[Opensim-dev] versioning question
Toni Alatalo
antont at kyperjokki.fi
Fri Sep 4 07:28:24 UTC 2009
Stefan Andersson kirjoitti:
> to me, that what the community wants is some sense of progression to compare
> two "official" git revisions to each others.
>
That seems to be the case indeed.
> We are already doing this by tagging releases with a version number, so
> either we
> a) start doing (tagging) more frequent official releases so most users very
> seldom feel the need to bother with individual git revisions.
> or
> b) Tag some kind of sequential number to each opensim commit, possibly by
> adding some git commit hooks to update a text file somewhere.
>
I doubt a) would suffice and that we'd easily want b) - the granularity
of individual commits.
On the surface post-commit hooks sound better to me than the technique
parsing and processing log output, which was done in the solution in
that one blog post .. the idea there seems nice though, but I still
don't even understand it fully so this view is to be taken with a grain
of salt. http://michaelandlaura.org.uk/~michael/blog/index.php?id=379
> That would mean, basically that we would have a show version output saying
> something like OpenSim 0.6.6.10201 (Git:6cf610) (Dev) (interface version 5)
>
Seems like something that could make everyone happy :)
> /Stefan
>
~Toni
>> -----Original Message-----
>> From: opensim-dev-bounces at lists.berlios.de [mailto:opensim-dev-
>> bounces at lists.berlios.de] On Behalf Of Toni Alatalo
>> Sent: den 3 september 2009 20:07
>> To: opensim-dev at lists.berlios.de
>> Subject: Re: [Opensim-dev] versioning question
>>
>> On Sep 3, 2009, at 7:30 PM, Justin Clark-Casey wrote:
>>
>>
>>>> The next revision to 6cf610 will be 6cf610 + 1 = 6cf611, then at
>>>> least there is some logic to it.
>>>>
>>> This is exactly what isn't the case - the identifiers are generated
>>> randomly without any connection between them. This
>>> seems to be one of the (few) downsides of git.
>>>
>> In a way it is what GIT, and distributed versioning control, is. I
>> think at least, am still just learning the practice of it..
>>
>> On the base system level, there is no sequence 'cause you can arrive to
>> the same situation in many ways.
>>
>> Commits are commits, and each commit/revision/patch has an identity. I
>> think this 'theory of patches' applies to GIT as well, the part about
>> 'Patch relationships' seems relevant
>> http://darcs.net/manual/node9.html#SECTION00940000000000000000 ..
>> besides sequential patches there are parallel patches: "There is one
>> other very useful relationship that two patches can have, which is to
>> be parallel patches, which means that the two patches have an identical
>> context (i.e. their representation applies to identical trees). Of
>> course, two patches may also have no simple relationship to one
>> another. In that case, if you want to do something with them, you'll
>> have to manipulate them with respect to other patches until they are
>> either in sequence or in parallel."
>>
>> So to have what grows as a sequence in the current opensim main thing,
>> something else besides the individual patch IDs are needed. The thing
>> in the blog post that has been referred to seemed promising, but don't
>> know much about that.
>>
>> ~Toni
>>
>>
>>> There was some talk on this list about somehow producing sequential
>>> numbers some time ago but I don't know the details.
>>>
>>>
>>>> If my hex intuitions serve me, 6cf610 < a82950 so does anyone know
>>>> how these version #s are to be interpreted or are they random
>>>> identifiers?
>>>> And if they are random, how do we track where we are? maybe, I
>>>>
>> should
>>
>>>> go back to the "10185 revision(s)" and use that #.
>>>>
>>>> -p
>>>>
>>>>
>>>>>> not sure how to interpret this hexadecimal indexing. Should I
>>>>>> consider
>>>>>> "10185" to be the number? The actual commit # is a long
>>>>>>
>> hexadecimal
>>
>>>>>> string. The beginning of it matches what you said:
>>>>>>
>>>>>> 6cf610 is the revision #? Do we need to bring out the old HP
>>>>>> calculators? :)
>>>>>> -paul
>>>>>>
>>>>>>
>>>>>>
>>>>>> Justin Clark-Casey wrote:
>>>>>>
>>>>>>
>>>>>>> Paul Fishwick wrote:
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>> Justin:
>>>>>>>>
>>>>>>>> Justin Clark-Casey wrote:
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>> The 'interface v5' part refers to the interface protocol
>>>>>>>>>
>> between
>>
>>>>>>>>> the region simulator and the grid (and defacto, between
>>>>>>>>> region simulators). It doesn't have anything at all to do with
>>>>>>>>> the OpenSim version (0.6.6).
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>> OK
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>> In principle, if you are running from a source repository
>>>>>>>>>
>> update
>>
>>>>>>>>> then the git version should be tacked on the end of
>>>>>>>>> OpenSim 0.6.6 (e.g. 0.6.6 239a1). This should also be true if
>>>>>>>>> you are updating from the SVN mirror (e.g. 0.6.6.10108).
>>>>>>>>> If this isn't the case then this is arguably a bug.
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>> Where does one find this build # (i.e., "239a1" or "10108") ? It
>>>>>>>> is not
>>>>>>>> produced through "show version". If that is a bug, I can file
>>>>>>>>
>> it.
