[Pharo-project] [keith_hodges@yahoo.co.uk: [squeak-dev] Monticello version numbers]
This came to my attention last night as I made a modification to MC1.5 in order to work with Pharo packages. ----- Forwarded message from Keith Hodges <keith_hodges@yahoo.co.uk> ----- To: The general-purpose Squeak developers list <squeak-dev@lists.squeakfoundation.org> From: Keith Hodges <keith_hodges@yahoo.co.uk> Date: Fri, 27 Jun 2008 17:18:56 +0100 Subject: [squeak-dev] Monticello version numbers Many do not know that Monticello 1.5 supports more complicated/natural version numbering schemes. Versions can be... 100 1.0.20 b523057e.606b.464b (abbreviated UUID with: $- replaced by $.) so.... this means... PLEASE DONT USE $. in your author initials when saving MC packages!!!!!! this has been a public service announcement Keith ----- End forwarded message ----- I don't know how or if it conflicts with Pharo dev policy. I don't see any requirement to use author names like lukas.renglli, but it happened. I don't think this version number scheme has been actually tested, despite what Keith said; it gave me a walkback when the version number was accidentally detected as 'renglli.271'. Does this interfere with some pharo policy? If so, any suggestions on a remedy? How do you envision MC package naming working once Author class is created? Keith's motivation behind the change: Seaside2.8 used version names with periods, but inside the package name. He made it more first-class UUID versions were a request by Chris Mueller -- Matthew Fulmer -- http://mtfulmer.wordpress.com/
I don't think this version number scheme has been actually tested, despite what Keith said; it gave me a walkback when the version number was accidentally detected as 'renglli.271'.
I never used these initials and I usually don't spell my own name wrong ;-) Cheers, Lukas -- Lukas Renggli http://www.lukas-renggli.ch
I don't think this version number scheme has been actually tested, despite what Keith said; it gave me a walkback when the version number was accidentally detected as 'renglli.271'.
I never used these initials and I usually don't spell my own name wrong ;-)
:) But this is good to know that we MC15 we can break most of my package. I hate those initials and prefer full names for traceabiility point of view. stef
On Fri, Jun 27, 2008 at 10:32:00PM +0200, St?phane Ducasse wrote:
:) But this is good to know that we MC15 we can break most of my package. I hate those initials and prefer full names for traceabiility point of view.
Ok Keith and I had a discussion and found that the only problem was that the sorter that handles the sorting in the various MC tools is not robust against either fancy author initials or fancy version strings. So, we had an intense pair-programming session and fixed both problems, using a much smarter sorting policy for MCVersionInfo. So now, both fancy author initials an fancy version strings work nicely in MC1.5. I was able to use your repository just fine. The new version is available via LPF. -- Matthew Fulmer -- http://mtfulmer.wordpress.com/
thanks. I leaving for RMLL in a couple of hours. Stef
:) But this is good to know that we MC15 we can break most of my package. I hate those initials and prefer full names for traceabiility point of view.
Ok Keith and I had a discussion and found that the only problem was that the sorter that handles the sorting in the various MC tools is not robust against either fancy author initials or fancy version strings. So, we had an intense pair-programming session and fixed both problems, using a much smarter sorting policy for MCVersionInfo. So now, both fancy author initials an fancy version strings work nicely in MC1.5. I was able to use your repository just fine. The new version is available via LPF.
-- Matthew Fulmer -- http://mtfulmer.wordpress.com/
_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
participants (3)
-
Lukas Renggli -
Matthew Fulmer -
Stéphane Ducasse