Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

fix: Increase the compatibility of id structure between the Databuilder and the Metadata Library #445

Merged
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 6 additions & 3 deletions databuilder/extractor/neptune_search_data_extractor.py
Original file line number Diff line number Diff line change
Expand Up @@ -142,7 +142,7 @@ def _user_search_query(graph: GraphTraversalSource, tag_filter: str) -> List[Dic

def _dashboard_search_query(graph: GraphTraversalSource, tag_filter: str) -> List[Dict]:
traversal = graph.V().hasLabel(DashboardMetadata.DASHBOARD_NODE_LABEL)
traversal = traversal.has('full_name')
traversal = traversal.has('name')
if tag_filter:
traversal = traversal.where('published_tag', tag_filter)

Expand Down Expand Up @@ -184,13 +184,16 @@ def _dashboard_search_query(graph: GraphTraversalSource, tag_filter: str) -> Lis
__.constant('')
)) # group_description
traversal = traversal.by(
__.out(DashboardMetadata.DASHBOARD_DASHBOARD_GROUP_RELATION_TYPE).values('group_url')
__.out(DashboardMetadata.DASHBOARD_DASHBOARD_GROUP_RELATION_TYPE).values('dashboard_group_url')
) # group_url
traversal = traversal.by('dashboard_url') # dashboard_url
traversal = traversal.by('key') # uri

traversal = traversal.by(
__.out('EXECUTED').has('key', TextP.endingWith('_last_successful_execution')).values('timestamp')
__.coalesce(
__.out('EXECUTED').has('key', TextP.endingWith('_last_successful_execution')).values('timestamp'),
__.constant('')
)
) # last_successful_run_timestamp
traversal = traversal.by(
__.out(DashboardQuery.DASHBOARD_QUERY_RELATION_TYPE).values('name').dedup().fold()
Expand Down
2 changes: 1 addition & 1 deletion databuilder/models/es_last_updated.py
Original file line number Diff line number Diff line change
Expand Up @@ -16,7 +16,7 @@ class ESLastUpdated(GraphSerializable):

LABEL = 'Updatedtimestamp'
KEY = 'amundsen_updated_timestamp'
LATEST_TIMESTAMP = 'latest_timestmap'
LATEST_TIMESTAMP = 'latest_timestamp'

def __init__(self,
timestamp: int,
Expand Down
2 changes: 1 addition & 1 deletion databuilder/models/table_metadata.py
Original file line number Diff line number Diff line change
Expand Up @@ -157,7 +157,7 @@ class ColumnMetadata:
COLUMN_NODE_LABEL = 'Column'
COLUMN_KEY_FORMAT = '{db}://{cluster}.{schema}/{tbl}/{col}'
COLUMN_NAME = 'name'
COLUMN_TYPE = 'type'
COLUMN_TYPE = 'col_type'
COLUMN_ORDER = 'sort_order'
COLUMN_DESCRIPTION = 'description'
COLUMN_DESCRIPTION_FORMAT = '{db}://{cluster}.{schema}/{tbl}/{col}/{description_id}'
Expand Down
8 changes: 4 additions & 4 deletions databuilder/models/table_stats.py
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,7 @@ class TableColumnStats(GraphSerializable):
"""
LABEL = 'Stat'
KEY_FORMAT = '{db}://{cluster}.{schema}' \
'/{table}/{col}/{stat_name}/'
'/{table}/{col}/{stat_type}/'
STAT_Column_RELATION_TYPE = 'STAT_OF'
Column_STAT_RELATION_TYPE = 'STAT'

Expand All @@ -40,7 +40,7 @@ def __init__(self,
self.start_epoch = start_epoch
self.end_epoch = end_epoch
self.cluster = cluster
self.stat_name = stat_name
self.stat_type = stat_name
self.stat_val = str(stat_val)
self._node_iter = iter(self.create_nodes())
self._relation_iter = iter(self.create_relation())
Expand All @@ -64,7 +64,7 @@ def get_table_stat_model_key(self) -> str:
schema=self.schema,
table=self.table,
col=self.col_name,
stat_name=self.stat_name)
stat_type=self.stat_type)

def get_col_key(self) -> str:
# no cluster, schema info from the input
Expand All @@ -84,7 +84,7 @@ def create_nodes(self) -> List[GraphNode]:
label=TableColumnStats.LABEL,
attributes={
'stat_val': self.stat_val,
'stat_name': self.stat_name,
'stat_type': self.stat_type,
'start_epoch': self.start_epoch,
'end_epoch': self.end_epoch,
}
Expand Down
61 changes: 47 additions & 14 deletions databuilder/serializers/neptune_serializer.py
Original file line number Diff line number Diff line change
Expand Up @@ -17,6 +17,7 @@
NEPTUNE_RELATIONSHIP_HEADER_TO = "~to"