>>
>>>>>>>> -paul
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>> If you execute 'git log' on the command line, the long string
>>>>>>> after the word "commit" is the build identifier, e.g. in
>>>>>>>
>>>>>>> <paste>
>>>>>>> commit 4841858600cf35903966d48451015b2be5e97d5c
>>>>>>> Author: Chris Hart <chris at reactiongrid.com>
>>>>>>> Date: Wed Sep 2 13:28:57 2009 +0100
>>>>>>>
>>>>>>> MSSQL patch to mirror mantis 4077 for MySQL to improve
>>>>>>>
>> region
>>
>>>>>>> load times
>>>>>>> </paste>
>>>>>>>
>>>>>>> 4841858600cf35903966d48451015b2be5e97d5c is the identifier. We
>>>>>>> often identify these only by the first selection of
>>>>>>> characters, e.g.
>>>>>>>
>>>>>>> 48418
>>>>>>>
>>>>>>> These identifiers are not sequential. There might be a shorter
>>>>>>> way to get this information from the command line but I
>>>>>>> don't know what it is.
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>>
>>>>>>>>> Gustavo Alberto Navarro Bilbao wrote:
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>> Yes, is very dark. In our server we are using the last Diva's
>>>>>>>>>> version
>>>>>>>>>> from git, than works in our case better than the 0.6.6
>>>>>>>>>> downladed from
>>>>>>>>>> the wiki, a service with 0.6.6 .10108, in OSGrid, an other
>>>>>>>>>>
>> with
>>
>>>>>>>>>> 0.6.6
>>>>>>>>>> .239a1 and other with modrex.
>>>>>>>>>>
>>>>>>>>>> The 0.6.6.6 in the console show the same "interfase v 5", very
>>>>>>>>>> criptic
>>>>>>>>>> and really dark.
>>>>>>>>>>
>>>>>>>>>> Albert
>>>>>>>>>>
>>>>>>>>>> 2009/9/3 Paul Fishwick <fishwick at cise.ufl.edu
>>>>>>>>>> <mailto:fishwick at cise.ufl.edu>>
>>>>>>>>>>
>>>>>>>>>> With the use of git or svn, what is the relationship
>>>>>>>>>> between the
>>>>>>>>>> Version and the Build #? Once upon a time, we would do a
>>>>>>>>>> "show version" in the console, and it was very clear which
>>>>>>>>>> specific
>>>>>>>>>> opensim version was being used. Now, it says:
>>>>>>>>>>
>>>>>>>>>> OpenSim 0.6.6 (Dev) (interface version 5)
>>>>>>>>>>
>>>>>>>>>> rather than something clearer (at least to me) like:
>>>>>>>>>>
>>>>>>>>>> OpenSim Version 10115
>>>>>>>>>>
>>>>>>>>>> The problem is that I don't know which build to refer to
>>>>>>>>>>
>> as
>>
>>>>>>>>>> I would
>>>>>>>>>> imagine
>>>>>>>>>> that the code base changes daily. Thanks.
>>>>>>>>>>
>>>>>>>>>> -p
>>>>>>>>>>
>>>>>>>>>> --
>>>>>>>>>> Paul Fishwick, PhD
>>>>>>>>>> Professor and Director, Digital Arts and Sciences Programs
>>>>>>>>>> University of Florida
>>>>>>>>>> Computer & Information Science and Eng. Dept.
>>>>>>>>>> Bldg. CSE, Room 301
>>>>>>>>>> P.O. Box 116120
>>>>>>>>>> Gainesville, FL 32611
>>>>>>>>>> Email: fishwick at cise.ufl.edu
>>>>>>>>>>
>> <mailto:fishwick at cise.ufl.edu>
>>
>>>>>>>>>> Phone: (352) 392-1414
>>>>>>>>>> Fax: (352) 392-1220
>>>>>>>>>> Web: http://www.cise.ufl.edu/~fishwick
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>>> Opensim-dev mailing list
>>>>>>>>>> Opensim-dev at lists.berlios.de
>>>>>>>>>> <mailto:Opensim-dev at lists.berlios.de>
>>>>>>>>>> https://lists.berlios.de/mailman/listinfo/opensim-dev
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> --------------------------------------------------------------
>>>>>>>>>>
>> --
>>
>>>>>>>>>> --------
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>>> Opensim-dev mailing list
>>>>>>>>>> Opensim-dev at lists.berlios.de
>>>>>>>>>> https://lists.berlios.de/mailman/listinfo/opensim-dev
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>
>>>>
>>> --
>>> justincc
>>> Justin Clark-Casey
>>> http://justincc.org
>>> _______________________________________________
>>> Opensim-dev mailing list
>>> Opensim-dev at lists.berlios.de
>>> https://lists.berlios.de/mailman/listinfo/opensim-dev
>>>
>> _______________________________________________
>> Opensim-dev mailing list
>> Opensim-dev at lists.berlios.de
>> https://lists.berlios.de/mailman/listinfo/opensim-dev
>>
>
> _______________________________________________
> Opensim-dev mailing list
> Opensim-dev at lists.berlios.de
> https://lists.berlios.de/mailman/listinfo/opensim-dev
>
More information about the Opensim-dev
mailing list