I am trying to install Pydoop (https://crs4.github.io/pydoop/installation.html) on CentOS 6 (Hortonworks HDP physical cluster, 4 nodes). I have Python 3.4 and Python 2.6 installed on my master. When I install it using Python 3 it shows syntax error like below:
[root@abc]# python3 setup.py build
File "setup.py", line 45
print 'using setuptools version', setuptools.__version__
^
SyntaxError: Missing parentheses in call to 'print'
When I install it using Python 2 it shows following error
[root@abc]# python setup.py build
using setuptools version 35.0.2
Traceback (most recent call last):
File "setup.py", line 65, in <module>
import pydoop
File "/root/pydoop/pydoop/__init__.py", line 42, in <module>
_HADOOP_INFO = _PATH_FINDER.find() # fill the cache ASAP
File "/root/pydoop/pydoop/hadoop_utils.py", line 624, in find
info[a] = getattr(self, a)()
File "/root/pydoop/pydoop/hadoop_utils.py", line 433, in hadoop_home
_hadoop_home_from_version_cmd() or
File "/root/pydoop/pydoop/hadoop_utils.py", line 394, in
_hadoop_home_from_version_cmd
output = sp.check_output([hadoop_exec, 'version'])
AttributeError: 'module' object has no attribute 'check_output'
I don't want to upgrade default python from 2.6 to 2.7 as it may break yum which uses python2.6 and upgrading it may cause problems. Any suggestions please?