[postgis-devel] Regarding Raster RFC2 edit (r16452)

[Sandro Santilli]

On Tue, Mar 13, 2018 at 06:47:50AM -0700, Bborie Park wrote:
> Apologies for not being more transparent as I changed the RFC. I was
> writing some unit tests for a separate python package built based
> upon the
> WKB RFC and just could not confirm correct behavior. After checking
> a few
> spots (including the old prototypes of WKT Raster and GDAL), i
> confirmed
> that the enum/index used in the RFC was wrong.

When a specification says something and code does something else, I
tend to think that the code is bogus, not that the specification is.
As this is debatable the correct procedure would be to start a
revision process to come to an agreement about what's wrong and what's
right.

On IRC I've read something about an unused PT_16BF pixel type, so
there seem to be more to discuss about ? Could you please formally
start a thread describing what discrepancy you found and which options
you see to fix it ?

Filing a blocker ticket targetting cleaning of this situation would
probably help too.

--strk;