[Kolab-devel] CVS to Mercurial conversion

[Gunnar Wrobel]

Zitat von Thomas Arendsen Hein <thomas at intevation.de>:

> * Gunnar Wrobel <wrobel at kolabsys.com> [20100901 09:56]:
>> The basic ideas behind my tag suggestions:
>>
>>  - Use the "server" prefix to identify the product. So far CVS just  
>> holds the
>>    "server" product. But it is likely that there will be variants of this
>>    product in the future (e.g. "hosted" or maybe "debian" or some such).
>>  - Use the version number in the tag.
>>
>> So the most recent release tag would be "server.2.2.4".
>
> I would suggest not using "server", because everything in this
> repository will be about the server.

Fine with me. I wasn't certain if there is going to be additional  
stuff in there.

> "hosted" or "debian" would be a
> server in the hosted or debian flavor.
>
> So I think these are appropriate tag names:
> 2.2.4-rc2
> hosted-2.2.4-rc2 or 2.2.4-rc2-hosted
> debian-2.2.4-rc2 or 2.2.4-rc2-debian

Definitely "hosted-2.2.4-rc2" as "hosted" as suffix looks like it  
might be part of the version number.

>
>> There was nothing specific to my branch name suggestions. I'm used to
>> capitals because of the Horde project but to be honest I don't know if
>> there exist specific conventions concerning branch names.
>
> I don't think we need to use capitals to avoid conflicts between
> tags and branch names.

Yes, looks like mercurial separates this clearly enough. In fact Horde  
does not use this in git anymore either.

>
>>  - Is the listed mail address the correct one?
>
> I will double-check with my information.
>
>>  - I tried identifying the most recent mail addresses for the persons with
>>    commit access. Is this what we want to use or should we use old
>> addresses if
>>    we are certain those were being used when the person was actively
>> working on
>>    the Kolab server?
>
> Hmm, if the old address is a company address, we might want to give
> credit to the company by using the old address, even if the
> developer no longer works for this company or the company no longer
> exists.
>
> Having a current address has benefits though, but matching the
> address used on kolab-devel at that time might be more valuable.

Fine with me.

>
>>  - Should we send a mail to all the people on the list and get  
>> verification of
>>    the addresses?
>
> I don't think so.

Okay.

>
>>  - Should we use Umlauts in the names or not?
>
> I just looked at the username settings of our users, half of the
> people having umlauts in their name use umlauts, the other half uses
> two replacement letters. I got rid of my umlaut by marriage :)
>
> At first I thought we should not use umlauts, but on a second
> thought we should use them for inactive contributors (Stuart Bingé)
> and ask current contributors (e.g. Marcus Hüwe) what they would use
> in the future.

As mentioned in my response to Mathieu I was just worried about any  
subsystems that might interact with hg and that might not correctly  
show the characters.

>
>> I felt like a human spam bot when compiling this list :)
>
> Spam, lovely Spam, wonderful Spam :)
>
>> Thomas: To which extent is it possible to correct any of the information
>> listed here *after* the conversion has been done?
>
> Tag names:
>   Absolutely no problem. The most current tag name will be used.
>

Okay.

> Branch names:
>   Create a new branch from the old one, close the old branch.
>   The old branch name will be visible when looking at old changesets
>   though.

Okay.

>
> Email addresses:
>   Not possible without doing a Mercurial->Mercurial conversion of
>   the repository, which in turn will change the changeset hashes of
>   all revisions following a changed email address.

Okay, so this list needs a final check before converting. You agree?

