On Mon, Jun 20, 2011 at 8:46 PM, Uwe Bonnes
<bon@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote:
>>>>>> "Xiaofan" == Xiaofan Chen <xiaofanc@xxxxxxxxx> writes:
> Xiaofan> Even then, the question may come back to what is the intention
> Xiaofan> of libftdi-1.0?
>
> Xiaofan> Is the intention to replace libftdi-0.1x, then it is probably a
> Xiaofan> better idea to put it in the same location. Right now only one
> Xiaofan> can exist in the same place (/usr or /usr/local) anyway -- they
> Xiaofan> can not coexist under /usr. I suspect this is the original
> Xiaofan> intention and then it is better to be compatible with
> Xiaofan> libftdi-0.1x and put the header in the same place.
>
> Xiaofan> Is the intention not to replace libftdi-0.1x but create a new
> Xiaofan> library, in that case, it is better to rename the header file
> Xiaofan> so that it can easily coexist with the old library.
>
> The intention is to have
> /usr/(local/)/include/ftdi.h for 0.1
> /usr/(local/)/include/libftdi-1.0/ftdi.h for 1.0
> /usr/(local/)/include/libftdi-x.x/ftdi.h for x.x for later versions
>
> and have
> #include <ftdi.h>
> #include<libftdi-1.0/ftdi.h>
> #include<libftdi-x.x/ftdi.h>
>
> For the two latter cases,
> #include<libftdi-1.0/libftdi.h>
> #include<libftdi-x.x/libftdi.h>
>
> would also be fine, but i.m.h.o. bring no advantage.
If 1.0 and 0.1 can not co-exist, I see no benefits of the
above. If this is to differentiate library version, then it
is actually much easier to have a new API called
libftdi_getversion() which return the version at runtime.
--
Xiaofan
--
libftdi - see http://www.intra2net.com/en/developer/libftdi for details.
To unsubscribe send a mail to libftdi+unsubscribe@xxxxxxxxxxxxxxxxxxxxxxx
|
