kicad-developers team mailing list archive

[Oliver Walters <oliver.henry.walters@xxxxxxxxx>]

Maurice,

This patch does *not* generate any models on the fly. It provides the
ability for users to re-use existing model data (which they are NOT forced
to use by any means). In the example of pin headers or terminal blocks,
where each unit is identical, there is no need to re-validate each model.
It is simply another means of achieving a goal.

On Tue, Nov 7, 2017 at 9:52 AM, easyw <easyw@xxxxxxxxxxxx> wrote:

> Oliver,
>
> just an other point against it IMO:
>
> If you develop a build on the fly 3D library, in case of update on OCE
> release, you would need to check and probably review all the models
> generated by your scripts...
> As I already pointed out on GitHub discussion, STEP are not just rendering
> models... they need to be done with a geometry check to guarantee a fine
> mechanical result.
>
> A static library is just as is... good a and reliable without any need to
> be re-checked or rebuilt once approved...
>
> Why just rework what is done till now and for which kind of advantage or
> optimization?
>
> Still my best option is to decide what to deploy with the Kicad installer
> and a dynamic downloader that will fetch and download only what is needed
> for the project.
>
>
> Maurice
>
>
> On 11/6/2017 11:34 PM, Oliver Walters wrote:
>
>> Seth, for the STEP export that would be a better idea, I can look into
>> how to implement that in the kicad2step tool.
>>
>> Wayne, if you are completely against this as a feature, then I shan't
>> spend any more time on it, but:
>>
>> 1. I think it is a useful feature to have even if it is not the "right"
>> way to solve the current 3D library issues. That was only one motivating
>> factor. I still think it is useful.
>> 2. In terms of "code clarity" - take a look at the patch set, I don't
>> think it will cause any issues for readability.
>> 3. There are some bug fixes and code cleanup that I have performed as
>> part of this patch set.
>> 4. If I can use the STEP format array_placement as Seth suggests, then
>> the internal representation of the exported STEP file will be improved
>>
>> Regarding STEP / WRL comparisons - That's a separate issue to what is
>> being discussed here. This patch works just as well for .wrl or .step
>> models.
>>
>>