[Caml-list] FORT (or other test frameworks)

[Caml-list] FORT (or other test frameworks)

[John L. Masson]

Hello list,

Is anyone here using FORT? Does anyone have any examples or
documentation for it? 

I've read, and I thought understood, the source, but I can't get a
testscript to run. 

If I just pass it to the 'fort' executable (i.e. 'fort testsuite.ml'),
the Fort module and its functions are not available to the script. 

I can compile the script, and link to fort.cmo and fortmain.cmo, and
this executable produces a report, but it shows no tests were run.

If I don't link fortmain.cmo, nothing happens at all.

What am I doing wrong?

-- 
John Masson 
Interactive Developer

Worth Media

-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] FORT (or other test frameworks)

[Patrick M Doane]

Hi John,

On 12 Sep 2001, John L. Masson wrote:

> Hello list,
> 
> Is anyone here using FORT? Does anyone have any examples or
> documentation for it? 
> 
> I've read, and I thought understood, the source, but I can't get a
> testscript to run. 
> 
> If I just pass it to the 'fort' executable (i.e. 'fort testsuite.ml'),
> the Fort module and its functions are not available to the script. 

I would expect this to work... have you installed the executable? The
'fort' executable will need to find the bytecode data for the Fort module
and will search in the OCAMLLIB directory.

Also, inside testsuite.ml you will need to refer to the Fort functions
with a qualified prefix (e.g. Fort.test) or include an 'open Fort;;'
directive at the top.

> I can compile the script, and link to fort.cmo and fortmain.cmo, and
> this executable produces a report, but it shows no tests were run.
> 
> If I don't link fortmain.cmo, nothing happens at all.

It may be worthwhile to think about linking Fort into an application, but
there is currently no way to separate the report generation from script
execution. 

> What am I doing wrong?

If you're still having problems, feel free to e-mail me.

Thanks!

Patrick


-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] FORT (or other test frameworks)

[Sven]

On Wed, Sep 12, 2001 at 12:09:26PM -0400, Patrick M Doane wrote:
> Hi John,
> 
> On 12 Sep 2001, John L. Masson wrote:
> 
> > Hello list,
> > 
> > Is anyone here using FORT? Does anyone have any examples or
> > documentation for it? 
> > 
> > I've read, and I thought understood, the source, but I can't get a
> > testscript to run. 
> > 
> > If I just pass it to the 'fort' executable (i.e. 'fort testsuite.ml'),
> > the Fort module and its functions are not available to the script. 
> 
> I would expect this to work... have you installed the executable? The
> 'fort' executable will need to find the bytecode data for the Fort module
> and will search in the OCAMLLIB directory.

should this not be $OCAMLLIB/fort ?

or is there still no consensus on a policy on this ?

Friendly,

Sven Luther
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr

Re: [Caml-list] FORT (or other test frameworks)

[Patrick M Doane]

On Thu, 13 Sep 2001, Sven wrote:

> On Wed, Sep 12, 2001 at 12:09:26PM -0400, Patrick M Doane wrote:
> > I would expect this to work... have you installed the executable? The
> > 'fort' executable will need to find the bytecode data for the Fort module
> > and will search in the OCAMLLIB directory.
> 
> should this not be $OCAMLLIB/fort ?
> 
> or is there still no consensus on a policy on this ?

I could modify the installation to default to a subdirectory, but I'm not
sure that it offers much benefit.  For large packages that install many
modules, it might help to be able to easily uninstall by deleting a
directory. 

Until Caml has real package support, installing into the OCAMLLIB
directory is probably the best policy.  This will find conflicts between
module names sooner rather than later and makes the use of the modules
easier.  Installing non-INRIA modules into a single subdirectory like
'contrib' might be a reasonable alternative though. 

Patrick

-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr