I haven't been offline, just too swamped to really *do* much
about this stuff recently.
> At 12:50 PM -0400 4/19/02, [EMAIL PROTECTED] wrote:
> >Craig Berry wrote:
> >
> >> Otherwise this will very likely not get fixed before 5.8.0.
> >> I think I'll go ahead and submit my patch that gets perldoc
> >> working since it makes things no worse than they already were.
> >
> >Your work is much appreciated. I have perl@16005 recently
> >(slowly) downloaded and would like to pitch in. So, let's
> >see the proposed patch. I take it then that Charles Lanes'
> >torture test is not a *.t file in the perl distribution?
> OK, there are actually 2, mutually exclusive possible patches here.
> Perhaps folks can help me figure out which is better. The idea
> behind both of them is that a child process created by system(),
> backticks, or opening a pipe may need to get to the SYS$COMMAND of
> the parent. This is the case with perldoc, where the pager (such as
> TYPE/PAGE) can't do its thing if there is no SYS$COMMAND or if
> SYS$COMMAND points to _NLA0: as we currently have it doing.
> The first approach is simpler and merely consists of omitting the
> input file argument to lib$spawn. This causes the child to inherit
> SYS$INPUT and SYS$COMMAND from the parent, but SYS$INPUT will be
> explicitly overridden in vmspipe.com. The docs say nothing about
> SYS$COMMAND but observation shows that this is what happens. This
> approach depends on trusting the docs when they say nothing will be
> read from the parent's input when the input file argument is omitted.
> I suspect this may not always be true or we wouldn't have been
> explicitly passing _NLA0: in the first place.
> The second approach passes SYS$COMMAND's equivalence name as a symbol
> to the subprocess where the symbol can then be used to define a
> logical. This is the same approach we are already using to define
> the other PPFs for the subprocess; we just never dealt with
> SYS$COMMAND before.
My bet is that the second approach is the better (and more flexible)
one. But this is not based on any detailed testing or code examination.
As for the "hangs", any idea what bit of the torture tests it's hanging
on?
--
Drexel University \V --Chuck Lane
======]---------->--------*------------<-------[===========
(215) 895-1545 _/ \ Particle Physics
FAX: (215) 895-5934 /\ /~~~~~~~~~~~ [EMAIL PROTECTED]
