1 Traceback (most recent call last):
File "tera2.py", line 5, in <module>
session = udaExec.connect(method="odbc", system="ZZZZ.COMP.com",username="USERNAME", password="PASSWORD");
.py" , line 137, in connect
.py" , line 129, in connect
py", line 263, in __init__
py", line 227, in init
py", line 204, in initOdbcLibrary
odbc = ctypes.cdll.LoadLibrary(odbcLibPath)
File "/usr/lib/python3.4/ctypes/__init__.py", line 429, in LoadLibrary
File "/usr/lib/python3.4/ctypes/__init__.py", line 351, in __init__
self._handle = _dlopen(self._name, mode)
OSError: No such file or directory
Is there a reason why you want to use Cygwin? Why wouldn't you just use "pip install teradata" from Windows command line prompt?
Windows ODBC driver works just fine under Cygwin. but you need to install pyodbc from pypi. Since pypi doesn't host binaries for Cygwin (or Linux for that matter), make sure you install toolchain under Cygwin to build pyodbc, that is gcc, make, automake, gcc-c++ etc. pip should automatically build the package for you. Note that pyodbc doesn't support the "new" Teradata types such as interval, period etc.
As far as people wanting to use Cygwin, I guess that's just a preference for people who are used to command-line. I use it all the time and find it powerful tool when combined with other Cygwing/*nix tools.
teradata package from pypi runs atop ODBC or REST, so just installing teradata package won't be enough. I wish they had been more creative with the name rather than calling the package 'teradata'. That's confusing for people who are just trying to make python work with teradata.