Changes between Version 1 and Version 2 of Ticket #16576, comment 10


Ignore:
Timestamp:
Feb 13, 2013, 2:00:43 PM (7 years ago)
Author:
Nick Fenwick
Comment:

Legend:

Unmodified
Added
Removed
Modified
  • Ticket #16576, comment 10

    v1 v2  
    33Now it would be nice to find a 'dijit' way of providing stopParser: true, perhaps by using a new mixin.
    44
    5 We also talked about the server-side-dijit blog post ..
    6 
    7 {{{
    8 <neekfenwick> i think his sentence "Monkey-patching the _TemplatedMixin prototype became the easiest solution." is where i've solved his problem
    9  my patch provides a clean solution to that problem .. but from there I lost the plot and bill had to point out the stopParser attribute to me (which I've now used in that tests file, and prelimiarily in our project code here, and it seems good)
    10 <neekfenwick> wildbill, at that point in the blog, he's saying that the default _TemplatedMixin does stuff he doesn't want (I presume, require a this.templateString and do the cache stringsubst bollocks) so he overrode it and provided a stripped down version
    11  i'm solving that by splitting the behaviour into a new mixin, and you get to choose which mixins you'll use in your scenario. _TemplatedMixin uses my new mixin too, so it's all working together nicely
    12 }}}
     5A further edit on this comment.  I incorrectly thought I'd 'solved' the problem James had with monkey patching _TemplatedMixin.  However, his problem was that he was instantiating `dijit/form/Button`, which via one of its bases extends _TemplatedMixin, and he was not in a position to use the new _AttachMixin if it had existed.  He must work with the Button dijit hierarchy as-is, and so must work around _TemplatedMixin's intention to replace the dom nodes with its own.