Re: Meson build updates - Mailing list pgsql-hackers

From Tristan Partin
Subject Re: Meson build updates
Date
Msg-id CTPDBT2W78PQ.2QUDVHFNR5MCS@gonk
Whole thread Raw
In response to Re: Meson build updates  (Andres Freund <andres@anarazel.de>)
Responses Re: Meson build updates
List pgsql-hackers
On Thu Jun 29, 2023 at 2:02 PM CDT, Andres Freund wrote:
> Hi,
>
> On 2023-06-13 14:56:36 -0500, Tristan Partin wrote:
> > I was thinking today. When you initially wrote the build, did you try
> > using the src/bin/meson.build file as the place where all the binaries
> > were built? As you say, most of the src/bin/xxx/meson.build files are
> > extrememly reptitive.
>
> > We had a similar-ish issue in my last project which I solved like:
> >
> > https://github.com/hse-project/hse/blob/master/tools/meson.build#L20-L405
> >
> > This is a pattern I used quite frequently in that project. One benefit
> > of this approach is that the binaries all end up next to each other in
> > the build tree which is eventually how they'll be laid out in the
> > install destination. The other benefit is of course reducing reptitive
> > code.
>
> I think the build directory and the source code directory not matching in
> structure would have made it considerably harder sell for people to migrate.
>
> I.e. I considered it, but due to meson's "no outputs outside of the current
> directory" rule, it didn't (and sadly still doesn't) really seem viable.

Yeah, I guess it is a matter if you like the layout being closer to the
installation or the source tree at the expense of repetition. I am
partial to the installation since it is less to type if you run a binary
from the build directory and less repetition, but all good. Maybe
something that could be reconsidered when autotools is dropped.

I still think the overrides are important, at the very least for libpq,
but I will defer to your aforementioned decision for now.

--
Tristan Partin
Neon (https://neon.tech)



pgsql-hackers by date:

Previous
From: "Tristan Partin"
Date:
Subject: Re: Meson build updates
Next
From: Andrew Dunstan
Date:
Subject: Re: pgsql: Fix search_path to a safe value during maintenance operations.