gEDA-bug: [Bug 701431] Re: Export and use accessors for getting / setting object visibility

[Peter TB Brett <701431@xxxxxxxxxxxxxxxxxx>]

Sorry, my previous comment was a bit bogus. I've had a look at the
gschem case a bit further, and it seems to me that all of the places
where you don't currently have w_current available, the user_data is
unused. Additionally, the Multiattrib structure keeps a w_current
pointer for you as well.  Haven't looked at gattrib yet, but there
*should* be a similar setup.

-- 
You received this bug notification because you are a member of gEDA Bug
Team, which is subscribed to gEDA.
https://bugs.launchpad.net/bugs/701431

Title:
  Export and use accessors for getting / setting object visibility

Status in GPL Electronic Design Automation tools:
  In Progress

Bug description:
  Spilling this follow-up from bug 700841, the solution to which which
  introduces a libgeda private getter and setter for object visibility.

  Copying from my last comment there:

  I'm surprised that the getter / setter changes only affect libgeda
  though, but in a way this is neater in some ways.

  What I propose is that we push them for the stable-1.6 branch as is,
  which will fix this bug.

  There is still plenty of direct use of the object->visibility flag in
  gschem and gattrib - It would be amazing if you could write us a
  follow-up patch which applies on top of those you have already
  written, which:

  1. Moves the getter and setters out of the libgeda private APIs, and into the exported API
  2. Substitutes all use of the old API in gschem and gattrib

  To ensure you catch all cases, what I would suggest is that you also
  rename the member varibable from "visibility" to "_visibility". This
  will be the last patch in the series after you have replaced all but
  the legal usage in libgeda.

  What I would do myself, is use "stgit", manage these pages:

  1. Make a patch which exports the API
  2. Make a patch which renames the member variable "visibility" to "_visibility", and fixes up its usage in the getter + setters (ONLY)
  3. On top of this, make two more patches - one to use the setter, one to use the getter - in the other parts of the suite.
  You know you're done when the whole suite compiles again ;)

  Then I'd push those two patches, and "float" the rename patch (No. 2
  to the top. This way, the series should compile and work at every
  commit - which is important.

  Lets see - should we leave this bug open until all usage is converted, or should this be a new bug?
  (New bug I think.. I'll start it off with this comment.




_______________________________________________
geda-bug mailing list
geda-bug@xxxxxxxxxxxxxx
http://www.seul.org/cgi-bin/mailman/listinfo/geda-bug