Develop Reference

Robotframework "Choose file" causes AttributeError: module 'base64' has no attribute 'encodestring' in docker

I'm trying to run a test in a docker container, which runs locally with no issues:
I want to upload a correct.csv file from 'correct' directory
*** Keyword ***
Upload file
[Arguments] ${directory} ${file}
Choose File ${choose_file_input} ${EXECDIR}/Files/${directory}/${file}
** Test case ***
Upload
Upload file correct correct.csv
But when running test in docker I get a FAIL with the AttributeError: module 'base64' has no attribute 'encodestring'. Is it because there is no GUI in docker? or the encoding needs to be fixed? Or eventually maybe there is another solution I can use for uploading files?
15:20:01.250 INFO Sending /App/Files/correct/correct.csv to browser.
15:20:01.251 DEBUG POST http://192.168.1.29:4444/wd/hub/session/4b6d453b394adaaa51bb4149e9ba8678/elements {"using": "xpath", "value": "//div[#id=\"upload\"]//input"}
15:20:01.252 DEBUG Starting new HTTP connection (1): 192.168.1.29:4444
15:20:01.305 DEBUG http://192.168.1.29:4444 "POST /wd/hub/session/4b6d453b394adaaa51bb4149e9ba8678/elements HTTP/1.1" 200 90
15:20:01.305 DEBUG Finished Request
15:20:01.618 FAIL AttributeError: module 'base64' has no attribute 'encodestring'
15:20:01.619 DEBUG Traceback (most recent call last):
File "/usr/local/lib/python3.9/site-packages/SeleniumLibrary/__init__.py", line 490, in run_keyword
return DynamicCore.run_keyword(self, name, args, kwargs)
File "/usr/local/lib/python3.9/site-packages/robotlibcore.py", line 103, in run_keyword
return self.keywords[name](*args, **(kwargs or {}))
File "/usr/local/lib/python3.9/site-packages/SeleniumLibrary/keywords/formelement.py", line 224, in choose_file
self.find_element(locator).send_keys(file_path)
File "/usr/local/lib/python3.9/site-packages/selenium/webdriver/remote/webelement.py", line 475, in send_keys
value = self._upload(local_file)
File "/usr/local/lib/python3.9/site-packages/selenium/webdriver/remote/webelement.py", line 695, in _upload
content = base64.encodestring(fp.getvalue())

Based on the traceback you have found this issue:
Selenium 3 is incompatible with Python 3.9
This is the fix for the issue: DeprecationWarning of base64.encodestring().
They won't back port this fix:
Thanks for the issue. We won't be releasing another version 3 as we're
heading to finishing off Selenium 4. It is a drop in replacement to
use Selenium 4.0.0.a5 so should work the same. There should not be any
breaking changes.
So you could upgrade selenium to Selenium 4.0.0.a5 or
Downgrade Python to 3.7 for example. I suppose locally you do not run 3.9.

We were running into this issue as well, but going back to an older version of Python was not an option due to incompatibilities with other libraries. If you find yourself in the same spot, you can re-create the alias like so:
import base64
base64.encodestring = base64.encodebytes
In whatever your entry-point is.

Pl\Python: issues when importing module

I have successfully setup plpyton3u extension in Postgresql 10 (64 bit) on my windows 10 (64 bit) machine. However, when i try to make a http request by calling requests module I am getting attribute error AttributeError: 'module' object has no attribute 'get' . Here is the code that I am using
CREATE OR REPLACE FUNCTION from_url(
-- The URL to download.
IN url text,
-- Should any errors (like HTTP transport errors)
-- throw an exception or simply return default_response
IN should_throw boolean DEFAULT true,
-- The default response if any errors are found.
-- Only used when should_throw is set to true
IN default_response text DEFAULT E''
)
RETURNS text
AS $$
# We will use traceback so we get decent error reporting.
import requests
import traceback
# Either throws an error or returns the defeault response
# depending on the should_throw parameter of the from_url() function
def on_error():
if should_throw:
# plpy.error() throws an exception which stops the current transaction
plpy.error("Error downloading '{0}'\n {1}".format(url, traceback.format_exc()))
else:
return default_response
try:
response = requests.get(url)
return response.data
except:
# Log and re-throw the error or return the default response
return on_error()
$$ LANGUAGE plpython3u VOLATILE;
select from_url(<SOME_DATA_FETCHING_URL>);
Where SOME_DATA_FETCHING_URL is the url of the server serving the data. When I run this code it throws following error
ERROR: plpy.Error: Error downloading SOME_DATA_FETCHING_URL
Traceback (most recent call last):
File "", line 16, in __plpython_procedure_from_url_24640
AttributeError: 'module' object has no attribute 'get'
CONTEXT: Traceback (most recent call last):
PL/Python function "from_url", line 19, in
return on_error()
PL/Python function "from_url", line 11, in on_error
plpy.error("Error downloading '{0}'\n {1}".format(url, traceback.format_exc()))
PL/Python function "from_url"
SQL state: XX000
my PYTHONPATH is set to C:\Python34\;C:\Python34\Scripts;C:\Python34\Lib;
I checked on python command prompt, I am able to import requests and run the get command successfully.
Am I missing any settings in PostGRESQL? which will allow me to run this code successfully?

