Re: Patch for missing shell commands in rc2

[J. David Bryan]

On 22 Mar 2006 at 20:57, Eli Zaretskii wrote:

> I don't have Windows 9x handy, so I cannot try this, but that was only
> an example (btw, `start' does exist on Windows 9x, albeit not as an
> internal command, IIRC). 

Right, "start" is an external command under Win9x.


> What I'm arguing is that running commands through the Windows shell may
> well produce different results.

The issue is restricted to the ten commands added to the list: assoc, 
color, endlocal, ftype, move, popd, pushd, setlocal, start, and title.  
None of these exist as internal commands in command.com, so under 9x, 
executing via the shell would cause external commands to be run.  I've seen 
no behavior to suggest that arguments passed via the shell to external 
programs somehow arrive differently than arguments passed directly (subject 
to the line-length limitation).

Your example is of a change in behavior caused by a shell command that 
masks an external command.  That is a valid concern, but I don't see that 
it has anything to do with parameter passing.


> Did you actually try this?

I just tried a quick test of a 500-character line on Win95.  Works.


> That is already enough to try to solve the problem in another way.

Fair enough.


> And I believe I suggested a better way, which is to defer the test to
> the run time. 

I believe you did.


> I believe the overhead will be negligible, since all you need is a
> bunch of string comparisons.

And some variable and environment lookups.


> Well, these aren't jobs for Make, but okay.

Sorry, I must have missed that part in the manual describing which jobs are 
permitted and which are not.  :-)

                                      -- Dave