def test_force_unicode(self):
self.assertEqual(force_unicode(b'Hello \xe2\x98\x83'), 'Hello ☃')
# Don't mangle, it's already Unicode.
self.assertEqual(force_unicode('Hello ☃'), 'Hello ☃')
self.assertEqual(force_unicode(1), '1', "force_unicode() should convert ints")
self.assertEqual(force_unicode(1.0), '1.0', "force_unicode() should convert floats")
self.assertEqual(force_unicode(None), 'None', 'force_unicode() should convert None')
def test_force_unicode(self):
self.assertEqual(force_unicode(b'Hello \xe2\x98\x83'), 'Hello ☃')
# Don't mangle, it's already Unicode.
self.assertEqual(force_unicode('Hello ☃'), 'Hello ☃')
self.assertEqual(force_unicode(1), '1', "force_unicode() should convert ints")
self.assertEqual(force_unicode(1.0), '1.0', "force_unicode() should convert floats")
self.assertEqual(force_unicode(None), 'None', 'force_unicode() should convert None')
def test_force_unicode(self):
self.assertEqual(force_unicode(b"Hello \xe2\x98\x83"), "Hello ☃")
# Don't mangle, it's already Unicode.
self.assertEqual(force_unicode("Hello ☃"), "Hello ☃")
self.assertEqual(force_unicode(1), "1", "force_unicode() should convert ints")
self.assertEqual(force_unicode(1.0), "1.0", "force_unicode() should convert floats")
self.assertEqual(force_unicode(None), "None", "force_unicode() should convert None")
def test_force_unicode(self):
self.assertEqual(force_unicode(b"Hello \xe2\x98\x83"), "Hello ☃")
# Don't mangle, it's already Unicode.
self.assertEqual(force_unicode("Hello ☃"), "Hello ☃")
self.assertEqual(force_unicode(1), "1",
"force_unicode() should convert ints")
self.assertEqual(force_unicode(1.0), "1.0",
"force_unicode() should convert floats")
self.assertEqual(force_unicode(None), "None",
"force_unicode() should convert None")
def test_safe_urlencode(self):
self.assertEqual(
force_unicode(unquote_plus(safe_urlencode({"test": "Hello ☃! Helllo world!"}))),
"test=Hello ☃! Helllo world!",
)
self.assertEqual(
force_unicode(unquote_plus(safe_urlencode({"test": ["Hello ☃!", "Helllo world!"]}, True))),
"test=Hello \u2603!&test=Helllo world!",
)
self.assertEqual(
force_unicode(unquote_plus(safe_urlencode({"test": ("Hello ☃!", "Helllo world!")}, True))),
"test=Hello \u2603!&test=Helllo world!",
)
def _to_python(self, value):
"""
Converts values from Solr to native Python values.
"""
if isinstance(value, (int, list, tuple, float, long, complex)):
return value
if value == 'true':
return True
elif value == 'false':
return False
is_string = False
if isinstance(value, str):
value = force_unicode(value)
if isinstance(value, basestring):
is_string = True
if is_string == True:
possible_datetime = DATETIME_REGEX.search(value)
if possible_datetime:
date_values = possible_datetime.groupdict()
for dk, dv in date_values.items():
date_values[dk] = int(dv)
return datetime.datetime(date_values['year'], date_values['month'], date_values['day'], date_values['hour'], date_values['minute'], date_values['second'])
return value
def put(self, data):
'''
Supports partial update of solr.