(In case anyone else stumbles on this question) I had a similar problem, in my case it turned out to be caused by inadequate access permissions for user 'postgres' to the directory containing my Python modules. One thing I have learnt is that to check a PL/Python problem using the Python command prompt, it's always necessary to run it as the same user as the Postgres server (in my case, 'postgres')

Python 3.6.3: Multiple "Exception ignored in: <generator object..." in flask app

I'm running a flask app, upgraded everything from Python 2.7 to 3 about 5 months ago.
Most things have gone smooth enough, other than this one that's consistently bugging me locally. I'm on a MacBook on OSX 10.12.6, and a brew install of Python 3.6.3 under a virtualenv.
When a request comes in from a page that seemingly has multiple static requests (.css, .js, and image files mainly), I seem to be able to get this error just about anywhere that's using generators anywhere in my code.
Some examples (request is a flask.request object):
A place that checks to see if a path starts with '/static' of '/admin/static' (my code),
any(request.path.startswith(k) for k in self._static_paths)
.
Exception ignored in: <generator object CustomPrincipal._is_static_route.<locals>.<genexpr> at 0x11450f3b8>
Traceback (most recent call last):
File "/Developer/repos/git/betapilibs/lbbsports/flask_monkeypatches.py", line 22, in <genexpr>
any(_checker(request.path, k) for k in self._static_paths)
SystemError: error return without exception set
If a url path is restricted, check if the logged in user has the proper permissions / role,
return role in (role.name for role in self.roles)
.
Exception ignored in: <generator object UserMixin.has_role.<locals>.<genexpr> at 0x1155a7e08>
Traceback (most recent call last):
File "/Developer/virtualenvs/lbb3/lib/python3.6/site-packages/flask_security/core.py", line 386, in <genexpr>
SystemError: error return without exception set
A custom bit of code to ensure their "sub" account id is valid,
(not any(ident == account_id for ident in account_ids))
.
Exception ignored in: <generator object CustomSession.get_set_accounts.<locals>.<genexpr> at 0x115ff4fc0>
Traceback (most recent call last):
File "/Developer/customflask/flasklogin.py", line 168, in <genexpr>
SystemError: error return without exception set
Now, nothing seems to break in the system, I just get these error messages, and not consistently, only sometimes. If I set a breakpoint anywhere these errors are being reported to be happening, they don't error any more.
If I do something like, in the first example, break it into request.path.startswith('/static') or request.path.startswith('/admin/static'), I no longer get the error message, and in general, I never have a problem using request all over the place in the rest of the app.

A thing that was wrong in my local development setup was that I was serving all the /static and /admin/static through the flask app, instead of serving them through the web-server (in my case, nginx). So for some of the urls I was hitting, there might have been 10 requests come in basically at the same time, with Flask in debug mode, and a debugger connected as well (via PyCharm).
When I went through the trouble to ensure that all '/static' and '/admin/static' get served from there, instead of via flask, and flask was only getting 1 request per url, this problem went away.
I won't mark this as the answer, because there is still an underlying issue, but in case others have the same problem as me, this was a solution for my situation.

Ambari server setup: 'NoneType' object has no attribute 'lower'

I'm trying to setup Hadoop on my EC2 instance using this tutorial. I'm trying to setup the ambari server when I get this error:
[root#ip-xxx-xxx-xxx-xxx ec2-user]# ambari-server setup
Using python /usr/bin/python2.6
Setup ambari-server
Checking SELinux...
WARNING: Could not run /usr/sbin/sestatus: OK
Ambari-server daemon is configured to run under user 'root'. Change this setting [y/n] (n)?
Adjusting ambari-server permissions and ownership...
Checking iptables...
Checking JDK...
JCE Policy archive already exists, using /var/lib/ambari-server/resources/jce_policy-6.zip
Completing setup...
Traceback (most recent call last):
File "/usr/sbin/ambari-server.py", line 4236, in <module>
main()
File "/usr/sbin/ambari-server.py", line 4055, in main
setup(options)
File "/usr/sbin/ambari-server.py", line 2089, in setup
retcode = configure_os_settings()
File "/usr/sbin/ambari-server.py", line 1909, in configure_os_settings
os_name = os_info[0].lower()
AttributeError: 'NoneType' object has no attribute 'lower'
I'm really unsure how this is happening and don't know what to do fix this. Does anyone know what I'm doing wrong?
Edit: I looked at the code at found this:
os_info = platform.linux_distribution(
None, None, None, ['SuSE', 'redhat' ], 0
)
os_name = os_info[0].lower()
It appears that platform.linux_distribution is creating an array with it's parameters and doing something else to it. I can't find the implementation of the function in the file and there are several files that are in its same directory, still not sure what should I do.

I solved the problem, I removed the 3 None objects and removed the 2-size array and made them their own parameters. However now Ambari is causing another problem.

errors with gae-sessions and nose

I'm running into a few problems with adding gae-sessions to a relatively mature GAE app. I followed the readme carefully and also looked at the demo.
First, just adding the gaesesions directory to my app causes the following error when running tests with nose and nose-gae:
Exception ImportError: 'No module named threading' in <bound method local.__del__ of <_threading_local.local object at 0x103e10628>> ignored
All the tests run fine so not a big problem but suggests that something isn't right.
Next, if I add the following two lines of code:
from gaesessions import get_current_session
session = get_current_session()
And run my tests, then I get the following error:
Traceback (most recent call last):
File "/Users/.../unit_tests.py", line 1421, in testParseFBRequest
data = tasks.parse_fb_request(sr)
File "/Users/.../tasks.py", line 220, in parse_fb_request
session = get_current_session()
File "/Users/.../gaesessions/__init__.py", line 36, in get_current_session
return _tls.current_session
File "/Library/.../python2.7/_threading_local.py", line 193, in __getattribute__
return object.__getattribute__(self, name)
AttributeError: 'local' object has no attribute 'current_session'
This error does not happen on the dev server.
Any suggestions on fixing the above would be greatly appreciated.

I ran into the same problem. The problem seems to be that the gae testbed behaves differently than the development server. I don't know the specifics but ended up solving it by adding
def setUp(self):
testbed.Testbed().activate()
# after activating the testbed:
from gaesessions import Session, set_current_session
set_current_session(Session())