Latest Contributions by KimOllivier

Sorry the NTv2http://en.wikipedia.org/wiki/NTv2 is National Transform version 2 developed by the Canadian mapping agency to use a grid to adjust for nonlinear distortions that a datum transform requires. It enables efficient coordinate adjustment that is fast and reliable. It has been adopted by many national agencies to upgrade coordinate systems. It was not relevant to your shift problem because you are not changing datums with a simple shift. I described the latest way to handle projection definitions in ArcGIS that was confusing. At 10.x Esri imbedded all the text file definitions (*.prj) files into a binary dictionary. I have forgotten the magic incantation for the earlier version to create a custom projection. At the worst all you need to do is copy a .prj file and edit with a text editor that handles unix line endings (eg wordpad not notepad). The basic idea still holds, create a custom projection with the offset you require as a false origin. ... View more

Am I missing the point? There was no diagram so maybe I am. Why not dissolve on the selected features before buffering? Then the adjacent boundaries will not be buffered. Are the adjacent polygons not touching? In that case there are other tools that can jump the gap in the Cartographic Toolset. eg AggregatePolygons Tool ... View more

Use a custom projection with a false origin of 360 degrees. We have just been doing shifts in NZ after major earthquakes which have all features moving different amounts over a lot of featureclasses. Too many for rubber sheeting to handle. The solution was to create a transform grid using the NTv2 tools and run a pseudo-projection by defining a custom projection. After the projection using transform we redefine the projection back to the original. In your case it is easier because you don't need a transform, just a shift. Get out a copy of the projection (from properties of a featureclass in ArcCatalog) and hit the copy_and_modify button. Add a false origin of 360 degrees for your custom projection. Then reproject to the new projection and afterwards override the new projection by redefining it back to the original projection. Merge the two. ... View more

If the source was a raster to polygons then there are often problems with the shape, size and closing of the polygons because they were really edges. Does your data have the jaggies? It could be that the precision of the coordinates has resulted in a lack of snapping. RepairGeometry will fix small problems, but if you still have a universal polygon you still have a problem. Snapping vertexes is hard to do well, but you can try Perhaps you have an enclosing polygon that we used to call the "Universal Polygon" that is everything that is not forest. This might have a million vertices and be self intersecting or drawn in the wrong direction. You will have to find the offending polygon and delete it before you can clean up. Look for the largest polygon and check the vertexcount. populate a field eg count = shape.pointCount in expression calculator. Dice is a tool to split oversize polygons. Having a very large polygon with lots of 'donut holes' will cause problems. Split the larger polygons so that they do not enclose other polygons. shape.partCount will expose those. Even though repair geometry should fix problems, try finding polygons with very small areas and delete those manually. Just sort the table by area. If the area is zero or null you should delete them immediately. ... View more

Describe will dynamically recalculate the extent of all the features in a featureclass after your changes. You cannot reset it, it is a property of the extent of all the features used for creating spatial indexing. Here is a test using the Python window in ArcMap arcpy.env.workspace = r'E:\project\forumhelp' >>> print arcpy.Describe("sourcearea").extent 2626051.8 185544.48 2626576.5 186112.94 NaN NaN NaN NaN >>> # moved a vertex to the left, saved >>> print arcpy.Describe("sourcearea").extent 2625867.4387207 185544.480102539 2626576.50012207 186112.940124512 NaN NaN NaN NaN I do not know what you are 'recalculating' in ArcMap. I cannot find a property in the layer dialog. Do you mean that you are setting the extent of the data frame for the full extent command to override the default extent? Once you have the extent object from Describe you can program the dataframe extent using the Mapping module and the DataFrame object. Or you could modify the extent object first to say add 10% before using it. ... View more

