def poll_sync(
self,
connector_id: str,
initial_last_sync_completion: datetime.datetime,
poll_interval: float = DEFAULT_POLL_INTERVAL,
poll_timeout: float = None,
) -> Dict[str, Any]:
"""
Given a Fivetran connector and the timestamp at which the previous sync completed, poll
until the next sync completes.
The previous sync completion time is necessary because the only way to tell when a sync
completes is when this value changes.
Args:
connector_id (str): The Fivetran Connector ID. You can retrieve this value from the
"Setup" tab of a given connector in the Fivetran UI.
initial_last_sync_completion (datetime.datetime): The timestamp of the last completed sync
(successful or otherwise) for this connector, prior to running this method.
poll_interval (float): The time (in seconds) that will be waited between successive polls.
poll_timeout (float): The maximum time that will waited before this operation is timed
out. By default, this will never time out.
Returns:
Dict[str, Any]: Parsed json data representing the API response.
"""
poll_start = datetime.datetime.now()
while True:
(
curr_last_sync_completion,
curr_last_sync_succeeded,
curr_sync_state,
) = self.get_connector_sync_status(connector_id)
self._log.info(f"Polled '{connector_id}'. Status: [{curr_sync_state}]")
if curr_last_sync_completion > initial_last_sync_completion:
break
if poll_timeout and datetime.datetime.now() > poll_start + datetime.timedelta(
seconds=poll_timeout
):
raise Failure(
f"Sync for connector '{connector_id}' timed out after {datetime.datetime.now() - poll_start}."
)
# Sleep for the configured time interval before polling again.
time.sleep(poll_interval)
connector_details = self.get_connector_details(connector_id)
if not curr_last_sync_succeeded:
raise Failure(
f"Sync for connector '{connector_id}' failed!",
metadata={
"connector_details": MetadataValue.json(connector_details),
"log_url": MetadataValue.url(get_fivetran_logs_url(connector_details)),
},
)
return connector_details
def the_solid(_context):
yield AssetObservation(
asset_key="foo",
metadata={
"text": "FOO",
"int": 22,
"url": MetadataValue.url("http://fake.com"),
"float": 0.1,
"python": MetadataValue.python_artifact(MetadataValue),
},
)
def the_solid(_context):
yield AssetMaterialization(
asset_key="foo",
metadata={
"text": "FOO",
"int": 22,
"url": MetadataValue.url("http://fake.com"),
"float": 0.1,
"path": MetadataValue.path(Path("/a/b.csv")),
"python": MetadataValue.python_artifact(MetadataValue),
},
)
def my_failure_metadata_op():
path = "/path/to/files"
my_files = get_files(path)
if len(my_files) == 0:
raise Failure(
description="No files to process",
metadata={
"filepath":
MetadataValue.path(path),
"dashboard_url":
MetadataValue.url("http://mycoolsite.com/failures"),
},
)
return some_calculation(my_files)
def sort_by_calories(context, cereals):
sorted_cereals = sorted(cereals,
key=lambda cereal: int(cereal["calories"]))
least_caloric = sorted_cereals[0]["name"]
most_caloric = sorted_cereals[-1]["name"]
logger = get_dagster_logger()
logger.info(f"Least caloric cereal: {least_caloric}")
logger.info(f"Most caloric cereal: {most_caloric}")
fieldnames = list(sorted_cereals[0].keys())
sorted_cereals_csv_path = os.path.abspath(
f"output/calories_sorted_{context.run_id}.csv")
os.makedirs(os.path.dirname(sorted_cereals_csv_path), exist_ok=True)
with open(sorted_cereals_csv_path, "w") as fd:
writer = csv.DictWriter(fd, fieldnames)
writer.writeheader()
writer.writerows(sorted_cereals)
