Latest Contributions by BlakeTerhune

SRID 4152 was the ticket! However, it only works in 10.2.2 SDE. When I run the same query in 10.0 it errors because it can't find the SRID in the spatial references table. Follow-up questions: Any idea on the best way to update/refresh the spatial references table in SDE so it has the right SRIDs? Do you know if there is a way to calculate latitude and longitude as separate fields instead of the CLOB returned from ST_AsText? ... View more

I just checked in our development instance of SDE that we upgraded to 10.2.2 and the SDE.ST_Spatial_References table is different. It actually has SRIDs that match the expected numbers like 4269 and 4326, but it also has the other "off" numbers like 0 and 48. SRID 6152 is not in there at all. I tried running the same query on the Dev 10.2.2 SDE database and I got it to work but the coordinates still seem wrong (there's no negative). ... View more

Both 6152 and 4269 still give the same error: SRID does not exist. The SRID column in the ST_Spatial_References table go from 0 to 56 and many of the values in SR_NAME are the same. Of the 56 records, there are only 8 distinct: GCS_WGS_1984 NAD_1927_StatePlane_Arizona_Central_FIPS_0202 NAD_1983_HARN_StatePlane_Arizona_Central_FIPS_0202 NAD_1983_HARN_StatePlane_Arizona_Central_FIPS_0202_Feet_Intl NAD_1983_HARN_StatePlane_Arizona_Central_FIPS_0202_IntlFeet NAD_1983_HARN_UTM_Zone_12N NAD_1983_UTM_Zone_12N Unknown_CoordSys Maybe our ST_Spatial_References table is incomplete? ... View more

As always, thanks for the reply, Joshua. I get what you're saying, but I get this error when I try to use that number: ORA-20005: srid 4326 does not exist in st_spatial_references table. ORA-06512: at "SDE.ST_GEOMETRY_OPERATORS", line 2556 When I look at the SDE.ST_Spatial_References table, the SRID column has smaller numbers and there are four records that have WGS1984 in the name. The CS_ID field has the 4326 number, but it doesn't appear that ST_Transform is looking at that field. ... View more

I think Subtypes are the best way to go here. I have used them once in the past and it is a little tricky to wrap your head around at first. Could you describe your data and the rules the domains must follow? ... View more

Well, I figured out the (HUGE CLOB) issue, but I still can't get it to convert from UTM to Latitude, Longitude. I get an error saying ORA-20603: Spatial References are not compatible. ORA-06512: at "SDE.ST_GEOMETRY_SHAPELIB_PKG", line 788 ORA-06512: at "SDE.ST_GEOMETRY_OPERATORS", line 2577 Here's the SQL SELECT EID, SHAPE, ST_X(SHAPE) as X, ST_Y(SHAPE) as Y,     sde.st_astext(sde.st_transform(SHAPE, 0)) AS TEXTCOORD FROM LIS.ADDRESS_PNT I was just testing the ST_X and ST_Y to see if any ST_Geometry operations would work (and they did). ... View more

I'm on 10.2.2 and just tested this with a new Personal GDB and did not have the same issue. The Rnd field was not automatically populated. ... View more

Oh, ok, now I see what you're doing with AddField. The version of the update cursor you are using is actually an old method and is being replaced by the one using the data access module (arcpy.da). It might be worth transitioning to arcpy.da.UpdateCursor to see if maybe that solves your issue. ... View more

There's some good information in this thread that might help you. To simply things, You could probably just look at using Calculate Field instead of an update cursor. Also, take another look at how you're adding your fields; Your syntax looks strange and I don't see where you specify the field type. Finally, since you are joining the dirpath and filename a lot, might be easier to just make a variable at the beginning for your full file path. I haven't tested this, but here's some code that hopefully portrays some of my ideas. I also created a dictionary (addFields) where the key is the new field name and the value is the possible search fields. addFields = {     "New_Name": ["ENG_NAME", "Name"],     "Another_New_Field": ["PossibleSearchField1", "AnotherPsblSrchFld", "MaybeAnotherPSF"],     "New_Fld": ["SimilarName", "Other_Existing_Fld"] } filepath = os.path.join(dirpath, filename) fieldsExisting = [f.name for f in arcpy.ListFields(filepath)] for addField, psf in addFields.items():     if addField not in fieldsExisting:         # Adjust AddField parameters as needed         arcpy.AddField_management(             filepath,  ## in_table             addField,  ## field_name             "TEXT",  ## field_type             "",  ## field_precision             "",  ## field_scale             12  ## field_length         )         psfMatch = set(fieldsExisting) & set(psf)  ## Compare lists and return matches in a set         if len(psfMatch) == 1:             matchField = next(iter(psfMatch))  ## Get field name from psfMatch set             arcpy.CalculateField_management(                 filepath,  ## in_table                 addField,  ## field                 "!{}!".format(matchField),  ## expression                 "PYTHON_9.3"  ## expression_type             )         else:             print "More than one possible search field match, unable to caclulate field." This doesn't attempt to convert field datatypes. I was just writing out some ideas. ... View more

Thanks for the reply's guys. I think the answer is that this can't be done. I've come across a few sources that claim it is not possible to send a PL/SQL record type table object through cx_Oracle. Discussion of the cx_Oracle Python interface to Oracle () ... View more

Sorry for the confusion with the first post. I was just trying to get my thoughts out into some psuedo code. Unfortunately, the procedure I'm calling does not use a refcursor; it has to be a table object of record datatype. ... View more

Yes, I believe I do need to create an array. I'd like to provide some more information about the procedure I'm calling. It has an input of a table type and an output of a table type. Both in and out tables are defined with a custom RECORD datatype. Here's how the input table is defined: TYPE meter_info_tab_type IS TABLE OF meter_info_rec_type INDEX BY BINARY_INTEGER; Here's a sample of what the input table's record type looks like: TYPE meter_info_rec_type IS RECORD (     c_esb_guid              VARCHAR2(50),     c_newserialnum          VARCHAR2(10),     c_remoteid              VARCHAR2(15),     d_install               DATE,     c_remotetype            VARCHAR2(2),     c_recordtype            VARCHAR2(1),     l_processed             NUMBER, ) Here's how the input and output tables are used in the procedure: PROCEDURE InsertMeterInfo (    p_error_data            OUT     error_tab_type,    p_meter_info_data       IN      meter_info_tab_type ) IS ... rest of procedure code... All the cx_oracle arrayvar() examples use VARCHAR2 or NUMBER datatypes. How do I define an arrayvar with a RECORD datatype so I can put in a list of lists or tuples for the rows? This page looked to have some promising information, but I'm still having trouble. ... View more

ListUsers() has the IsDirectConnection property. I think If it's not direct connect, then it's application server. ... View more