#151 closed defect (fixed)
[wktraster] rt_ prefixes brake seemless integration
Reported by: | pracine | Owned by: | pracine |
---|---|---|---|
Priority: | medium | Milestone: | WKTRaster 0.1.6 |
Component: | raster | Version: | master |
Keywords: | Cc: | mloskot |
Description
We must rename every function with the RT_ prefix with a ST_ prefix to ensure seemless integration with PostGIS geometry functions.
Change History (15)
comment:1 by , 16 years ago
Milestone: | 1.4.1 |
---|
comment:2 by , 16 years ago
Milestone: | → wktraster 0.1.6 |
---|---|
Owner: | changed from | to
Reporter: | changed from | to
Summary: | rt_ prefix brake WKT Raster seemless integration → rt_ prefixes brake WKT Raster seemless integration |
comment:3 by , 16 years ago
Status: | new → assigned |
---|
comment:4 by , 16 years ago
Resolution: | → fixed |
---|---|
Status: | assigned → closed |
comment:5 by , 16 years ago
Resolution: | fixed |
---|---|
Status: | closed → reopened |
Is it intentional that rt_lib_version() function has been renamed to st_lib_version() ? I believe this name is unclear and ambiguous, so I've reopened the ticket.
comment:6 by , 16 years ago
Would postgis_raster_lib_version() & postgis_raster_lib_build_date() be better?
PostGIS names his own ones postgis_lib_version() & postgis_lib_build_date()
comment:8 by , 16 years ago
Resolution: | → fixed |
---|---|
Status: | reopened → closed |
comment:9 by , 16 years ago
Fixed by Pierre in r4146
I'd suggest to take care of two details while resolving tickets:
- Include ticket number in log message:
svn ci -m "Fixed bug ... (#151)
what tells Trac to replace #151 with link to ticket 151 wherever log message is displayed.
- Include revision number of resolving changeset in ticket's comment, while closing a ticket. Example of comment:
Fixed in trunk (r4146)
what tells Trac to replace r4146 with link to changeset.
It saves a lot of time to find out what has changed, who fixed what and how it's been fixed.
comment:11 by , 16 years ago
Guys I just realized a possibly small problem with using ST_ prefix. While writing book and complaining about why relying on autocasts is bad and you should explictly use ST_GeomFromText or CAST or what have you. It occurred to me that the downside of using ST_ for Raster, is that you can never declare a Text auto cast for it.
Auto casts I think are bad anyway. The point though —
In Postgis — you can do this ST_Intersects('POINT(….)', 'LINESTRING(…)');
and PostgreSQL will happily cast both to geometry because there is an auto cast in place.
However if you introduced an auto cast that takes WKT Raster text representation —> RASTER — it will break everything.
So I guess the point is if you plan to create CASTS that convert text to RASTER — make sure they are not declared automatic — otherwise it would cause an ambiguous function name error thingy.
follow-up: 13 comment:12 by , 16 years ago
Thanks for the warning. I'm happy it is not a too big problem. We won't implement automatic CAST from text…
comment:14 by , 15 years ago
Summary: | rt_ prefixes brake WKT Raster seemless integration → [wktraster] rt_ prefixes brake seemless integration |
---|
comment:15 by , 15 years ago
Version: | → trunk |
---|
Pierre,
Are you going to update function names in rtpostgis.sql or you want me to do it?