[postgis-devel] Enabling wagyu by default

Greg Troxel gdt at lexort.com
Fri May 31 05:01:25 PDT 2019


Raúl Marín Rodríguez <rmrodriguez at carto.com> writes:

>> I'm more concerned about the logistics of defaulting it.  If it's
>> like it uses it if it finds it available in system like other things,
>> I'm okay with that but if it then increases the compilation or
>> downloads additional etc, I'm a bit uneasy about that.
>
> It's included in the tarball, so no downloads required. In my machine, building
> the wagyu code without cache takes ~8 seconds (blame C++ templates).

As a packager, my usual plea is that the build should be determined by
the options passed, and fail if that doesn't work, rather than
detect/adapt, to the extent reasonable.

This means that if libfoo is optional, then the user can pass
--enable-foo to use it (and fail the build if not present) or
--disable-foo (to ignore it even if present).  That requires a bit more
thought, but I think the reduction in un-understood side effects is
worth it.

Here, I think you are talking about changing a default (which is not
objectionable), and there won't be any kind of different outcome based
on what is installed.

>> Seems like the risks of using Wagyu are (if it is not maintained by the original developers):
>
>> - compilation problems going forward (since PostGIS is vendoring it)
>> - functional limitations as MVT requirements evolve
>> - finding issues which are showstoppers
>
> The one I most concerced about is the first one; that's why I want to make
> it a default now so there is enough time to fix any build issue.

Presumably this does not change the required version of C++, and would
not cause postgis to fail to build anyplace it builds now.  Or I am
guessing that is a hypothesis and it is of course not really known.

> As I don't see any big arguments against it, I'll try enabling it as soon
> as I can and see how bots behave, ideally before the alpha2 release.

I would say that if the default is going to change, there definitely
needs to be an alpha with it, with two weeks of alpha, before a beta,
and once a beta is out this sort of change is not ok.  (I do not mean to
object now; I don't understand the details well enough to have an
opinion - I am merely pointing out what beta means :-), and cautioning
because a number of projects have an unfortunate history of big changes
just before a release.)


More information about the postgis-devel mailing list