possible pyconuk sprint and things to get done

[Gary Martin <ga...@physics.org>]

Hi everyone,

As usual things have slipped a bit. However, I am currently on holiday and at PyCon UK.

So, one thing that I can do here is suggest a sprint tomorrow to work on the ideas and possibly create more code for the django based core bloodhound and maybe more depending on what people are interested in doing here.

One pitfall in this is that, certainly for large code contributions, we would need ICLAs. For the purpose of the sprint, it will probably be best to use a fork of a repo so to avoid hindering progress on the day and see if I can convince people to go through the ICLA process.

So, any thoughts from anyone?

Meanwhile, there are probably some tasks that we could do with performing that are a bit overdue:

 * there appeared to be good support for developing our experimental bloodhound based on django with git for source control. This should probably just be done soon.
 * we could do with checking on the main project page and check that it is up to any required standards for apache projects
 * for the experimental branch (whether on a git repo or in svn) we need to make sure the licensing is up to date and corrected to correctly adress whatever is actually there.
 * reason about the appropriate content of the core project (at the moment things like the manage.py and various other items are committed to the branch and there may be better ways to structure the project - that also leads to questions about documenting general project setup)

I will try to raise tickets around these things soon.

Cheers,
    Gary

Re: possible pyconuk sprint and things to get done

[Gary Martin <ga...@physics.org>]

Hi again,

While I did not hear from anyone in time, I thought it would be a good idea to continue with the idea of a sprint. There were many sprints suggested at pyconuk alongside my suggestion and perhaps not so many people to go round but I managed to interest someone in helping out. I don't think this will result in a new committer but they were a great help to me in working out a number of things and providing some guidance to areas of django that I have not played with until now.

The first thing that came up was whether the core model may be a bit difficult to understand for a newcomer. I am certainly keen to make our models relatively easy to understand while retaining flexibility but it is something to ponder on. Perhaps it will be enough to document the concept properly in the long run.

Next, the suggestion of helping out through adding in the Django Rest Framework (http://www.django-rest-framework.org) came up and so I was given some good advice to start out with that. I was also interested in working out whether, having put together a rest api, there might be a way to get the api to self-describe itself. This prompted Mark to point me at Swagger and, in particular Django Rest Swagger (https://django-rest-swagger.readthedocs.io/) which gives a surprising amount of features for relatively little configuration.

Anyway, Mark was also interested in working on a personal project of his so I ended up giving Mark what advice I could to him - essentially it became a mutual 'rubber duck' session for the remainder of the sprint.

At the moment I am on the train home back from Cardiff, continuing to tweak the code to add features to the api like urls for list items in the api responses.

I will get the changes merged to the experimental branch on our apache svn repo so others can play with what has been created but I thought everyone might like to know how things went today!

Cheers,
    Gary


On Tue, 18 Sep 2018, at 3:59 PM, Gary Martin wrote:
> Hi everyone,
> 
> As usual things have slipped a bit. However, I am currently on holiday 
> and at PyCon UK.
> 
> So, one thing that I can do here is suggest a sprint tomorrow to work on 
> the ideas and possibly create more code for the django based core 
> bloodhound and maybe more depending on what people are interested in 
> doing here.
> 
> One pitfall in this is that, certainly for large code contributions, we 
> would need ICLAs. For the purpose of the sprint, it will probably be 
> best to use a fork of a repo so to avoid hindering progress on the day 
> and see if I can convince people to go through the ICLA process.
> 
> So, any thoughts from anyone?
> 
> Meanwhile, there are probably some tasks that we could do with 
> performing that are a bit overdue:
> 
>  * there appeared to be good support for developing our experimental 
> bloodhound based on django with git for source control. This should 
> probably just be done soon.
>  * we could do with checking on the main project page and check that it 
> is up to any required standards for apache projects
>  * for the experimental branch (whether on a git repo or in svn) we need 
> to make sure the licensing is up to date and corrected to correctly 
> adress whatever is actually there.
>  * reason about the appropriate content of the core project (at the 
> moment things like the manage.py and various other items are committed 
> to the branch and there may be better ways to structure the project - 
> that also leads to questions about documenting general project setup)
> 
> I will try to raise tickets around these things soon.
> 
> Cheers,
>     Gary