'''
if data is None or len(data) == 0:
return
# Update Solr: (Mostly from pysolr.Solr code.)
# Generate the exact update command in xml -
# <add>
# <doc>
# <field name="id">1</field>
# <field name="memory_used" update="set">832</field>
# </doc>
# </add>
data_xml = ET.Element('add')
for doc_update in data:
doc_element = ET.Element('doc')
id_field = ET.Element('field', **{'name':'id'})
id_field.text = str(doc_update['id'])
doc_element.append(id_field)
for field in doc_update['fields']:
field_xml = ET.Element('field', **{'name':field['name'], 'update':field['command']})
field_xml.text = str(field['value'])
doc_element.append(field_xml)
data_xml.append(doc_element)
# This returns a bytestring.
data_xml_str = ET.tostring(data_xml, encoding='utf-8')
# Convert back to Unicode.
data_xml_str = pysolr.force_unicode(data_xml_str)
try:
solr = session.get_solr_interface(self.solr_url)
solr._update(data_xml_str)
except:
LOG.exception('Failed to add to solr.')
raise
def test_safe_urlencode(self):
self.assertEqual(
force_unicode(
unquote_plus(safe_urlencode({'test':
'Hello ☃! Helllo world!'}))),
'test=Hello ☃! Helllo world!')
self.assertEqual(
force_unicode(
unquote_plus(
safe_urlencode({'test': ['Hello ☃!', 'Helllo world!']},
True))),
"test=Hello \u2603!&test=Helllo world!")
self.assertEqual(
force_unicode(
unquote_plus(
safe_urlencode({'test': ('Hello ☃!', 'Helllo world!')},
True))),
"test=Hello \u2603!&test=Helllo world!")
def test__build_doc_with_sets(self):
doc = {"id": "doc_1", "title": "Set test doc", "tags": {"alpha", "beta"}}
doc_xml = force_unicode(
ElementTree.tostring(self.solr._build_doc(doc), encoding="utf-8")
)
self.assertIn('<field name="id">doc_1</field>', doc_xml)
self.assertIn('<field name="title">Set test doc</field>', doc_xml)
self.assertIn('<field name="tags">alpha</field>', doc_xml)
self.assertIn('<field name="tags">beta</field>', doc_xml)
self.assertEqual(len(doc_xml), 144)
def test__build_doc(self):
doc = {
'id': 'doc_1',
'title': 'Example doc ☃ 1',
'price': 12.59,
'popularity': 10,
}
doc_xml = force_unicode(ElementTree.tostring(self.solr._build_doc(doc), encoding='utf-8'))
self.assertTrue('<field name="title">Example doc ☃ 1</field>' in doc_xml)
self.assertTrue('<field name="id">doc_1</field>' in doc_xml)
self.assertEqual(len(doc_xml), 152)
def test__build_doc(self):
doc = {
'id': 'doc_1',
'title': 'Example doc ☃ 1',
'price': 12.59,
'popularity': 10,
}
doc_xml = force_unicode(ElementTree.tostring(self.solr._build_doc(doc), encoding='utf-8'))
self.assertTrue('<field name="title">Example doc ☃ 1</field>' in doc_xml)
self.assertTrue('<field name="id">doc_1</field>' in doc_xml)
self.assertEqual(len(doc_xml), 152)
def test__build_doc(self):
doc = {
"id": "doc_1",
"title": "Example doc ☃ 1",
"price": 12.59,
"popularity": 10,
}
doc_xml = force_unicode(
ElementTree.tostring(self.solr._build_doc(doc), encoding="utf-8"))
self.assertIn('<field name="title">Example doc ☃ 1</field>', doc_xml)
self.assertIn('<field name="id">doc_1</field>', doc_xml)
self.assertEqual(len(doc_xml), 152)
def test_safe_urlencode(self):
self.assertEqual(
force_unicode(
unquote_plus(safe_urlencode({"test":
"Hello ☃! Helllo world!"}))),
"test=Hello ☃! Helllo world!",
)
self.assertEqual(
force_unicode(
unquote_plus(
safe_urlencode({"test": ["Hello ☃!", "Helllo world!"]},
True))),
"test=Hello \u2603!&test=Helllo world!",
)
self.assertEqual(
force_unicode(
