[petsc-dev] reminder never use #include "mylocalinclude.h" in PETSc source

[Barry Smith]

On May 3, 2010, at 11:31 AM, Jed Brown wrote:

> On Mon, 3 May 2010 15:25:04 -0300, Lisandro Dalcin  
> <dalcinl at gmail.com> wrote:
>> On 3 May 2010 09:59, Matthew Knepley <knepley at gmail.com> wrote:
>>> I had the same understanding as Jed. What was the problem which  
>>> prompted the
>>> fix?
>>
>> Perhaps the issue is not related to C compilation, but generation of
>> hyper-linked html documentation?
>
> That's possible (and what I asked in my earlier message), but would it
> really be that hard for the documentation tools to obey the same  
> include
> resolution rules as the compilers?
>
> Jed

    This has been the PETSc style for several years. It is followed  
well (Matt follows it :-). There was only a few bad spots due to Jed  
and Shri. Perhaps I should have just fixed them and not sent the  
email. It has nothing to do with generation of documentation or  
Fortran stubs.

    The reason for it is to disconnect the location of a source file  
from "its" include files. Otherwise you have the implicit assumption  
that a source file is compiled in the directory that it is sitting it.  
I want the freedom to compile source code with NO assumption as to  
where it is sitting. That is the freedom to compile code in a  
different way then the traditional recursive make that PETSc uses to  
build its libraries. (One simple reason is that the traditional  
recursive make is damn slow).

    Barry