Analyze API and related test suite using the external yaml definition
In the future the yaml files should be in separate repo.
This commit is contained in:
Honza Kral
+48
-1
@@ -85,9 +85,56 @@ class NamespacedClient(object):
|
||||
return self.client.transport
|
||||
|
||||
class ClusterClient(NamespacedClient):
|
||||
pass
|
||||
@query_params('level', 'local', 'master_timeout', 'timeout', 'wait_for_active_shards', 'wait_for_nodes', 'wait_for_relocating_shards', 'wait_for_status')
|
||||
def health(self, index=None, params=None):
|
||||
"""
|
||||
The cluster health API allows to get a very simple status on the health of the cluster.
|
||||
http://elasticsearch.org/guide/reference/api/admin-cluster-health/
|
||||
|
||||
:arg index: Limit the information returned to a specific index
|
||||
:arg level: Specify the level of detail for returned information, default u'cluster'
|
||||
:arg local: Return local information, do not retrieve the state from master node (default: false)
|
||||
:arg master_timeout: Explicit operation timeout for connection to master node
|
||||
:arg timeout: Explicit operation timeout
|
||||
:arg wait_for_active_shards: Wait until the specified number of shards is active
|
||||
:arg wait_for_nodes: Wait until the specified number of nodes is available
|
||||
:arg wait_for_relocating_shards: Wait until the specified number of relocating shards is finished
|
||||
:arg wait_for_status: Wait until cluster is in a specific state, default None
|
||||
"""
|
||||
status, data = self.transport.perform_request('GET', _make_path('_cluster', 'health', index), params=params)
|
||||
return data
|
||||
|
||||
class InidicesClient(NamespacedClient):
|
||||
@query_params('analyzer', 'field', 'filters', 'format', 'index', 'prefer_local', 'text', 'tokenizer')
|
||||
def analyze(self, index=None, body=None, params=None):
|
||||
"""
|
||||
Performs the analysis process on a text and return the tokens breakdown of the text.
|
||||
http://www.elasticsearch.org/guide/reference/api/admin-indices-analyze/
|
||||
|
||||
:arg index: The name of the index to scope the operation
|
||||
:arg analyzer: The name of the analyzer to use
|
||||
:arg field: The name of the field to
|
||||
:arg filters: A comma-separated list of filters to use for the analysis
|
||||
:arg index: The name of the index to scope the operation
|
||||
:arg prefer_local: With `true`, specify that a local shard should be used if available, with `false`, use a random shard (default: true)
|
||||
:arg text: The text on which the analysis should be performed (when request body is not used)
|
||||
:arg tokenizer: The name of the tokenizer to use for the analysis
|
||||
"""
|
||||
status, data = self.transport.perform_request('GET', _make_path(index, '_analyze'), params=params, body=body)
|
||||
return data
|
||||
|
||||
@query_params('ignore_indices')
|
||||
def refresh(self, index=None, params=None):
|
||||
"""
|
||||
The refresh API allows to explicitly refresh one or more index, making all operations performed since the last refresh available for search.
|
||||
http://www.elasticsearch.org/guide/reference/api/admin-indices-refresh/
|
||||
|
||||
:arg index: A comma-separated list of index names; use `_all` or empty string to perform the operation on all indices
|
||||
:arg ignore_indices: When performed on multiple indices, allows to ignore `missing` ones, default u'none'
|
||||
"""
|
||||
status, data = self.transport.perform_request('POST', _make_path(index, '_refresh'), params=params)
|
||||
return data
|
||||
|
||||
@query_params('timeout')
|
||||
def create(self, index, body=None, params=None):
|
||||
"""
|
||||
|
||||
@@ -0,0 +1,52 @@
|
||||
---
|
||||
"Analyze API setup":
|
||||
- do:
|
||||
indices.create:
|
||||
index: test
|
||||
body:
|
||||
mappings:
|
||||
test:
|
||||
properties:
|
||||
text:
|
||||
type: string
|
||||
analyzer: whitespace
|
||||
|
||||
---
|
||||
"Analyze API text format":
|
||||
- do:
|
||||
indices.analyze:
|
||||
format: text
|
||||
text: tHE BLACK and white! AND red
|
||||
- is:
|
||||
tokens: "[black:4->9:<ALPHANUM>]\n\n4: \n[white:14->19:<ALPHANUM>]\n\n6: \n[red:25->28:<ALPHANUM>]\n"
|
||||
|
||||
---
|
||||
"Analyze API JSON format":
|
||||
- do:
|
||||
indices.analyze:
|
||||
text: Foo Bar
|
||||
- length: { tokens: 2 }
|
||||
- is: { tokens.0.token: foo }
|
||||
- is: { tokens.1.token: bar }
|
||||
|
||||
---
|
||||
"Analyze API JSON format - tokenizer and filter":
|
||||
- do:
|
||||
indices.analyze:
|
||||
filters: lowercase
|
||||
text: Foo Bar
|
||||
tokenizer: keyword
|
||||
- length: { tokens: 1 }
|
||||
- is: { tokens.0.token: foo bar }
|
||||
|
||||
---
|
||||
"Analyze API JSON format - index and field":
|
||||
- do:
|
||||
indices.analyze:
|
||||
field: text
|
||||
index: test
|
||||
text: Foo Bar!
|
||||
- length: { tokens: 2 }
|
||||
- is: { tokens.0.token: Foo }
|
||||
- is: { tokens.1.token: Bar! }
|
||||
|
||||
@@ -0,0 +1,129 @@
|
||||
"""
|
||||
Dynamically generated set of TestCases based on set of yaml files decribing
|
||||
some integration tests. These files are shared among all official Elasticsearch
|
||||
clients.
|
||||
"""
|
||||
from os import listdir
|
||||
from os.path import dirname, abspath, join
|
||||
import yaml
|
||||
from unittest import TestCase, SkipTest
|
||||
|
||||
from elasticsearch import Elasticsearch
|
||||
|
||||
|
||||
class InvalidActionType(SkipTest):
|
||||
pass
|
||||
|
||||
|
||||
class YamlTestCase(TestCase):
|
||||
def run_code(self, test):
|
||||
""" Execute an instruction based on it's type. """
|
||||
for action in test:
|
||||
self.assertEquals(1, len(action))
|
||||
action_type, action = action.items()[0]
|
||||
|
||||
if hasattr(self, 'run_' + action_type):
|
||||
getattr(self, 'run_' + action_type)(action)
|
||||
else:
|
||||
raise InvalidActionType(action_type)
|
||||
|
||||
def run_do(self, action):
|
||||
""" Perform an api call with given parameters. """
|
||||
self.assertEquals(1, len(action))
|
||||
|
||||
method, args = action.items()[0]
|
||||
|
||||
# locate api endpoint
|
||||
api = self.client
|
||||
for m in method.split('.'):
|
||||
self.assertTrue(hasattr(api, m))
|
||||
api = getattr(api, m)
|
||||
|
||||
self.last_response = api(**args)
|
||||
|
||||
def run_length(self, action):
|
||||
self.run_is(action, len)
|
||||
|
||||
def run_is(self, action, transform=None):
|
||||
""" Match part of last response to test data. """
|
||||
self.assertEquals(1, len(action))
|
||||
path, expected = action.items()[0]
|
||||
|
||||
# fetch the possibly nested value from last_response
|
||||
value = self.last_response
|
||||
for step in path.split('.'):
|
||||
if step.isdigit():
|
||||
step = int(step)
|
||||
self.assertIsInstance(value, list)
|
||||
self.assertGreater(len(value), step)
|
||||
else:
|
||||
self.assertIn(step, value)
|
||||
value = value[step]
|
||||
|
||||
# sometimes we need to transform the json value before comparing
|
||||
if transform:
|
||||
value = transform(value)
|
||||
|
||||
# compare target value
|
||||
self.assertEquals(expected, value)
|
||||
|
||||
def tearDown(self):
|
||||
# clean up everything
|
||||
self.client.indices.delete()
|
||||
|
||||
|
||||
|
||||
def construct_case(filename, name):
|
||||
"""
|
||||
Parse a definition of a test case from a yaml file and construct the
|
||||
TestCase subclass dynamically transforming the individual tests into test
|
||||
methods. Always use the first one as `setUp`.
|
||||
"""
|
||||
def get_test_method(name, test):
|
||||
def test_(self):
|
||||
self.run_code(test)
|
||||
|
||||
# remember the name as docstring so it will show up
|
||||
test_.__doc__ = name
|
||||
return test_
|
||||
|
||||
|
||||
def get_setUp(name, definition):
|
||||
def setUp(self):
|
||||
self.client = Elasticsearch(['localhost:9900'])
|
||||
self.last_response = None
|
||||
self.run_code(definition)
|
||||
|
||||
# make sure the cluster is ready
|
||||
self.client.cluster.health(wait_for_status='yellow')
|
||||
self.client.indices.refresh()
|
||||
|
||||
setUp.__doc__ = name
|
||||
return setUp
|
||||
|
||||
with open(filename) as f:
|
||||
tests = list(yaml.load_all(f))
|
||||
|
||||
|
||||
# take the first test as setUp method
|
||||
attrs = {'setUp' : get_setUp(*tests.pop(0).items()[0])}
|
||||
# create test methods for the rest
|
||||
for i, test in enumerate(tests):
|
||||
if not test:
|
||||
continue
|
||||
attrs['test_%d' % i] = get_test_method(*test.items()[0])
|
||||
|
||||
return type(name, (YamlTestCase, ), attrs)
|
||||
|
||||
|
||||
current_dir = abspath(dirname(__file__))
|
||||
# find all the test definitions in yaml files ...
|
||||
for filename in listdir(current_dir):
|
||||
if not filename.endswith('.yaml'):
|
||||
continue
|
||||
|
||||
# ... parse them
|
||||
name = 'Test' + filename.rsplit('.', 1)[0][3:].title()
|
||||
# and insert them into locals for test runner to find them
|
||||
locals()[name] = construct_case(join(current_dir, filename), name)
|
||||
|
||||
Reference in New Issue
Block a user