#
# This library is free software; you can redistribute it and/or
# modify it under the terms of the GNU Lesser General Public
# License as published by the Free Software Foundation; either
# version 2.1 of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful, but
# WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
# Lesser General Public License for more details.
#
# You should have received a copy of the GNU Lesser General Public
# License along with this program; if not, write to the Free Software
# Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
#
"""
**Experimental:** *New in pywbem 0.9 as experimental.*
Operation recorder interface and implementations.
"""
from __future__ import print_function, absolute_import
from collections import namedtuple
try:
from collections import OrderedDict # pylint: disable=import-error
except ImportError:
from ordereddict import OrderedDict # pylint: disable=import-error
from datetime import datetime, timedelta
import logging
import six
import yaml
from yaml.representer import RepresenterError
from .cim_obj import CIMInstance, CIMInstanceName, CIMClass, CIMClassName, \
CIMProperty, CIMMethod, CIMParameter, CIMQualifier, \
CIMQualifierDeclaration, NocaseDict, _ensure_unicode
from .cim_types import CIMInt, CIMFloat, CIMDateTime
from .exceptions import CIMError
from ._logging import LOGGER_API_CALLS_NAME, LOGGER_HTTP_NAME
if six.PY2:
import codecs # pylint: disable=wrong-import-order
__all__ = ['BaseOperationRecorder', 'TestClientRecorder',
'LogOperationRecorder',
'OpArgs', 'OpResult', 'HttpRequest', 'HttpResponse']
if six.PY2:
_Longint = long # noqa: F821
else:
_Longint = int
OpArgsTuple = namedtuple("OpArgsTuple", ["method", "args"])
def _represent_ordereddict(dump, tag, mapping, flow_style=None):
"""PyYAML representer function for OrderedDict.
This is needed for yaml.safe_dump() to support OrderedDict.
Courtesy:
http://blog.elsdoerfer.name/2012/07/26/make-pyyaml-output-an-ordereddict/
"""
value = []
node = yaml.MappingNode(tag, value, flow_style=flow_style)
if dump.alias_key is not None:
dump.represented_objects[dump.alias_key] = node
best_style = True
if hasattr(mapping, 'items'):
mapping = mapping.items()
for item_key, item_value in mapping:
node_key = dump.represent_data(item_key)
node_value = dump.represent_data(item_value)
if not (isinstance(node_key, yaml.ScalarNode) and
not node_key.style):
best_style = False # pylint: disable=bad-indentation
if not (isinstance(node_value, yaml.ScalarNode) and
not node_value.style):
best_style = False # pylint: disable=bad-indentation
value.append((node_key, node_value))
if flow_style is None:
if dump.default_flow_style is not None:
node.flow_style = dump.default_flow_style
else:
node.flow_style = best_style
return node
yaml.SafeDumper.add_representer(
OrderedDict,
lambda dumper, value:
_represent_ordereddict(dumper, u'tag:yaml.org,2002:map', value))
# Tag for CIMDateTime serialization in yaml files
CIMDATETIME_TAG = '!CIMDateTime'
def _cimdatetime_representer(dumper, cimdatetime):
"""
PyYAML representer function for CIMDateTime objects.
This is needed for yaml.safe_dump() to support CIMDateTime.
"""
cimdatetime_str = str(cimdatetime)
node = dumper.represent_scalar(CIMDATETIME_TAG, cimdatetime_str)
return node
yaml.SafeDumper.add_representer(CIMDateTime, _cimdatetime_representer)
def _cimdatetime_constructor(loader, node):
"""
PyYAML constructor function for CIMDateTime objects.
This is needed for yaml.safe_load() to support CIMDateTime.
"""
cimdatetime_str = loader.construct_scalar(node)
cimdatetime = CIMDateTime(cimdatetime_str)
return cimdatetime
yaml.SafeLoader.add_constructor(CIMDATETIME_TAG, _cimdatetime_constructor)
# Some monkey-patching for better diagnostics:
def _represent_undefined(self, data):
"""Raises flag for objects that cannot be represented"""
raise RepresenterError("cannot represent an object: %s of type: %s; "
"yaml_representers: %r, "
"yaml_multi_representers: %r" %
(data, type(data), self.yaml_representers.keys(),
self.yaml_multi_representers.keys()))
yaml.SafeDumper.represent_undefined = _represent_undefined
[docs]class OpArgs(OpArgsTuple):
"""
**Experimental:** *New in pywbem 0.9 as experimental.*
A named tuple representing the name and input arguments of the invocation
of a :class:`~pywbem.WBEMConnection` method, with the following named fields
and attributes:
Attributes:
~OpArgs.method (:term:`unicode string`):
Name of the :class:`~pywbem.WBEMConnection` method.
