Opened 12 years ago
Last modified 3 years ago
#2103 new enhancement
Ability to load topogeoms using shp2pgsql
Reported by: | robe | Owned by: | strk |
---|---|---|---|
Priority: | medium | Milestone: | PostGIS Fund Me |
Component: | topology | Version: | master |
Keywords: | gsoc, usability | Cc: |
Description
In writing chapters of the book, one exercise is to load the data with shp2pgsql and then a second step to load into a specific topology.
Aside from speed issues, I can't see why we can't just go straight to a topogeom column much like we do with geography is -G is chosen.
This would require another switch for topology name and name of topology. The only tricky part is that for the toTopogeom calls we need the generated layer id, which would require a connection to the database to get unless we have a lookup function of some sort that can lookup the layerid based on topo column table/name.
Change History (11)
comment:1 by , 12 years ago
comment:2 by , 12 years ago
This would make a good google summer of code project. Added to page -- http://trac.osgeo.org/postgis/wiki/GoogleSummerCode2012
comment:3 by , 12 years ago
Also added link to http://wiki.osgeo.org/wiki/Google_Summer_of_Code_2012_Ideas#The_ideas_pages
comment:4 by , 12 years ago
Keywords: | gsoc added |
---|
comment:5 by , 12 years ago
Milestone: | PostGIS 2.1.0 → PostGIS 2.2.0 |
---|
comment:6 by , 10 years ago
Milestone: | PostGIS 2.2.0 → PostGIS Future |
---|
comment:8 by , 3 years ago
We have a lookup function now: https://trac.osgeo.org/postgis/changeset/429bf56aa47e55a13555a15d20ef08a1c14d2aba/git (undocumented though)
comment:9 by , 3 years ago
shp2pgsql switches -t and -T are already taken, ideas on what switch to use for this ?
comment:10 by , 3 years ago
Keywords: | usability added |
---|
I've been actually thinking that AddTopoGeometryColumn could be modified (or overloaded) to also accept an existing column for the purpose of registering it when missing (think CopyTopology). The same function could maybe return an already-registered layer_id too. I like the idea of reducing signatures ...