David Landgren (d...@landgren.net)

Without looking at the code, that would be a test for a bugfix I added
recently. It will be a simple matter to revert that to the two-arg form.

I'll get this wrapped up tonight.

Thanks for the report,
David

Actually I do, I run podchecker. Alas, it did not consider that to be an
error. One thing that annoys me at the moment at some point I upgraded
my POD tools, and now pod2text produces the blightful ``quotes'' instead
of just "quotes". Recipes to fix that gratefully received.

I fixed the problem on use.perl, but the correcting the mail message is
now quite beyond my reach.

David

You know, that would be a very good reason why it wasn't in Jerry's list
in the first place. A commendable decision, it's very 5.8-friendly!

David

File::Path 2.04 too (*pimp pimp*)

I don't think it can go any further. The only remaining possibilities
would be the utterly Huffman'ed 'elf' and 'ef'. But that would be sick.

David

I wasn't complaining, and in fact managed to finish the summary tonight.
Which is just as well, because I'll be offline until Monday night.

> (Apart from writing fewer e-mails, of course.)

Yes, that would be a good idea. Starting from now, you all have an
allowance of two messages per week. Use them wisely.

David

Can't you just bundle the appropriate module that fixes your problem
along with the rest of the application? How exactly is it packaged?

David

Minor quibble (I'm way behind on sumamrising)

> + # Test with non-existing class (used to segfault)

I thought the general movement these days was to embed the bug number in
the test

> +    eval { Qux->foo() };
> + is($@, '', "->next::can on non-existing package name");

so

is($@, '', "->next::can on non-existing package name (RT bug #51092)");

David

Minor nit with the patch:

@@ -647,12 +657,14 @@

   if(socketpair(AF_UNIX, SOCK_STREAM, 0, fd)) {
       perror("Debug leaking scalars socketpair failed");
+        MAIN_EXIT_PROBE();
       abort();
   }


The perl codebase, for better or worse, contains tabs (interpreted as 8
spaces). You will improve the chance of getting this integrated (and I
think it would be great to have it in) by redoing your patch to indent
cleanly (less work for the pumpking).

Thanks,
David

The lag is less than an hour, give or take.

> 2) Does the message have to get read by a moderator first?

If you're not a direct subscriber, yes.

I just checked, and you're not subscribed to the list, at least, not
from this gmail account that you sent the message from in January.

> 3) Did I somehow get blocked from the list by sending those first 3
> erroneous messages?

In any case, it made it to the list; I remember summarising the thread :)

David

In Enrico's defense, H.Merijn did zrite kurilla with two ells. Kurila
doesn't have sufficient exposure for Google to suggest "perhaps you
meant kurila?" in the presence of such a typo.

David

Thanks, fixed.

> * David Landgren <david@landgren.net> [2008-01-11 14:40]:
>> Rafaël killed the v-string portability warning in 5.10, declaring that
>>   it would no longer be present in 5.10.1.
>>
>>     no-one shed a tear
>> 500 Can't connect to metamark.net:80 (connect: Operation not permitted)
>
> and you probably want it to die screaming instead of silently
> replacing URLs with error messages.

I blame Ask :o)

And yes, the code will now die screaming. Thanks for the spot.
Interested parties are invited to amend local copies of the summary by
replacing the above error message with the following link

   http://xrl.us/bev6c

David "tearing hair out" Landgren

Jerry D. Hedden wrote:
>> Change 32999 by [email protected: nic...@nicholas-bouvard] on 2008/01/18 14:13:30
>>
>>         A target for sperl.i

What is sperl anyway? A static build? Setuid?

Thanks,
DAvid