Front page | perl.perl5.porters |
Postings from December 2004
At 10:43 PM -0500 12/29/04, Michael G Schwern wrote:
>On Wed, Dec 29, 2004 at 12:09:58PM -0600, Craig A. Berry wrote:
>> But still not that close on VMS. For starters, here are some things I
>> noticed:
>>
>> 1.) There are several instances of the string:
>>
>> bin/instmodsh
>>
>> in various targets, though EXE_FILES does now show instmodsh in native
>> syntax.
>
>I'll investigate this.
>
>
>> 2.) The ci target has a single command that is too long for the command
>> buffer.
>
>This code hasn't been touched in over a year so I'm not going to worry about
>it for this release. Post it up on rt.cpan.org please.
>
>PS You're not actually using that, are you?
No, just reading through looking for things that might cause trouble.
> > 3.) The IGNORE prefix needs to have space after it and before the
>> command it prefixes. This is the same as for the SILENT prefix. However,
>> when both prefixes appear together, there must be no space between the
>> two prefixes but there must be space before the command they prefix. See
>> section 2.6.4 of your MMS manual.
>
>Yeah, it says that but neither MMS nor MMK have a problem with it as
>evidenced by the realclean and clean targets. However, when I changed it
>from "$(IGNORE)$(FOO)" to "$(IGNORE) $(FOO)" bsdmake broke. What I have
>now seems to work for everybody so I'm inclined to leave it.
>
>Are you seeing problems?
Yes, MMS and MMK both definitely have a problem with not leaving
space between the prefixes. Take the following descrip.mms:
$ type descrip.mms
bar : foo
write sys$output "OK"
foo :
- @ type nosuchfile.dat
[end of file]
and run them like so:
$ mmk
%TYPE-W-SEARCHFAIL, error searching for D0:[CRAIG.FOO]NOSUCHFILE.DAT;
-RMS-E-FNF, file not found
%MMK-F-ERRUPD, error status %X10951238 occurred when updating target FOO
$
$ mms
@ type nosuchfile.dat
%DCL-E-OPENIN, error opening D21:[FOO]TYPE.COM; as input
-RMS-E-FNF, file not found
write sys$output "OK"
OK
What's happened here is that MMK has successfully processed the
silent prefix but ignored the ignore prefix. It apparently takes
whichever one it comes to last when there is space between them.
MMS, on the other hand, has successfully processed the ignore prefix
but has treated the silent prefix as part of the command; it
apparently takes whichever one it comes to first.
Furthermore, I can't see any way to make the ignore prefix work as a
macro. The decision about whether the success or failure of a
command will be taken into account apparently happens before macro
substitution in both MMK and MMS.
--
________________________________________
Craig A. Berry
mailto:craigberry@mac.com
"... getting out of a sonnet is much more
difficult than getting in."
Brad Leithauser
Thread Previous
|
Thread Next