Geographic Information Systems Stack Exchange is a question and answer site for cartographers, geographers and GIS professionals. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I'm searching how the output of one program (text file with some writings, but also with data as numbers, dates, times, coordinates, all separated with commas) can be transformed to a Feature Dataset.

1) .txt -> .csv -> gp.TabletodBase and then it's nearly done.. But get problems writing csv with Excel dialect - all row is writen in one cell. So, I create the delimiter ';' to get the data into cell by cell.

f= open(rawtable, 'r')
f2 = open(outtable, 'wb')
csv.register_dialect('dotcoma', delimiter=';')
writer=csv.writer(f2, dialect='dotcoma')

for line in itertools.islice(f, 6, 7): #reading labels that are in 6'th row
    labels= line.strip('\n').split(",")

But well, then Arcgis does not recognize the table.. I have been trying to cast the data, like integers, floats, strings.. but that did not help..

2) .txt -> .shp... ----> THAT WORKS!!!

The shapefile does not have a type of time.. So was skiping it.

EDITED However tried the Pyshp library. But after creating fields blindly I record the tuple of all data in a row and I'm getting an error: IndexError: tuple index out of range..

I simply write fields from the list of data, depending to which group they go:

for fieldname in labelkeys:
    if fieldname in t_int:
        w.field(fieldname, 'N', 8, 0) #1probl was here as I made '8' vs. 8 ! stpd
        w.field (fieldname,'C', 25)

Another problem with coordinates. I was writing floats... and it did not like it.. So I had to write integers. Plus one cannot pass a tuple to w.record(). Thanks to StackExchange I've learned the benefits of asterix :) and one simply has to write w.record(*mylist).

3) .txt -> feature class... Here I have problems writing the entire row at once..(like the list of values) into a NewRow. Since field names are made automatically I can not for each make like row.fieldname=somevariable. I have tried while looping to SetValues row.setValue(labelkeys[unit], data[unit]) and then at the end of a list rows.InsertRow(row) , but it seemed that it sets just one value. Though I tried by hand and got an error in data casting..

The error comes that I read data like string '34', cast it to int(34) and try to write into field with type 'DOUBLE' (since I grouped some data and wanted to make more generic types).

if labelkeys[unit] in t_int:
    if data[unit]:
        row.setValue(labelkeys[unit], int(data[unit]))

So for now.. nothing is really working..

Maybe there is some other way I have not though about?

If not, could anyone help to get the things work?...

share|improve this question
Similar to… – Chad Cooper Oct 16 '12 at 16:07
Also, I think you mean featureclass and not feature dataset. – Chad Cooper Oct 16 '12 at 16:08
@ChadCooper indeed! – najuste Oct 16 '12 at 19:54
up vote 2 down vote accepted

In version 9.3 there is a toolbox called Create Features from Text File, which can do what you require. The help files are quite detailed and are found here.

The following code sample should be simple enough to adapt to reflect your dataset:

# Create geoprocessing dispatch object
import arcgisscripting
gp = arcgisscripting.create()

# Set up inputs to tool
inTxt = r"C:\temp\StreamPoints.txt"
inSep = "." #the symbol used to indicate a decimal point
output = r"C:\temp\Streams.shp"

# Run tool
gp.CreateFeaturesFromTextFile(inTxt, inSep, output, "#")


If you require numerous attributes of a range of types to be written to a feature dataset from a text file or csv file, it my be easiest to use pyshp. I wrote up an answer to a similar question about pyshp a few days ago and will reproduce the relevant parts of this answer below.

The pyshp/xbase datatyping has always been tricky for me until I found this user guide for the xbase format and as a result of this question I have written a small note on my blog regarding the relevant pyshp datatypes, part of which I have pasted below:

  • C is ASCII characters
  • N is a double precision integer limited to around 18 characters in length
  • D is for dates in the YYYYMMDD format, with no spaces or hyphens between the sections.
  • F is for floating point numbers with the same length limits as N
  • L is for logical data which is stored in the shapefile's attribute table as a short integer as a 1 (true) or a 0 (false). The values it can receive are 1, 0, y, n, Y, N, T, F or the python builtins True and False

There is a DateTime type as well which should allow you to store correctly formatted time data as well using the same syntax as the above datatypes.

Here is an example python script showing the use of the various pyshp datatypes being read in from lists, which could be populated with your data using the csv module:

import shapefile as shp

#Input parameters
out_file = 'GPS_Pts.shp'

#Set up blank lists for data
date,notes = ['20121012','20121012','20121011'],['This','is a','string']
logic = [True,False,True]

#Set up shapefile writer and create empty fields
w = shp.Writer(shp.POINT)
w.autoBalance = 1 #ensures gemoetry and attributes match
w.field('Y','F',10,5) #float - needed for coordinates
w.field('Date','D') #date
w.field('Notes','C',50) #string
w.field('ID_N','N',12) #double

#loop through the data and write the shapefile
for j,k in enumerate(x):
    w.point(k,y[j]) #write the geometry
    w.record(k,y[j],date[j], notes[j], id_no[j], logic[j]) #write the attributes

#Save shapefile
share|improve this answer
This tool is used just to create a Features using the coordinates. What I am searching first of all is creating a table with a lot of features.. especially DATE, TIME and so on – najuste Oct 16 '12 at 9:10
Ah. My apologies, I read the question too fast. I'll update my answer. – sgrieve Oct 16 '12 at 10:27
Wow, that's a loooong explanation. Thanks a lot for such explicity, will take time to read and test things :) Just a small rremark about DateTime type. I need Time as well and so .shp does not have it.. and so pyshp I guess it's just for .shp. That stays a minus, but just one, compared with so many I have for now.. – najuste Oct 16 '12 at 10:51
Yeah, I just did some more reading and you're right, a third party driver is needed to support datetime in xbase files which the shapefiles cannot use. The date portion does work however, so it may be possible to parse out times and dates into 2 fields, a date field and a double precision integer and use unix time ( to record a precise time which can be converted back to a human readable format within the script. All the best. – sgrieve Oct 16 '12 at 11:07
Unfortnatelly, getting an error while wirting a record: 'IndexError: tuple index out of range'. Since I do not know exact field names, just have them in a list, I create the list of casted attributes from the line I read from txt, and simply make it tuple in order to write the whole line at once. Is it possible without saving shp (as I can't that either) to check how many fields were created? – najuste Oct 17 '12 at 14:09

Have you tested MakeXYEventLayer? I believe it will create a Feature Layer from a variety of formats, csv, txt, xls, etc. You can then run Copy_features to save the features to Shapefile or file geodatabase. If the field data types aren't interpreted correctly, you can create a blank feature class with data types you require. You can add or remove fields too. Finally, run Append to push your data into the final schema. Identical Field names are automatically mapped by the field mapper.

share|improve this answer
First, thought that MakeXYEventLayer is available just manually! Will try a bit later, since it worked out with Pyshp – najuste Oct 18 '12 at 12:30

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.