When I load data from PostgreSQL into Stata some of the data has unexpected characters appended. How can I avoid this?
Here is the Stata code I am using:
odbc query mydatabase, schema $odbc
odbc load, exec("SELECT * FROM my_table") $odbc allstring
Here is an example of the output I see:
198734/0 one/0/r April/0/0/0
893476/0 two/0/r May/0/0/0
324192/0 three/0/r June/0/0/0
In Postgres the data is:
198734 one April
893476 two May
324192 three June
I see this in mostly in larger tables and with fields of all datatypes in PostgreSQL. If I export the data to a csv there are no trailing characters.
The odbci.ini file I am using looks like this:
[ODBC Data Sources]
mydatabase = PostgreSQL
[mydatabase]
Debug = 1
CommLog = 1
ReadOnly = no
Driver = /usr/lib64/psqlodbcw.so
Servername = myserver
Servertype = postgres
FetchBufferSize = 99
Port = 5432
Database = mydatabase
[Default]
Driver = /usr/lib64/psqlodbcw.so
I am using odbc version unixODBC 2.3.1 and PostgreSQL version 9.4.9 with server encoding UTF8 and Stata version 14.1.
What is causing the unexpected characters in the data imported into Stata? I know that I can clean the data once it’s in Stata but I would like to avoid this.
I was able to fix this by adding the line
set odbcdriver ansi
to the Stata code.