I wanted to revisit this again... Planning to kill this setting (with
API deprecation) to reduce branching throughout the code. Please voice
your objections if you have any.
Andrus
On Jan 25, 2006, at 1:04 AM, Andrus Adamchik wrote:
> Hmm... now that I am going through the places where it is used, it
> doesn't look that bad anymore (partially due to the
> GenericSelectQuery -> QueryMetadata refactoring I've done in the
> last few days). I guess I'll keep it for now, as it may become more
> important once we implement other types of inheritance in addition
> to the single-table variety.
>
> Sorry for the noise :-)
>
> Andrus
>
>
> On Jan 24, 2006, at 5:33 PM, Andrus Adamchik wrote:
>
>> Great. There is no objections. I'll go ahead with it.
>>
>> Andrus
>>
>>
>> On Jan 23, 2006, at 3:39 PM, Cris Daniluk wrote:
>>
>>> I've never used it, but I *have* been annoyed by reading through the
>>> spaghetti code it spawns :)
>>>
>>> On 1/23/06, Gentry, Michael (Contractor) <michael_gentr..anniemae.com
>>> > wrote:
>>>> I've never used it ...
>>>>
>>>> -----Original Message-----
>>>> From: Andrus Adamchik [mailto:andru..bjectstyle.org]
>>>> Sent: Saturday, January 21, 2006 4:40 PM
>>>> To: Cayenne Devel
>>>> Subject: Deprecating "isResolvingInherited"
>>>>
>>>>
>>>> Anybody ever set "isResolvingInherited" to FALSE?
>>>>
>>>> Supposedly this can buy some performance improvements, but the
>>>> resulting object graph ends up being inconsistent, besides we
>>>> need to
>>>> pass this flag around in many places, resulting in spaghetti code.
>>>>
>>>> I am thinking of deprecating this flag. Any objections?
>>>>
>>>> Andrus
>>>>
>>>
>>
>>
>
>
This archive was generated by hypermail 2.0.0 : Wed Nov 19 2008 - 02:28:28 EST