METADATA_KEY_PROPERTY_NAME = 'key:String(single)'
# last seen property names
NEPTUNE_LAST_EXTRACTED_AT_NODE_PROPERTY_NAME = "last_extracted_datetime"
NEPTUNE_LAST_EXTRACTED_AT_NODE_PROPERTY_NAME_BULK_LOADER_FORMAT = "{name}:Date(single)".format(
Expand Down Expand Up @@ -46,30 +47,33 @@ def convert_relationship(relationship: Optional[GraphRelationship]) -> List[Dict
if relationship.start_key == '' or relationship.end_key == '':
return []

relation_id = "{from_vertex_id}_{to_vertex_id}_{label}".format(
from_vertex_id=relationship.start_key,
to_vertex_id=relationship.end_key,
label=relationship.type
neptune_start_key = "{label}:{key}".format(
label=relationship.start_label,
key=relationship.start_key
)
relation_id_reverse = "{from_vertex_id}_{to_vertex_id}_{label}".format(
from_vertex_id=relationship.end_key,
to_vertex_id=relationship.start_key,
label=relationship.reverse_type
neptune_end_key = "{label}:{key}".format(
label=relationship.end_label,
key=relationship.end_key
)
relation_id = get_forward_relationship_id(relationship)
relation_id_reverse = get_reverse_relationship_id(relationship)
current_string_time = datetime.utcnow().strftime('%Y-%m-%dT%H:%M:%S')
forward_relationship_doc = {
NEPTUNE_HEADER_ID: relation_id,
NEPTUNE_RELATIONSHIP_HEADER_FROM: relationship.start_key,
NEPTUNE_RELATIONSHIP_HEADER_TO: relationship.end_key,
METADATA_KEY_PROPERTY_NAME: relation_id,
NEPTUNE_RELATIONSHIP_HEADER_FROM: neptune_start_key,
NEPTUNE_RELATIONSHIP_HEADER_TO: neptune_end_key,
NEPTUNE_HEADER_LABEL: relationship.type,
NEPTUNE_LAST_EXTRACTED_AT_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT: current_string_time,
NEPTUNE_CREATION_TYPE_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT: NEPTUNE_CREATION_TYPE_JOB
NEPTUNE_CREATION_TYPE_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT: NEPTUNE_CREATION_TYPE_JOB,

}

reverse_relationship_doc = {
NEPTUNE_HEADER_ID: relation_id_reverse,
NEPTUNE_RELATIONSHIP_HEADER_FROM: relationship.end_key,
NEPTUNE_RELATIONSHIP_HEADER_TO: relationship.start_key,
METADATA_KEY_PROPERTY_NAME: relation_id_reverse,
NEPTUNE_RELATIONSHIP_HEADER_FROM: neptune_end_key,
NEPTUNE_RELATIONSHIP_HEADER_TO: neptune_start_key,
NEPTUNE_HEADER_LABEL: relationship.reverse_type,
NEPTUNE_LAST_EXTRACTED_AT_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT: current_string_time,
NEPTUNE_CREATION_TYPE_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT: NEPTUNE_CREATION_TYPE_JOB
Expand All @@ -90,6 +94,26 @@ def convert_relationship(relationship: Optional[GraphRelationship]) -> List[Dict
]


def get_forward_relationship_id(relationship: GraphRelationship) -> str:
return "{label}:{from_vertex_label}:{from_vertex_id}_{to_vertex_label}:{to_vertex_id}".format(
from_vertex_id=relationship.start_key,
from_vertex_label=relationship.start_label,
to_vertex_id=relationship.end_key,
to_vertex_label=relationship.end_label,
label=relationship.type
)