context.log_event(
AssetMaterialization(
asset_key="sorted_cereals_csv",
description="Cereals data frame sorted by caloric content",
metadata={
"sorted_cereals_csv_path":
MetadataValue.path(sorted_cereals_csv_path)
},
))
def _ge_validation_fn(context, dataset):
data_context = context.resources.ge_data_context
validator_kwargs = {
"datasource_name": datasource_name,
"data_connector_name": data_connector_name,
"data_asset_name": datasource_name or data_asset_name,
"runtime_parameters": {
runtime_method_type: dataset
},
"batch_identifiers": batch_identifiers,
"expectation_suite_name": suite_name,
**extra_kwargs,
}
validator = data_context.get_validator(**validator_kwargs)
run_id = {
"run_name": datasource_name + " run",
"run_time": datetime.datetime.utcnow(),
}
results = validator.validate(run_id=run_id)
validation_results_page_renderer = ValidationResultsPageRenderer(
run_info_at_end=True)
rendered_document_content_list = validation_results_page_renderer.render(
validation_results=results)
md_str = "".join(
DefaultMarkdownPageView().render(rendered_document_content_list))
meta_stats = MetadataEntry("Expectation Results",
value=MetadataValue.md(md_str))
yield ExpectationResult(
success=bool(results["success"]),
metadata_entries=[meta_stats],
)
yield Output(results.to_json_dict())
def my_metadata_materialization_op(context):
df = read_df()
remote_storage_path = persist_to_storage(df)
context.log_event(
AssetMaterialization(
asset_key="my_dataset",
description="Persisted result to storage",
metadata={
"text_metadata":
"Text-based metadata for this event",
"path":
MetadataValue.path(remote_storage_path),
"dashboard_url":
MetadataValue.url("http://mycoolsite.com/url_for_my_data"),
"size (bytes)":
calculate_bytes(df),
},
))
return remote_storage_path
def raw_file_op(_context):
yield AssetMaterialization(
asset_key="table_info",
metadata={
"table_path":
MetadataValue.path("/path/to/{}.raw".format(name))
},
)
yield do_expectation(_context, name)
yield Output(name)
def read_s3_key(context):
s3_key = context.solid_config["s3_key"]
bucket = context.solid_config["bucket"]
path = f"s3://{bucket}/{s3_key}"
context.log.info(f"Found file {path}")
yield AssetMaterialization(
asset_key=AssetKey(["log_s3", path]),
metadata={"S3 path": MetadataValue.url(path)},
)
yield Output(path)
def test_fs_io_manager():
with tempfile.TemporaryDirectory() as tmpdir_path:
io_manager = fs_io_manager.configured({"base_dir": tmpdir_path})
pipeline_def = define_pipeline(io_manager)
result = execute_pipeline(pipeline_def)
assert result.success
handled_output_events = list(
filter(lambda evt: evt.is_handled_output, result.event_list))
assert len(handled_output_events) == 2
filepath_a = os.path.join(tmpdir_path, result.run_id, "solid_a",
"result")
result_metadata_entry_a = handled_output_events[
0].event_specific_data.metadata_entries[0]
assert result_metadata_entry_a.label == "path"
assert result_metadata_entry_a.value == MetadataValue.path(filepath_a)
assert os.path.isfile(filepath_a)
with open(filepath_a, "rb") as read_obj:
assert pickle.load(read_obj) == [1, 2, 3]
loaded_input_events = list(
filter(lambda evt: evt.is_loaded_input, result.event_list))
input_metadata_entry_a = loaded_input_events[
0].event_specific_data.metadata_entries[0]
assert input_metadata_entry_a.label == "path"
assert input_metadata_entry_a.value == MetadataValue.path(filepath_a)
assert len(loaded_input_events) == 1
assert "solid_a" == loaded_input_events[
0].event_specific_data.upstream_step_key
