tempest/common/rest_client.py

# Copyright 2012 OpenStack Foundation
# Copyright 2013 IBM Corp.
# All Rights Reserved.
#
#    Licensed under the Apache License, Version 2.0 (the "License"); you may
#    not use this file except in compliance with the License. You may obtain
#    a copy of the License at
#
#         http://www.apache.org/licenses/LICENSE-2.0
#
#    Unless required by applicable law or agreed to in writing, software
#    distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
#    WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
#    License for the specific language governing permissions and limitations
#    under the License.

import collections
import inspect
import json
from lxml import etree
import re
import time

import jsonschema

from tempest.common import http
from tempest import config
from tempest import exceptions
from tempest.openstack.common import log as logging
from tempest.services.compute.xml import common

CONF = config.CONF

# redrive rate limited calls at most twice
MAX_RECURSION_DEPTH = 2
TOKEN_CHARS_RE = re.compile('^[-A-Za-z0-9+/=]*$')

# All the successful HTTP status codes from RFC 2616
HTTP_SUCCESS = (200, 201, 202, 203, 204, 205, 206)


class RestClient(object):

    TYPE = "json"

    # This is used by _parse_resp method
    # Redefine it for purposes of your xml service client
    # List should contain top-xml_tag-names of data, which is like list/array
    # For example, in keystone it is users, roles, tenants and services
    # All of it has children with same tag-names
    list_tags = []

    # This is used by _parse_resp method too
    # Used for selection of dict-like xmls,
    # like metadata for Vms in nova, and volumes in cinder
    dict_tags = ["metadata", ]

    LOG = logging.getLogger(__name__)

    def __init__(self, auth_provider):
        self.auth_provider = auth_provider

        self.endpoint_url = None
        self.service = None
        # The version of the API this client implements
        self.api_version = None
        self._skip_path = False
        self.build_interval = CONF.compute.build_interval
        self.build_timeout = CONF.compute.build_timeout
        self.general_header_lc = set(('cache-control', 'connection',
                                      'date', 'pragma', 'trailer',
                                      'transfer-encoding', 'via',
                                      'warning'))
        self.response_header_lc = set(('accept-ranges', 'age', 'etag',
                                       'location', 'proxy-authenticate',
                                       'retry-after', 'server',
                                       'vary', 'www-authenticate'))
        dscv = CONF.identity.disable_ssl_certificate_validation
        self.http_obj = http.ClosingHttp(
            disable_ssl_certificate_validation=dscv)

    def _get_type(self):
        return self.TYPE

    def get_headers(self, accept_type=None, send_type=None):
        if accept_type is None:
            accept_type = self._get_type()
        if send_type is None:
            send_type = self._get_type()
        return {'Content-Type': 'application/%s' % send_type,
                'Accept': 'application/%s' % accept_type}

    def __str__(self):
        STRING_LIMIT = 80
        str_format = ("config:%s, service:%s, base_url:%s, "
                      "filters: %s, build_interval:%s, build_timeout:%s"
                      "\ntoken:%s..., \nheaders:%s...")
        return str_format % (CONF, self.service, self.base_url,
                             self.filters, self.build_interval,
                             self.build_timeout,
                             str(self.token)[0:STRING_LIMIT],
                             str(self.get_headers())[0:STRING_LIMIT])

    def _get_region(self, service):
        """
        Returns the region for a specific service
        """
        service_region = None
        for cfgname in dir(CONF._config):
            # Find all config.FOO.catalog_type and assume FOO is a service.
            cfg = getattr(CONF, cfgname)
            catalog_type = getattr(cfg, 'catalog_type', None)
            if catalog_type == service:
                service_region = getattr(cfg, 'region', None)
        if not service_region:
            service_region = CONF.identity.region
        return service_region

