[petsc-dev] error with build

[Barry Smith]

On Nov 15, 2010, at 4:43 PM, Lisandro Dalcin wrote:

> On 15 November 2010 18:53, Barry Smith <bsmith at mcs.anl.gov> wrote:
>> 
>>   Here is that part of the makefile. So I guess the answer is, yes it should all just work but I don't know why it doesn't.
>> 
> 
> That "makefile" is NOT intended to be petsc4py buildsystem. I use it
> as a convenience for my everyday hacking.

   Yuck, yuck, yuck. If the "true" buildsystem is too clunky to use for everyday hacking then the "true" buildsystem should be improved. Having a clunky "true" buildsystem plus a "good" buildsystem for everyday hacking is a recipe for very annoyed people.

   Barry

> 
>> Or maybe it does. It would help if petsc4py was properly in the nightly builds! Or is it?
>> 
> 
> Not sure if BuildSystem handles the hg pull of petsc4py-dev ....
> 
>>   Barry
>> 
>> 
>> 
>> CY_SRC_PXD = $(wildcard src/include/petsc4py/*.pxd)
>> CY_SRC_PXI = $(wildcard src/PETSc/*.pxi)
>> CY_SRC_PYX = $(wildcard src/PETSc/*.pyx)
>> src/PETSc.c: src/petsc4py.PETSc.c
>> src/petsc4py.PETSc.c: ${CY_SRC_PXD} ${CY_SRC_PXI} ${CY_SRC_PYX}
>>        ${PYTHON} ./conf/cythonize.py
>> 
>> 
>> 
>> On Nov 15, 2010, at 3:44 PM, Jed Brown wrote:
>> 
>>> On Mon, Nov 15, 2010 at 22:21, Barry Smith <bsmith at mcs.anl.gov> wrote:
>>> It will run cython the FIRST time after a clone but not after a hg pull. Which is wrong.  I would like to see this code check if it is a .hg repository (for example) and always run cython in that case.
>>> 
>>> Is it somehow not possible to write correct Make dependencies for cython?
>>> 
>>> Jed
>> 
>> 
> 
> 
> 
> -- 
> Lisandro Dalcin
> ---------------
> CIMEC (INTEC/CONICET-UNL)
> Predio CONICET-Santa Fe
> Colectora RN 168 Km 472, Paraje El Pozo
> Tel: +54-342-4511594 (ext 1011)
> Tel/Fax: +54-342-4511169