~OpArgs.args (:class:`py:dict`):
Dictionary of input arguments (both positional and keyword-based).
"""
__slots__ = ()
def __repr__(self):
return "OpArgs(method={s.method!r}, args={s.args!r})".format(s=self)
OpResultTuple = namedtuple("OpResultTuple", ["ret", "exc"])
[docs]class OpResult(OpResultTuple):
"""
**Experimental:** *New in pywbem 0.9 as experimental.*
A named tuple representing the result of the invocation of a
:class:`~pywbem.WBEMConnection` method, with the following named fields
and attributes:
Attributes:
~OpResult.ret (:class:`py:object`):
Return value, if the method returned.
`None`, if the method raised an exception.
Note that `None` may be a legitimate return value, so the test for
exceptions should be done based upon the :attr:`exc` variable.
~OpResult.exc (:exc:`~py:exceptions.Exception`):
Exception object, if the method raised an exception.
`None`, if the method returned.
"""
__slots__ = ()
def __repr__(self):
return "OpResult(ret={s.ret!r}, exc={s.exc!r})".format(s=self)
HttpRequestTuple = namedtuple("HttpRequestTuple",
["version", "url", "target", "method", "headers",
"payload"])
[docs]class HttpRequest(HttpRequestTuple):
"""
**Experimental:** *New in pywbem 0.9 as experimental.*
A named tuple representing the HTTP request sent by the WBEM client, with
the following named fields and attributes:
Attributes:
~HttpRequest.version (:term:`number`):
HTTP version from the request line (10 for HTTP/1.0, 11 for HTTP/1.1).
~HttpRequest.url (:term:`unicode string`):
URL of the WBEM server (e.g. 'https://myserver.acme.com:15989').
~HttpRequest.target (:term:`unicode string`):
Target URL segment as stated in request line (e.g. '/cimom').
~HttpRequest.method (:term:`unicode string`):
HTTP method as stated in the request line (e.g. "POST").
~HttpRequest.headers (:class:`py:dict`):
A dictionary of all HTTP header fields:
* key (:term:`unicode string`): Name of the header field
* value (:term:`unicode string`): Value of the header field
~HttpRequest.payload (:term:`unicode string`):
HTTP payload, i.e. the CIM-XML string.
"""
__slots__ = ()
def __repr__(self):
return "HttpRequest(version={s.version!r}, url={s.url!r}, " \
"target={s.target!r}, method={s.method!r}, " \
"headers={s.headers!r}, payload={s.payload!r})" \
.format(s=self)
HttpResponseTuple = namedtuple("HttpResponseTuple",
["version", "status", "reason", "headers",
"payload"])
[docs]class HttpResponse(HttpResponseTuple):
"""
**Experimental:** *New in pywbem 0.9 as experimental.*
A named tuple representing the HTTP response received by the WBEM client,
with the following named fields and attributes:
Attributes:
~HttpResponse.version (:term:`number`):
HTTP version from the response line (10 for HTTP/1.0, 11 for HTTP/1.1).
~HttpResponse.status (:term:`number`):
HTTP status code from the response line (e.g. 200).
~HttpResponse.reason (:term:`unicode string`):
HTTP reason phrase from the response line (e.g. "OK").
~HttpResponse.headers (:class:`py:dict`):
A dictionary of all HTTP header fields:
* key (:term:`unicode string`): Name of the header field
* value (:term:`unicode string`): Value of the header field
~HttpResponse.payload (:term:`unicode string`):
HTTP payload, i.e. the CIM-XML string.
"""
__slots__ = ()
def __repr__(self):
return "HttpResponse(version={s.version!r}, status={s.status!r}, " \
"reason={s.reason!r}, headers={s.headers!r}, " \
"payload={s.payload!r})".format(s=self)
[docs]class BaseOperationRecorder(object):
# pylint: disable=too-many-instance-attributes
"""
**Experimental:** *New in pywbem 0.9 as experimental.*
Abstract base class defining the interface to an operation recorder,
that records the WBEM operations executed in a connection to a WBEM
server.
An operation recorder can be added to a connection via the
:meth:`~pywbem.WBEMConnection.add_operation_recorder` method. The operation
recorders of a connection can be retrieved via the
:attr:`~pywbem.WBEMConnection.operation_recorders` property.
Each operation that is executed on a connection will cause the
:meth:`record` method of those operation recorders of the connection to be
called, that are enabled.
After being added to a connection, an operation recorder is enabled. It can
be disabled and re-enabled using the
:meth:`~pywbem.BaseOperationRecorder.disable` and
:meth:`~pywbem.BaseOperationRecorder.enable` methods, respectively.
This can be used to temporarily pause the recorder.