def get_reverse_relationship_id(relationship: GraphRelationship) -> str:
return "{label}:{from_vertex_label}:{from_vertex_id}_{to_vertex_label}:{to_vertex_id}".format(
to_vertex_id=relationship.start_key,
to_vertex_label=relationship.start_label,
from_vertex_id=relationship.end_key,
from_vertex_label=relationship.end_label,
label=relationship.reverse_type
)


def convert_node(node: Optional[GraphNode]) -> Dict[str, Any]:
if node is None:
return {}
Expand All @@ -98,8 +122,10 @@ def convert_node(node: Optional[GraphNode]) -> Dict[str, Any]:
return {}

current_string_time = datetime.utcnow().strftime('%Y-%m-%dT%H:%M:%S')
node_id = get_node_id(node)
node_dict = {
NEPTUNE_HEADER_ID: node.key,
NEPTUNE_HEADER_ID: node_id,
METADATA_KEY_PROPERTY_NAME: node_id,
NEPTUNE_HEADER_LABEL: node.label,
NEPTUNE_LAST_EXTRACTED_AT_NODE_PROPERTY_NAME_BULK_LOADER_FORMAT: current_string_time,
NEPTUNE_CREATION_TYPE_NODE_PROPERTY_NAME_BULK_LOADER_FORMAT: NEPTUNE_CREATION_TYPE_JOB
Expand All @@ -117,6 +143,13 @@ def convert_node(node: Optional[GraphNode]) -> Dict[str, Any]:
return node_dict


def get_node_id(node: GraphNode) -> str:
return "{label}:{key}".format(
label=node.label,
key=node.key
)


def _get_neptune_type_for_value(value: Any) -> Optional[str]:
if isinstance(value, six.string_types):
return "String"
Expand Down
38 changes: 25 additions & 13 deletions tests/unit/models/dashboard/test_dashboard_chart.py
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@
)
from databuilder.serializers import neo4_serializer, neptune_serializer
from databuilder.serializers.neptune_serializer import (
NEPTUNE_CREATION_TYPE_JOB, NEPTUNE_CREATION_TYPE_NODE_PROPERTY_NAME_BULK_LOADER_FORMAT,
METADATA_KEY_PROPERTY_NAME, NEPTUNE_CREATION_TYPE_JOB, NEPTUNE_CREATION_TYPE_NODE_PROPERTY_NAME_BULK_LOADER_FORMAT,
NEPTUNE_CREATION_TYPE_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT, NEPTUNE_HEADER_ID, NEPTUNE_HEADER_LABEL,
NEPTUNE_LAST_EXTRACTED_AT_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT, NEPTUNE_RELATIONSHIP_HEADER_FROM,
NEPTUNE_RELATIONSHIP_HEADER_TO,
Expand Down Expand Up @@ -44,7 +44,8 @@ def test_create_nodes(self) -> None:
'LABEL': 'Chart'
}
neptune_expected = {
'~id': '_dashboard://gold.dg_id/d_id/query/q_id/chart/c_id',
'~id': 'Chart:_dashboard://gold.dg_id/d_id/query/q_id/chart/c_id',
METADATA_KEY_PROPERTY_NAME: 'Chart:_dashboard://gold.dg_id/d_id/query/q_id/chart/c_id',
'~label': 'Chart',
NEPTUNE_LAST_EXTRACTED_AT_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT: ANY,
NEPTUNE_CREATION_TYPE_NODE_PROPERTY_NAME_BULK_LOADER_FORMAT: NEPTUNE_CREATION_TYPE_JOB,
Expand Down Expand Up @@ -77,7 +78,8 @@ def test_create_nodes(self) -> None:
'url': 'http://gold.foo.bar/'
}
neptune_expected2 = {
'~id': '_dashboard://gold.dg_id/d_id/query/q_id/chart/c_id',
'~id': 'Chart:_dashboard://gold.dg_id/d_id/query/q_id/chart/c_id',
METADATA_KEY_PROPERTY_NAME: 'Chart:_dashboard://gold.dg_id/d_id/query/q_id/chart/c_id',
'~label': 'Chart',
'id:String(single)': 'c_id',
NEPTUNE_LAST_EXTRACTED_AT_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT: ANY,
Expand Down Expand Up @@ -112,26 +114,36 @@ def test_create_relation(self) -> None:
}