filepath_b = os.path.join(tmpdir_path, result.run_id, "solid_b",
"result")
result_metadata_entry_b = handled_output_events[
1].event_specific_data.metadata_entries[0]
assert result_metadata_entry_b.label == "path"
assert result_metadata_entry_b.value == MetadataValue.path(filepath_b)
assert os.path.isfile(filepath_b)
with open(filepath_b, "rb") as read_obj:
assert pickle.load(read_obj) == 1
def my_asset_key_materialization_op(context):
df = read_df()
remote_storage_path = persist_to_storage(df)
yield AssetMaterialization(
asset_key=AssetKey(["dashboard", "my_cool_site"]),
description="Persisted result to storage",
metadata={
"dashboard_url":
MetadataValue.url("http://mycoolsite.com/dashboard"),
"size (bytes)": calculate_bytes(df),
},
)
yield Output(remote_storage_path)
def asset_metadata(_context, model_info):
config = dict(SHARED_SNOWFLAKE_CONF)
config["schema"] = model_info["schema"]
with connect_snowflake(config=config) as con:
df = pd.read_sql(f"SELECT * FROM {model_info['name']} LIMIT 5",
con=con)
num_rows = con.execute(
f"SELECT COUNT(*) FROM {model_info['name']}").fetchone()
return {
"Data sample": MetadataValue.md(df.to_markdown()),
"Rows": num_rows[0]
}
def _columns_to_metadata(
columns: Mapping[str, Any]) -> Optional[Mapping[str, Any]]:
return ({
"schema":
MetadataValue.table_schema(
TableSchema(columns=[
TableColumn(
name=name,
type=metadata.get("data_type") or "?",
description=metadata.get("description"),
) for name, metadata in columns.items()
]))
} if len(columns) > 0 else None)
def _table_data_to_materialization(
fivetran_output: FivetranOutput,
asset_key_prefix: List[str],
schema_name: str,
table_data: Dict[str, Any],
) -> AssetMaterialization:
table_name = table_data["name_in_destination"]
asset_key = asset_key_prefix + [schema_name, table_name]
if not table_data["enabled"]:
return None
metadata = {
"connector_url": MetadataValue.url(
get_fivetran_connector_url(fivetran_output.connector_details)
)
}
if table_data.get("columns"):
metadata["column_info"] = MetadataValue.json(table_data.get("columns"))
return AssetMaterialization(
asset_key=asset_key,
description=f"Table generated via Fivetran sync: {schema_name}.{table_name}",
metadata=metadata,
)
def result_to_materialization(
result: Dict[str, Any],
asset_key_prefix: Optional[List[str]] = None,
docs_url: Optional[str] = None,
) -> Optional[AssetMaterialization]:
"""
This is a hacky solution that attempts to consolidate parsing many of the potential formats
that dbt can provide its results in. This is known to work for CLI Outputs for dbt versions 0.18+,
as well as RPC responses for a similar time period, but as the RPC response schema is not documented
nor enforced, this can become out of date easily.
"""
asset_key_prefix = check.opt_list_param(asset_key_prefix, "asset_key_prefix", of_type=str)
# status comes from set of fields rather than "status"
if "fail" in result:
success = not result.get("fail") and not result.get("skip") and not result.get("error")
else:
success = result["status"] == "success"
if not success:
return None
# all versions represent timing the same way
metadata = {"Execution Time (seconds)": result["execution_time"]}
metadata.update(_timing_to_metadata(result["timing"]))
# working with a response that contains the node block (RPC and CLI 0.18.x)
if "node" in result:
unique_id = result["node"]["unique_id"]
metadata.update(_node_result_to_metadata(result["node"]))
else:
unique_id = result["unique_id"]
id_prefix = unique_id.split(".")
