Re: [COMMITTERS] pgsql: Sync tzload() and tzparse() APIs with IANA release tzcode2016c. - Mailing list pgsql-hackers

From Christian Ullrich
Subject Re: [COMMITTERS] pgsql: Sync tzload() and tzparse() APIs with IANA release tzcode2016c.
Date
Msg-id 56FA4AA5.4030205@chrullrich.net
Whole thread Raw
In response to Re: [COMMITTERS] pgsql: Sync tzload() and tzparse() APIs with IANA release tzcode2016c.  (Christian Ullrich <chris@chrullrich.net>)
Responses Re: Re: [COMMITTERS] pgsql: Sync tzload() and tzparse() APIs with IANA release tzcode2016c.
List pgsql-hackers
* Christian Ullrich wrote:

> * Tom Lane wrote:

>> Christian Ullrich <chris@chrullrich.net> writes:

>>> zic aborts somewhere between writing Etc/UTC and UTC.
>>
>> Huh ... I would not have guessed that.  Can you track down exactly
>> where it's failing?
>
> I'd love to, but with 656ee84 I cannot reproduce on my Windows 10
> system. I can try on the animals where it actually failed, but now that
> there's a fix, that won't be necessary, right?

Weird. vcregress check works, install fails. Perhaps the directories are
precreated for check?

Anyway, I think Michael's fix is wrong. The bug is that the Win32
version of link() (at the bottom of zic.c) does not set errno if its
attempt to copy the file fails, so what dolink() puts into link_errno is
bogus.

The additional mkdirs() call just papers over the actual bug; the
existing one in line 802 will do nicely once it actually runs.

Patch attached.

--
Christian


Attachment

pgsql-hackers by date:

Previous
From: Sridhar N Bamandlapally
Date:
Subject: pg_largeobject
Next
From: Artur Zakirov
Date:
Subject: Re: IF (NOT) EXISTS in psql-completion