    def _get_endpoint_type(self, service):
        """
        Returns the endpoint type for a specific service
        """
        # If the client requests a specific endpoint type, then be it
        if self.endpoint_url:
            return self.endpoint_url
        endpoint_type = None
        for cfgname in dir(CONF._config):
            # Find all config.FOO.catalog_type and assume FOO is a service.
            cfg = getattr(CONF, cfgname)
            catalog_type = getattr(cfg, 'catalog_type', None)
            if catalog_type == service:
                endpoint_type = getattr(cfg, 'endpoint_type', 'publicURL')
                break
        # Special case for compute v3 service which hasn't its own
        # configuration group
        else:
            if service == CONF.compute.catalog_v3_type:
                endpoint_type = CONF.compute.endpoint_type
        return endpoint_type

    @property
    def user(self):
        return self.auth_provider.credentials.get('username', None)

    @property
    def tenant_name(self):
        return self.auth_provider.credentials.get('tenant_name', None)

    @property
    def password(self):
        return self.auth_provider.credentials.get('password', None)

    @property
    def base_url(self):
        return self.auth_provider.base_url(filters=self.filters)

    @property
    def token(self):
        return self.auth_provider.get_token()

    @property
    def filters(self):
        _filters = dict(
            service=self.service,
            endpoint_type=self._get_endpoint_type(self.service),
            region=self._get_region(self.service)
        )
        if self.api_version is not None:
            _filters['api_version'] = self.api_version
        if self._skip_path:
            _filters['skip_path'] = self._skip_path
        return _filters

    def skip_path(self):
        """
        When set, ignore the path part of the base URL from the catalog
        """
        self._skip_path = True

    def reset_path(self):
        """
        When reset, use the base URL from the catalog as-is
        """
        self._skip_path = False

    def expected_success(self, expected_code, read_code):
        assert_msg = ("This function only allowed to use for HTTP status"
                      "codes which explicitly defined in the RFC 2616. {0}"
                      " is not a defined Success Code!").format(expected_code)
        assert expected_code in HTTP_SUCCESS, assert_msg

        # NOTE(afazekas): the http status code above 400 is processed by
        # the _error_checker method
        if read_code < 400 and read_code != expected_code:
                pattern = """Unexpected http success status code {0},
                             The expected status code is {1}"""
                details = pattern.format(read_code, expected_code)
                raise exceptions.InvalidHttpSuccessCode(details)

    def post(self, url, body, headers=None):
        return self.request('POST', url, headers, body)

    def get(self, url, headers=None):
        return self.request('GET', url, headers)

    def delete(self, url, headers=None, body=None):
        return self.request('DELETE', url, headers, body)

    def patch(self, url, body, headers=None):
        return self.request('PATCH', url, headers, body)

    def put(self, url, body, headers=None):
        return self.request('PUT', url, headers, body)

    def head(self, url, headers=None):
        return self.request('HEAD', url, headers)

    def copy(self, url, headers=None):
        return self.request('COPY', url, headers)

    def get_versions(self):
        resp, body = self.get('')
        body = self._parse_resp(body)
        versions = map(lambda x: x['id'], body)
        return resp, versions

    def _find_caller(self):
        """Find the caller class and test name.

        Because we know that the interesting things that call us are
        test_* methods, and various kinds of setUp / tearDown, we
        can look through the call stack to find appropriate methods,
        and the class we were in when those were called.
        """
        caller_name = None
        names = []
        frame = inspect.currentframe()
        is_cleanup = False
        # Start climbing the ladder until we hit a good method
        while True:
            try:
                frame = frame.f_back
                name = frame.f_code.co_name
                names.append(name)
                if re.search("^(test_|setUp|tearDown)", name):
                    cname = ""
                    if 'self' in frame.f_locals:
                        cname = frame.f_locals['self'].__class__.__name__
                    if 'cls' in frame.f_locals:
                        cname = frame.f_locals['cls'].__name__
                    caller_name = cname + ":" + name
                    break
                elif re.search("^_run_cleanup", name):
                    is_cleanup = True
                else:
                    cname = ""
                    if 'self' in frame.f_locals:
                        cname = frame.f_locals['self'].__class__.__name__
                    if 'cls' in frame.f_locals:
                        cname = frame.f_locals['cls'].__name__