# only generate materializations for models
if id_prefix[0] != "model":
return None
if docs_url:
metadata["docs_url"] = MetadataValue.url(f"{docs_url}#!/model/{unique_id}")
return AssetMaterialization(
description=f"dbt node: {unique_id}",
metadata=metadata,
asset_key=asset_key_prefix + id_prefix,
)
def my_metadata_output(context):
df = get_some_data()
yield Output(
df,
metadata={
"text_metadata":
"Text-based metadata for this event",
"dashboard_url":
MetadataValue.url("http://mycoolsite.com/url_for_my_data"),
"raw_count":
len(df),
"size (bytes)":
calculate_bytes(df),
},
)
def many_table_materializations(_context):
with open(file_relative_path(__file__, MARKDOWN_EXAMPLE), "r") as f:
md_str = f.read()
for table in raw_tables:
yield AssetMaterialization(
asset_key="table_info",
metadata={
"table_name":
table,
"table_path":
MetadataValue.path(f"/path/to/{table}"),
"table_data": {
"name": table
},
"table_name_big":
MetadataValue.url(f"https://bigty.pe/{table}"),
"table_blurb":
MetadataValue.md(md_str),
"big_int":
29119888133298982934829348,
"float_nan":
float("nan"),
},
)
def file_handle_to_s3(context, file_handle):
bucket = context.solid_config["Bucket"]
key = context.solid_config["Key"]
with context.resources.file_manager.read(file_handle, "rb") as fileobj:
context.resources.s3.upload_fileobj(fileobj, bucket, key)
s3_file_handle = S3FileHandle(bucket, key)
yield AssetMaterialization(
asset_key=s3_file_handle.s3_path,
metadata={
last_key(key): MetadataValue.path(s3_file_handle.s3_path)
},
)
yield Output(value=s3_file_handle, output_name="s3_file_handle")
def _ge_validation_fn(context, dataset):
data_context = context.resources.ge_data_context
if validation_operator_name is not None:
validation_operator = validation_operator_name
else:
data_context.add_validation_operator(
"ephemeral_validation",
{
"class_name": "ActionListValidationOperator",
"action_list": []
},
)
validation_operator = "ephemeral_validation"
suite = data_context.get_expectation_suite(suite_name)
final_batch_kwargs = batch_kwargs or {"dataset": dataset}
if "datasource" in batch_kwargs:
context.log.warning(
"`datasource` field of `batch_kwargs` will be ignored; use the `datasource_name` "
f"parameter of the {decorator_name} factory instead.")
final_batch_kwargs["datasource"] = datasource_name
batch = data_context.get_batch(final_batch_kwargs, suite)
run_id = {
"run_name": datasource_name + " run",
"run_time": datetime.datetime.utcnow(),
}
results = data_context.run_validation_operator(
validation_operator, assets_to_validate=[batch], run_id=run_id)
res = convert_to_json_serializable(
results.list_validation_results())[0]
validation_results_page_renderer = ValidationResultsPageRenderer(
run_info_at_end=True)
rendered_document_content_list = (
validation_results_page_renderer.render_validation_operator_result(
results))
md_str = " ".join(
DefaultMarkdownPageView().render(rendered_document_content_list))
meta_stats = MetadataEntry("Expectation Results",
value=MetadataValue.md(md_str))
yield ExpectationResult(
success=res["success"],
metadata_entries=[
meta_stats,
],
)
yield Output(res)
def my_metadata_expectation_op(context, df):
df = do_some_transform(df)
context.log_event(
ExpectationResult(
success=len(df) > 0,
description="ensure dataframe has rows",
metadata={
"text_metadata":
"Text-based metadata for this event",
"dashboard_url":
MetadataValue.url("http://mycoolsite.com/url_for_my_data"),
"raw_count":
len(df),
"size (bytes)":
calculate_bytes(df),
},
))
return df
def test_table_metadata_value_schema_inference():
table_metadata_entry = MetadataEntry(
"foo",
value=MetadataValue.table(
records=[
TableRecord(name="foo", status=False),
TableRecord(name="bar", status=True),
],
),
)
schema = table_metadata_entry.entry_data.schema # type: ignore
assert isinstance(schema, TableSchema)
assert schema.columns == [
TableColumn(name="name", type="string"),
