[Chandler-dev] New conflicts API

Morgen Sagen morgen at osafoundation.org
Mon Feb 12 09:22:49 PST 2007


On Feb 11, 2007, at 10:18 PM, Philippe Bossut wrote:

> Hi Morgen,
>
> This sounds great! Hope Reid will chime in if he has issues. I just  
> have one question:
>
> Morgen Sagen wrote:
>>    field -- a string indicating which field was changed
>>
>
> What's the granularity of that "field"? I understand it's not  
> mapping attributes 1:1 .

The granularity of "field" can be a spectrum ranging from the entire  
item, down to an individual EIM field.  At the moment, a Conflict's  
field granularity *is* the entire item, but soon it will be an EIM  
field.  The currently defined EIM fields essentially map 1:1 to  
Chandler attributes.  There are several EIM record types defined (in  
sharing/model.py) which correspond to the various Stamps:

ItemRecord:
	- title (displayName)
	- triageStatus
	- triageStatusChanged
	- lastModifiedBy
	- createdOn

NoteRecord
	- body
	- icalUid
	- reminderTime

TaskRecord
	(Doesn't add any fields)

EventRecord
	- dtstart
	- dtend
	- location
	- rrule
	- exrule
	- rdate
	- exdate
	- status (transparency)

MailMessageRecord
	(TBD)

The records representing event modifications/exceptions are also  
being discussed.

I said that the granularity of a Conflict object's "field" is a  
spectrum -- that's because if we need to, we could add the ability  
for the code behind the conflict API to "link" two or more fields  
together such that they can only be rendered/applied/discarded as a  
unit.  We're not sure we have any fields which require this yet.

~morgen


More information about the chandler-dev mailing list