Paul Gillard
Not a member
Paul Gillard's latest activity
Wednesday, June 23 2010
-
Paul Gillard commented at 9:38 PMSecond patch to remove incorrect comment re-based against master and attached.
-
Paul Gillard commented at 11:02 AMSorry for the confusion. My comments were simply to explain to Kane how my patch works. Its not an indication that there is a problem with the patch. I'm happy ...
-
Paul Gillard commented at 9:16 AMThe foreign key attribute doesn't get cleared just the instance variable on the object which contains the associated object. An (imagined) IRB session might be:
...
Friday, June 18 2010
-
- Assigned user set to josh
Josh, I hope you don't mind me assigning this to you? Koz assigned a previous ticket I worked on regarding cloning to you (#3164).
-
- Tag changed from patch activerecord, clone to patch activerecord, clone, patch
I attach two patches. One which clears the association cache when cloning and so returning cloning to being genuinely shallow. The other which removes an incorr...
-
Paul Gillard created the ticket at 12:01 AMProblem
Cloned Active Record objects receive the original object's associations.
Explanation
Roughly speaking Ruby clones objects by copying instance variables ...
Tuesday, October 06 2009
-
Paul Gillard commented at 6:17 PMIt would seem my second point is not a valid one. #initialize_copy only gets called through #dup and #clone and neither accept a block.
Tuesday, September 08 2009
-
Paul Gillard created the ticket at 10:49 AMAs discussed here I'm attaching a patch which overrides #initialize_copy rather than #clone in order to implement cloning of AR objects. This will allow AR clas...
Tuesday, August 04 2009
-
Paul Gillard commented at 8:08 PMI've reworked the changes into two patches against master as requested.
The one thing I don't like about it is that there are now two ways in which
method_missi...
Wednesday, July 29 2009
-
Paul Gillard commented at 6:39 PMI wish I'd read my above comment once more. I meant your comment made sense to me rather than I thought the
attribute_reset!name made sense.reset_attribute!i...