Date: Thu, 19 Aug 1999 11:38:16 +0300 (IDT)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: Esa A E Peuha <peuha AT cc DOT helsinki DOT fi>
cc: djgpp-workers AT delorie DOT com, "Paul D. Smith" <pausmith AT nortelnetworks DOT com>
Subject: Re: Make 3.78 is in pretest
In-Reply-To: <Pine.OSF.4.03.9908190941140.22739-100000@sirppi.helsinki.fi>
Message-ID: <Pine.SUN.3.91.990819112915.9461E-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: djgpp-workers AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk


On Thu, 19 Aug 1999, Esa A E Peuha wrote:

> That's the result I get with the executable you sent me, except for the
> following tests (they work for me):
> 
> > features/echoing
> > targets/clean

These sometimes fail, sometimes don't, even in my case.  The exact result 
depends on several factors:

	- what does $SHELL point to;
	- whether PATH uses forward- or back-slashes;
	- whether there are other files in the work directory.

(there might be other factors).

> I didn't realize Windows doesn't allow `>' in file names.

Windows is not Unix...  None of the redirection/pipe characters is 
allowed, and wildcard characters `*' and `?' are also disallowed.

> > 1) features/default_names
> 
> >    The problem here is that Makefile and makefile map to the same file
> >    on DOS/Windows.  The solution I suggest is to generate each
> >    standard makefile just before it is used, as opposed to generating
> >    them all and then running them all:
> 
> But that would partly defeat the purpose of the test, which is to make
> sure that Make searches for the standard makefiles in specific order.  I
> think this should first test whether Makefile and makefile map to
> different files, and then skip creating and testing makefile if they
> don't.  (Even on DOS it's reasonable to test that Make prefers GNUMakefile
> over Makefile.)

Can you send a patch?  (But before that, please look at the new version 
of the test suite, which has some Windows-related changes already.)

> > 14) targets/clean
> 
> >    This uses "...", see features/echoing above.  Here's the patch:
> 
> How come these don't work for you?

The expansion of "..." depends on whether there are subdirectories when 
the command runs; if there are no subdirectories, it expands to itself.  
Therefore, the results depend on how many other tests failed.

Also, if you don't have Sh-utils installed, you don't have echo.exe, so 
the command internal to COMMAND.COM gets invoked instead, and the latter 
doesn't know about "...".

In any case, using "..." alone is dangerous.

> Anyway, there's no
> reason why the scripts should run echo if the command line is changed as
> you suggest.

I will try with the original version as well.