binary compatibility in kabc

[Ian Reinhart Geiser]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Saturday 07 June 2003 11:31 am, Dirk Mueller wrote:
> Hi,
>
> kabc broke in HEAD binary compatibility multiple times already compared to
> 3_1_BRANCH.
>
> Is there any reason why the same BC rules like for the rest of kdelibs
> don't apply to kabc?
Probably because they broke BC between 3.0 and 3.1.   I have gotten burned 
pretty well on this from clients who made the mistake to trying out KAB for 
smaller projects.   Is it time we explore a BC breakage script for checkins, 
so we can see this happen in a more real time?

I know this came up on IRC a few times, and no-one could think of really good 
tests that could be done on the fly.  One option (im talking out of my hat 
here on this one because i forget if you can access a diff in a checkin 
script)  would be to check for changed virtuals order, added/removed 
privates, added/removed virtuals etc... Not perfect, but would cause a commit 
message to maby alert an unsuspecting developer that they are breaking binary 
compat.

Just my two very frustrated cents, as Ive seem to be a good lightning rod for 
BC breakage in KDE. Then again, Im trying to make a living by getting 
companies to use it for development or end use.

Cheers
	-ian reinhart geiser
- -- 
- --:Ian Reinhart Geiser <geiseri at yahoo.com>
- --:Public Key: http://geiseri.myip.org/publickey.asc
- --:Public Calender: http://geiseri.myip.org/publicevents.ics
- --:Jabber: geiseri at geiseri.myip.org
- --:Be an optimist -- at least until they start moving animals in 
- --:   pairs to Cape Canaveral. ~ Source Unknown
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQE+4gruPy62TRm8dvgRAv6pAKCEAK9JMZ0ZPRQJRHs2+5t7frAUeQCg3Omv
3UWQwZq6ep+5PPyYbhghSr4=
=st0d
-----END PGP SIGNATURE-----