unquote_plus(
safe_urlencode({"test": ("Hello ☃!", "Helllo world!")},
True))),
"test=Hello \u2603!&test=Helllo world!",
)
def test__build_doc_with_sets(self):
doc = {
'id': 'doc_1',
'title': 'Set test doc',
'tags': set(['alpha', 'beta']),
}
doc_xml = force_unicode(ElementTree.tostring(self.solr._build_doc(doc), encoding='utf-8'))
self.assertTrue('<field name="id">doc_1</field>' in doc_xml)
self.assertTrue('<field name="title">Set test doc</field>' in doc_xml)
self.assertTrue('<field name="tags">alpha</field>' in doc_xml)
self.assertTrue('<field name="tags">beta</field>' in doc_xml)
self.assertEqual(len(doc_xml), 144)
def test__build_doc_with_sets(self):
doc = {
'id': 'doc_1',
'title': 'Set test doc',
'tags': set(['alpha', 'beta']),
}
doc_xml = force_unicode(ElementTree.tostring(self.solr._build_doc(doc), encoding='utf-8'))
self.assertTrue('<field name="id">doc_1</field>' in doc_xml)
self.assertTrue('<field name="title">Set test doc</field>' in doc_xml)
self.assertTrue('<field name="tags">alpha</field>' in doc_xml)
self.assertTrue('<field name="tags">beta</field>' in doc_xml)
self.assertEqual(len(doc_xml), 144)
def add(solr,
docs,
dsId,
commit=True,
boost=None,
commitWithin="1000",
waitFlush=None,
waitSearcher=None):
"""
Adds or updates documents.
Requires ``docs``, which is a list of dictionaries. Each key is the
field name and each value is the value to index.
Optionally accepts ``commit``. Default is ``True``.
Optionally accepts ``boost``. Default is ``None``.
Optionally accepts ``commitWithin``. Default is ``None``.
Optionally accepts ``waitFlush``. Default is ``None``.
Optionally accepts ``waitSearcher``. Default is ``None``.
Usage::
solr.add([
{
"id": "doc_1",
"title": "A test document",
},
{
"id": "doc_2",
"title": "The Banana: Tasty or Dangerous?",
},
])
"""
start_time = time.time()
#self.log.debug("Starting to build add request...")
message = ET.Element('add')
if commitWithin:
message.set('commitWithin', commitWithin)
for doc in docs:
message.append(solr._build_doc(doc, boost=boost))
# This returns a bytestring. Ugh.
m = ET.tostring(message, encoding='utf-8')
# Convert back to Unicode please.
m = pysolr.force_unicode(m)
#print "Indexing to: " + dsId
end_time = time.time()
#self.log.debug("Built add request of %s docs in %0.2f seconds.", len(message), end_time - start_time)
return update(solr,
m,
dsId,
commit=commit,
waitFlush=waitFlush,
waitSearcher=waitSearcher)
def test__build_xml_doc_with_empty_values(self):
doc = {
"id": "doc_1",
"title": "",
"price": None,
"tags": [],
}
doc_xml = force_unicode(
ElementTree.tostring(self.solr._build_xml_doc(doc), encoding="utf-8")
)
self.assertNotIn('<field name="title" />', doc_xml)
self.assertNotIn('<field name="price" />', doc_xml)
self.assertNotIn('<field name="tags" />', doc_xml)
self.assertIn('<field name="id">doc_1</field>', doc_xml)
self.assertEqual(len(doc_xml), 41)
def add(solr, docs, dsId, commit=True, boost=None, commitWithin="1000", waitFlush=None, waitSearcher=None):
"""
Adds or updates documents.
Requires ``docs``, which is a list of dictionaries. Each key is the
field name and each value is the value to index.
Optionally accepts ``commit``. Default is ``True``.
Optionally accepts ``boost``. Default is ``None``.
Optionally accepts ``commitWithin``. Default is ``None``.
Optionally accepts ``waitFlush``. Default is ``None``.
Optionally accepts ``waitSearcher``. Default is ``None``.
