Latest Contributions by JoshuaBixby

Regarding Google, it is interesting/odd that ArcGIS 10.1 seems to be the higher-ranked search results.  With ArcGIS 10.1 and 10.2, you could just manually change the URL, which was easy enough.  With 10.3, not only did Esri change the subdomain for the online help, it changed the name from "Help" to "Documentation."  The fact it was done with no notice to customers was a bit annoying. Here is the ArcGIS 10.3 Desktop overview of the Conflation toolset.  If you dive into the links on that page, there is some good information on edge matching, feature matching, rubber sheeting, etc....  I would say start by reading the documentation and work through the examples.  Beyond that, I don't see any offerings in the Esri Training Catalog that are specific to conflation. ... View more

The Conflation toolset was introduced in ArcGIS 10.2.1.  There really hasn't been any changes or updates to it since then, as the lack of references to any such changes or updates in the documentation implies. A big part of why you have gotten no response is likely the appearance you have done nothing yourself to try and answer the question first.  Maybe the What's New lists were checked as well as the Help files; if so, none of that came across in the original question. ... View more

ArcPy versions are really tied to the underlying ArcGIS for Desktop, Engine, Server, etc... application that is installed.  The GetInstallInfo (arcpy) function will let you know which application the currently loaded arcpy package is pointing to along with a bunch of other information, including version. >>> import arcpy >>> arcpy.GetInstallInfo()['Version'] u'10.3' Looking at the ArcGIS 10.3 Issues Addressed List, nothing jumps out at me for CalclateField_management or ArcPy that would seem related to this issue.  That said, sometimes bugs get addressed indirectly as the result of a different bug being addressed directly, i.e., not every issue that gets addressed with a new release ends up in the Issues Addressed List. Looking at your StackExchange post, it appears you are using Enthought Canopy for Python package deployment.  I am also wondering if it might be a configuration issue tied to how iPython is being pushed your machine with ArcGIS installed. ... View more

What version of iPython are you using?  I am running iPython 3.0.0 with ArcGIS 10.3, and I am able to use arcpy.CalculateField_management without any errors.  There were 11 flush-related issues closed in iPython 3.0.0, maybe this was related to one of them. ... View more

If you have the C:\Python27\ArcGISx6410.2 directory on your machine, then you installed 64-bit Background Geoprocessing (not part of the standard installation) or ArcGIS for Server.  Assuming the latter isn't the case, the former is what put the extra Python interpreter installation on your machine. With regard to the OP's situation, 64-bit Background Geooprocessing wasn't introduced until ArcGIS 10.1 SP1.  That said, the information provided is useful for others that install 64-bit Background Geoprocessing and get confused over which Python interpreter they are using when. ... View more

FeatureClassToNumpyArray, nah, that would be too easy.   I got tunnel vision on showing the mechanics of building the numpy array, but FeatureClassToNumPyArray is the more ArcPy approach. ... View more

This question was originally cross posted to StackExchange, where it received feedback, some of which was marked best answer. Geographic Information Systems Stack Exchange >  How to get a search cursor on a particular version in arcpy ... View more

Thanks for uploading data, it helps. If I use your data as-is, and choose Intersect_3D instead of Intersect, the polygon in question gets selected for a total of 1358.  If I project your data from WGS84 to a North American continental projection like Lamberts Conformal Conic, I get the same result. Since using the 3D intersect option and projecting your data explicitly gets the same, expected result; I can only surmise something is going on with default values or tolerances when using geographic coordinates and a 2D intersect.  Maybe the projection-on-the-fly is having an issue, I am not sure. ... View more

If I understand you correctly, it is one polygon on or near the periphery that is not being handled the way you think.  A few questions. Is the missing polygon a single-part or multi-part polygon? You appear to be working at continental map scale, i.e., 1:60,000,000.  What spatial reference are you using? Which part specifically isn't working?  Does the arcpy.SelectLayerByLocation_management tool miss the polygon or does that step work and it is the search cursor that is missing it? Regarding cursors, you should really consider using the newer ArcPy Data Access (arcpy.da) cursors instead of the older/original ArcPy (arcpy) cursors. ... View more

Regarding spatial reference, yes, it is always a good idea to specify a spatial reference when creating ArcPy geometry objects.  There are numerous, known precision issues that crop up with ArcPy geometry objects when spatial references aren't specified.  Is lack of a spatial reference the issue here?  We will know soon enough when you test it. When you are talking about a "shared point with the extent rectangle," I am a bit uncertain whether that is what you mean.  "Shared point" could be taken to mean something very specific, like a polygon in question must share a vertex with the extent rectangle.  Try adding the spatial reference, and then report back.  Also, are the differences you are finding only along the periphery of the extent object?  Are any polygons more in the interior being handled different. There are 15+ different types of overlaps one can specify.  I picked Intersect in this case because it is the default and most common.  It could be the CopyFeatures_management tool uses a different overlap type, or it could be a bug in one tool or the other. ... View more

arcpy.env.workspace is not a Python special method, it is an ArcPy environment setting exposed as a property on ArcPy's env class.  When ArcPy is instantiated, the environment settings get defined and stored as a set in one of the environment object's internal or private attributes, _environments to be specific.  You can use the Python dir() built-in function to see those environment settings, but you also see lots of other class attributes as well.  The following code snippet shows only the environment settings: import arcpy for env in arcpy.env._environments:     print env The interactive console is retrieving those environment settings and making them available for code completion to the user.  The set of environment settings doesn't exist until an instance of ArcPy is created, so a user doesn't see them for code completion in the editor window. I can't say I am doing complete justice with this explanation, but I find it hard to get much more in depth without getting into the weeds of Python class structure and how properties are defined and managed. I think Esri would argue one should use the online documentation to understand ArcPy and not rely on code completion in a code editor or interactive window.  Is Esri's implementation of the ArcPy Env class frustrating because of how it impacts code completion in editor windows, I think so; but I also think this factor is low on the list when considering how to structure ArcPy. ... View more