neptune_forward_expected = {
NEPTUNE_HEADER_ID: "{from_vertex_id}_{to_vertex_id}_{label}".format(
from_vertex_id=start_key,
to_vertex_id=end_key,
NEPTUNE_HEADER_ID: "{label}:{from_vertex_id}_{to_vertex_id}".format(
from_vertex_id="Query:" + start_key,
to_vertex_id="Chart:" + end_key,
label='HAS_CHART'
),
NEPTUNE_RELATIONSHIP_HEADER_FROM: start_key,
NEPTUNE_RELATIONSHIP_HEADER_TO: end_key,
METADATA_KEY_PROPERTY_NAME: "{label}:{from_vertex_id}_{to_vertex_id}".format(
from_vertex_id="Query:" + start_key,
to_vertex_id="Chart:" + end_key,
label='HAS_CHART'
),
NEPTUNE_RELATIONSHIP_HEADER_FROM: "Query:" + start_key,
NEPTUNE_RELATIONSHIP_HEADER_TO: "Chart:" + end_key,
NEPTUNE_HEADER_LABEL: 'HAS_CHART',
NEPTUNE_LAST_EXTRACTED_AT_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT: ANY,
NEPTUNE_CREATION_TYPE_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT: NEPTUNE_CREATION_TYPE_JOB
}

neptune_reversed_expected = {
NEPTUNE_HEADER_ID: "{from_vertex_id}_{to_vertex_id}_{label}".format(
from_vertex_id=end_key,
to_vertex_id=start_key,
NEPTUNE_HEADER_ID: "{label}:{from_vertex_id}_{to_vertex_id}".format(
from_vertex_id="Chart:" + end_key,
to_vertex_id="Query:" + start_key,
label='CHART_OF'
),
METADATA_KEY_PROPERTY_NAME: "{label}:{from_vertex_id}_{to_vertex_id}".format(
from_vertex_id="Chart:" + end_key,
to_vertex_id="Query:" + start_key,
label='CHART_OF'
),
NEPTUNE_RELATIONSHIP_HEADER_FROM: end_key,
NEPTUNE_RELATIONSHIP_HEADER_TO: start_key,
NEPTUNE_RELATIONSHIP_HEADER_FROM: "Chart:" + end_key,
NEPTUNE_RELATIONSHIP_HEADER_TO: "Query:" + start_key,
NEPTUNE_HEADER_LABEL: 'CHART_OF',
NEPTUNE_LAST_EXTRACTED_AT_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT: ANY,
NEPTUNE_CREATION_TYPE_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT: NEPTUNE_CREATION_TYPE_JOB
Expand Down
37 changes: 24 additions & 13 deletions tests/unit/models/dashboard/test_dashboard_last_modified.py
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@
)
from databuilder.serializers import neo4_serializer, neptune_serializer
from databuilder.serializers.neptune_serializer import (
NEPTUNE_CREATION_TYPE_JOB, NEPTUNE_CREATION_TYPE_NODE_PROPERTY_NAME_BULK_LOADER_FORMAT,
METADATA_KEY_PROPERTY_NAME, NEPTUNE_CREATION_TYPE_JOB, NEPTUNE_CREATION_TYPE_NODE_PROPERTY_NAME_BULK_LOADER_FORMAT,
NEPTUNE_CREATION_TYPE_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT, NEPTUNE_HEADER_ID, NEPTUNE_HEADER_LABEL,
NEPTUNE_LAST_EXTRACTED_AT_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT, NEPTUNE_RELATIONSHIP_HEADER_FROM,
NEPTUNE_RELATIONSHIP_HEADER_TO,
Expand Down Expand Up @@ -55,7 +55,8 @@ def test_neptune_dashboard_timestamp_nodes(self) -> None:
actual = self.dashboard_last_modified.create_next_node()
actual_neptune_serialized = neptune_serializer.convert_node(actual)
neptune_expected = {
NEPTUNE_HEADER_ID: self.expected_ts_key,
NEPTUNE_HEADER_ID: 'Timestamp:' + self.expected_ts_key,
METADATA_KEY_PROPERTY_NAME: 'Timestamp:' + self.expected_ts_key,
NEPTUNE_HEADER_LABEL: 'Timestamp',
NEPTUNE_LAST_EXTRACTED_AT_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT: ANY,
NEPTUNE_CREATION_TYPE_NODE_PROPERTY_NAME_BULK_LOADER_FORMAT: NEPTUNE_CREATION_TYPE_JOB,
Expand Down Expand Up @@ -87,31 +88,41 @@ def test_dashboard_owner_relations_neptune(self) -> None:
actual = self.dashboard_last_modified.create_next_relation()
actual_serialized = neptune_serializer.convert_relationship(actual)
neptune_forward_expected = {
NEPTUNE_HEADER_ID: "{from_vertex_id}_{to_vertex_id}_{label}".format(
from_vertex_id=self.expected_dashboard_key,
to_vertex_id=self.expected_ts_key,
NEPTUNE_HEADER_ID: "{label}:{from_vertex_id}_{to_vertex_id}".format(
from_vertex_id='Dashboard:' + self.expected_dashboard_key,
to_vertex_id='Timestamp:' + self.expected_ts_key,
label='LAST_UPDATED_AT'
),
NEPTUNE_RELATIONSHIP_HEADER_FROM: self.expected_dashboard_key,
NEPTUNE_RELATIONSHIP_HEADER_TO: self.expected_ts_key,
METADATA_KEY_PROPERTY_NAME: "{label}:{from_vertex_id}_{to_vertex_id}".format(
from_vertex_id='Dashboard:' + self.expected_dashboard_key,
to_vertex_id='Timestamp:' + self.expected_ts_key,
label='LAST_UPDATED_AT'
),
NEPTUNE_RELATIONSHIP_HEADER_FROM: 'Dashboard:' + self.expected_dashboard_key,
NEPTUNE_RELATIONSHIP_HEADER_TO: 'Timestamp:' + self.expected_ts_key,
NEPTUNE_HEADER_LABEL: 'LAST_UPDATED_AT',
NEPTUNE_LAST_EXTRACTED_AT_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT: ANY,
NEPTUNE_CREATION_TYPE_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT: NEPTUNE_CREATION_TYPE_JOB
}

