On Tue, Apr 02, 2002 at 06:04:39AM +0900, Dan Kogai wrote:
> On Tuesday, April 2, 2002, at 05:43 , Jarkko Hietaniemi wrote:
> > On Mon, Apr 01, 2002 at 02:38:46PM -0600, Craig A. Berry wrote:
> >> This is a not entirely portable revision of a completely unportable
> >> method
> >> of determining whether enc2xs is being run during a core build, but it
> >> does
> >> allow the build to proceed on VMS. Can't the extensions that use
> >> enc2xs
> >> pass it a parameter of PERL_CORE=$(PERL_CORE) or similar?
> >>
> >> --- ext/Encode/bin/enc2xs;-0 Sun Mar 31 15:05:56 2002
> >> +++ ext/Encode/bin/enc2xs Mon Apr 1 14:20:56 2002
> >> @@ -1,7 +1,7 @@
> >> #!./perl
> >> BEGIN {
> >> - # fiddle with @INC iff I am a part of perl dist
> >> - if ($^X =~ m/\bminiperl$/o){
> >> + # fiddle with @INC if I am a part of perl dist
> >> + if ($^X =~ m/miniperl(?:\.exe\;\d+)?$/o) {
> >
> > Ahhh, this is what probably tripped cygwin, too. $Config{_exe} would
> > be more portable than .exe.
>
> Thank you folks. How about the one below ? Does it look
>
> --- bin/enc2xs 2002/03/31 21:00:50 1.10
> +++ bin/enc2xs 2002/04/01 20:59:58
> @@ -1,7 +1,8 @@
> #!./perl
> BEGIN {
> # fiddle with @INC iff I am a part of perl dist
> - if ($^X =~ m/\bminiperl$/o){
> + require Config; import Config;
> + if ($^X =~ m/\bminiperl$Config{_exe}$/){
As Schwern pointed out, none of this should be necessary, if the
Encode $ENV{PERL_CORE} changes he sent are applied. (This, of course,
will require revisiting when and if Encode is backported to older
Perls, but that's VERY post-5.8 worry.)
> warn "Fixing \@INC for perl core.\n";
> unshift @INC, qw(../../lib ../../../lib ../../../../lib);
> $ENV{PATH} .= ';../..;../../..;../../../..' if $^O eq 'MSWin32';
>
> Dan the Encode Maintainer
--
$jhi++; # http://www.iki.fi/jhi/
# There is this special biologist word we use for 'stable'.
# It is 'dead'. -- Jack Cohen
