kuphal at dls
Sep 24, 2006, 5:36 AM
Post #2 of 3
(731 views)
Permalink
|
|
Re: [mythtv-commits] mythtv commit: r11281 - in branches/release-0-20-fixes/mythtv/libs by bjm
[In reply to]
|
|
Martin Ebourne wrote:
> On Sun, 24 Sep 2006 02:00:29 +0000, mythtv wrote:
>
>> Author: bjm
>> Date: 2006-09-24 02:00:27 +0000 (Sun, 24 Sep 2006)
>> New Revision: 11281
>> Changeset: http://cvs.mythtv.org/trac/changeset/11281
>>
>> Modified:
>>
>> branches/release-0-20-fixes/mythtv/libs/libmyth/mythcontext.h
>> branches/release-0-20-fixes/mythtv/libs/libmythtv/programinfo.cpp
>> branches/release-0-20-fixes/mythtv/libs/libmythtv/programinfo.h
>>
>> Log:
>>
>> Add parentid to the proginfo StringLists so that info passed to and from
>> slaves have the correct ID. This bumps NUMPROGRAMLINES to 42 and
>> MYTH_PROTO_VERSION to "31".
>>
>
> Is it really a good idea to change the protocol version on the stable
> branch?
>
> Many projects have a policy where binary compatibility, database
> compatibility, and especially protocol level compatibility must be kept
> at all times on stable branches, and I suspect that users have an
> expectation that minor versions will work together. Are there any such
> guidelines for MythTV?
>
> In myth's case it could be even more of a problem because the packagers
> tend to release off the head of the stable branch rather than waiting for
> point releases, so sooner or later people are going to uprade 0.20-6 to
> 0.20-7 or some such on one machine and be very surprised when it all
> breaks.
>
>
Not to speak for Isaac, but it has pretty much been MythTV policy that
all systems must run the same revision so someone updating one machine
to 0.20-7 while another is 0.20-6 is unsupported.
Kevin
_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev
|
