Version Numbers & Branches

Topics: General Topics, SharpMap Project, SharpMap v0.9 / v1.x, SharpMap v2.0
Nov 30, 2012 at 8:45 PM

Can someone give me a brief summary of the main differences between the versions/branches in the SharpMap project (or a link to this discussion)? It has been a while since I used SharpMap and I am not sure what has changes in the last year or so. I last used version 0.9 and I believe that was more or less represented in the trunk at that time. What does the trunk represent right now? And, what version/branch would you recommend as the most stable and robust lineage for "common" GIS tasks? 

Dec 3, 2012 at 6:39 AM

Hello atv,

The version in trunk will be released as v1.0 ASAP. The v1.0 branch was an earlier attempt to get there, but has been abandoned. I'll delete it.
The version in the trunk has full topology support because of NTS/GeoAPI usage. We refactored the "Web" components that prevented us using ClientProfile to a new SharpMap.Web project.

The only relevant branch is v0.9.5-DeltaShell, that you can consider as a source of inspiration for further work.

Hth FObermaier

Dec 4, 2012 at 2:00 AM

Thanks FObermaier for the information. What about version 2...was it not very successful? I am specifically looking for read & write support for PostGIS right now (create tables, edit geometries etc) and was wondering if any functionality was written for that in V2?

Dec 4, 2012 at 7:01 AM

Yes, atv, version 2 was not very successful indeed. Nonetheless it has code to read and write postgres/postgis tables.

It should be straight forward to create and implement

interface IUpdatableProvider : IProvider
    IUpdateableProvider Create(FeatureDataTable table, string schema, string name);
    void Insert(IList<FeatureDataRow>);
    void Upate(IList<FeatureDataRow>);
    void Delete(IList<FeatureDataRow>);
for v1.0

You can find the code in the v2 trunk. You might want to consider using NetTopologySuite.IO.PostGis project for writing the geometries, as that way you don't need the conversion to and from Well-known-binary.

Hth FObermaier