Re: [gitg] Question about gitg bahaviour in "All branches" mode

[Marcin Zajączkowski <mszpak wp pl>]

Dnia 18-11-2010 o godz. 22:29 Jesse van den Kieboom napisał(a):
> On 11/18/2010 09:56 PM, Marcin Zajączkowski wrote:
> > Dnia 12-11-2010 o godz. 20:37 Marcin Zajączkowski napisał(a):
> >   
> >> Dnia 12-11-2010 o godz. 9:02 Guilhem Bonnefille napisał(a):
> >>     
> >>> Hi,
> >>>
> >>> 2010/11/11 Marcin Zajączkowski <mszpak wp pl>:
> >>>       
> >>>> I would like to consult one behaviour of gitg on my git repository I
> >>>> don't understand. In "All branches" mode my master point with an arror
> >>>> to some past commit [1] while it all should be in one line (in master
> >>>> only view it looks ok). How it should be interpreted?
> >>>>         
> >>> The arrow indicates that branch life line is really long and an
> >>> ellipsis is better to clean the display. You can control this feature
> >>> via preferences options.
> >>>
> >>> Interpretation of the current display is that master and appinfo-0.5.0
> >>> are two long living parallel branches.
> >>>       
> >> They shouldn't.
> >> When I manually check parents it looks ok. The problem could be that
> >> "[maven-release-plugin] prepare release appinfo-0.5.0" and
> >> "[maven-release-plugin] prepare for next development iteration" have the
> >> same commit date. gitg displays it in reverse order (regarding to parent
> >> chain).
> >>     
> > Do you think the same date could be a reason to display those two
> > commits in a wrong order?
> >   
> The order is not 'wrong'. I think it's simply the difference between
> showing the revisions in topological order or not. There is a preference
> for this in the preferences.

Described behavior was with topological order disabled. When enabled 
mentioned commits are in ascending order and there no additional arrow - 
everything is as an one line (in all branches view).
I don't know why topological order should matter in mentioned situation. 
Two commits were made one by one in the same line (in the first commit 
new tag was created, but I don't know if it should be a reason).

anyway, it's not a big problem for me. I asked on the mailing list 
because I suspected that something wrong is with my repository 
(everything seems to be fine after check in the meanwhile) or there is 
some bug in gitg. But if you think it's something from category "it's a 
feature" then I won't dig it deeper.


Thanks for given explanations
Marcin



> >
> >   
> >> <QOUTE>
> >> commit c652c5721f24d7f9788cacbef0d0d975703bc522
> >> 3d7f18d60d98f8643a33e8eed71080362cd7877e
> >> Author: Marcin Zajaczkowski <szpak users sourceforge net>
> >> Date:   Thu Nov 4 23:24:09 2010 +0100
> >>
> >>     docs clean up do 0.5.0
> >>
> >> commit 3d7f18d60d98f8643a33e8eed71080362cd7877e
> >> 5a4866fac3129b5dc78bb7a7cd4e3a2a9fb5da0b
> >> Author: Marcin Zajaczkowski <szpak users sourceforge net>
> >> Date:   Thu Nov 4 22:09:17 2010 +0100
> >>
> >>     [maven-release-plugin] prepare for next development iteration
> >>
> >> commit 5a4866fac3129b5dc78bb7a7cd4e3a2a9fb5da0b
> >> 5617ad5a24a040f2f28de1ac1dc9a816446ef75c
> >> Author: Marcin Zajaczkowski <szpak users sourceforge net>
> >> Date:   Thu Nov 4 22:09:17 2010 +0100
> >>
> >>     [maven-release-plugin] prepare release appinfo-0.5.0
> >>
> >> commit 5617ad5a24a040f2f28de1ac1dc9a816446ef75c
> >> 1c33c279c912242bd1ed67881e1935493ac20214
> >> Author: Marcin Zajaczkowski <szpak users sourceforge net>
> >> Date:   Thu Nov 4 22:05:23 2010 +0100
> >>
> >>     revert version in pom to 0.5.0-SNAPSHOT
> >> </QUOTE>
> >>
> >>
> >>     
> >>> I you have doubt about the understanding of the history made by gitg,
> >>> I suggest you to use another Git browser (like gitk). By this way, you
> >>> will be able to double check your history.
> >>>       
> >> Today I checked out mentioned repo [1] using TortoiseGit and there is
> >> everything in one line in "all branches mode":
> >> http://picturepush.com/+IvY8
> >>
> >>
> >> Could it be an issue with the same date?
> >>
> >>
> >> [1] - git://appinfo.git.sourceforge.net/gitroot/appinfo/appinfo
> >>
> >>
> >> Regards
> >> Marcin
> >>



--
Ad...