End of 2016 update on PyKF5 bindings

[Stephen Kelly]

Shaheed Haque wrote:

> The three main areas I know I need to cover before I can declare the
> rules database formats good enough that the risk of change seems small
> are:
> 
> - Almost arbitrary edits to all structures that clang emits.
> - Automated support for templated types such as QMap and QHash
> (%MappedType et. al.).
> - See what, if anything, can be done to automate %ConvertToSubclassCode.
> 
> The first is in very good shape as of a few days ago. The second (as
> of this morning) produces code that can be imported by Python. The
> third is not yet started (though Phil from PyQt-land has pointed me at
> some sample code he has). 

Sounds great.

Here is the state of things from my point of view:

 * All of the bindings for individual frameworks repos that I had so far in 
my github are now in the kde repos.
 * The bindings in the frameworks repos are getting released and packaged, 
which means they will get to users
 * As far as I know, the most pressing packaging related issues are 
resolved: https://phabricator.kde.org/T5016

However, some of the bindings discard certain functions because the sip 
generator in ECM is missing some features. For example, KCompletion has 
rules like

 ["KCompletionBase", "keyBindingMap", ".*", ".*", ".*", 
rules_engine.function_discard],

because that method uses QMap etc. So, your further work on the type 
mapping, once it is in ECM, should allow reinstating those functions.

Another example is that the kconfig bindings are in, but the 
ConvertToSubclassCode stuff you mentioned above is not present, so perhaps 
the behavior on the python side is not everything a user would expect. I'm 
just guessing about that. If anything from your investigation there should 
go to ECM, then that can be done too to improve the bindings.

The repo at https://github.com/steveire/frameworks-bindings is now using 
only kde clones and tests all frameworks which have bindings so far.

So, with the goal of getting more bindings into users hands and making them 
more useful, I think there are two orthogonal things to do:

 * Add rules to more frameworks repos as I have been doing
 * Extend the set of features in the ECM repo to make better bindings 
possible

If you have suggestions for additional frameworks repos to add bindings to, 
I'd be interested to know.

Thanks,

Steve.