This is a combination of 2 commits. The first commit's
message is:

Prevent MemoryError when logging response bodies

Response bodies are loaded into memory prior to
being logged.

Loading huge response bodies may result in a
MemoryError.

This patch proposes that only JSON and TEXT
responses be logged, i.e when the Content-Type
header is application/json or application/text.

Responses that do not include or have a different
Content-Type header will have their body omitted.

This is a sort of backport of the fix for
keystoneauth sessions, see
I93b6fff73368c4f58bdebf8566c4948b50980cee

Co-Authored-By: Samuel de Medeiros Queiroz <samueldmq@gmail.com>

Closes-bug: 1616105

Change-Id: I8f43eee3a0b35041c6cf672e476f8151cf2f8d14
(cherry-picked from: 3e56e0d7e5e1a76d806a3bc1f6d5ef9070f95771)

Only log application/json in session to start

When whitelisting content types to debug print from session we chose
application/json and application/text. application/text is not a real
mime type, text is typically text/plain.

Rather than guess at mime types only print application/json to start
with, but make it easy for additional types to be added later.

Adapted from keystoneauth: Ica5fee076cdab8b1d5167161d28af7313fad9477
Related-Bug: 1616105

Change-Id: Ieaa8fb3ea8d25e09b89498f23b70b18c0f6153f1
(cherry-picked from: 51d16fa344829aadf454faf5e0c4535a8f96a7c8)

Do not log binary data during debug logging of a session.
Replace the binary data with the string <binary_data> instead.

sort of a backport of: I5184002f3a21c5e0ee510b21b9a7884c8dccd1e3

Change-Id: I07ddbc3967f297597542f1975004d94c490f6e6b
Related-Bug: 1616105
(cherry picked from commit af770f17b705a66bd4292b2a54df46ec5fdaa12b)

Currently, logs display the hash values of X-Auth-Token,
Authorization, and X-Subject-Token, but not the value of
the X-Service-Token.  This patch set adds the X-Service-Token
to the list of header fields to be hashed for logging purposes.

Change-Id: Iaa3a27f4b6c3baf964fa0c71328ffe9df43b2c0a
Closes-Bug: #1654847
(cherry picked from commit 56af8c90ecbb3cb5d29036151108b1e4e7a69bcc)

This reverts commit 870be44c0ed5564101f9cf9d53e8085fb01680c8.

This change breaks end users of the library. See:

http://logs.openstack.org/50/285450/1/check/gate-shade-dsvm-functional-keystone2/d1093b5/console.html#_2016-02-26_20_49_32_928

For an example of a consumer of the library being broken.

Change-Id: I1912003afb89579eb869767db7a411c451bc9806

Create a custom list with flag `truncated` to support corresponding
response from the identity service.

This is wanted by Horizon, that wants to know that the list returned
from keystone is not full and that more strict filters need to be
applied.

The previous attempt in commit c28d40814962b3a8ccb81e5e7d7f832c8f0a3c9a
was reverted by d20b300589863bcf165945beb129ebcc3621a14f because it
broke other code. This commit changes the way the flag is added and
verifies that existing code will not break.

Change-Id: Ia86cfd91110adae6d7ab86ff1f152a8f9be27837
Closes-Bug: 1520244

Add support for the domain_id attribute in the role entity.

Partially Implements: blueprint domain-specific-roles
Change-Id: I06af7647e15aa742609b3fe1b9b222fbeaeb5735

While the entity for an inference rule should be thought of as a
resource, the rules are essentially relationships between roles.
The `implied_role` API is linked with the role API, and thus the
client functions are part of v3/role.py.  However, it does not
map completely cleanly to the Crud baseclass, and requires
some custom URL generation.

Change-Id: I80a40e88b571fe9b0eca3af8b705ea79f28eb904