GeometryEngine.Project() Arguments

JensBuchta

What do you think about this API method?

<SPAN class="keyword token">public</SPAN> <SPAN class="keyword token">static</SPAN> Geometry GeometryEngine<SPAN class="punctuation token">.</SPAN><SPAN class="token function">Project</SPAN><SPAN class="punctuation token">(</SPAN>Geometry geometry<SPAN class="punctuation token">,</SPAN> SpatialReference spatialReference<SPAN class="punctuation token">)</SPAN><SPAN class="line-numbers-rows"><SPAN>‍</SPAN></SPAN>

In my opinion, this method is to generic and involves to many casts when it's used, e.g.

<SPAN class="keyword token">var</SPAN> mapPoint <SPAN class="operator token">=</SPAN> <SPAN class="keyword token">new</SPAN> <SPAN class="token class-name">MapPoint</SPAN><SPAN class="punctuation token">(</SPAN><SPAN class="punctuation token">.</SPAN><SPAN class="punctuation token">.</SPAN><SPAN class="punctuation token">.</SPAN><SPAN class="punctuation token">)</SPAN>
<SPAN class="keyword token">var</SPAN> projectedPoint <SPAN class="operator token">=</SPAN> <SPAN class="punctuation token">(</SPAN>MapPoint<SPAN class="punctuation token">)</SPAN>GeometryEngine<SPAN class="punctuation token">.</SPAN><SPAN class="token function">Project</SPAN><SPAN class="punctuation token">(</SPAN>mapPoint<SPAN class="punctuation token">,</SPAN> anySpatialReference<SPAN class="punctuation token">)</SPAN><SPAN class="punctuation token">;</SPAN><SPAN class="line-numbers-rows"><SPAN>‍</SPAN><SPAN>‍</SPAN></SPAN>

If we had a finite list of supported overloads, like:

<SPAN class="keyword token">public</SPAN> <SPAN class="keyword token">static</SPAN> MapPoint GeometryEngine<SPAN class="punctuation token">.</SPAN><SPAN class="token function">Project</SPAN><SPAN class="punctuation token">(</SPAN>MapPoint mapPoint<SPAN class="punctuation token">,</SPAN> SpatialReference sr<SPAN class="punctuation token">)</SPAN> <SPAN class="punctuation token">{</SPAN><SPAN class="punctuation token">}</SPAN>
<SPAN class="keyword token">public</SPAN> <SPAN class="keyword token">static</SPAN> LineSegment GeometryEngine<SPAN class="punctuation token">.</SPAN><SPAN class="token function">Project</SPAN><SPAN class="punctuation token">(</SPAN>LineSegment lineSegment<SPAN class="punctuation token">,</SPAN> SpatialReference sr<SPAN class="punctuation token">)</SPAN> <SPAN class="punctuation token">{</SPAN><SPAN class="punctuation token">}</SPAN>
<SPAN class="keyword token">public</SPAN> <SPAN class="keyword token">static</SPAN> Polygon GeometryEngine<SPAN class="punctuation token">.</SPAN><SPAN class="token function">Project</SPAN><SPAN class="punctuation token">(</SPAN>Polygon polygon<SPAN class="punctuation token">,</SPAN> SpatialReference sr<SPAN class="punctuation token">)</SPAN> <SPAN class="punctuation token">{</SPAN><SPAN class="punctuation token">}</SPAN>
<SPAN class="punctuation token">.</SPAN><SPAN class="punctuation token">.</SPAN><SPAN class="punctuation token">.</SPAN><SPAN class="line-numbers-rows"><SPAN>‍</SPAN><SPAN>‍</SPAN><SPAN>‍</SPAN><SPAN>‍</SPAN></SPAN>

...use of the API would be easier and more clear.

What do you think about this?

SomeGISGuy

Thank you for the feedback. However there is actually no guarantee that projecting a geometry of a certain type will return that same type. There are cases where a polygon can collapse to a line or a point, or a line to a point, and Envelopes will almost always return as Polygons.

Projecting points will however always return a point or null, so I could see an argument for adding a point-specific overload.