@smpeters - Status changed from 'open' to 'stalled'

From @Hugmeir

On Sun Feb 18 01​:05​:26 2001, Peter@​PSDT.com wrote​:

How?

The RT System itself - Status changed from 'stalled' to 'open'

From @pjscott

On May 5, 2012, at 4​:03 AM, "Brian Fraser via RT" <perlbug-followup@​perl.org> wrote​:

On Sun Feb 18 01​:05​:26 2001, Peter@​PSDT.com wrote​:

How?

Er, you have me at a disadvantage. Neither my memory nor my mail spool goes back 11 years. I am sure that I thought I was including a body in my message at the time but why there is none now I cannot say. Bit decay, perhaps.

I think that it is likely - at least at the time - that setting PERL5OPT affected building perl from source, and that I thought it shouldn't, since it happened to me by accident. Opinions may differ.

From @tonycoz

On Sat, May 05, 2012 at 04​:03​:17AM -0700, Brian Fraser via RT wrote​:

On Sun Feb 18 01​:05​:26 2001, Peter@​PSDT.com wrote​:

How?

Anything that loads a non in-lib core module?

# just an example, Imager isn't suitable for PERL5OPT -M
$ ./Configure -des -Dusedevel && PERL5OPT=-MImager make -j3
...
cc -fstack-protector -L/usr/local/lib -o miniperl \
  perlmini.o opmini.o miniperlmain.o gv.o toke.o perly.o pad.o regcomp.o dump.o util.o mg.o reentr.o mro.o keywords.o hv.o av.o run.o pp_hot.o sv.o pp.o scope.o pp_ctl.o pp_sys.o doop.o doio.o regexec.o utf8.o taint.o deb.o universal.o globals.o perlio.o perlapi.o numeric.o mathoms.o locale.o pp_pack.o pp_sort.o -lnsl -ldl -lm -lcrypt -lutil -lc
./miniperl -w -Ilib -MExporter -e '<?>' || make minitest
Can't locate Imager.pm in @​INC (@​INC contains​: lib .).
BEGIN failed--compilation aborted.

Perhaps there's a less ridiculous option that causes a build failure.

Tony

From @nwc10

On Tue, May 08, 2012 at 10​:02​:55AM +1000, Tony Cook wrote​:

On Sat, May 05, 2012 at 04​:03​:17AM -0700, Brian Fraser via RT wrote​:

On Sun Feb 18 01​:05​:26 2001, Peter@​PSDT.com wrote​:

How?

Anything that loads a non in-lib core module?

# just an example, Imager isn't suitable for PERL5OPT -M
$ ./Configure -des -Dusedevel && PERL5OPT=-MImager make -j3
...
cc -fstack-protector -L/usr/local/lib -o miniperl \
perlmini.o opmini.o miniperlmain.o gv.o toke.o perly.o pad.o regcomp.o dump.o util.o mg.o reentr.o mro.o keywords.o hv.o av.o run.o pp_hot.o sv.o pp.o scope.o pp_ctl.o pp_sys.o doop.o doio.o regexec.o utf8.o taint.o deb.o universal.o globals.o perlio.o perlapi.o numeric.o mathoms.o locale.o pp_pack.o pp_sort.o -lnsl -ldl -lm -lcrypt -lutil -lc
./miniperl -w -Ilib -MExporter -e '<?>' || make minitest
Can't locate Imager.pm in @​INC (@​INC contains​: lib .).
BEGIN failed--compilation aborted.

Perhaps there's a less ridiculous option that causes a build failure.

Likely. But your example suggests to me that *mini*perl should ignore all of
these, as they aren't* used by the build process, and that the invocations for
./perl should explicitly clear them all.

(Which isn't too hard actually, as ./perl is, or should be, invoked
consistently throughout the Makefile with a macro that sets up
"LD_LIBRARY_PATH" etc)

If this idea seems sane, an no-one else beats me to it, I'll try to do it in
the next couple of months.

Nicholas Clark

* I should check this :-)

From @hvds

Nicholas Clark <nick@​ccl4.org> wrote​:
:> Perhaps there's a less ridiculous option that causes a build failure.
:
:Likely. But your example suggests to me that *mini*perl should ignore all of
:these, as they aren't* used by the build process, and that the invocations for
:./perl should explicitly clear them all.
:
:(Which isn't too hard actually, as ./perl is, or should be, invoked
:consistently throughout the Makefile with a macro that sets up
:"LD_LIBRARY_PATH" etc)
:
:If this idea seems sane, an no-one else beats me to it, I'll try to do it in
:the next couple of months.

I think by "should ignore all of these" you're saying "ignore anything in
PERL5OPT" (or possibly "PERL5OPT and friends").

If so, please consider whether it is worth providing a mechanism to specify
the value to use for PERL5OPT (resp. "... and friends") for the miniperl
build as a get-out clause.

I do not have a use for this myself, so I do not request it; but I imagine,
should it be of use, it would be easier to make that change now than to try
and restore the facility in a user-controllable form later.

Hugo

From @jkeenan

On Tue May 08 12​:02​:40 2012, hv wrote​:

Nicholas Clark <nick@​ccl4.org> wrote​:
:> Perhaps there's a less ridiculous option that causes a build
failure.
:
:Likely. But your example suggests to me that *mini*perl should ignore
all of
:these, as they aren't* used by the build process, and that the
invocations for
:./perl should explicitly clear them all.
:
:(Which isn't too hard actually, as ./perl is, or should be, invoked
:consistently throughout the Makefile with a macro that sets up
:"LD_LIBRARY_PATH" etc)
:
:If this idea seems sane, an no-one else beats me to it, I'll try to
do it in
:the next couple of months.