"""
def __init__(self):
self._enabled = True
self._conn_id = None
self.reset()
[docs] def enable(self):
"""
*New in pywbem 0.10.*
Enable the recorder.
"""
self._enabled = True
[docs] def disable(self):
"""
*New in pywbem 0.10.*
Disable the recorder.
"""
self._enabled = False
@property
def enabled(self):
"""
*New in pywbem 0.10.*
Indicate whether the recorder is enabled.
"""
return self._enabled
[docs] @staticmethod
def open_file(filename, file_mode='w'):
"""
*New in pywbem 0.10.*
A static convenience function that performs the open of the recorder
file correctly for different versions of Python. This covers the
issue where the file should be opened in text mode but that is
done differently in Python 2 and Python 3.
The returned file-like object must be closed by the caller.
Parameters:
filename(:term:`string`):
Name of the file where the recorder output will be written
file_mode(:term:`string`):
Optional file mode. The default is 'w' which overwrites any
existing file. if 'a' is used, the data is appended to any
existing file.
Returns:
File-like object.
Example::
recorder = TestClientRecorder(...)
recorder_file = recorder.open_file('recorder.log')
. . . # Perform WBEM operations using the recorder
recorder_file.close()
"""
if six.PY2:
# Open with codecs to define text mode
return codecs.open(filename, mode=file_mode, encoding='utf-8')
return open(filename, file_mode, encoding='utf8')
[docs] def reset(self, pull_op=None):
"""Reset all the attributes in the class. This also allows setting
the pull_op attribute that defines whether the operation is to be
a traditional or pull operation.
This does NOT reset _conn.id as that exists through the life of
the connection.
"""
self._pywbem_method = None
self._pywbem_args = None
self._pywbem_result_ret = None
self._pywbem_result_exc = None
self._http_request_version = None
self._http_request_url = None
self._http_request_target = None
self._http_request_method = None
self._http_request_headers = None
self._http_request_payload = None
self._http_response_version = None
self._http_response_status = None
self._http_response_reason = None
self._http_response_headers = None
self._http_response_payload = None
self._pull_op = pull_op
[docs] def stage_wbem_connection(self, wbem_connection):
"""
Stage information about the connection. Used only by
LogOperationRecorder.
"""
pass
[docs] def stage_pywbem_args(self, method, **kwargs):
"""
Set requst method and all args.
Normally called before the cmd is executed to record request
parameters
"""
# pylint: disable=attribute-defined-outside-init
self._pywbem_method = method
self._pywbem_args = kwargs
[docs] def stage_pywbem_result(self, ret, exc):
""" Set Result return info or exception info"""
# pylint: disable=attribute-defined-outside-init
self._pywbem_result_ret = ret
self._pywbem_result_exc = exc
[docs] def stage_http_request(self, conn_id, version, url, target, method, headers,
payload):
"""Set request HTTP information including url, headers, etc."""
# pylint: disable=attribute-defined-outside-init
self._http_request_version = version
self._http_request_conn_id = conn_id
self._http_request_url = url
self._http_request_target = target
self._http_request_method = method
self._http_request_headers = headers
self._http_request_payload = payload
# pylint: disable=unused-argument
[docs] def stage_http_response1(self, conn_id, version, status, reason, headers):
"""Set response http info including headers, status, etc.
conn_id unused here. Used in log"""
# pylint: disable=attribute-defined-outside-init
self._http_response_version = version
self._http_response_status = status
self._http_response_reason = reason
self._http_response_headers = headers
[docs] def stage_http_response2(self, payload):
"""Stage second part of http response, the payload"""
# pylint: disable=attribute-defined-outside-init
self._http_response_payload = payload
[docs] def record_staged(self):
"""Encode staged information on request and result to output"""
if self.enabled:
pwargs = OpArgs(
self._pywbem_method,
self._pywbem_args)
pwresult = OpResult(
self._pywbem_result_ret,
self._pywbem_result_exc)
httpreq = HttpRequest(
self._http_request_version,
self._http_request_url,
self._http_request_target,
self._http_request_method,
self._http_request_headers,
self._http_request_payload)
httpresp = HttpResponse(
self._http_response_version,
self._http_response_status,
self._http_response_reason,
self._http_response_headers,
self._http_response_payload)
self.record(pwargs, pwresult, httpreq, httpresp)
[docs] def record(self, pywbem_args, pywbem_result, http_request, http_response):
"""
Function that is called to record a single WBEM operation, i.e. the
invocation of a single :class:`~pywbem.WBEMConnection` method.
This function is called only when the recorder is enabled, i.e. it
does not need to check for recorder enablement.
