From: | Andrew Dunstan <andrew(at)dunslane(dot)net> |
---|---|
To: | Andres Freund <andres(at)anarazel(dot)de> |
Cc: | pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: [RFC] building postgres with meson |
Date: | 2022-02-10 17:00:16 |
Message-ID: | f7c473d7-2484-ca9d-7333-38bbbf7d885d@dunslane.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On 2/6/22 15:57, Andrew Dunstan wrote:
> On 2/6/22 13:39, Andres Freund wrote:
>> Hi,
>>
>> On 2022-02-06 12:06:41 -0500, Andrew Dunstan wrote:
>>> Here's a patch. I've tested the perl piece on master and it works fine.
>>> It applies cleanly down to 9.4, which is before we got transform modules
>>> (9.5) which fail if we just omit doing this platform-specific piece.
>> Given https://postgr.es/m/34e972bc-6e75-0754-9e6d-cde2518773a1%40dunslane.net
>> wouldn't it make sense to simply remove the pexports/gendef logic instead of
>> moving to gendef?
>>
> I haven't found a way to fix the transform builds if we do that. So
> let's leave that as a separate exercise unless you have a solution for
> that - this patch is really trivial.
>
>
Any objection to my moving ahead with this? My current workaround is this:
cat > /usr/bin/pexports <<EOF
#!/bin/sh
/ucrt64/bin/gendef - "$@"
EOF
chmod +x /usr/bin/pexports
(gendef is available in the ucrt64/mingw-w64-ucrt-x86_64-tools-git
package on msys2)
cheers
andrew
--
Andrew Dunstan
EDB: https://www.enterprisedb.com
From | Date | Subject | |
---|---|---|---|
Next Message | Andrey Borodin | 2022-02-10 17:33:56 | Re: Race condition in TransactionIdIsInProgress |
Previous Message | Swaha Miller | 2022-02-10 16:53:15 | Re: support for CREATE MODULE |