Your data is fine. It all opens for me with ArcGIS 10.2 without conversion. I have found to my surprise that ArcMap can now read E00 files natively without translation. The 'coverages' are represented as a dataset containing simple point lines and polygons. Just use the AddData button. Similarly the tables in the Access database will load as tables but without any geometry. Not all the geometry layers have a projection defined but they all overlay anyway because they are in the same coordinate system, including the TIF image. You could export the layers to a current format eg shape file or featureclass in a new geodatabase and define the projection. To make the SOURCEAREA fill you need to run FeatureToPolygon Tool to rebuild polygons from the arc layer. The WELLS point layer has a key LOC_ID which can be used to join the access tables for labelling, symbolising and analysis. ... View more

If you run the tool from the toolbox a dialog will allow you to fill in the parameters and validate them before you can press OK. An invalid parameter suggests you are missing a quote or something in the command. If it runs successfully, you can copy the command from the results as a Python snippet to be pasted into a script. You should be able to read the E00 file in a text editor. It will look strange, but is a text file. The format allowed for splitting the file so you could have E00, E01, E02.... files. (one on each 1.44MB floppy disk) Are you sure you have all the files? It should have and EOF at the end. E00 files have been obsolete since 1999, so it could be a third party product creating an invalid file. If it is small you could attach it for me to look at. The name has changed because ARC/INFO is no longer installed with ArcGIS at version 10 for the original tool to use. ... View more

Use the Describe object to find the extent of the dataset. A single featureclass can be its own dataset. print arcpy.Describe('seacoast').extent 1089354.4464 4747978.9305 2092003.9868 6223163.9395 NaN NaN NaN NaN ... View more

The name of the tool seems to have changed to arcpy.ImportFromE00_conversion() import arcpy arcpy.env.workspace = "C:/data" arcpy.ImportFromE00_conversion("citylim.e00", "C:/output", "citylim") ... View more

I've been working on getting our parcel locator to return close matches on the same street (eg. input 100 main street, mytown which doesn't exist and have 101 main street, mytown returned instead of 100 main street, someothertown) and ran across some very strange behaviour.  The behaviour is 100% repeatable and I have found the pattern.  I presume you are editing the XML configuration file with a different wildcard pattern. Be careful what you ask for! Perhaps post the mod you have made to the number pattern? Numbers are really strings as far as pattern matching goes so it will not find adjacent numbers eg from 99 to 101. I have noticed similar approximate matching from odd to even numbers which is very annoying because it matches across the road instead of the neighbour. ... View more

At 10.2.1 Esri include two Python modules XLRD and XLWT which provide an API to read and write Excel spreadsheets, so it is unlikely that the ExcelToTable will be fixed because we can do it ourselves. What should have happened is that the developer used these tools to read the structure and metadata in the Excel file and present a dialog to choose the sheets, named ranges and also a fieldmapping to alter the terrible defaults of 255 characters and floating point numbers for everything. ... View more

I despaired trying to get reports to work, the output was too complex. So I switched to exporting to a spreadsheet using the Excel COM interface and a template. Another radical alternative might be to export directly to a PDF using reportlab. Reportlab has an easy module to pour data into tables and graphs. Or maybe use matplotlib? ... View more

Then when I chose the rematch bottom on the geocoding toolbar, I get an error that says "There was an error reading the geocoding information from this dataset". So it looks like the 'rematch' button silently fails. At least when you use the geocoding toolbar you get some feedback. ...But what does it mean? I cannot find any error message like that in the Help. I have seen it myself and it does not have an error number either. Also what does accessing the original locator do? Because I do have access to it. You cannot rematch if you cannot see the locator. But if you have just used it, then it must be available. Maybe post the field schema you are geocoding, or say the first few records? ... View more

Maybe don't use a spreadsheet as a data source because all the text fields will be 255 character wide and all numbers floating point? Use Microsoft schema.ini to load from a csv dump of the sheet or use FME to load with valid field names and normal widths. Many people have trouble with the default loading of spreadsheets for any geoprocessing, not just locating. Putting the source on a local drive is a good idea, but if it is a spreadsheet it's still a bad format to use in data processing. ... View more