RE: [HACKERS] CVS

Thanks all for this info.  I'm presuming that, because no-one has been rude
yet, that a lot of this is not in the developers manual, or FAQ yet.  Would
it be worthwhile putting it there?  Just a quick paragraph with the latest
settings, and default options for particular types of developers (e.g.:
somebody who wants to hack the source and contribute, somebody who wants the
latest patch tree, somebody who wants access to the latest source, but no
contributions, etc).

MikeA


>> -----Original Message-----
>> From: Tom Lane [mailto:tgl@sss.pgh.pa.us]
>> Sent: Monday, July 19, 1999 5:41 PM
>> To: Ansley, Michael
>> Cc: 'pgsql-hackers@postgresql.org'
>> Subject: Re: [HACKERS] CVS 
>> 
>> 
>> "Ansley, Michael" <Michael.Ansley@intec.co.za> writes:
>> >>> If there is any further activity in the 6.5 branch, it'd be to
>> >>> produce a 6.5.2 bug-fix release.  We don't generally do 
>> that except
>> >>> for really critical bugs, since double-patching a bug in both the
>> >>> tip and a branch is a pain.
>> 
>> > Double-patching is a pain, but I thought that that was the 
>> point of using
>> > CVS to do your branching.  AFAIK, CVS will merge the 
>> bug-fixes in, say, the
>> > 6.5.1 branch back into the main branch.  Because you want 
>> to fix the bugs in
>> > 6.5 into 6.5.1, without having to double-patch, but new 
>> development must
>> > only go into the main branch.  So, when 6.5.1 is released, 
>> it is merged back
>> > into the main branch to pass the fixes over, and also 
>> carries on to 6.5.2 in
>> > a continuation of the existing branch.
>> 
>> The trouble is that the tip usually diverges fast enough 
>> that mechanical
>> application of the same diffs to tip and stable branch doesn't work
>> very well :-(.
>> 
>> Also, our usual practice is to prove out a bugfix in the tip and then
>> consider whether to apply it to stable branches.  I'm not 
>> sure whether
>> CVS supports that as easily as merging a branch to the tip, but I'd
>> be *really* wary of mechanical diff transfer to stable branches...
>> if the diff extracts too little or too much of the changes in the
>> tip file, you might not find out till too late.
>> 
>>             regards, tom lane
>>