I guess this is fixed in the trunk, but I think the highlighting of the first END in the heredoc should be corrected as well.

-- 
Cheers,
Bozhidar

On Thursday, September 5, 2013 at 2:45 PM, Dmitry Gutov wrote:

Bozhidar Batsov <bozhidar.batsov@gmail.com> writes:

Code like the following breaks font-locking logic of ruby-mode:

expected_source = <<-END.strip_indent
each do |x|
some_method
other_method
end
END

The second END gets highlighted as a constant, instead of a string. I
guess the two consecutive ends cause the problem.


WFM: http://i.imgur.com/YPsjoDS.png

Have you tried it in the current trunk?

The first END highlighting looks out of place, but it's caused by the
"Perl-ish keywords" rule.