← Back to team overview

kicad-developers team mailing list archive

Re: Pulling mac 5.0.2...

 

If you use `git am` to apply the patches, you wont get -dirty appended
to the version string.  However, you will end up with -###-gcommithash
appended so you still wont end up with 5.0.2 as the version string which
is what I'm gessing Adam is looking for.

On 1/18/19 2:33 PM, Nick Østergaard wrote:
> Mmm, what is wrong wit the git describe when the patches are applied
> with git am?
> 
> fre. 18. jan. 2019 20.11 skrev Wayne Stambaugh <stambaughw@xxxxxxxxx
> <mailto:stambaughw@xxxxxxxxx>>:
> 
>     This is a result of modifications to the repo.  The --dirty option of
>     `git describe` checks to see anything is modified and appends -dirty to
>     the version string.  This way we know if someone modified the source for
>     a given commit.  You could change the command in
>     CreateGitVersionHeader.cmake to `git descibe` to drop -dirty from the
>     version string.
> 
>     Wayne
> 
>     On 1/18/19 1:40 PM, Adam Wolf wrote:
>     > Hi Wayne!
>     >
>     > I have since fixed the ngspice build race condition on macOS. I have
>     > modified the build scripts so that patches occur via patch, not git
>     > itself.  Unfortunately, now the git version info shows
>     5.0.2-dirty.  Is
>     > this how it shows on the Windows builds too?
>     >
>     > Adam Wolf
>     >
>     > On Tue, Jan 8, 2019 at 9:05 AM Adam Wolf
>     <adamwolf@xxxxxxxxxxxxxxxxxxxx <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx>
>     > <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx
>     <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx>>> wrote:
>     >
>     >     Alright.  Those changes are made.  I am doing builds now. 
>     They are
>     >     going to be 5.0.2-5 in order to ... reduce confusion.
>     >
>     >     After builds, I need to upload them to testing/, download
>     them, run
>     >     them through a test procedure that's in the README, and then
>     see if
>     >     this fixes the issues for users.  If so, I will move the 5.0.2-5
>     >     packages to stable/, and adjust the website.
>     >
>     >     In the meantime, I am going to move the 5.0.2 packages out of the
>     >     way on the server.  I'm getting multiple tickets a day for the
>     same
>     >     issue and we don't need more people downloading the bad macOS
>     >     package.  Once we have the new package up, we can put it back for
>     >     posterity's sake or whatever.
>     >
>     >     Adam
>     >
>     >     On Tue, Jan 8, 2019 at 8:15 AM Adam Wolf
>     >     <adamwolf@xxxxxxxxxxxxxxxxxxxx
>     <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx>
>     >     <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx
>     <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx>>> wrote:
>     >
>     >         Thanks Wayne.  Will do.  I appreciate your fast response.
>     >
>     >         Adam
>     >
>     >         On Tue, Jan 8, 2019 at 7:58 AM Wayne Stambaugh
>     >         <stambaughw@xxxxxxxxx <mailto:stambaughw@xxxxxxxxx>
>     <mailto:stambaughw@xxxxxxxxx <mailto:stambaughw@xxxxxxxxx>>> wrote:
>     >
>     >             Hey Adam,
>     >
>     >             Rather than committing the macos build patches to the git
>     >             repo, why not
>     >             just run patch from the build script to apply them?  This
>     >             way you don't
>     >             taint the git repo commit log and the version string
>     will be
>     >             5.0.2
>     >             assuming that it the branch that you are building. 
>     This is
>     >             how we have
>     >             done this in the past on other platforms.
>     >
>     >             Cheers,
>     >
>     >             Wayne
>     >
>     >             On 1/8/2019 8:41 AM, Adam Wolf wrote:
>     >             > Hi Wayne,
>     >             >
>     >             > I need a judgement call, and it's a little urgent.
>     >             >
>     >             > The current Mac packages call themselves 5.0.2, but the
>     >             version
>     >             > information inside is Version: (5.0.2-4-g3082e92af),
>     >             release build. 
>     >             > This is because there are 4 patches applied to the 5.0.2
>     >             source during
>     >             > packaging.  These are exclusively for packaging changes,
>     >             mostly to get
>     >             > Python to work the way it needs to be
>     redistributable for
>     >             macOS. Without
>     >             > these patches, kicad works, but it does not work in a
>     >             redistributable
>     >             > way.  If they were included in upstream, they
>     wouldn't work at
>     >             > all--unless folks also used the rest of the macOS build
>     >             script.  Because
>     >             > of this, I deemed it reasonable to include the
>     patches in
>     >             the macOS
>     >             > build script process.
>     >             >
>     >             > I need to rerelease the 5.0.2 packages.  Normally,
>     the way
>     >             this would be
>     >             > done on other projects is that I would make a 5.0.2-2
>     >             release.  I
>     >             > actually built those this weekend, but the problem
>     is that
>     >             folks could
>     >             > definitely think they are already running 5.0.2-4.
>     >             >
>     >             > I thought, oh, ok! For release builds, we should
>     override
>     >             the git
>     >             > version string, and burn in what we're building, so that
>     >             would just say
>     >             > Version: (5.0.2), release build.  It appears we cannot
>     >             override the git
>     >             > generated part, only append to the end.
>     >             >
>     >             > Normally I'd throw this to the list and wait a while for
>     >             consensus, but
>     >             > many Mac users are reporting issues with the current
>     5.0.2
>     >             package and
>     >             > it needs to be replaced as soon as we can.
>     >             >
>     >             > One option would be to "fix it for this release" by
