Parser goes weird on complex regexp in code

[zverok]

Steps to reproduce

Minimal code I was able to create for reproducing the issue:

class BaseWrapper
  def description(text)
    @object.description =
      text
      .gsub(/^[ \t]+/, '')      # remove spaces at a beginning of each line
      .gsub(/\A\n|\n\s*\Z/, '')
  end

  class PostProcessProxy
  end
end

What's going on after this is instead of PostProcessProxy there is ostProcessProxy in docs (and all the other entity names after problematic method is screwed this or that way).

Specifically, problem is created by those two lines:

      .gsub(/^[ \t]+/, '')      # remove spaces at a beginning of each line
      .gsub(/\A\n|\n\s*\Z/, '')

If I'll remove the second with complicated regexp -- everything becomes ok. If I remove the comment from the first of them -- everything becomes ok. But in current state of those lines everything is screwed up.

Environment details:

• OS: Ubuntu 14.04
• Ruby version (ruby -v): ruby 2.2.4p230 (2015-12-16 revision 53155) [i686-linux]
• YARD version (yard -v): yard 0.9.5

[iarkhanhelsky]

There is no problem with regex. Less code to reproduce same issue:

class BreakParser
  def break_parser
    obj # This comment breaks all.
      .bar
  end
end

class AffectedClass
end

calling yard --list will produce:

yard/t.rb:1: BreakParser
yard/t.rb:2: BreakParser#break_parser
yard/t.rb:8: ffectedClass

No matter how many methods after break_parser will be inserted in BreakParser class. All of them will be fine. Only next class will be affected.

The problem in AstNode#source_range it maps to"ffectedClass " (note trailing space) instead of "AffectedClass".

Issue causes totally unreadable documentation in more complex examples.

[lsegal]

Thanks for reporting-- sorry this took so long to track down!