Re: [dpdk-dev] [PATCH v2 1/4] build: add single source of DPDK version number

[Bruce Richardson <bruce.richardson@intel.com>]

On Wed, Mar 13, 2019 at 12:23:16PM +0100, Thomas Monjalon wrote:
> 13/03/2019 12:17, Bruce Richardson:
> > On Wed, Mar 13, 2019 at 12:06:38PM +0100, Thomas Monjalon wrote:
> > > 07/03/2019 14:34, Bruce Richardson:
> > > > Add a new file DPDK_VERSION to hold the current DPDK version number.
> > > > Have meson use this file for it's project version, and have make use
> > > > it for reporting out "showversion" and "showversionum".
> > > > 
> > > > Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
> > > > Acked-by: Luca Boccassi <bluca@debian.org>
> > > [...]
> > > > --- /dev/null
> > > > +++ b/DPDK_VERSION
> > > 
> > > Why not VERSION ?
> > > 
> > 
> > No particular reason. I just picked a name I thought reasonable, but I'm
> > fine with just VERSION on its own.
> > 
> > The one advantage of having DPDK in the title is that if one has a clone to
> > a directory called e.g. "temp", a quick "ls" will show the forgetful what
> > product it was that they cloned to a directory with such an unhelpful name.
> 
> Good try of justification :)
>
Worth a shot! :-)
 
> I think I prefer just VERSION.
> 
Ok, will change in V2.

> > > [...]
> > > > --- a/mk/rte.sdkconfig.mk
> > > > +++ b/mk/rte.sdkconfig.mk
> > > >  showversion:
> > > > -	@set -- \
> > > > -		$$(sed -rne 's,^#define RTE_VER_[A-Z_]*[[:space:]]+([0-9]+).*,\1,p' \
> > > > -			-e 's,^#define RTE_VER_SUFFIX[[:space:]]+"(.*)",\1,p' \
> > > > -			$(RTE_SRCDIR)/lib/librte_eal/common/include/rte_version.h) ;\
> > > > -		printf '%d.%02d.%d' "$$1" "$$2" "$$3"; \
> > > > -		if [ -z "$$5" ]; then echo; \
> > > > -		else printf '%s' "$$4"; \
> > > > -			if [ $$5 -lt 16 ] ; then echo $$5; \
> > > > -			else echo $$(($$5 - 16)); fi; \
> > > > -		fi
> > > > +	@cat $(RTE_SRCDIR)/DPDK_VERSION
> > > 
> > > I'm a bit sad about removing this complex command ;)
> > > 
> > Yes, but don't worry, I'm proposing a bit of an addition of sed and awk
> > to other makefile commands to compensate, though sadly not quite as
> > unintelligable to the masses as this. :-)
> 
> I can teach you :)
>
Thanks, but I think I'll pass on that opportunity! :-)