visit_children with :recurse doesn't visit every child

[camertron]

I must be missing something. Using the code example in the readme on a small snippet of C++ code doesn't print anything, and trying on a larger code sample only prints a very small subset of the nodes in the AST. Here's the small chunk of C++ code I'm trying to parse:

ofPtr< ABTest > ABTest::create()
{
  ofPtr< ABTest > create( new ABTest() );
  create->init( create, ABTest::getValueStoreType() );
  if (create->_this)
  {
    return create;
  }
  return ofPtr< ABTest >();
}

Here's the ruby code that's trying to parse it:

require 'ffi/clang'

include FFI::Clang

index = Index.new(false)
translation_unit = index.parse_translation_unit("/Users/cameron/Desktop/test.cpp")
cursor = translation_unit.cursor

cursor.visit_children do |cursor, parent|
  puts "#{cursor.kind} #{cursor.spelling.inspect}"
  next :recurse
end

This code prints nothing. Is it possible next :recurse isn't working for some reason? Thanks for your help :)

[ioquatix]

Is that the entire file?

[camertron]

It's part of a larger file. The same ruby code I posted above only prints a small subset if the nodes in the AST when applied to it. I can't post the entire file since it's part of our closed-source codebase.

[ioquatix]

Please fork and submit a complete failing test case so we can investigate further.

[camertron]

Ok, I've worked up a minimal test case that shows that method calls inside methods defined in classes don't get visited. Is this the kind of test case you had in mind? I really don't know what I'm doing here...

[ioquatix]

Awesome, yeah, that is good. Can you make a PR? Then, we'll proceed from there.

[camertron]

Ok cool, created PR #36. Thanks for your help on this :)

[ioquatix]

In your example can you check translation_unit.diagnostics?

[camertron]

As in print out the return value?

[ioquatix]

Well if that is non-empty it will indicate the parser had some problems with the input and the cursor may not find what you expect. I'm really suggesting it so that you could try for yourself to see what is wrong

[camertron]

Ah ok I see. It's been long enough since I initially filed this bug that I don't have a need to parse C++ code anymore :( But thanks for circling back all the same.