[p4] Building without changelists

Peter Janes peterj at liberate.com
Mon Oct 1 06:12:05 PDT 2001


Thanks, Stephen and Robert, this has helped immensely.  Now the trick is to 
get people to stop asking for this kind of build in the first place. :)

Peter J.

Robert Cowham wrote:

> Just a reminder to check out the Branching FAQs at
> http://public.perforce.com/public/perforce/faq/branching.html
> 
> There's some good stuff in there, particularly in this instance the one
> about "We tried to integrate a single change from one branch to another but
> in the resolve step we saw hundreds of deltas that weren't in this change.
> Why was this? "....
> 
> Robert
> 
> 
>>-----Original Message-----
>>From: perforce-user-admin at perforce.com
>>[mailto:perforce-user-admin at perforce.com]On Behalf Of Stephen Vance
>>Sent: Sunday, September 30, 2001 05:36
>>To: Peter Janes; perforce-user at perforce.com
>>Subject: Re: [p4] Building without changelists
>>
>>
>>At 04:19 PM 9/28/2001 -0400, Peter Janes wrote:
>>
>>>I've run into a case where I need to make a test build of some of our
>>>source, but excluding a non-contiguous set of changelists.
>>>
>>After reading
>>
>>>Tech Note 057, it looks like I need to create a branch and integrate the
>>>changes.  However, it's not clear whether I can do the full
>>>
>>integration in
>>
>>>one step, without intervening "p4 submit" commands.  This may become a
>>>somewhat regular operation, so I'm interested in any comments on
>>>
>>why this
>>
>>>is a bad idea or how to manage it better, too.
>>>
>>>Should the following work to avoid the modifications in
>>>
>>changelists 3 and 7?
>>
>>Not really.  First off, if you are specifying a branch spec, you don't
>>specify both source and target the same way you do without a branch
>>spec.  You only need to specify the target if it is different from the
>>intersection of the branch spec target and your current client spec.  You
>>specify the source with the -s option.  Examples below, assuming
>>"target/..." is equivalent to the intersection of the branch spec target
>>and the client spec.
>>
>>Second, you should
>>
>>
>>># p4 branch myBranch
>>># p4 integ -b myBranch source/... at 2 target/...
>>>
>># p4 integ -b myBranch -s source/... at 2
>>
>>If source/... is the source of the branch spec, then you can just do
>>
>># p4 integ -b myBranch -s @2
>>
>>
>>># p4 integ -b myBranch source/... at 4, at 6 target/...
>>>
>>Under the last set of assumptions
>>
>># p4 integ -b myBranch @4, at 6
>>
>>
>>># p4 integ -b myBranch source/... at 7, at now target/...
>>>
>>The ranges are inclusive, so this would capture change 7, as
>>well.  Assuming there is a change 8,
>>
>># p4 integ -b myBranch @8, at now
>>
>>or
>>
>># p4 integ -b myBranch @8,#head
>>
>>
>>># p4 resolve
>>># p4 submit
>>>
>>If you submit after all integrations, you will get a dirty merge.  If you
>>submit after each, you will get a set of clean merges (assuming that they
>>are individually clean).  If the terms "dirty" and "clean" merge
>>don't make
>>sense, suffice it to say that clean merges are handled better as they
>>relate to future merges using the branch spec.
>>
>>
>>
>>>Our server is currently unavailable, so I'm unable to try this myself.
>>>Previous attempts (before reading TN057, so probably with bad syntax or
>>>filespecs) have always reported "all revisions are integrated" after the
>>>first "p4 integ", so I have to "p4 submit" after each "p4 integ".
>>>
>>If the source was really the source of your branch spec, then the
>>intersection of the specified source and the branch spec's target would
>>have been the empty set.  The second file spec argument would have been
>>used, but if there had already been an integration, there may have been
>>nothing more to integrate.
>>
>>
>>Stephen Vance
>>mailto:steve at vance.com
>>http://www.vance.com/
>>
>>_______________________________________________
>>perforce-user mailing list  -  perforce-user at perforce.com
>>http://maillist.perforce.com/mailman/listinfo/perforce-user
>>
>>
> 
> _______________________________________________
> perforce-user mailing list  -  perforce-user at perforce.com
> http://maillist.perforce.com/mailman/listinfo/perforce-user
> 





More information about the perforce-user mailing list