                    # the fact that we are running cleanups is indicated pretty
                    # deep in the stack, so if we see that we want to just
                    # start looking for a real class name, and declare victory
                    # once we do.
                    if is_cleanup and cname:
                        if not re.search("^RunTest", cname):
                            caller_name = cname + ":_run_cleanups"
                            break
            except Exception:
                break
        # prevents frame leaks
        del frame
        if caller_name is None:
            self.LOG.debug("Sane call name not found in %s" % names)
        return caller_name

    def _get_request_id(self, resp):
        for i in ('x-openstack-request-id', 'x-compute-request-id'):
            if i in resp:
                return resp[i]
        return ""

    def _log_request(self, method, req_url, resp, secs=""):
        # if we have the request id, put it in the right part of the log
        extra = dict(request_id=self._get_request_id(resp))
        # NOTE(sdague): while we still have 6 callers to this function
        # we're going to just provide work around on who is actually
        # providing timings by gracefully adding no content if they don't.
        # Once we're down to 1 caller, clean this up.
        if secs:
            secs = " %.3fs" % secs
        self.LOG.info(
            'Request (%s): %s %s %s%s' % (
                self._find_caller(),
                resp['status'],
                method,
                req_url,
                secs),
            extra=extra)

    def _parse_resp(self, body):
        if self._get_type() is "json":
            body = json.loads(body)

            # We assume, that if the first value of the deserialized body's
            # item set is a dict or a list, that we just return the first value
            # of deserialized body.
            # Essentially "cutting out" the first placeholder element in a body
            # that looks like this:
            #
            #  {
            #    "users": [
            #      ...
            #    ]
            #  }
            try:
                # Ensure there are not more than one top-level keys
                if len(body.keys()) > 1:
                    return body
                # Just return the "wrapped" element
                first_key, first_item = body.items()[0]
                if isinstance(first_item, (dict, list)):
                    return first_item
            except (ValueError, IndexError):
                pass
            return body
        elif self._get_type() is "xml":
            element = etree.fromstring(body)
            if any(s in element.tag for s in self.dict_tags):
                # Parse dictionary-like xmls (metadata, etc)
                dictionary = {}
                for el in element.getchildren():
                    dictionary[u"%s" % el.get("key")] = u"%s" % el.text
                return dictionary
            if any(s in element.tag for s in self.list_tags):
                # Parse list-like xmls (users, roles, etc)
                array = []
                for child in element.getchildren():
                    array.append(common.xml_to_json(child))
                return array

            # Parse one-item-like xmls (user, role, etc)
            return common.xml_to_json(element)

    def response_checker(self, method, url, headers, body, resp, resp_body):
        if (resp.status in set((204, 205, 304)) or resp.status < 200 or
                method.upper() == 'HEAD') and resp_body:
            raise exceptions.ResponseWithNonEmptyBody(status=resp.status)
        # NOTE(afazekas):
        # If the HTTP Status Code is 205
        #   'The response MUST NOT include an entity.'
        # A HTTP entity has an entity-body and an 'entity-header'.
        # In the HTTP response specification (Section 6) the 'entity-header'
        # 'generic-header' and 'response-header' are in OR relation.
        # All headers not in the above two group are considered as entity
        # header in every interpretation.

        if (resp.status == 205 and
            0 != len(set(resp.keys()) - set(('status',)) -
                     self.response_header_lc - self.general_header_lc)):
                        raise exceptions.ResponseWithEntity()
        # NOTE(afazekas)
        # Now the swift sometimes (delete not empty container)
        # returns with non json error response, we can create new rest class
        # for swift.
        # Usually RFC2616 says error responses SHOULD contain an explanation.
        # The warning is normal for SHOULD/SHOULD NOT case

        # Likely it will cause an error
        if not resp_body and resp.status >= 400:
            self.LOG.warning("status >= 400 response with empty body")

    def _request(self, method, url, headers=None, body=None):
        """A simple HTTP request interface."""
        # Authenticate the request with the auth provider
        req_url, req_headers, req_body = self.auth_provider.auth_request(
            method, url, headers, body, self.filters)

        # Do the actual request, and time it
        start = time.time()
        resp, resp_body = self.http_obj.request(
            req_url, method, headers=req_headers, body=req_body)
        end = time.time()
        self._log_request(method, req_url, resp, secs=(end - start))
        # Verify HTTP response codes
        self.response_checker(method, url, req_headers, req_body, resp,
                              resp_body)

        return resp, resp_body

    def request(self, method, url, headers=None, body=None):
        retry = 0

        if headers is None:
            # NOTE(vponomaryov): if some client do not need headers,
            # it should explicitly pass empty dict
            headers = self.get_headers()

        resp, resp_body = self._request(method, url,
                                        headers=headers, body=body)

        while (resp.status == 413 and
               'retry-after' in resp and
                not self.is_absolute_limit(
                    resp, self._parse_resp(resp_body)) and
                retry < MAX_RECURSION_DEPTH):
            retry += 1
            delay = int(resp['retry-after'])
            time.sleep(delay)
            resp, resp_body = self._request(method, url,
                                            headers=headers, body=body)
        self._error_checker(method, url, headers, body,
                            resp, resp_body)
        return resp, resp_body

    def _error_checker(self, method, url,
                       headers, body, resp, resp_body):

        # NOTE(mtreinish): Check for httplib response from glance_http. The
        # object can't be used here because importing httplib breaks httplib2.
        # If another object from a class not imported were passed here as
        # resp this could possibly fail
        if str(type(resp)) == "<type 'instance'>":
            ctype = resp.getheader('content-type')
        else:
            try:
                ctype = resp['content-type']
            # NOTE(mtreinish): Keystone delete user responses doesn't have a
            # content-type header. (They don't have a body) So just pretend it
            # is set.
            except KeyError:
                ctype = 'application/json'

        # It is not an error response
        if resp.status < 400:
            return

        JSON_ENC = ['application/json', 'application/json; charset=utf-8']
        # NOTE(mtreinish): This is for compatibility with Glance and swift
        # APIs. These are the return content types that Glance api v1
        # (and occasionally swift) are using.
        TXT_ENC = ['text/plain', 'text/html', 'text/html; charset=utf-8',
                   'text/plain; charset=utf-8']
        XML_ENC = ['application/xml', 'application/xml; charset=utf-8']

        if ctype.lower() in JSON_ENC or ctype.lower() in XML_ENC:
            parse_resp = True
        elif ctype.lower() in TXT_ENC:
            parse_resp = False
        else:
            raise exceptions.InvalidContentType(str(resp.status))

        if resp.status == 401 or resp.status == 403:
            raise exceptions.Unauthorized()

        if resp.status == 404:
            raise exceptions.NotFound(resp_body)

        if resp.status == 400:
            if parse_resp:
                resp_body = self._parse_resp(resp_body)
            raise exceptions.BadRequest(resp_body)

        if resp.status == 409:
            if parse_resp:
                resp_body = self._parse_resp(resp_body)
            raise exceptions.Conflict(resp_body)

        if resp.status == 413:
            if parse_resp:
                resp_body = self._parse_resp(resp_body)
            if self.is_absolute_limit(resp, resp_body):
                raise exceptions.OverLimit(resp_body)
            else:
                raise exceptions.RateLimitExceeded(resp_body)

        if resp.status == 422:
            if parse_resp:
                resp_body = self._parse_resp(resp_body)
            raise exceptions.UnprocessableEntity(resp_body)

        if resp.status in (500, 501):
            message = resp_body
            if parse_resp:
                try:
                    resp_body = self._parse_resp(resp_body)
                except ValueError:
                    # If response body is a non-json string message.
                    # Use resp_body as is and raise InvalidResponseBody
                    # exception.
                    raise exceptions.InvalidHTTPResponseBody(message)
                else:
                    if isinstance(resp_body, dict):
                        # I'm seeing both computeFault
                        # and cloudServersFault come back.
                        # Will file a bug to fix, but leave as is for now.
                        if 'cloudServersFault' in resp_body:
                            message = resp_body['cloudServersFault']['message']
                        elif 'computeFault' in resp_body:
                            message = resp_body['computeFault']['message']
                        elif 'error' in resp_body:  # Keystone errors
                            message = resp_body['error']['message']
                            raise exceptions.IdentityError(message)
                        elif 'message' in resp_body:
                            message = resp_body['message']
                    else:
                        message = resp_body

            raise exceptions.ServerFault(message)

        if resp.status >= 400:
            raise exceptions.UnexpectedResponseCode(str(resp.status))

    def is_absolute_limit(self, resp, resp_body):
        if (not isinstance(resp_body, collections.Mapping) or
                'retry-after' not in resp):
            return True
        if self._get_type() is "json":
            over_limit = resp_body.get('overLimit', None)
            if not over_limit:
                return True
            return 'exceed' in over_limit.get('message', 'blabla')
        elif self._get_type() is "xml":
            return 'exceed' in resp_body.get('message', 'blabla')

    def wait_for_resource_deletion(self, id):
        """Waits for a resource to be deleted."""
        start_time = int(time.time())
        while True:
            if self.is_resource_deleted(id):
                return
            if int(time.time()) - start_time >= self.build_timeout:
                raise exceptions.TimeoutException
            time.sleep(self.build_interval)

    def is_resource_deleted(self, id):
        """
        Subclasses override with specific deletion detection.
        """
        message = ('"%s" does not implement is_resource_deleted'
                   % self.__class__.__name__)
        raise NotImplementedError(message)

    @classmethod
    def validate_response(cls, schema, resp, body):
        # Only check the response if the status code is a success code
        # TODO(cyeoh): Eventually we should be able to verify that a failure
        # code if it exists is something that we expect. This is explicitly
        # declared in the V3 API and so we should be able to export this in
        # the response schema. For now we'll ignore it.
        if str(resp.status).startswith('2'):
            response_code = schema['status_code']
            if resp.status not in response_code:
                msg = ("The status code(%s) is different than the expected "
                       "one(%s)") % (resp.status, response_code)
                raise exceptions.InvalidHttpSuccessCode(msg)
            response_schema = schema.get('response_body')
            if response_schema:
                try:
                    jsonschema.validate(body, response_schema)
                except jsonschema.ValidationError as ex:
                    msg = ("HTTP response body is invalid (%s)") % ex
                    raise exceptions.InvalidHTTPResponseBody(msg)
            else:
                if body:
                    msg = ("HTTP response body should not exist (%s)") % body
                    raise exceptions.InvalidHTTPResponseBody(msg)


class NegativeRestClient(RestClient):
    """
    Version of RestClient that does not raise exceptions.
    """
    def _error_checker(self, method, url,
                       headers, body, resp, resp_body):
        pass

    def send_request(self, method, url_template, resources, body=None):
        url = url_template % tuple(resources)
        if method == "GET":
            resp, body = self.get(url)
        elif method == "POST":
            resp, body = self.post(url, body)
        elif method == "PUT":
            resp, body = self.put(url, body)
        elif method == "PATCH":
            resp, body = self.patch(url, body)
        elif method == "HEAD":
            resp, body = self.head(url)
        elif method == "DELETE":
            resp, body = self.delete(url)
        elif method == "COPY":
            resp, body = self.copy(url)
        else:
            assert False

        return resp, body