Parameters:
pywbem_args (:class:`~pywbem.OpArgs`):
The name and input arguments of the :class:`~pywbem.WBEMConnection`
method that is recorded.
pywbem_result (:class:`~pywbem.OpResult`):
The result (return value or exception) of the
:class:`~pywbem.WBEMConnection` method that is recorded.
http_request (:class:`~pywbem.HttpRequest`):
The HTTP request sent by the :class:`~pywbem.WBEMConnection` method
that is recorded.
`None`, if no HTTP request had been sent (e.g. because an exception
was raised before getting there).
http_response (:class:`~pywbem.HttpResponse`):
The HTTP response received by the :class:`~pywbem.WBEMConnection`
method that is recorded.
`None`, if no HTTP response had been received (e.g. because an
exception was raised before getting there).
"""
raise NotImplementedError
[docs]class LogOperationRecorder(BaseOperationRecorder):
"""
**Experimental:** *New in pywbem 0.11 and redesigned in pywbem 0.12 as
experimental.*
A recorder that logs certain aspects of the WBEM operations driven by
pywbem users to Python loggers.
This recorder supports two Python loggers:
* API logger (Python logger name: `'pywbem.api'`) - Logs API calls and
returns, for the :class:`~pywbem.WBEMConnection` methods that drive WBEM
operations (see :ref:`WBEM operations`). This logs the API parameters and
results, including CIM objects/exceptions. It also logs the creation of
:class:`~pywbem.WBEMConnection` objects to capture connection information
in order to determine the connection to which a particular log record
belongs.
* HTTP logger (Python logger name: `'pywbem.http'`) - Logs HTTP requests
and responses between the pywbem client and WBEM server. This logs the
HTTP request data and response data including HTTP headers and CIM-XML
payload.
All logging calls are at the :attr:`py:logging.DEBUG` logging level.
"""
def __init__(self, conn_id, detail_levels=None):
"""
Parameters:
conn_id (:term:`connection id`):
String that represents an id for a particular connection
that is used to build the name of each logger. The logger names
are uniqueqly idenfified by the conn_id suffix to the logger name
(ex.pywbem.ops.1-2343) so that each WBEMConnection could be
logged with a separate logger.
detail_levels (:class:`py:dict`):
Dictionary identifying the detail level for one or both of
the loggers.
Key: Simple logger name (e.g. 'api').
Value: Detail level, either a string from
:data:`~pywbem._logging.LOG_DETAIL_LEVELS`, or an integer that
specifies the maximum size of each log record.
"""
super(LogOperationRecorder, self).__init__()
self._conn_id = conn_id
self.detail_levels = {}
# logging only occurs if the corresponding detail level is not None
self.api_detail_level = None
self.http_detail_level = None
self.api_maxlen = None
self.http_maxlen = None
self.set_detail_level(detail_levels)
# build name for logger for this connection
if conn_id:
self.apilogger = logging.getLogger(
'%s.%s' % (LOGGER_API_CALLS_NAME, conn_id))
self.httplogger = logging.getLogger(
'%s.%s' % (LOGGER_HTTP_NAME, conn_id))
else:
self.apilogger = logging.getLogger(LOGGER_API_CALLS_NAME)
self.httplogger = logging.getLogger(LOGGER_HTTP_NAME)
[docs] def set_detail_level(self, detail_levels):
"""
Sets the detail levels from the input dictionary in detail_levels.
"""
if detail_levels is None:
return
self.detail_levels = detail_levels
if 'api' in detail_levels:
self.api_detail_level = detail_levels['api']
if 'http' in detail_levels:
self.http_detail_level = detail_levels['http']
if isinstance(self.api_detail_level, int):
self.api_maxlen = self.api_detail_level
if isinstance(self.http_detail_level, int):
self.http_maxlen = self.http_detail_level
[docs] def stage_wbem_connection(self, wbem_connection):
"""
Log connection information. This includes the connection id (conn_id)
that is output with the log entry. This entry is logged if either
http or api loggers are enable. It honors both the logger and
detail level of either api logger if defined or http logger if defined.
If the api logger does not exist, the output shows this as an http
loggger output since we do not want to create an api logger for this
specific output
"""
self._conn_id = wbem_connection.conn_id
if self.enabled:
if self.api_detail_level is not None:
logger = self.apilogger
detail_level = self.api_detail_level
max_len = self.api_maxlen
elif self.http_detail_level is not None:
logger = self.httplogger
detail_level = self.http_detail_level
max_len = self.http_maxlen
else:
return
if logger.isEnabledFor(logging.DEBUG):
conn_data = str(wbem_connection) if detail_level == 'summary' \
else repr(wbem_connection)
if max_len and (len(conn_data) > max_len):
conn_data = conn_data[:max_len] + '...'
logger.debug('Connection:%s %s', self._conn_id, conn_data)
[docs] def stage_pywbem_args(self, method, **kwargs):
"""
Log request method and all args.
