Re: [incquery-dev] QE "attach" feature, bugzilla email storm

[Ujhelyi Zoltán <ujhelyiz@xxxxxxxxxx>]

Hi,

after todays work (when I tried to close quite a few tickets), I still have 33 open issues assigned to me for 0.8 (and have a lot of other stuff going on until that). So realistically, if some of the use cases mentioned in these tickets is especially irritating, I can try to provide some hotfix, but I will not have time to do it correctly. In this sense, the 2.4/2.5 support has no influence on my stance, as I will not have time to rewrite the required stuff.

I only started this conversation as I _know_ these issues are really problematic, so we should at least know what are we pushing before us. Maybe we could try to fix _some_ of them with quick fixes (at least for some common cases, etc.), but they would be far from a correct solution.

The same goes for performance: if we can find hot spots, and they are easy to work around, let's to that for now, and continue this conversation later.

To tell the truth, the only big thing I am trying to fill in until 0.8 is the replacement of the PDE-related stuff from the generator, that e.g. breaks our Xcore integration, but even that is at risk of being too late.

Cheers,
Zoli
-- Zoltán Ujhelyi
https://www.inf.mit.bme.hu/en/members/ujhelyiz

Fault Tolerant Systems Research Group
Budapest University of Technology and Economics

On 2014.04.08., at 22:19, Istvan Rath <rath@xxxxxxxxxx> wrote:

> In the original sense of your suggestion, are there any type provider-related issues that we can _realistically_ address within 0.8?
> 
> This type provider seems to be a major source of headache, as the performance-related problems seem to stem from here too, see my comments on https://bugs.eclipse.org/bugs/show_bug.cgi?id=428015.
> 
> Is this also related to Xtext version changes? I.e. does our decision to support 2.4/2.5 for 0.8 have any influence on this?
> 
> 
> On 08 Apr 2014, at 10:29, Ujhelyi Zoltán <ujhelyiz@xxxxxxxxxx> wrote:
> 
>> Hi,
>> 
>> we are talking about the following issues:
>> 
>> * 419114: Ambiguous variable type definitions: error even on internal variables, https://bugs.eclipse.org/bugs/show_bug.cgi?id=419114
>> * Quite complex issue, needs checking all possible cases
>> * 412972: parameter type is incorrect in generated code, https://bugs.eclipse.org/bugs/show_bug.cgi?id=412972
>> * Needs a full restructuring of the type provider; no test case available; depends on pattern ordering :(
>> * 410823: Method parameters in Matcher generated based on inconsistent type warning, https://bugs.eclipse.org/bugs/show_bug.cgi?id=410823
>> * Similar to previous; a bit more important, but basically the same issue
>> * 404652: Wrong type parameter in the generated evaluateXExpression() method for EJavaClass<T>, https://bugs.eclipse.org/bugs/show_bug.cgi?id=404652
>> * Only appears when using generics in the EMF metamodel
>> 
>> Three of them relates to ambiguous type definitions; they are very bad, but very complex issues. Basically, all three should be solved together, but it would require a large effort. Sadly, I am reasonably sure, I won't have time for this, although they can be really problematic issues.
>> 
>> The fourth is an easier decision: if we do not support EMF generics, it can be avoided.
>> 
>> Cheers,
>> Zoli
>> -- Zoltán Ujhelyi
>> https://www.inf.mit.bme.hu/en/members/ujhelyiz
>> 
>> Fault Tolerant Systems Research Group
>> Budapest University of Technology and Economics
>> 
>> On 2014.04.03., at 20:51, Istvan Rath <rath@xxxxxxxxxx> wrote:
>> 
>>> 
>>> On 03 Apr 2014, at 20:46, Ujhelyi Zoltán <ujhelyiz@xxxxxxxxxx> wrote:
>>> 
>>>> On the other hand, this means, we should focus only on critical type inference issues in 0.8, and push forward the other ones. Of course, what is critical, is more-or-less open to interpretation.
>>> 
>>> I generally agree, can you identify the ones that you consider critical? The rest should be pushed to the Future.
>>> 
>>> thanks
>>> Istvan
>>> 
>>> Istvan Rath, PhD
>>> Research Fellow
>>> Fault Tolerant Systems Research Group
>>> Budapest University of Technology and Economics
>>> rath@xxxxxxxxxx
>>> 
>>> 
>>> 
>>> _______________________________________________
>>> incquery-dev mailing list
>>> incquery-dev@xxxxxxxxxxx
>>> https://dev.eclipse.org/mailman/listinfo/incquery-dev
>> 
>> _______________________________________________
>> incquery-dev mailing list
>> incquery-dev@xxxxxxxxxxx
>> https://dev.eclipse.org/mailman/listinfo/incquery-dev
> 
> Istvan Rath, PhD
> Research Fellow
> Fault Tolerant Systems Research Group
> Budapest University of Technology and Economics
> rath@xxxxxxxxxx
> 
> 
> 
> _______________________________________________
> incquery-dev mailing list
> incquery-dev@xxxxxxxxxxx
> https://dev.eclipse.org/mailman/listinfo/incquery-dev