def execute(self, function_context: FlinkFunctionContext,
input_list: List[Table]) -> List[Table]:
t_env = function_context.get_table_env()
statement_set = function_context.get_statement_set()
table = input_list[0]
Popen('rm -rf /root/debug', shell=True)
t_env.register_function(
"build_index",
udf(BuildIndexUDF(self.path, self.element_type, self.dimension),
[DataTypes.STRING(), DataTypes.STRING()], DataTypes.STRING()))
dummy_output_path = '/tmp/indexed_key'
if os.path.exists(dummy_output_path):
if os.path.isdir(dummy_output_path):
shutil.rmtree(dummy_output_path)
else:
os.remove(dummy_output_path)
t_env.connect(FileSystem().path(dummy_output_path)) \
.with_format(OldCsv()
.field('key', DataTypes.STRING())) \
.with_schema(Schema()
.field('key', DataTypes.STRING())) \
.create_temporary_table('train_sink')
statement_set.add_insert(
"train_sink", table.select("build_index(uuid, feature_data)"))
return []
def execute(self, function_context: FlinkFunctionContext) -> Table:
example_meta: af.ExampleMeta = function_context.get_example_meta()
t_env = function_context.get_table_env()
t_env.connect(FileSystem().path(example_meta.batch_uri)) \
.with_format(OldCsv()
.field('word', DataTypes.STRING())) \
.with_schema(Schema()
.field('word', DataTypes.STRING())) \
.create_temporary_table('mySource')
return t_env.from_path('mySource')
def execute(self, function_context: FlinkFunctionContext, input_table: Table) -> None:
example_meta: af.ExampleMeta = function_context.get_example_meta()
output_file = example_meta.batch_uri
if os.path.exists(output_file):
os.remove(output_file)
t_env = function_context.get_table_env()
statement_set = function_context.get_statement_set()
t_env.connect(FileSystem().path(output_file)) \
.with_format(OldCsv()
.field_delimiter('\t')
.field('word', DataTypes.STRING())
.field('count', DataTypes.BIGINT())) \
.with_schema(Schema()
.field('word', DataTypes.STRING())
.field('count', DataTypes.BIGINT())) \
.create_temporary_table('mySink')
statement_set.add_insert('mySink', input_table)
def execute(self, function_context: FlinkFunctionContext,
input_list: List[Table]) -> List[Table]:
t_env = function_context.get_table_env()
table = input_list[0]
t_env.register_function(
"search",
udf(SearchUDTF(self.path, self.element_type), DataTypes.STRING(),
DataTypes.STRING()))
return [table.select("face_id, search(feature_data) as near_id")]
def execute(self, function_context: FlinkFunctionContext,
input_list: List[Table]) -> List[Table]:
t_env = function_context.get_table_env()
table = input_list[0]
Popen('rm -rf /root/test', shell=True)
t_env.register_function(
"search",
udf(SearchUDTF3(self.path, self.element_type), DataTypes.STRING(),
DataTypes.INT()))
return [
table.select("face_id, device_id, search(feature_data) as near_id")
]