>     adding
>     >             yet another
>     >             > patch that makes it so the gitversion can be overridden,
>     >             make a 5.0.2-5
>     >             > release, and get those patches upstreamed behind some
>     >             conditionals
>     >             > before the next release, so that the gitversion, next
>     >             time, will be 5.1
>     >             > or 5.0.3 or whatever, and then I could append a -2 or
>     >             whatever if this
>     >             > happens in the future...
>     >             >
>     >             > Thoughts?
>     >             >
>     >             > Adam Wolf
>     >             >
>     >             >
>     >             > On Mon, Jan 7, 2019 at 7:42 PM Adam Wolf
>     >             <adamwolf@xxxxxxxxxxxxxxxxxxxx
>     <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx>
>     >             <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx
>     <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx>>
>     >             > <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx
>     <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx>
>     >             <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx
>     <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx>>>> wrote:
>     >             >
>     >             >     It looks like there's something wrong with the
>     shared
>     >             library
>     >             >     references of just the 5.0.2 packages.  They were
>     >             generated using
>     >             >     the build script, but not 100% automatically.  I've
>     >             set Jenkins up
>     >             >     to build those too, which should help reduce human
>     >             error next time.
>     >             >
>     >             >     This is assuming I fatfingered something in the
>     build.
>     >             >
>     >             >     The nighties and 5.0.1 seem fine.
>     >             >
>     >             >     I have a contract delivery this week, and things
>     are a
>     >             little
>     >             >     frantic, but I should still be able to get this
>     fixed.
>     >             >
>     >             >     Adam
>     >             >
>     >             >     On Mon, Jan 7, 2019, 5:46 PM Andy Peters
>     >             <devel@xxxxxxxxx <mailto:devel@xxxxxxxxx>
>     <mailto:devel@xxxxxxxxx <mailto:devel@xxxxxxxxx>>
>     >             >     <mailto:devel@xxxxxxxxx <mailto:devel@xxxxxxxxx>
>     <mailto:devel@xxxxxxxxx <mailto:devel@xxxxxxxxx>>> wrote:
>     >             >
>     >             >
>     >             >
>     >             >         > On Jan 7, 2019, at 3:20 PM, Adam Wolf
>     >             >         <adamwolf@xxxxxxxxxxxxxxxxxxxx
>     <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx>
>     >             <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx
>     <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx>>
>     >             >         <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx
>     <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx>
>     >             <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx
>     <mailto:adamwolf@xxxxxxxxxxxxxxxxxxxx>>>> wrote:
>     >             >         >
>     >             >         > Hi folks!
>     >             >         >
>     >             >         > Just a heads up, the macos 5.0.2 packages are
>     >             gross for some
>     >             >         reason.  I am regenerating them and we'll see
>     >             what's going on.
>     >             >         >
>     >             >         > (I am regenerating them at 5.0.2-2)
>     >             >
>     >             >         Gross in what way? I haven’t pulled down a
>     nightly
>     >             in a couple
>     >             >         of weeks.
>     >             >
>     >             >         -a
>     >             >         _______________________________________________
>     >             >         Mailing list:
>     https://launchpad.net/~kicad-developers
>     >             >         Post to     :
>     kicad-developers@xxxxxxxxxxxxxxxxxxx
>     <mailto:kicad-developers@xxxxxxxxxxxxxxxxxxx>
>     >             <mailto:kicad-developers@xxxxxxxxxxxxxxxxxxx
>     <mailto:kicad-developers@xxxxxxxxxxxxxxxxxxx>>
>     >             >         <mailto:kicad-developers@xxxxxxxxxxxxxxxxxxx
>     <mailto:kicad-developers@xxxxxxxxxxxxxxxxxxx>
>     >             <mailto:kicad-developers@xxxxxxxxxxxxxxxxxxx
>     <mailto:kicad-developers@xxxxxxxxxxxxxxxxxxx>>>
>     >             >         Unsubscribe :
>     https://launchpad.net/~kicad-developers
>     >             >         More help   :
>     https://help.launchpad.net/ListHelp
>     >             >
>     >             >
>     >             > _______________________________________________
>     >             > Mailing list: https://launchpad.net/~kicad-developers
>     >             > Post to     : kicad-developers@xxxxxxxxxxxxxxxxxxx
>     <mailto:kicad-developers@xxxxxxxxxxxxxxxxxxx>
>     >             <mailto:kicad-developers@xxxxxxxxxxxxxxxxxxx
>     <mailto:kicad-developers@xxxxxxxxxxxxxxxxxxx>>
>     >             > Unsubscribe : https://launchpad.net/~kicad-developers
>     >             > More help   : https://help.launchpad.net/ListHelp
>     >             >
>     >
>     >             _______________________________________________
>     >             Mailing list: https://launchpad.net/~kicad-developers
>     >             Post to     : kicad-developers@xxxxxxxxxxxxxxxxxxx
>     <mailto:kicad-developers@xxxxxxxxxxxxxxxxxxx>
>     >             <mailto:kicad-developers@xxxxxxxxxxxxxxxxxxx
>     <mailto:kicad-developers@xxxxxxxxxxxxxxxxxxx>>
>     >             Unsubscribe : https://launchpad.net/~kicad-developers
>     >             More help   : https://help.launchpad.net/ListHelp
>     >
> 
>     _______________________________________________
>     Mailing list: https://launchpad.net/~kicad-developers
>     Post to     : kicad-developers@xxxxxxxxxxxxxxxxxxx
>     <mailto:kicad-developers@xxxxxxxxxxxxxxxxxxx>
>     Unsubscribe : https://launchpad.net/~kicad-developers
>     More help   : https://help.launchpad.net/ListHelp
> 


Follow ups

References