Normally called before the cmd is executed to record request
parameters.
This method does not support the summary detail_level because
that seems to add little info to the log that is not also in the
response.
"""
# pylint: disable=attribute-defined-outside-init
self._pywbem_method = method
if self.enabled and self.api_detail_level is not None and \
self.apilogger.isEnabledFor(logging.DEBUG):
# TODO: future bypassed code to only ouput name and method if the
# detail is summary. We are not doing this because this is
# effectively the same information in the response so the only
# additional infomation is the time stamp.
# if self.api_detail_level == summary:
# self.apilogger.debug('Request:%s %s', self._conn_id, method)
# return
# Order kwargs. Note that this is done automatically starting
# with python 3.6
kwstr = ', '.join([('{0}={1!r}'.format(key, kwargs[key]))
for key in sorted(six.iterkeys(kwargs))])
if self.api_maxlen and (len(kwstr) > self.api_maxlen):
kwstr = kwstr[:self.api_maxlen] + '...'
# pylint: disable=bad-continuation
self.apilogger.debug('Request:%s %s(%s)', self._conn_id, method,
kwstr)
[docs] def stage_pywbem_result(self, ret, exc):
"""
Log result return or exception parameter. This method provides varied
type of formatting based on the detail_level parameter and the
data in ret.
"""
def format_result(ret, max_len):
""" format ret as repr while clipping it to max_len if
max_len is not None.
"""
# format the 'summary' and 'paths' detail_levels
if self.api_detail_level == 'summary':
if isinstance(ret, list):
if ret:
ret_type = type(ret[0]).__name__ if ret else ""
return 'list of %s; count=%s' % (ret_type, len(ret))
return 'Empty'
ret_type = type(ret).__name__
if hasattr(ret, 'classname'):
name = ret.classname
elif hasattr(ret, 'name'):
name = ret.name
else:
name = ""
return '%s %s' % (ret_type, name)
elif self.api_detail_level == 'paths':
if isinstance(ret, list):
if ret:
if hasattr(ret[0], 'path'):
result_fmt = ', '.join([str(p.path) for p in ret])
else:
result_fmt = '%s' % ret
else:
result_fmt = ''
else:
if hasattr(ret, 'path'):
result_fmt = '%s' % ret.path
else:
result_fmt = '%s' % ret
else:
result_fmt = '{0!r}'.format(ret)
if max_len and (len(result_fmt) > max_len):
result_fmt = result_fmt[:max_len] + '...'
return result_fmt
if self.enabled and self.api_detail_level is not None and \
self.apilogger.isEnabledFor(logging.DEBUG):
if exc: # format exception
# exceptions are always either all or reduced length
result = format_result(
'%s(%s)' % (exc.__class__.__name__, exc), self.api_maxlen)
return_type = 'Exception'
else: # format result
# test if type is tuple (subclass of tuple but not type tuple)
# pylint: disable=unidiomatic-typecheck
qrc = ""
# format open/pull response
if isinstance(ret, tuple) and \
type(ret) is not tuple: # pylint: disable=C0123
try: # test if field instances or paths
rtn_data = ret.instances
data_str = 'instances'
except AttributeError:
rtn_data = ret.paths
data_str = 'paths'
try: # test for query_result_class
qrc = ', query_result_class=%s' % \
ret.query_result_class
except AttributeError:
pass
result = '{0}(context={1}, eos={2}{3}, {4}={5})' \
.format(type(ret).__name__, ret.context, ret.eos, qrc,
data_str, format_result(rtn_data,
self.api_maxlen))
# format enumerate response except not open/pull
elif isinstance(ret, list):
try: # test for query_result_class
qrc = ', query_result_class=%s' % ret.query_result_class
except AttributeError:
pass
ret_fmtd = format_result(ret, self.api_maxlen)
result = '%s%s' % (qrc, ret_fmtd)
# format single return object
else:
result = format_result(ret, self.api_maxlen)
return_type = 'Return'
self.apilogger.debug('%s:%s %s(%s)', return_type, self._conn_id,
self._pywbem_method, result)
[docs] def stage_http_request(self, conn_id, version, url, target, method, headers,
payload):
"""Log request HTTP information including url, headers, etc."""
if self.enabled and self.http_detail_level is not None and \
self.httplogger.isEnabledFor(logging.DEBUG):
# pylint: disable=attribute-defined-outside-init
# if Auth header, mask data
if 'Authorization' in headers:
authtype, cred = headers['Authorization'].split(' ')
headers['Authorization'] = '%s %s' % (authtype, 'X' * len(cred))
header_str = ' '.join('{0}:{1!r}'.format(k, v)
for k, v in headers.items())
if self.http_detail_level == 'summary':
upayload = ""
elif isinstance(payload, six.binary_type):
upayload = payload.decode('utf-8')
else:
upayload = payload
if self.http_maxlen and (len(payload) > self.http_maxlen):
upayload = upayload[:self.http_maxlen] + '...'
