kicad-lib-committers team mailing list archive

[jp charras <jp.charras@xxxxxxxxxx>]

Le 25/08/2013 10:12, Kerusey Karyu a écrit :
> Hello.
>
> I would like to split the atmel.lib into several smaller libraries. This
> would be a similar division to the families which already exists in the
> Microchip MCU libraries.
>
> My proposals are as follows:
> - atmel_avrlegacymcu.lib - where the symbols of Classic AVR family have
> been moved
> - atmel_avrtinymcu.lib - where the symbols of AVR Tiny family have been
> moved
> - atmel_avrmegamcu.lib - where the symbols of AVR Mega family have been
> moved
> - atmel_51mcu.lib - where the symbols of Atmel’s MCS-51 family have been
> moved
> - atmel_avrxmegamcu.lib - where the symbols of AVR Xmega family have
> been moved
>
> Later this same method can be used to create specific libraries for
> other processors families, like: SAM3, SAM4, SAMD20 (new Atmel’s Cortex
> M0+ devices), UC32 or many other non-MCU devices: memories, touch
> sensors, RF, etc.
>
> After the library split, atmel.lib could be removed to avoid duplicates.
> However, backward compatibility problems may face in the way in the
> existing users projects. (I can already see the tons of messages “Where
> is my library?” on the user mailing list.)
>
>
> Since this is a big change and may have some sort of influences, first I
> would like to know the opinion of other library committers and eventual
> acceptance of those changes. Including the removal of atmel.lib.
>
>
> Regards
> Kerusey Karyu

Generally speaking, when a library grows and/or some components become 
obsolete, splitting it is a good practice.
Users cannot expect to have more and more components in libs, and do not 
have any change.

Components are not removed, they are moved.
Removing atmel.lib and adding few atmel_xxx.lib is self-explanory.
Therefore, only the config of eeschema/pcbnew should be modified.
This not a lot of work for users.

Frankly, I do not care about users who complain.

By the way, thanks you for your hard work on libraries. There is not a 
lot of volunteers to maintain them.

--
Jean-Pierre CHARRAS