Usage::
solr.add([
{
"id": "doc_1",
"title": "A test document",
},
{
"id": "doc_2",
"title": "The Banana: Tasty or Dangerous?",
},
])
"""
start_time = time.time()
#self.log.debug("Starting to build add request...")
message = ET.Element('add')
if commitWithin:
message.set('commitWithin', commitWithin)
for doc in docs:
message.append(solr._build_doc(doc, boost=boost))
# This returns a bytestring. Ugh.
m = ET.tostring(message, encoding='utf-8')
# Convert back to Unicode please.
m = pysolr.force_unicode(m)
#print "Indexing to: " + dsId
end_time = time.time()
#self.log.debug("Built add request of %s docs in %0.2f seconds.", len(message), end_time - start_time)
return update(solr, m, dsId, commit=commit, waitFlush=waitFlush, waitSearcher=waitSearcher)
def test__build_xml_doc_with_empty_values_and_field_updates(self):
doc = {
"id": "doc_1",
"title": "",
"price": None,
"tags": [],
}
fieldUpdates = {
"title": "set",
"tags": "set",
}
doc_xml = force_unicode(
ElementTree.tostring(
self.solr._build_xml_doc(doc, fieldUpdates=fieldUpdates),
encoding="utf-8",
)
)
self.assertIn('<field name="title" null="true" update="set" />', doc_xml)
self.assertNotIn('<field name="price" />', doc_xml)
self.assertIn('<field name="tags" null="true" update="set" />', doc_xml)
self.assertIn('<field name="id">doc_1</field>', doc_xml)
self.assertEqual(len(doc_xml), 134)
def _send_request(self,
method,
path='',
body=None,
headers=None,
files=None):
"""
Copy and paste of the base (pysolr version 3.2.0) _send_request()
method except for the resp = requests_method() line, which
passes along the auth information.
"""
url = self._create_full_url(path)
method = method.lower()
log_body = body
if headers is None:
headers = {}
if log_body is None:
log_body = ''
elif not isinstance(log_body, str):
log_body = repr(body)
self.log.debug("Starting request to '%s' (%s) with body '%s'...", url,
method, log_body[:10])
start_time = time.time()
try:
requests_method = getattr(self.session, method, 'get')
except AttributeError:
err = "Unable to send HTTP method '{0}.".format(method)
raise pysolr.SolrError(err)
try:
bytes_body = body
if bytes_body is not None:
bytes_body = pysolr.force_bytes(body)
resp = requests_method(url,
data=bytes_body,
headers=headers,
files=files,
timeout=self.timeout,
auth=self.auth)
except requests.exceptions.Timeout as err:
error_message = "Connection to server '%s' timed out: %s"
self.log.error(error_message, url, err, exc_info=True)
raise pysolr.SolrError(error_message % (url, err))
except requests.exceptions.ConnectionError as err:
error_message = "Failed to connect to server at '%s', are you " \
"sure that URL is correct? Checking it in a " \
"browser might help: %s"
params = (url, err)
self.log.error(error_message, *params, exc_info=True)
raise pysolr.SolrError(error_message % params)
end_time = time.time()
self.log.info("Finished '%s' (%s) with body '%s' in %0.3f seconds.",
url, method, log_body[:10], end_time - start_time)
if int(resp.status_code) != 200:
error_message = self._extract_error(resp)
data = {
'data': {
'headers': resp.headers,
'response': resp.content
}
}
self.log.error(error_message, extra=data)
raise pysolr.SolrError(error_message)
return pysolr.force_unicode(resp.content)
def _send_request(self, method, path='', body=None, headers=None,
files=None):
"""
Copy and paste of the base (pysolr version 3.2.0) _send_request()
method except for the resp = requests_method() line, which
passes along the auth information.
