Re: Building with MinGW - Mailing list pgsql-hackers

From Andrew Dunstan
Subject Re: Building with MinGW
Date
Msg-id 56086A90.4050605@dunslane.net
Whole thread Raw
In response to Building with MinGW  (Jeff Janes <jeff.janes@gmail.com>)
List pgsql-hackers

On 09/27/2015 02:11 PM, Jeff Janes wrote:
> Has anyone had success following the instructions at 
> https://wiki.postgresql.org/wiki/Building_With_MinGW#Installing_Git 
> recently?
>
> I've followed the instructions to set up the build environment on a 
> Windows box, and I can't build from git.  I can from the nightly 
> tarball.  So I think the most recent environment you get when 
> following these instructions has suffered some regression in the 
> bison, yak, or something else involved in building from git but not 
> from tarball.


The Msys DTK contains a bison and flex that are adequate for our 
purposes. Are you sure you installed the developer tools?

I just followed this recipe. However, I had to do the following 
additional small steps on a fresh Windows Server 2012 instances on Amazon:
 * the git installer had to be run as Administrator or it got   permissions violations * I had to add this to the end
ofthe MSys PATH: ":/c/prog/git/cmd" (I   installed it in c:/prog/git) * I had to add "C:/mingw /mingw" to the fstab
 


Given that, I was able to clone the code, build it and run the 
regression set without incident. This was using the gcc that comes with 
the mingw installer, not the separate one from mingw-w64.


>
> Also, is there anyway to know the exact git commit which the nightly 
> tarballs are build against?
>
>
>

Not to my knowledge. Perhaps the tarball generator should create a file 
reporting the head commit on the branch.

cheers

andrew




pgsql-hackers by date:

Previous
From: Thomas Munro
Date:
Subject: Re: BRIN indexes for MAX, MIN, ORDER BY?
Next
From: Tom Lane
Date:
Subject: Re: BRIN indexes for MAX, MIN, ORDER BY?