TableColumn(name="status", type="bool"),
]
def _materialization_for_stream(
name: str,
stream_schema_props: Dict[str, Any],
stream_stats: Dict[str, Any],
asset_key_prefix: List[str],
) -> AssetMaterialization:
return AssetMaterialization(
asset_key=asset_key_prefix + [name],
metadata={
"schema": MetadataValue.table_schema(
TableSchema(
columns=[
TableColumn(name=name, type=str(info.get("type", "unknown")))
for name, info in stream_schema_props.items()
]
)
),
**{k: v for k, v in stream_stats.items() if v is not None},
},
)
def read_file(context):
relative_filename = context.op_config["filename"]
directory = context.op_config["directory"]
filename = os.path.join(directory, relative_filename)
try:
fstats = os.stat(filename)
context.log.info("Found file {}".format(relative_filename))
yield AssetMaterialization(
asset_key=AssetKey(["log_file", relative_filename]),
metadata={
"path": MetadataValue.path(filename),
"File status": {
"size": fstats.st_size,
"ctime": fstats.st_ctime,
"mtime": fstats.st_mtime,
},
},
)
yield Output(relative_filename)
except FileNotFoundError:
context.log.error("No file found: {}".format(relative_filename))
def build_component_top_stories(model: TruncatedSVD,
user_story_matrix: IndexedCooMatrix,
story_titles: DataFrame):
"""
For each component in the collaborative filtering model, finds the titles of the top stories
it's associated with.
"""
n_stories = 10
components_column = []
titles_column = []
story_titles = story_titles.set_index("id")
for i in range(model.components_.shape[0]):
component = model.components_[i]
top_story_indices = component.argsort()[-n_stories:][::-1]
top_story_ids = user_story_matrix.col_index[top_story_indices]
top_story_titles = story_titles.loc[top_story_ids]
for title in top_story_titles["title"]:
components_column.append(i)
titles_column.append(title)
component_top_stories = DataFrame({
"component_index":
Series(components_column),
"title":
Series(titles_column)
})
yield Output(
component_top_stories,
metadata={
"Top component top stories":
MetadataValue.md(
top_components_to_markdown(component_top_stories)),
},
)
def many_materializations_and_passing_expectations(_context):
tables = [
"users",
"groups",
"events",
"friends",
"pages",
"fans",
"event_admins",
"group_admins",
]
for table in tables:
yield AssetMaterialization(
asset_key="table_info",
metadata={
"table_path": MetadataValue.path(f"/path/to/{table}.raw"),
},
)
yield ExpectationResult(
success=True,
label="{table}.row_count".format(table=table),
description="Row count passed for {table}".format(table=table),
)
def _materialization_for_stream(
name: str,
stream_info: Dict[str, Any],
stream_stats: Dict[str, Any],
asset_key_prefix: List[str],
) -> AssetMaterialization:
return AssetMaterialization(
asset_key=asset_key_prefix + [name],
metadata={
"schema":
MetadataValue.table_schema(
TableSchema(columns=[
TableColumn(name=name, type=str(info["type"]))
for name, info in stream_info["stream"]["jsonSchema"]
["properties"].items()
])),
"columns":
",".join(name for name in stream_info["stream"]["jsonSchema"]
["properties"].keys()),
**{k: v
for k, v in stream_stats.items() if v is not None},
},
)
def metadata_for_actions(df):
return {
"min_score": int(df["score"].min()),
"max_score": int(df["score"].max()),
"sample rows": MetadataValue.md(df[:5].to_markdown()),
}
def daily_top_action(_, df1, df2):
df = pd.concat([df1, df2]).nlargest(1, "score")
return Output(df, metadata={"data": MetadataValue.md(df.to_markdown())})
def _best_n_actions(_, df):
df = df.nlargest(n, "score")
return Output(
df,
metadata={"data": MetadataValue.md(df.to_markdown())},
)
def handle_output(self, context, obj):
super().handle_output(context, obj)
# can pretend this actually came from a library call
yield MetadataEntry(label="num rows written to db",
description=None,
entry_data=MetadataValue.int(len(obj)))