"""
url = self._create_full_url(path)
method = method.lower()
log_body = body
if headers is None:
headers = {}
if log_body is None:
log_body = ''
elif not isinstance(log_body, str):
log_body = repr(body)
self.log.debug("Starting request to '%s' (%s) with body '%s'...", url,
method, log_body[:10])
start_time = time.time()
try:
requests_method = getattr(self.session, method, 'get')
except AttributeError:
err = "Unable to send HTTP method '{0}.".format(method)
raise pysolr.SolrError(err)
try:
bytes_body = body
if bytes_body is not None:
bytes_body = pysolr.force_bytes(body)
resp = requests_method(url, data=bytes_body, headers=headers,
files=files, timeout=self.timeout,
auth=self.auth)
except requests.exceptions.Timeout as err:
error_message = "Connection to server '%s' timed out: %s"
self.log.error(error_message, url, err, exc_info=True)
raise pysolr.SolrError(error_message % (url, err))
except requests.exceptions.ConnectionError as err:
error_message = "Failed to connect to server at '%s', are you " \
"sure that URL is correct? Checking it in a " \
"browser might help: %s"
params = (url, err)
self.log.error(error_message, *params, exc_info=True)
raise pysolr.SolrError(error_message % params)
end_time = time.time()
self.log.info("Finished '%s' (%s) with body '%s' in %0.3f seconds.",
url, method, log_body[:10], end_time - start_time)
if int(resp.status_code) != 200:
error_message = self._extract_error(resp)
data = {'data': {'headers': resp.headers, 'response': resp.content}}
self.log.error(error_message, extra=data)
raise pysolr.SolrError(error_message)
return pysolr.force_unicode(resp.content)
def _to_python(self, value):
"""
Converts values from Solr to native Python values.
"""
if value is None:
return value
if isinstance(value, (int, float, complex)):
return value
is_list = isinstance(value, (list, tuple))
values_processed = []
values_to_process = []
if isinstance(value, (list, tuple)):
# Clone the value
values_to_process = value[:]
else:
values_to_process.append(value)
for value in values_to_process:
if value == "true":
values_processed.append(True)
continue
elif value == "false":
values_processed.append(False)
continue
is_string = False
if IS_PY3:
if isinstance(value, bytes):
value = force_unicode(value)
if isinstance(value, str):
is_string = True
else:
if isinstance(value, str):
value = force_unicode(value)
if isinstance(value, string_types):
is_string = True
if is_string:
possible_datetime = DATETIME_REGEX.search(value)
if possible_datetime:
date_values = possible_datetime.groupdict()
for dk, dv in date_values.items():
date_values[dk] = int(dv)
values_processed.append(
datetime.datetime(
date_values["year"],
date_values["month"],
date_values["day"],
date_values["hour"],
date_values["minute"],
date_values["second"],
)
)
continue
# elif ObjectId.is_valid(value):
# values_processed.append(value)
# continue
elif is_valid_uuid(value, version=4):
values_processed.append(value)
continue
elif is_valid_uuid(value, version=3):
values_processed.append(value)
continue
elif is_valid_uuid(value, version=2):
values_processed.append(value)
continue
elif is_valid_uuid(value, version=1):
values_processed.append(value)
continue
try:
# This is slightly gross but it's hard to tell otherwise what
# the string's original type might have been.
values_processed.append(ast.literal_eval(value))
except (ValueError, SyntaxError):
# If it fails, continue on.
pass
values_processed.append(value)
return values_processed if is_list else values_processed[0]
def test__build_doc(self):
doc = {"id": "doc_1", "title": "Example doc ☃ 1", "price": 12.59, "popularity": 10}
doc_xml = force_unicode(ET.tostring(self.solr._build_doc(doc), encoding="utf-8"))
self.assertTrue('<field name="title">Example doc ☃ 1</field>' in doc_xml)
self.assertTrue('<field name="id">doc_1</field>' in doc_xml)
self.assertEqual(len(doc_xml), 152)
def test_safe_urlencode(self):
self.assertEqual(force_unicode(unquote_plus(safe_urlencode({'test': 'Hello ☃! Helllo world!'}))), 'test=Hello ☃! Helllo world!')
self.assertEqual(force_unicode(unquote_plus(safe_urlencode({'test': ['Hello ☃!', 'Helllo world!']}, True))), "test=Hello \u2603!&test=Helllo world!")
self.assertEqual(force_unicode(unquote_plus(safe_urlencode({'test': ('Hello ☃!', 'Helllo world!')}, True))), "test=Hello \u2603!&test=Helllo world!")
