Can we ship 5.28.0 with cygwin broken?

5.28.0 has been delayed enough that it is affecting 5.29/5.30.  We also 
traditionally have it out by the time TPC-NA happens.

We have two blocking tickets open for 5.28.0.

1) https://rt.perl.org/Ticket/Display.html?id=133220  perl_parse() 
return value breaks PostgreSQL pl/perl

blead already contains a revert that should allow this to be removed 
from the list.

2) https://rt.perl.org/Ticket/Display.html?id=133152 Cygwin build failure

There is this patch:

From: Tony Cook <tony@develop-help.com>
Date: Wed, 2 May 2018 11:15:30 +1000
Subject: [PATCH] (perl #133152) nostdio.h is broken, don't use it

Is there a reason not to apply this patch, or to ship 5.28.0 now with an 
expectation of a soon 5.28.1?  I think we used to expect the dot 1 to 
come out like a month after dot 0.
0
public
6/4/2018 2:32:13 PM
perl.perl5.porters 47363 articles. 0 followers. Follow

2 Replies
51 Views

Similar Articles

[PageSpeed] 42

Karl Williamson writes:
> 2) https://rt.perl.org/Ticket/Display.html?id=133152 Cygwin build failure
>
> There is this patch:
>
> From: Tony Cook <tony@develop-help.com>
> Date: Wed, 2 May 2018 11:15:30 +1000
> Subject: [PATCH] (perl #133152) nostdio.h is broken, don't use it
>
> Is there a reason not to apply this patch, or to ship 5.28.0 now with
> an expectation of a soon 5.28.1?  I think we used to expect the dot 1
> to come out like a month after dot 0.

I don't have any plans of moving Cygwin from 5.26 any time soon, so from
the Cygwin perspective I'm OK with quickfixing (or even not fixing) this
particular bug.


Regards,
Achim.
-- 
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

DIY Stuff:
http://Synth.Stromeko.net/DIY.html
0
Stromeko
6/4/2018 7:09:54 PM
On Mon, Jun 4, 2018 at 2:09 PM, Achim Gratz <Stromeko@nexgo.de> wrote:
> Karl Williamson writes:
>> 2) https://rt.perl.org/Ticket/Display.html?id=133152 Cygwin build failure
>>
>> There is this patch:
>>
>> From: Tony Cook <tony@develop-help.com>
>> Date: Wed, 2 May 2018 11:15:30 +1000
>> Subject: [PATCH] (perl #133152) nostdio.h is broken, don't use it
>>
>> Is there a reason not to apply this patch, or to ship 5.28.0 now with
>> an expectation of a soon 5.28.1?  I think we used to expect the dot 1
>> to come out like a month after dot 0.
>
> I don't have any plans of moving Cygwin from 5.26 any time soon, so from
> the Cygwin perspective I'm OK with quickfixing (or even not fixing) this
> particular bug.

A safer quick fix might be to add Cygwin next to EBCDIC inside of
nostdio.h as a condition for skipping the redefines, though that would
not exclude the redefinition of FILE.

As far as I can tell, the purpose of nostdio.h is to prevent people
from using stdio calls in core without explicitly acknowledging that
they are mixing stdio and PerlIO by setting PERLIO_NOT_STDIO to zero.
Do we really not need that anymore?  Isn't it still possible we could
unwilttingly apply a patch that uses stdio calls and seems to work ok
but causes less immediately obvious problems by bypassing the layers?
0
craig
6/4/2018 8:24:01 PM
Reply: