
require 'active_support/core_ext/kernel/singleton_class'
+if RUBY_ENGINE == 'ruby' && RUBY_VERSION == '1.9.3' && RUBY_PATCHLEVEL == 0
+ # This is a hack to work around a bug in Ruby 1.9.3p0:
+ # http://redmine.ruby-lang.org/issues/5564
+ #
+ # Basically, at runtime we may need to perform some encoding conversions on the templates,
+ # but if the converter hasn't been loaded by Ruby beforehand (i.e. now), then it won't be
+ # able to find it (due to a bug).
+ #
+ # However, we don't know what conversions we may need to do a runtime. So we load up a
+ # marshal-dumped structure which contains a pre-generated list of all the possible conversions,
+ # and we load all of them.
+ #
+ # In my testing this increased the process size by about 3.9 MB (after the conversions array
+ # is GC'd) and took around 170ms to run, which seems acceptable for a workaround.
+ #
+ # The script to dump the conversions is: https://gist.github.com/1342729
+
+ filename = File.join(File.dirname(__FILE__), 'data', 'encoding_conversions.dump')
+ conversions = Marshal.load(File.read(filename))
+ conversions.each do |from, to_array|
+ to_array.each do |to|
+ Encoding::Converter.new(from, to)
+ end
+ end
+end
+
module ActionView
# = Action View Template
class Template

0 comments on commit fc98811

Please sign in to comment.