self.httplogger.debug('Request:%s %s %s %s %s %s\n %s',
conn_id, method, target, version, url,
header_str, upayload)
[docs] def stage_http_response1(self, conn_id, version, status, reason, headers):
"""Set response http info including headers, status, etc. """
# pylint: disable=attribute-defined-outside-init
self._http_response_version = version
self._http_response_status = status
self._http_response_reason = reason
self._http_response_headers = headers
self._http_response_conn_id = conn_id
[docs] def stage_http_response2(self, payload):
"""Log complete http response, including response1 and payload"""
# required because http code uses sending all None to reset
# parameters. We ignore that
if not self._http_response_version and not payload:
return
if self.enabled and self.http_detail_level is not None and \
self.httplogger.isEnabledFor(logging.DEBUG):
if self._http_response_headers:
header_str = \
' '.join('{0}:{1!r}'.format(k, v)
for k, v in self._http_response_headers.items())
else:
header_str = ''
if self.http_detail_level == 'summary':
upayload = ""
elif self.http_maxlen and (len(payload) > self.http_maxlen):
upayload = (_ensure_unicode(payload[:self.http_maxlen]) +
'...')
else:
upayload = _ensure_unicode(payload)
self.httplogger.debug('Response:%s %s:%s %s %s\n %s',
self._http_response_conn_id,
self._http_response_status,
self._http_response_reason,
self._http_response_version,
header_str, upayload)
[docs] def record_staged(self):
"""
Not used for logging. The logs are output in the various
stage... methods methods
"""
pass
[docs] def record(self, pywbem_args, pywbem_result, http_request, http_response):
"""Not used for logging"""
pass
[docs]class TestClientRecorder(BaseOperationRecorder):
"""
**Experimental:** *New in pywbem 0.9 as experimental.*
An operation recorder that generates test cases for each recorded
operation. The test cases are in the YAML format suitable for the
`test_client` unit test module of the pywbem project.
"""
# HTTP header fields to exclude when creating the testcase
# (in lower case)
EXCLUDE_REQUEST_HEADERS = [
'authorization',
'content-length',
'content-type',
]
EXCLUDE_RESPONSE_HEADERS = [
'content-length',
'content-type',
]
# Dummy server URL and credentials for use in generated test case
TESTCASE_URL = 'http://acme.com:80'
TESTCASE_USER = 'username'
TESTCASE_PASSWORD = 'password'
def __init__(self, fp):
"""
Parameters:
fp (file):
An open file that each test case will be written to. This file
should have been opened in text mode.
Since there are differences between python 2 and 3 in opening
files in text mode, the static method
:meth:`~pywbem.BaseOperationRecorder.open_file`
can be used to open the file or python 2/3 compatible open::
from io import open
f = open('blah.log', encoding='utf-8')
Example::
recorder = TestClientRecorder(
BaseOperationRecorder.open_file('recorder.log'))
"""
super(TestClientRecorder, self).__init__()
self._fp = fp
[docs] def record(self, pywbem_args, pywbem_result, http_request, http_response):
"""
Function that records the invocation of a single
:class:`~pywbem.WBEMConnection` method, by appending a corresponding
test case to the file.
Parameters: See :meth:`pywbem.BaseOperationRecorder.record`.
