On Fri, Jun 14, 2024 at 12:25 PM Andres Freund <andres@anarazel.de> wrote:
Hi,
On 2024-06-14 11:11:38 -0400, Andrew Dunstan wrote: > On 2024-06-14 Fr 11:09, Andrew Dunstan wrote: > > Over at [1] Andres expressed enthusiasm for enabling TAP tests to call > > LibPQ directly via FFI, and there was some support from others as well. > > Attached is a very rough POC for just that.There are two perl modules, > > one which wraps libpq (or almost all of it) in perl, and another which > > uses that module to create a session object that can be used to run SQL.
What are your current thoughts about a fallback for this? It seems possible to implement the session module ontop of BackgroundPsql.pm, if necessary. But I suspect we'll eventually get to a point where that gets less and less convenient.
I guess it's a question of how widely available FFI::Platypus is. I know it's available pretty much out of the box on Strawberry Perl and Msys2' ucrt perl. It works fine on my Ubuntu ARM64 instance. On my Mac I had to install it via cpan, but that worked fine. For the moment CYgwin has me beat, but I believe it's possible to make it work - at least the docs suggest it is. Not sure about other platforms.
I agree with you that falling back on BackgroundPsql is not a terribly satisfactory solution.
How much of a dependency is FFI::Platypus, compared to requiring perl headers to be installed? In case FFI::Platypus is a complicted dependency, a small XS wrapper could be an alternative.
Sure we could look at it. I might need to enlist some assistance there :-). Using FFI is really nice because it does so much of the work for you.
