pearu / f2py

Automatically exported from code.google.com/p/f2py
Other
55 stars 39 forks source link

a callcaller tree generator #18

Open GoogleCodeExporter opened 9 years ago

GoogleCodeExporter commented 9 years ago
I wrote a small program to generate callcaller tree for Fortran codes.
 f_calltree.F file1 file2,...  >callcaller.dat 2>callcaller.err
Generate call caller for fortran codes.
Put this just above f2py directory.

I hope you will include this in your fparser package if possible.
(Warning: this cause an error if computer codes are gramatically wrong 
 when we comment out preprocessor lines. Further, include is not treated well).

Another point.--
fparser uses yellowtext and so, and ubyte in numpy.
So people need to install numpy only in order to use fparser.
I think this is a little too demanding for people who just want 
to use fparser only. So I hope fparser works without numpy.

Original issue reported on code.google.com by TakaoKot...@gmail.com on 2 Apr 2010 at 7:40

Attachments:

GoogleCodeExporter commented 9 years ago
I am trying to develop a little better version. Analysis on
-----------
 1.program units. 
   *Decleared variables.
   *Contained program units.
   *Identification of I/O of variables. I/O of files.
 2.Relations between the units. 
-----------
are important things. 

Original comment by TakaoKot...@gmail.com on 10 Apr 2010 at 9:12

GoogleCodeExporter commented 9 years ago
The yellowtext, etc can be copied from numpy to f2py but I think
currently the most important usage of numpy within f2py is with
setup.py scripts. I find it more convenient to use numpy.distutils
for packages that contains many standalone subpackages.
May be in future, when fparser becomes more stable, we can remove
numpy dependency, but for now, I think, having numpy around is convenient,
especially because the f2py project will contain in addition to fparser
subpackage also, for example, wrapper generation package that will
require numpy.

About additions to fparser. Please look at the small programs in
fparser/scripts directory, it would be easier to include your scripts
to fparser if you would use the same convention for writing scripts
as these example scripts.

Original comment by pearu.peterson on 15 Apr 2010 at 7:24

GoogleCodeExporter commented 9 years ago
Thank you for your reply.
Since I had installed numpy and so in my machine (ubuntu latest),
no problem to use fparser. 

However,I had a problem when I ask my collabolator to use fparser.
He needed to spend a day or so(he is skillful enough), 
since he had little knowledge on python.

Now fparser use very limited part of numpy.
Maybe 
 1 yellow_text, red_text...
 2. ubyte.
So, it is probably easy to set a branch when numpy can not find.

Original comment by TakaoKot...@gmail.com on 15 Apr 2010 at 10:53