[Axiom-developer] RE: Automation & algebra lattice

[Bill Page]

David,

Here are the Perl and Bash scripts. Their usage should be
fairly obvious but if anything is unclear, please ask.

I put all of these scripts in the mnt/linux/bin directory.
Then I run them with the axiom root as the current directory,
specifying the set of spad files I want to analyzed like
this

  [axiom]% depends int/algebra/*.spad

The output goes to standard out. Because it takes a
while to execute (several hours unless you have a very
fast machine) you might prefer to do something like
the following;

  [axiom]% depends int/algebra/a*.spad int/algebra/A*.spad | tee a.dep
  [axiom]% depends int/algebra/b*.spad int/algebra/B*.spad | tee b.dep
  ...
  [axiom]% sort -u *.dep | unique
  [axiom]% cat *.spad | awk '{print $1" " $7;}' | sort -u | tsort 2>&1 |
tac

The "heuristic" algorithm that I used to find the dependecies
takes the list of exported functions from each domain in the
specified list of spad files and searches for these names
and the associated domains in all of the available spad and
pamphlet source files. I did it this way (sort of backwards)
in order to avoid having to really parse the spad source. I
think it produces fairly reliable results but perhaps not
100%.

Would you like me also to send you the complete dependency
file?

I am thinking of uploading these files to the savannah site
(not CVS, just auxillary related files) but I am not too
familiar how to do this. Your help and suggestions would be
appreciated.

Cheers,
Bill Page.

> -----Original Message-----
> From: David Mentré [mailto:address@hidden 
> Sent: Wednesday, August 20, 2003 9:18 AM
> To: Bill Page
> Cc: address@hidden
> Subject: Re: Automation & algebra lattice
> 
> 
> Hello Bill,
> 
> "Bill Page" <address@hidden> writes:
> 
> > About "parsing .spad files to write dependencies". I wrote
> > a Perl script that does this last week. Unfortunately the code is 
> > sitting on a machine that is still in the black-out zone following 
> > last week's power failure. I do have a print out and it is quite 
> > short, so if you are interested I can re-key it and send it. The 
> > parsing is a bit "heuristic" in places because of the macro 
> > flexibility in spad, but I think that it is at least 95% accurate.
> 
> Yes, I would be very interested in such a script (or the 
> dependencies in a file). There's no hurry however, I can wait 
> a few days that the machine reboots. :)
> 
> Yours,
> d.
> -- 
>  address@hidden
>

unique
Description: Binary data

depends
Description: Binary data

index
Description: Binary data

order
Description: Binary data