>
>> The suggested mercurial tag name is listed after the colon. This is
>> just my suggestion. Feel free to suggest an alternative.
>>
>> -TAGS---------------------------
>> V1-0-BETA-2:server.1.0.beta2
>> kdab: server.2.0.pre_kolabd.1
>> kolab-server-2-0-4:server.2.0.4
>> kolab-server-2-1-0:server.2.1.0
>> kolab-server-2-1-beta-2:server.2.1.beta2
>> kolab-server-2-1-beta-3:server.2.1.beta3
>> kolab-server-2-1-beta-4:server.2.1.beta4
>> kolab-server-2-1-rc-1:server.2.1.rc1
>> kolab-server-2-1-rc-2:server.2.1.rc2
>> kolab-server-2-2-0:server.2.2.0
>> kolab-server-2-2-1:server.2.2.1
>> kolab-server-2-2-1-beta-1:server.2.2.1.beta1
>> kolab-server-2-2-1-rc-1:server.2.2.1.rc1
>> kolab-server-2-2-2:server.2.2.2
>> kolab-server-2-2-3:server.2.2.3
>> kolab-server-2-2-3-rc-1:server.2.2.3.rc1
>> kolab-server-2-2-4:server.2.2.4
>> kolab-server-2-2-beta-2:server.2.2.beta2
>> kolab-server-2-2-beta-3:server.2.2.beta3
>> kolab-server-2-2-rc-1:server.2.2.rc1
>> kolab-server-2-2-rc-2:server.2.2.rc2
>> kolab-server-2-2-rc-3:server.2.2.rc3
>> kolab2:server.2.0.pre_kolabd.2
>> kolab_2_0_tag:server.2.0
>> kroupware:server.1.0
>> pre_autoperl:server.2.0.pre_perlkolab
>> proko2_m1:server.1.0.proko2.m1
>> proko2_m2c:server.1.0.proko2.m2
>> proko2_m3::server.1.0.proko2.m3
>> --------------------------------
>
> I found additional tags:
> AFTER_KONSEC_ADMIN
> BEFORE_KONSEC_ADMIN
> start (but I think we will remove that anyway)
>
> And I "kdab" seems to be a vendor branch, not a tag.
>
> Since tag names will be adjusted after the real conversion,
> discussing the names will not defer the conversion.

Okay so we'd correct my original suggestion from

  kolab-server-2-2-rc-1:server.2.2.rc1

to

  kolab-server-2-2-rc-1:2.2-rc1

We use a dash as the separator between the final number part of the  
version number and the "rc1", "beta1" etc. marker, right? In any case  
as mentioned in my response to Mathieu I'd also like to have these  
tags look like the release version numbers.
>
>> The suggested mercurial branch name is listed after the colon. This is
>> just my suggestion. Feel free to suggest an alternative.
>>
>> -BRANCHES-----------------------
>> KOLAB_1_0: SERVER_1_0
>> KOLAB_1_1: SERVER_1_1
>> kolab_2_0_branch: SERVER_2_0
>> kolab_2_1_branch: SERVER_2_1
>> kolab_2_2_branch: SERVER_2_3
>> suc_branch: SERVER_CUST_SUC
>> --------------------------------
>
> I don't know about the purpose of the KOLAB_1_0 and Kolab_1_x (where
> is KOLAB_1_1?) branches, but the kolab_2_[012]_branch branches were
> used for stable development, so I would name them:
> 2.0-stable
> 2.1-stable
> 2.2-stable
>
> The suc_branch evolved as a feature branch from 2.2-stable, so I
> would say:
> 2.2-feature-suc

Fine with me.

>
> Additional branches I found:
> kdab (vendor branch)
> steffen (vendor branch)
>
> I recommend dropping both completely.
>
>> Concatenated the authors from the different modules by:
>
> In the first step, only the server module will be converted.

Okay.

>
>> Finally search different resources on the net for the most probable
>> commiter names and mail addresses. Remarks added in [].
>>
>> -START----------------------------------------------------------------------------------------------------
>> achim=Achim Frank <achim.frank at erfrakon.de>
>> bernhard=Bernhard Reiter <bernhard at intevation.de>
>> bh=Bernhard Herzog <bh at intevation.de>
>> bo=Bo Thorsen <bo at fioniasoftware.dk>
>
> historical: bo=Bo Thorsen <bo at klaralvdalens-datakonsult.se>
>
> But maybe klaralvdalens-datakonsult.se (KDAB) already has enough
> credits to warrant using the new address if Bo wants.

Do we still have contact with him? If not I'd suggest using the old  
one as "bo at fioniasoftware.dk" is something I grabbed from the net. I  
didn't know he was working for KDAB at that time though.

>
>> danimo=Daniel Molkentin <daniel at molkentin.de>
>
> historical: danimo=Daniel Molkentin <molkentin at kde.org>

Use the old one then.