"""
testcase = OrderedDict()
testcase['name'] = pywbem_args.method
testcase['description'] = 'Generated by TestClientRecorder'
tc_pywbem_request = OrderedDict()
tc_pywbem_request['url'] = TestClientRecorder.TESTCASE_URL
tc_pywbem_request['creds'] = [TestClientRecorder.TESTCASE_USER,
TestClientRecorder.TESTCASE_PASSWORD]
tc_pywbem_request['namespace'] = 'root/cimv2'
tc_pywbem_request['timeout'] = 10
tc_pywbem_request['debug'] = False
tc_operation = OrderedDict()
tc_operation['pywbem_method'] = pywbem_args.method
for arg_name in pywbem_args.args:
tc_operation[arg_name] = self.toyaml(pywbem_args.args[arg_name])
tc_pywbem_request['operation'] = tc_operation
testcase['pywbem_request'] = tc_pywbem_request
tc_pywbem_response = OrderedDict()
if pywbem_result.ret is not None:
yaml_txt = 'pullresult' if self._pull_op else 'result'
tc_pywbem_response[yaml_txt] = self.toyaml(pywbem_result.ret)
if pywbem_result.exc is not None:
exc = pywbem_result.exc
if isinstance(exc, CIMError):
tc_pywbem_response['cim_status'] = self.toyaml(exc.status_code)
else:
tc_pywbem_response['exception'] = self.toyaml(
exc.__class__.__name__)
testcase['pywbem_response'] = tc_pywbem_response
tc_http_request = OrderedDict()
if http_request is not None:
tc_http_request['verb'] = http_request.method
tc_http_request['url'] = TestClientRecorder.TESTCASE_URL
if http_request.target:
tc_http_request['url'] += http_request.target
tc_request_headers = OrderedDict()
if http_request.headers is not None:
for hdr_name in http_request.headers:
if hdr_name.lower() not in \
TestClientRecorder.EXCLUDE_REQUEST_HEADERS:
tc_request_headers[hdr_name] = \
http_request.headers[hdr_name]
tc_http_request['headers'] = tc_request_headers
if http_request.payload is not None:
data = http_request.payload.decode('utf-8')
data = data.replace('><', '>\n<').strip()
else:
data = None
tc_http_request['data'] = data
testcase['http_request'] = tc_http_request
tc_http_response = OrderedDict()
if http_response is not None:
tc_http_response['status'] = http_response.status
tc_response_headers = OrderedDict()
if http_response.headers is not None:
for hdr_name in http_response.headers:
if hdr_name.lower() not in \
TestClientRecorder.EXCLUDE_RESPONSE_HEADERS:
tc_response_headers[hdr_name] = \
http_response.headers[hdr_name]
tc_http_response['headers'] = tc_response_headers
if http_response.payload is not None:
data = http_response.payload.decode('utf-8')
data = data.replace('><', '>\n<').strip()
else:
data = None
tc_http_response['data'] = data
else:
tc_http_response['exception'] = "# Change this to a callback " \
"function that causes this " \
"condition."
testcase['http_response'] = tc_http_response
testcases = []
testcases.append(testcase)
# The file is open in text mode, so we produce a unicode string
data = yaml.safe_dump(testcases, encoding=None, allow_unicode=True,
default_flow_style=False, indent=4)
data = data.replace('\n\n', '\n') # YAML dump duplicates newlines
self._fp.write(data)
self._fp.flush()
[docs] def toyaml(self, obj):
"""
Convert any allowable input argument to or return value from an
operation method to an object that is ready for serialization into
test_client yaml format.
"""
# namedtuple is subclass of tuple so it is instance of tuple but
# not type tuple. Cvt to dictionary and cvt dict to yaml.
# pylint: disable=unidiomatic-typecheck
if isinstance(obj, tuple) and type(obj) is not tuple:
ret_dict = obj._asdict()
return self.toyaml(ret_dict)
if isinstance(obj, (list, tuple)):
ret = []
# This does not handle namedtuple
for item in obj:
ret.append(self.toyaml(item))
return ret
elif isinstance(obj, (dict, NocaseDict)):
ret_dict = OrderedDict()
for key in obj.keys(): # get keys in original case
ret_dict[key] = self.toyaml(obj[key])
return ret_dict
elif obj is None:
return obj
elif isinstance(obj, six.binary_type):
return obj.decode("utf-8")
elif isinstance(obj, six.text_type):
return obj
elif isinstance(obj, CIMInt):
# CIMInt is _Longint and therefore may exceed the value range of
# int in Python 2. Therefore, we convert it to _Longint.
return _Longint(obj)
elif isinstance(obj, six.integer_types):