I think by "should ignore all of these" you're saying "ignore anything
in
PERL5OPT" (or possibly "PERL5OPT and friends").

If so, please consider whether it is worth providing a mechanism to
specify
the value to use for PERL5OPT (resp. "... and friends") for the
miniperl
build as a get-out clause.

I do not have a use for this myself, so I do not request it; but I
imagine,
should it be of use, it would be easier to make that change now than
to try
and restore the facility in a user-controllable form later.

Hugo

Nicholas, do you think we should proceed with this?

Thank you very much.
Jim Keenan

From @Leont

On Mon May 07 17​:03​:26 2012, tonyc wrote​:

Anything that loads a non in-lib core module?

# just an example, Imager isn't suitable for PERL5OPT -M
$ ./Configure -des -Dusedevel && PERL5OPT=-MImager make -j3
...
cc -fstack-protector -L/usr/local/lib -o miniperl \
perlmini.o opmini.o miniperlmain.o gv.o toke.o perly.o
pad.o regcomp.o dump.o util.o mg.o reentr.o mro.o keywords.o hv.o
av.o run.o pp_hot.o sv.o pp.o scope.o pp_ctl.o pp_sys.o doop.o
doio.o regexec.o utf8.o taint.o deb.o universal.o globals.o
perlio.o perlapi.o numeric.o mathoms.o locale.o pp_pack.o pp_sort.o
-lnsl -ldl -lm -lcrypt -lutil -lc
./miniperl -w -Ilib -MExporter -e '<?>' || make minitest
Can't locate Imager.pm in @​INC (@​INC contains​: lib .).
BEGIN failed--compilation aborted.

Perhaps there's a less ridiculous option that causes a build failure.

Can't this just be solved by adding a «PERL5OPT=» line in the Makefile?
That way, the environmental variable is overridden but miniperl can
still use PERL5OPT if required.

Leon

From @jkeenan

On Wed, 18 Sep 2013 23​:50​:23 GMT, jkeenan wrote​:

On Tue May 08 12​:02​:40 2012, hv wrote​:

Nicholas Clark <nick@​ccl4.org> wrote​:
:> Perhaps there's a less ridiculous option that causes a build
failure.
:
:Likely. But your example suggests to me that *mini*perl should ignore
all of
:these, as they aren't* used by the build process, and that the
invocations for
:./perl should explicitly clear them all.
:
:(Which isn't too hard actually, as ./perl is, or should be, invoked
:consistently throughout the Makefile with a macro that sets up
:"LD_LIBRARY_PATH" etc)
:
:If this idea seems sane, an no-one else beats me to it, I'll try to
do it in
:the next couple of months.

I think by "should ignore all of these" you're saying "ignore anything
in
PERL5OPT" (or possibly "PERL5OPT and friends").

If so, please consider whether it is worth providing a mechanism to
specify
the value to use for PERL5OPT (resp. "... and friends") for the
miniperl
build as a get-out clause.

I do not have a use for this myself, so I do not request it; but I
imagine,
should it be of use, it would be easier to make that change now than
to try
and restore the facility in a user-controllable form later.

Hugo

Nicholas, do you think we should proceed with this?

Thank you very much.
Jim Keenan

The discussion in this ticket petered out more than 3 years ago -- and we never had a clear statement of the problem to be solved in the first place.

So I recommend closing this ticket (and will do so within 7 days unless someone objects).

Thank you very much.

--
James E Keenan (jkeenan@​cpan.org)

From @jkeenan

On Tue, 27 Dec 2016 22​:44​:59 GMT, jkeenan wrote​:

On Wed, 18 Sep 2013 23​:50​:23 GMT, jkeenan wrote​:

On Tue May 08 12​:02​:40 2012, hv wrote​:

Nicholas Clark <nick@​ccl4.org> wrote​:
:> Perhaps there's a less ridiculous option that causes a build
failure.
:
:Likely. But your example suggests to me that *mini*perl should
ignore
all of
:these, as they aren't* used by the build process, and that the
invocations for
:./perl should explicitly clear them all.
:
:(Which isn't too hard actually, as ./perl is, or should be,
invoked
:consistently throughout the Makefile with a macro that sets up
:"LD_LIBRARY_PATH" etc)
:
:If this idea seems sane, an no-one else beats me to it, I'll try
to
do it in
:the next couple of months.

I think by "should ignore all of these" you're saying "ignore
anything
in
PERL5OPT" (or possibly "PERL5OPT and friends").

If so, please consider whether it is worth providing a mechanism to
specify
the value to use for PERL5OPT (resp. "... and friends") for the
miniperl
build as a get-out clause.

I do not have a use for this myself, so I do not request it; but I
imagine,
should it be of use, it would be easier to make that change now
than
to try
and restore the facility in a user-controllable form later.

Hugo

Nicholas, do you think we should proceed with this?

Thank you very much.
Jim Keenan

The discussion in this ticket petered out more than 3 years ago -- and
we never had a clear statement of the problem to be solved in the
first place.

So I recommend closing this ticket (and will do so within 7 days
unless someone objects).

Thank you very much.

Closing as per schedule.

Thank you very much.

--
James E Keenan (jkeenan@​cpan.org)

@jkeenan - Status changed from 'open' to 'rejected'