neptune_reversed_expected = {
NEPTUNE_HEADER_ID: "{from_vertex_id}_{to_vertex_id}_{label}".format(
from_vertex_id=self.expected_ts_key,
to_vertex_id=self.expected_dashboard_key,
NEPTUNE_HEADER_ID: "{label}:{from_vertex_id}_{to_vertex_id}".format(
from_vertex_id='Timestamp:' + self.expected_ts_key,
to_vertex_id='Dashboard:' + self.expected_dashboard_key,
label='LAST_UPDATED_TIME_OF'
),
NEPTUNE_RELATIONSHIP_HEADER_FROM: self.expected_ts_key,
NEPTUNE_RELATIONSHIP_HEADER_TO: self.expected_dashboard_key,
METADATA_KEY_PROPERTY_NAME: "{label}:{from_vertex_id}_{to_vertex_id}".format(
from_vertex_id='Timestamp:' + self.expected_ts_key,
to_vertex_id='Dashboard:' + self.expected_dashboard_key,
label='LAST_UPDATED_TIME_OF'
),
NEPTUNE_RELATIONSHIP_HEADER_FROM: 'Timestamp:' + self.expected_ts_key,
NEPTUNE_RELATIONSHIP_HEADER_TO: 'Dashboard:' + self.expected_dashboard_key,
NEPTUNE_HEADER_LABEL: 'LAST_UPDATED_TIME_OF',
NEPTUNE_LAST_EXTRACTED_AT_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT: ANY,
NEPTUNE_CREATION_TYPE_RELATIONSHIP_PROPERTY_NAME_BULK_LOADER_FORMAT: NEPTUNE_CREATION_TYPE_JOB
}

self.maxDiff = None
assert actual is not None
self.assertDictEqual(actual_serialized[0], neptune_forward_expected)
self.assertDictEqual(actual_serialized[1], neptune_reversed_expected)
Expand Down
Loading