# This case must be after CIMInt, because CIMInt is _Longint and
# would match a long value in Python 2.
# We don't convert six.integer_types to _Longint, because the value
# fits into the provided type, and there is no need to convert it.
# Note that in Python 2 (where that would make a difference), int
# and long do not inherit from each other, so it is likely best if
# we just don't convert.
return obj
elif isinstance(obj, bool):
return obj
elif isinstance(obj, CIMFloat):
return float(obj)
elif isinstance(obj, CIMDateTime):
return str(obj)
elif isinstance(obj, datetime):
return CIMDateTime(obj)
elif isinstance(obj, timedelta):
return CIMDateTime(obj)
elif isinstance(obj, CIMInstance):
ret_dict = OrderedDict()
ret_dict['pywbem_object'] = 'CIMInstance'
ret_dict['classname'] = self.toyaml(obj.classname)
ret_dict['properties'] = self.toyaml(obj.properties)
ret_dict['path'] = self.toyaml(obj.path)
return ret_dict
elif isinstance(obj, CIMInstanceName):
ret_dict = OrderedDict()
ret_dict['pywbem_object'] = 'CIMInstanceName'
ret_dict['classname'] = self.toyaml(obj.classname)
ret_dict['namespace'] = self.toyaml(obj.namespace)
ret_dict['keybindings'] = self.toyaml(obj.keybindings)
return ret_dict
elif isinstance(obj, CIMClass):
ret_dict = OrderedDict()
ret_dict['pywbem_object'] = 'CIMClass'
ret_dict['classname'] = self.toyaml(obj.classname)
ret_dict['superclass'] = self.toyaml(obj.superclass)
ret_dict['properties'] = self.toyaml(obj.properties)
ret_dict['methods'] = self.toyaml(obj.methods)
ret_dict['qualifiers'] = self.toyaml(obj.qualifiers)
return ret_dict
elif isinstance(obj, CIMClassName):
ret_dict = OrderedDict()
ret_dict['pywbem_object'] = 'CIMClassName'
ret_dict['classname'] = self.toyaml(obj.classname)
ret_dict['host'] = self.toyaml(obj.host)
ret_dict['namespace'] = self.toyaml(obj.namespace)
return ret_dict
elif isinstance(obj, CIMProperty):
ret_dict = OrderedDict()
ret_dict['pywbem_object'] = 'CIMProperty'
ret_dict['name'] = self.toyaml(obj.name)
ret_dict['value'] = self.toyaml(obj.value)
ret_dict['type'] = self.toyaml(obj.type)
ret_dict['reference_class'] = self.toyaml(obj.reference_class)
ret_dict['embedded_object'] = self.toyaml(obj.embedded_object)
ret_dict['is_array'] = self.toyaml(obj.is_array)
ret_dict['array_size'] = self.toyaml(obj.array_size)
ret_dict['class_origin'] = self.toyaml(obj.class_origin)
ret_dict['propagated'] = self.toyaml(obj.propagated)
ret_dict['qualifiers'] = self.toyaml(obj.qualifiers)
return ret_dict
elif isinstance(obj, CIMMethod):
ret_dict = OrderedDict()
ret_dict['pywbem_object'] = 'CIMMethod'
ret_dict['name'] = self.toyaml(obj.name)
ret_dict['return_type'] = self.toyaml(obj.return_type)
ret_dict['class_origin'] = self.toyaml(obj.class_origin)
ret_dict['propagated'] = self.toyaml(obj.propagated)
ret_dict['parameters'] = self.toyaml(obj.parameters)
ret_dict['qualifiers'] = self.toyaml(obj.qualifiers)
return ret_dict
elif isinstance(obj, CIMParameter):
ret_dict = OrderedDict()
ret_dict['pywbem_object'] = 'CIMParameter'
ret_dict['name'] = self.toyaml(obj.name)
ret_dict['type'] = self.toyaml(obj.type)
ret_dict['reference_class'] = self.toyaml(obj.reference_class)
ret_dict['is_array'] = self.toyaml(obj.is_array)
ret_dict['array_size'] = self.toyaml(obj.array_size)
ret_dict['qualifiers'] = self.toyaml(obj.qualifiers)
return ret_dict
elif isinstance(obj, CIMQualifier):
ret_dict = OrderedDict()
ret_dict['pywbem_object'] = 'CIMQualifier'
ret_dict['name'] = self.toyaml(obj.name)
ret_dict['value'] = self.toyaml(obj.value)
ret_dict['type'] = self.toyaml(obj.type)
ret_dict['propagated'] = self.toyaml(obj.propagated)
ret_dict['tosubclass'] = self.toyaml(obj.tosubclass)
ret_dict['toinstance'] = self.toyaml(obj.toinstance)
ret_dict['overridable'] = self.toyaml(obj.overridable)
ret_dict['translatable'] = self.toyaml(obj.translatable)
return ret_dict
elif isinstance(obj, CIMQualifierDeclaration):
ret_dict = OrderedDict()
ret_dict['pywbem_object'] = 'CIMQualifierDeclaration'
ret_dict['name'] = self.toyaml(obj.name)
ret_dict['type'] = self.toyaml(obj.type)
ret_dict['value'] = self.toyaml(obj.value)
ret_dict['is_array'] = self.toyaml(obj.is_array)
ret_dict['array_size'] = self.toyaml(obj.array_size)
ret_dict['scopes'] = self.toyaml(obj.scopes)
ret_dict['tosubclass'] = self.toyaml(obj.tosubclass)
ret_dict['toinstance'] = self.toyaml(obj.toinstance)
ret_dict['overridable'] = self.toyaml(obj.overridable)
ret_dict['translatable'] = self.toyaml(obj.translatable)
return ret_dict
else:
raise TypeError("Invalid type in TestClientRecorder.toyaml(): "
"%s %s" % (obj.__class__.__name__, type(obj)))