Switch to pure ADAL

[lmazuel]

Switch to ADAL implementation. Will require a msrestazure 0.5.0, since there is some internal changes that I can't assume no one is using.
Deprecate at the same time some old stuff, like china=True

Needs AzureAD/azure-activedirectory-library-for-python#142 to be iso-functional

Edit: Since adal 0.6.0 I have everything I need to migrate in a iso-functional way

[lmazuel]

@yugangw-msft It's done, testing is ok, I'd appreciate if you can take a look :)

I didn't tested SDK yet, doing it right now.

[lmazuel]

ChangeLog:

Breaking changes

These breaking changes applies to ServicePrincipalCredentials, UserPassCredentials, AADTokenCredentials

• Remove "auth_uri" attribute and parameter. This was unused.
• Remove "state" attribute. This was unused.
• Remove "client" attribute. This was exposed by mistake and should have been internal. No replacement is possible.
• Remove "token_uri" attribute and parameter. Use "cloud_environment" and "tenant" to impact the login url now.

Features

• Implementation is now using ADAL and not request-oauthlib. This allows more AD scenarios (like federated)
• Expose ADAL token cache system as API

[yugangw-msft]

Please test it with real credentials. If works, then all good.

[prburgu]

I think, I'm running into this bug with the following code. Can this change be hotfixed on the latest release? It appears this PR may take some time.

AUTHORITY_URL = "https://login.microsoftonline.com" + '/' + os.environ["TENANT"]
auth_context = adal.AuthenticationContext(AUTHORITY_URL, api_version=None)
resource = "https://vault.azure.net"
client_id = os.environ["CLIENT_ID"]
client_secret = os.environ["CLIENT_SECRET"]
token = auth_context.acquire_token_with_client_credentials(resource, client_id, client_secret)
credentials = AADTokenCredentials(token, client_id)
print("Got one token. Trying to get the token again")
token = auth_context.acquire_token_with_client_credentials(resource, client_id, client_secret)
credentials = AADTokenCredentials(token, client_id)
print("Got one token second time as well")

The second print never gets printed and I get this exception.

  File "/Users/prburgu/Downloads/test1/lib/python2.7/site-packages/dateutil/parser/_parser.py", line 76, in __init__
    '{itype}'.format(itype=instream.__class__.__name__))
TypeError: Parser must be a string or character stream, not float

[lmazuel]

Hi @prburgu
Sounds plausible, but could I get the complete stacktrace please? Not just the last line.

Also, this PR will be merged as 0.5.0, but this creates some conflicts right now with some packages that requires 0.4.x. I hope all will be merged and released by this summer.

[prburgu]

Actually, I've worked around this issue in my code, so there is no urgent need for hotfix and I'll leave it up to you on that. Here is the complete stacktrace.

Traceback (most recent call last):
  File "adal_test.py", line 13, in <module>
    token = auth_context.acquire_token_with_client_credentials(resource, client_id, client_secret)
  File "/Users/prburgu/Downloads/test1/lib/python2.7/site-packages/adal/authentication_context.py", line 179, in acquire_token_with_client_credentials
    return self._acquire_token(token_func)
  File "/Users/prburgu/Downloads/test1/lib/python2.7/site-packages/adal/authentication_context.py", line 128, in _acquire_token
    return token_func(self)
  File "/Users/prburgu/Downloads/test1/lib/python2.7/site-packages/adal/authentication_context.py", line 177, in token_func
    return token_request.get_token_with_client_credentials(client_secret)
  File "/Users/prburgu/Downloads/test1/lib/python2.7/site-packages/adal/token_request.py", line 306, in get_token_with_client_credentials
    token = self._find_token_from_cache()
  File "/Users/prburgu/Downloads/test1/lib/python2.7/site-packages/adal/token_request.py", line 128, in _find_token_from_cache
    return self._cache_driver.find(cache_query)
  File "/Users/prburgu/Downloads/test1/lib/python2.7/site-packages/adal/cache_driver.py", line 201, in find
    is_resource_tenant_specific)
  File "/Users/prburgu/Downloads/test1/lib/python2.7/site-packages/adal/cache_driver.py", line 167, in _refresh_entry_if_necessary
    expiry_date = parser.parse(entry[TokenResponseFields.EXPIRES_ON])
  File "/Users/prburgu/Downloads/test1/lib/python2.7/site-packages/dateutil/parser/_parser.py", line 1356, in parse
    return DEFAULTPARSER.parse(timestr, **kwargs)
  File "/Users/prburgu/Downloads/test1/lib/python2.7/site-packages/dateutil/parser/_parser.py", line 645, in parse
    res, skipped_tokens = self._parse(timestr, **kwargs)
  File "/Users/prburgu/Downloads/test1/lib/python2.7/site-packages/dateutil/parser/_parser.py", line 721, in _parse
    l = _timelex.split(timestr)         # Splits the timestr into tokens
  File "/Users/prburgu/Downloads/test1/lib/python2.7/site-packages/dateutil/parser/_parser.py", line 207, in split
    return list(cls(s))
  File "/Users/prburgu/Downloads/test1/lib/python2.7/site-packages/dateutil/parser/_parser.py", line 76, in __init__
    '{itype}'.format(itype=instream.__class__.__name__))
TypeError: Parser must be a string or character stream, not float

[lmazuel]

I was looking for the _find_token_from_cache and it's there :). You're right, that's the correct fix.
I'd rather wait for this PR to be merged if it's ok for you.

[lmazuel]

Merging in the 0.6.0 branch

[lmazuel]

For some reason I called the branch 0.6.0 and 0.5.0.... I renamed the branch.