GSoC Project Idea

[anubhav joshi]

Earlier I posted this on as a reply on the query conversation that I had started earlier but then I didn't get any response there so I am reposting this here.
I originally wanted to work on aggregates/annotation improvement in GSoC and still wish the same. The issues that I want to consider are:

Implementing arithmetic operations on aggregation/annotation.
Implementing conditional aggregates.
Fixing the errors related to the working eg. default values for aggregations/annotations, problem in using nested statements like filter() with them, etc
Adding datetime grouping to aggregation.

I have started working on the above issues. Regarding the errors part I have been able to figure out the problems.

Also I did some surfing the net regarding implementing arithmetic operations on aggregation.. I have found few ways:
1.)Adding it directly.
2.)Changing the complete implementation to f() or q() objects.

I have yet to work on the remaining issues which I surely will.

I want to know whether this will make up a good GSoC project.


Anubhav Joshi

[anubhav joshi]

I want to know whether this will make up a good GSoC project as I am also interested in 'improving the error messages part', therefore I want to make a final choice between the two topics.
Awaiting replies and suggestions.

Anubhav Joshi

[Josh Smeaton]

As I mentioned in the last thread, I've already made some progress on improving aggregates and annotations. I've submitted this PR for review https://github.com/django/django/pull/2184 and discussion here: https://groups.google.com/forum/#!topic/django-developers/8vEAwSwJGMc

Again, I'm unsure if the work I've done will eventually be committed because everyone is quite busy at the moment, though I'm hopeful. I've already completed arithmetic/F() expressions, and integration with .filter() etc. I've also done a Proof of Concept branch on Conditional Aggregates that is linked to the pull request above. I'm unsure what you mean when you say adding datetime grouping to aggregation though, would you mind expanding on that?

From my point of view, it'd be a waste to spend a full GSoC project on aggregates unless you came up with a significantly different implementation and could explain the relative benefits. That said, my PR hasn't been reviewed yet, so there may be an opportunity there.

Regards,

Josh

[anubhav joshi]

On Monday, February 24, 2014 4:21:15 PM UTC+5:30, Josh Smeaton wrote:

As I mentioned in the last thread, I've already made some progress on improving aggregates and annotations. I've submitted this PR for review https://github.com/django/django/pull/2184 and discussion here: https://groups.google.com/forum/#!topic/django-developers/8vEAwSwJGMc

Again, I'm unsure if the work I've done will eventually be committed because everyone is quite busy at the moment, though I'm hopeful. I've already completed arithmetic/F() expressions, and integration with .filter() etc. I've also done a Proof of Concept branch on Conditional Aggregates that is linked to the pull request above. I'm unsure what you mean when you say adding datetime grouping to aggregation though, would you mind expanding on that?

From my point of view, it'd be a waste to spend a full GSoC project on aggregates unless you came up with a significantly different implementation and could explain the relative benefits. That said, my PR hasn't been reviewed yet, so there may be an opportunity there.

Regards,

Josh

That is the very thing I want to ask whether is there a opportunity for me there or I should consider 'improving error reporting' for GSoC. I want to make that choice and knowing well about the work done by you, I want an opinion of the core-dev as soon as possible.