Re: [Kicad-developers] [PATCH] Changes to python interface enabling net->pads access.

2017-11-17 Thread Maciej Suminski
Hi Miles, On 11/17/2017 02:14 PM, miles mccoo wrote: > I see. Let me summarize to see if I understand. > > The python wrapper around wx was rewritten. A move to SIP is one of (the > main) differences that the new version brings with it. I imagine this > involves more than a simple recompile. At l

Re: [Kicad-developers] [PATCH] Changes to python interface enabling net->pads access.

2017-11-17 Thread Maciej Suminski
On 17/11/2017 9:08 AM, miles mccoo wrote: Another thing, there will be no more wxPython releases. The project has been renamed to wxPython and AFAIR nobody has tried to use it with KiCad. It may turn out that our SWIG work might be dropped/converted or we will stay stuck with wx3

Re: [Kicad-developers] [PATCH] Changes to python interface enabling net->pads access.

2017-11-16 Thread Maciej Suminski
Hi Miles, I think the silence you get is not caused by the message length, but two other factors: - We lack (at least I do) deep SWIG knowledge required to assess the patch. More or less, I understand what does your patch change, but I cannot tell if this is the right way to do it. The typemap tri

Re: [Kicad-developers] [PATCH] Changes to python interface enabling net->pads access.

2017-11-16 Thread miles mccoo
Actually, forget everything I wrote in my previous mail. Too long, no one wants to read all that and most of it's not relevant to my patch. My patch doesn't change core kicad in any way. It only exposes the existing connectivity class to python. The only deviation from the existing c++ interface

Re: [Kicad-developers] [PATCH] Changes to python interface enabling net->pads access.

2017-11-16 Thread miles mccoo
Interesting thread. Lots of stuff that I hadn't stumbled across before. There's a lot in my answer to cover the ground raised by everyone. For those that don't make it to the end, I'll say that there are no clear next steps for me. I'd like to be able to get from a net to that net's pads. I'd like

Re: [Kicad-developers] [PATCH] Changes to python interface enabling net->pads access.

2017-11-15 Thread Tomasz Wlostowski
On 15/11/17 18:38, Wayne Stambaugh wrote: > It looks like the PCB_COLLECTOR objects are not swigged yet but that > would be the way to do it. Properly swigged, there would be nothing > preventing you from collecting all objects connected to an net or any > other collection criteria for that matter

Re: [Kicad-developers] [PATCH] Changes to python interface enabling net->pads access.

2017-11-15 Thread Wayne Stambaugh
It looks like the PCB_COLLECTOR objects are not swigged yet but that would be the way to do it. Properly swigged, there would be nothing preventing you from collecting all objects connected to an net or any other collection criteria for that matter. That's what the collector object is for. Let's

Re: [Kicad-developers] [PATCH] Changes to python interface enabling net->pads access.

2017-11-15 Thread Tomasz Wlostowski
On 15/11/17 15:37, miles mccoo wrote: > Hello, > Hello Miles, Thanks for the patch. My comments below: > > A couple code review type observations on the existing code. > > 1. The new connectivity stuff is in connectivity.h. Perhaps it should > be called class_connectivity.h to match the n

[Kicad-developers] [PATCH] Changes to python interface enabling net->pads access.

2017-11-15 Thread miles mccoo
Hello, This is my fourth patch to Kicad. Each of those instances have had process errors, hopefully I'm not repeating any of those. some time ago, the connectivity model of pcbnew was reworked to improve ratsnest calculations. One of the side effects of that change was that the Pads() api on net