Hi,

I'm currently preparing the DStress test suite (http://dstress.kuehne.cn/www/dstress.html) for Windows.

The test code for the "simple" test cases is now in one C file instead of several shell lines in the Makefile. http://dstress.kuehne.cn/dstress.c [No, this isn't a code beauty contests ;)]

e.g.
dstress run some_source_file.d
dstress nocompile test/another_source.html 2> detailed.log 1> results.txt

As far as I'm aware only the simple "loadFile" functions needs a Windows counterpart.

Anybody willing?

Thomas

Thomas Kuehne wrote:
> Hi,
> 
> I'm currently preparing the DStress test suite (http://dstress.kuehne.cn/www/dstress.html) for Windows.
> 
> The test code for the "simple" test cases is now in one C file instead of several shell lines in the Makefile.  http://dstress.kuehne.cn/dstress.c [No, this isn't a code beauty contests ;)]

Did you write it in C to make sure that a DMD regression doesn't stop you from running the suite?  :-)

> e.g.
> dstress run some_source_file.d
> dstress nocompile test/another_source.html 2> detailed.log 1> results.txt
> 
> As far as I'm aware only the simple "loadFile" functions needs a
> Windows counterpart.
> 
> Anybody willing?

You mean willing to write the Windows counterpart?  I suppose I could have a go....

(Just looking at it, it appears that, given a file of size -1, it will allocate a zero-length buffer and set the byte just after the end of it to '\0', which I guess isn't what you meant?  What is a file of size -1, anyway?)

Stewart.

-- 
My e-mail is valid but not my primary mailbox.  Please keep replies on
the 'group where everyone may benefit.

In article <ctb7bu$2k90$1@digitaldaemon.com>, Thomas Kuehne says...
>
>Hi,
>
>I'm currently preparing the DStress test suite (http://dstress.kuehne.cn/www/dstress.html) for Windows.
>
>The test code for the "simple" test cases is now in one C file instead of several shell lines in the Makefile. http://dstress.kuehne.cn/dstress.c [No, this isn't a code beauty contests ;)]
>
>e.g.
>dstress run some_source_file.d
>dstress nocompile test/another_source.html 2> detailed.log 1> results.txt
>
>As far as I'm aware only the simple "loadFile" functions needs a Windows counterpart.
>
>Anybody willing?
>
>Thomas
>
>

I think this does it (see attached file).
It's based on std.file.read

----------------
Carlos Santander

Stewart Gordon schrieb in news:ctbb33$2pi1$1@digitaldaemon.com :
> Thomas Kuehne wrote:
> > Hi,
> >
> > I'm currently preparing the DStress test suite (http://dstress.kuehne.cn/www/dstress.html) for Windows.
> >
> > The test code for the "simple" test cases is now in one C file instead of several shell lines in the Makefile. http://dstress.kuehne.cn/dstress.c [No, this isn't a code beauty contests ;)]
>
> Did you write it in C to make sure that a DMD regression doesn't stop you from running the suite?  :-)

How could I write the test tool for the very compiler I am going to scrutinise?

> (Just looking at it, it appears that, given a file of size -1, it will allocate a zero-length buffer and set the byte just after the end of it to '\0', which I guess isn't what you meant?  What is a file of size -1, anyway?)

Mhh...


Thomas

Carlos Santander schrieb in news:ctbbr2$2qgd$1@digitaldaemon.com :
> In article <ctb7bu$2k90$1@digitaldaemon.com>, Thomas Kuehne says...
> >
> >Hi,
> >
> >I'm currently preparing the DStress test suite (http://dstress.kuehne.cn/www/dstress.html) for Windows.
> >
> >The test code for the "simple" test cases is now in one C file instead of several shell lines in the Makefile. http://dstress.kuehne.cn/dstress.c [No, this isn't a code beauty contests ;)]
> >
> >e.g.
> >dstress run some_source_file.d
> >dstress nocompile test/another_source.html 2> detailed.log 1> results.txt
> >
> >As far as I'm aware only the simple "loadFile" functions needs a Windows counterpart.
> >
> >Anybody willing?
>
> I think this does it (see attached file).
> It's based on std.file.read

Thanks. I have applied a small patch for "system"'s return value and put it online.

Thomas