[petsc-dev] Updating MOAB version during download

[Barry Smith]

On Jun 16, 2014, at 7:03 PM, Satish Balay <balay at mcs.anl.gov> wrote:

> On Mon, 16 Jun 2014, Barry Smith wrote:
> 
>> 
>> On Jun 16, 2014, at 5:28 PM, Satish Balay <balay at mcs.anl.gov> wrote:
>> 
> 
>>>> 2) It does not give me access to the branch so that I can make changes. Say I am working on feature-dmmoab in PETSc and see a little bug in the moab branch that (indirectly only since I am at some stupid headless commit-hash instead on a branch) I am pointing to, that if I quickly fix I can push and make life easier for my entire team of eight developers. I need to manual figure out what branch corresponds to the commit-hash thing I had checked out, change to that branch in moab, fix the branch in moab, push it and then comeback and edit moab.py in PETSc to point to the new commit-hash beasty of the moab branch.
>>> 
>>> We don't that luxuary of finding a bug in petsc [from nightly builds]
>>> and quickly fixing it in the appropriate branch anyway. We have to run
>>> a couple of git commands to do the appropriate thing. I would expect a
>>> smilar thing with moab would be fine. [its just that its more of a
>>> black-box to us petsc users wrt branch org]. But I don't see why
>>> --downlaod-package should be burdened with keeping track of 'git
>>> branches' which git doesn't track anyway.
>> 
>>   Because I sure as hell am not going to do something manually that can be done automatically. The reason I added —download-xxxx was NOT actually for end users (though they benefit from it greatly) BUT because __I__ refuse to keep downloading and installing over and over again over the years the same damn package as it evolves. I haven’t installed hypre in 10+ years (15?) manually, yet at least once a month I use —download-hypre image the wasted time if I still did it manually.
> 
> For one - I don't see the corelation between the above bug fix
> procedure I mentioned with the --download-package automation you are
> implying.

   I want to be able to use the —download-xxx beasty just as iff I had manually done a git clone ; get checkout to the right branch ; blah blah blah

    Jed seems to be saying, "if you just want to use the package for compiling then use —download-xxx otherwise go through the manual process of getting the damn package yourself and go through all the manual steps”. I say that is silly since —download-xxx already exists and can do 95% of what is needed already I never want to go through the manual steps even if I am contributing to that other package.

   Barry



> 
> Perhaps you are refering to the manually adding commit id to moab.py. For which I responded:
> 
>>> If automatic tracking is critical - then we should look at 'git submodule'
> 
> Perhaps there are others ways of automating this step [if thats the primary stumbling block.
> 
> Satish