[petsc-dev] Fwd: [petsc-maint #119133] petsc-dev configure crash

[Barry Smith]

On Jun 6, 2012, at 11:41 AM, Sean Farley wrote:

>>   Did I ever say a mechanism to "run arbitrary code"?  I do not believe I did, nor did I even hint at running arbitrary code. What I want is a mechanism to run another hg command, in fact a specific hg command. Not "arbitrary code".
> 
> Which boils down to arbitrary code :-) Your type of proposal has
> rejected repeatedly and no one has proposed a safe way to do it. If
> you feel this is easy, then by all means submit a patch to mercurial.
> You might as well train people to use 'hg pull -u' and have
> ./configure add the hook but Satish maintains that this doesn't solve
> all problems (which he is right) so it's useless (which I disagree
> with). Oh well. What do you propose, Barry?

   I propose what we do currently. It is not perfect but at worst requires an email of "pull BuildSystem". 

   Jed and Sean's email bitched about the current system (which is fine to bitch) but did not have specific suggestions on a new system.  So how do you propose to (1) not have the "pull BuildSystem" problem and (2) have bisection work? 

    Once we know the proposal we can determine the downside to the proposal and decide whether to change. But without a specific proposal (state it) you all are just flapping your lips.

   Barry