>
>> david=David Faure <dfaure at kdab.com> [only commited in module "doc"]
>> emanuel=Emanuel Schuetze <emanuel at intevation.de> [Umlaut: Schütze,  
>> only commited in module "doc"]
>> greve=Georg Greve <greve at kolabsys.com> [only commited in module "doc"]
>
> These three are not part of the current conversion.
>
>> gunnar=Gunnar Wrobel <wrobel at pardus.de>
>
> I really hope you got this one right :)
>
>> jan=Jan-Oliver Wagener <jan-oliver.wagner at intevation.de> [most  
>> likely candidate]
>
> Yes, "jan@" would be fine, too, but he mostly uses the long form.
>
>> khruskowski=Karl-Heinz Ruskowski <khruskowski at intevation.de>
>
>> ljohansen= [?, no clue who that might be, only commited in module "doc"]
>> ludwig=Ludwig Reiter <ludwig.reiter at intevation.de> [only commited  
>> in module "utils"]
>> marc=Marc Mutz <marc at kdab.com> [most likely candidate, only  
>> commited in module "doc"]
>
> These three are not part of the current conversion.
> ljohansen is Lennart Johansen (was an intern at Intevation)
>
>> marcus=Marcus Huewe <suse-tux at gmx.de> [Umlaut: Marcus Hüwe]
>
> Needs reply for the umlaut (I think with umlaut is fine)

Use the umlaut in case he does not respond to your request on IRC. I  
don't think this would warrant any delay.

>
>> martin=Martin Konold <martin.konold at erfrakon.de>
>> mathieu=Mathieu Parent <math.parent at gmail.com>
>> michel=Michel Boyer de la Giroday <michel at kdab.com> [only commited  
>> in module "doc"]
>
> michel is not part of the current conversion.
>
>> richard=Richard Bos <ml at radoeka.nl>
>> romain=Romain Pokrzywka <romain at kdab.com>
>> schreiber=Arne Schreiber<arne at kdab.com>
>
> schreiber is not part of the current conversion.
>
>> steffen=Steffen Hansen <steffen at kdab.com>
>> stephan=Stephan Buys <main at stephanbuys.com> [Is this the current  
>> mail address?]
>
> historical: stephan=Stephan Buys <s.buys at codefusion.co.za>

Use the old one then.

>
>> stuart=Stuart Binge <s.binge at codefusion.co.za> [Umlaut: Bingé]
>
> I think with umlaut is better here.

Okay. And also okay with the umlaut correction you mentioned in the  
later mail. I knew it was something more special than é.

>
>> tassilo=Tassilo Erlewein <tassilo.erlewein at erfrakon.de>
>> thomas=Thomas Arendsen Hein  <thomas at intevation.de>
>
> s/  / /
> thomas=Thomas Arendsen Hein <thomas at intevation.de>
>
>> till=Till Adam <till at kdab.net>
>
> All others are @kdab.com, we should use the same here.
> Oh, and not part of the current conversion.
>
>> tobias= [?, no clue who that might be, only commited in module "doc"]
>
> tobias=Tobias Larsson <tobias at klaralvdalens-datakonsult.se>
> (but not part of the current conversion)
>
>> torsten=Torsten Irlaender <torsten.irlaender at intevation.de> [only  
>> commited in module "utils" and "doc"]
>
> torsten is not part of the current conversion.
>
>> wilde=Sascha Wilde  <wilde at intevation.de>
>
> s/  / /
> wilde=Sascha Wilde <wilde at intevation.de>
>
>> -END-------------------------------------------------------------------------------------------------------
>
> *phew*

So I think if we use the old mail addresses and assume that Markus  
would be fine with using "ü" we'd have that list completed. Could you  
resend what you'd use then in the conversion?

Thanks!

Gunnar

>
> Many thanks for the work on this.
>
> Regards,
> Thomas
>
> --
> thomas at intevation.de - http://intevation.de/~thomas/ - OpenPGP key:  
> 0x5816791A
> Intevation GmbH, Neuer Graben 17, 49074 Osnabrueck - AG Osnabrueck,  
> HR B 18998
> Geschaeftsfuehrer: Frank Koormann, Bernhard Reiter, Dr. Jan-Oliver Wagner
>
> _______________________________________________
> Kolab-devel mailing list
> Kolab-devel at kolab.org
> https://kolab.org/mailman/listinfo/kolab-devel
>



--
Gunnar Wrobel
Developer, Kolab Systems AG

e: wrobel at kolabsys.com
t: +49 700 6245 0000
w: http://www.kolabsys.com

pgp: 9703 43BE

----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.