diff options
| author | ser <ser@b2dd03c8-39d4-4d8f-98ff-823fe69b080e> | 2008-01-20 04:31:57 +0000 |
|---|---|---|
| committer | ser <ser@b2dd03c8-39d4-4d8f-98ff-823fe69b080e> | 2008-01-20 04:31:57 +0000 |
| commit | 66aeb2f7080dea92703f10546fb3cbcc946f6fa3 (patch) | |
| tree | 4dfe958e610386eb752d60ad26a8f1702e1eb650 /lib/rexml/formatters | |
| parent | 00190701e0b8cc9ce5dbe9c836e9584d1c1caeb8 (diff) | |
| download | ruby-66aeb2f7080dea92703f10546fb3cbcc946f6fa3.tar.gz | |
r1479@bean: ser | 2008-01-19 14:26:31 -0500
r1483@bean: ser | 2008-01-19 14:47:23 -0500
Sam's fixes:
* Don't blow up on empty documents
* Add a test case for sorted attributes
* Making the output predictable simplifies unit tests, and doesn't cost
much given that most xml element have few attributes
* Ruby 1.9 revision 14922 is more strict
* Complete Ticket #134
* Fix for ticket #121
* Fix for ticket #124
* Fix for ticket #128
* Fix ticket #133
* Ticket #131 (Support Ruby 1.9)
* Fix for ticket #127
* Fix for ticket #123
* Add missing data needed by test case
r1481@bean (orig r1303): ser | 2008-01-19 17:22:32 -0500
Tagged for release
r1482@bean (orig r1304): ser | 2008-01-19 17:27:10 -0500
Version bump
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@15141 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Diffstat (limited to 'lib/rexml/formatters')
| -rw-r--r-- | lib/rexml/formatters/default.rb | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/lib/rexml/formatters/default.rb b/lib/rexml/formatters/default.rb index 77381bdf84..b4d63bc5b5 100644 --- a/lib/rexml/formatters/default.rb +++ b/lib/rexml/formatters/default.rb @@ -63,7 +63,7 @@ module REXML def write_element( node, output ) output << "<#{node.expanded_name}" - node.attributes.each_attribute do |attr| + node.attributes.to_a.sort_by {|attr| attr.name}.each do |attr| output << " " attr.write( output ) end unless node.attributes.empty? |