Code review comment for lp:~mzanetti/unity-api/surfaceCount-property
Revision history for this message
| Daniel d'Andrada (dandrader) wrote | # |
© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
0e1f616
(Get the code!)
On 19/05/2016 08:08, Michael Zanetti wrote:
>> As I explained in the unity8 MP, I think this Application:
>> property is redundant.
>>
>> As http://
>> pips/revision/2398 shows.
>
> Just because you think your approach is better and you're on a mission to put a needs fixing on each and every merge proposal that's not yours? As I said, I strongly disagree on apis that require people to capture pointers to objects with potentially different lifetime in lambdas (I can't even count how many crashes we had in qtmir because of this bad practice).
>
> Api design is not just about naming signals not like slots. It's about providing an api that is fail-safe to use. Requiring the user to have knowledge about lifetime of other objects than the one he connects to is bad.
>
We will have to agree to disagree in that topic.
So let me get just one more reviewer/opinion so we can move on and
resolve that standoff.
I apologize for putting back to "need review" an already approved branch
just because of API style (I see that now that it can be seen as an
offensive and agressive move), but would appreciate if the message
exchange remained on the topic and didn't degrade to personal attacks.