Custom Query (18300 matches)

Filters
 
Or
 
  
 
Columns

Show under each result:


Results (181 - 183 of 18300)

Ticket Resolution Summary Owner Reporter
#3851 fixed 0.9: Updates to wire for two bugs + 2 new simple demos. Jared Jurkiewicz Jared Jurkiewicz
Description

0.9: Updates to wire for two bugs + 2 new simple demos.

This fixes an issue with invoking events from a topic publish, as well as adds in two new demos showing simple action chaining and topic publish/subscribe.

#3050 fixed 0.9: Widget cleanup and destroy functions bill Jared Jurkiewicz
Description

When looking at widget cleanup and destroy functions for the widget, it is not obvious how to go about doing a custom destroy function that will ensure specially allocated things, such as topic subscriptions or programatically created composite child widgets are properly cleaned up.

In 0.4.2, you would do so by implementing destroy() on your widget, in 0.9, destroy exists and does things like cleaning up connected events should you connect via the widget connect() function. But, in looking further at manager.js, when destroyAll is called, it only calls _destroy(), and _destroy does not call destroy(). So, if you put all your cleanup in destroy over-ride, and you do a page reload, manager isn't going to invoke it and your cleanup just sin't going to happen (as far as I can tell).

So, I guess the question/bug here is, is should manager.js' destroyAll() be calling destroy instead of _destroy() (Since destroy's default impl called _destroy anyway())? Is it expected that widget implementors should over-ride both destory and _destroy? If so, why is _destroy private?

In other words, what is the expected pattern for custom cleanup and is there a problem here in dijit? At a first glance, it looks like there is a problem to me.

#3550 duplicate 0.9: XML handling severely broken in IE 6 alex Jared Jurkiewicz
Description

0.9: XML handling severely broken in IE 6

This happened in the past couple days. Something with how xhrGet handles the parameter handleAs: "xml" has severely broken on IE 6. Namely, it passes the deferred callback a completely unusable object. This is breaking the dojox.data.OpmlStore? and dojox.data.XmlStore? at a minimum. anyone using xhrGet on IE 6 and handleAs xml are broken. The same testcase works fine on Firefox 2.0, Safari beta 3, Opera 9.2, etc.

Output from Firefox 2.0 in console:

XML object is type: object XML object toString is: [object XMLDocument] XML content is: <root> <child/> <child>Text here</child> <foo/> </root> Trying to dump attribute i: load with value function load() { [native code] } Trying to dump attribute i: nodeName with value #document Trying to dump attribute i: nodeValue with value null .. .

Output on IE 6 for the same code:

XML object is type: object XML object toString is: XML content is: undefined On Error attribute: name value is: TypeError? On Error attribute: message value is: Object doesn't support this action On Error attribute: number value is: -2146827843 On Error attribute: description value is: Object doesn't support this action

The errors come from the part of the test where it tries to iterate over the attributes of the object and emit the values.

Note: See TracQuery for help on using queries.