1

I am trying to import a CSV into a DBF using ArcCatalog (10.1). In the field map, pictured below, I can only set an ID field to Long, Text, Double, etc. But there is no Numeric type as in DBF and shapefile attributes.

If I import the ID field as a Long, values will be truncated. If I import it as Text, it will be different type than Numeric. In either case, I won't be able to join the imported DBF to a shapefile with the ID.

How should I choose the type so that the ID in the imported DBF can be joined with a 'Numeric' ID in shapefiles?

enter image description here

PolyGeo
  • 65,136
  • 29
  • 109
  • 338
tinlyx
  • 11,057
  • 18
  • 71
  • 119
  • Can you edit your question to include an example of an ID field that gets truncated as Long, please? When I look at a shapefile I see IDs that have a Number Format of Numeric but their Data Type is Long so I am struggling to understand the precise steps you have taken to reach a problem with this. A more detailed description would be best to start with the first row of your CSV (just a couple of fields). – PolyGeo Nov 09 '14 at 08:36
  • Also, if you have not already reviewed Preparing CSV files for use in ArcGIS for Desktop?, I think that would be wise to do. – PolyGeo Nov 09 '14 at 08:44
  • @PolyGeo Thanks for the pointers. I looked again, and actually the data wasn't truncated. It's just that some values are bigger (1477701) and others are smaller (203071). I thought I had a truncation issue when I saw the small values in DBF and the large values from the shapefile they are supposed to match. – tinlyx Nov 09 '14 at 08:44
  • @PolyGeo Also, I ended up importing the field as Double and it just showed in DBF as Numeric. I will delete my question after you see this. – tinlyx Nov 09 '14 at 08:46
  • Rather than delete your question I would recommend that you write a short (about a paragraph) answer to summarize the solution. Self-deletion of questions can be done but they can be viewed negatively by some of the site's algorithms. – PolyGeo Nov 09 '14 at 08:53

1 Answers1

1

I took a closer look and actually the data wasn't truncated. It's just that some values are bigger (1477701) and others are smaller (203071). I thought I had a truncation issue when I saw the smaller values in the first page of the DBF and the large values from a segment of the attribute table of the shapefile they are supposed to match.

On a different note regarding the "No numeric field" issue, I know that DBFs save numerical values in decimal text forms of the Numeric type, and I got worried when I didn't see this option in ArcCatalog table-to-table importer's field types (i.e. when right click on CSV and export). What I did eventually is to import the CSV field as a Double as in the OP picture, and the field ended up in DBF as a Numeric field in ArcGIS. And this imported Numeric field matches OK with the Numeric field in the target shapefile it is supposed to match.

I suppose this is a good solution as importing integers as Double didn't introduce any imprecision like 3.000000001 and the imported values in DBF compare well with the values in shapefiles.

tinlyx
  • 11,057
  • 18
  • 71
  • 119