def filter_columns(self, identifier: str, columns: List[int], names: List[str], datastore: Datastore) -> VizualApiResult: """Dataset projection operator. Returns a copy of the dataset with the given identifier that contains only those columns listed in columns. The list of names contains optional new names for the filtered columns. A value of None in names indicates that the name of the corresponding column is not changed. Raises ValueError if no dataset with given identifier exists or if any of the filter columns are unknown. Parameters ---------- identifier: string Unique dataset identifier columns: list(int) List of column identifier for columns in the result. names: list(string) Optional new names for filtered columns. datastore : vizier.datastore.fs.base.FileSystemDatastore Datastore to retireve and update datasets Returns ------- vizier.engine.packages.vizual.api.VizualApiResult """ # Get dataset. Raise exception if dataset is unknown dataset = datastore.get_dataset(identifier) if dataset is None: raise ValueError('unknown dataset \'' + identifier + '\'') # The schema of the new dataset only contains the columns in the given # list. A column might need to be renamed. schema = list() column_mapping = list() col_list = [] for i in range(len(columns)): col_idx = get_index_for_column(dataset, columns[i]) col = dataset.columns[col_idx] if not names[i] is None: if not is_valid_name(names[i]): raise ValueError('invalid column name \'' + str(names[i]) + '\'') schema.append( MimirDatasetColumn(identifier=col.identifier, name_in_dataset=names[i], name_in_rdb=names[i])) else: schema.append(col) column_mapping.append({ "columns_column": col_idx, "columns_name": schema[-1].name }) col_list.append(col.name_in_rdb) command = {"id": "projection", "columns": column_mapping} response = mimir.vizualScript(dataset.identifier, command) return VizualApiResult.from_mimir(response)
def insert_column(self, identifier, position, name, datastore): """Insert column with given name at given position in dataset. Raises ValueError if no dataset with given identifier exists, if the specified column position is outside of the current schema bounds, or if the column name is invalid. Parameters ---------- identifier: string Unique dataset identifier position: int Index position at which the column will be inserted name: string, optional New column name datastore : vizier.datastore.fs.base.FileSystemDatastore Datastore to retireve and update datasets Returns ------- vizier.engine.packages.vizual.api.VizualApiResult """ # Raise ValueError if given colum name is invalid if not is_valid_name(name): raise ValueError('invalid column name \'' + str(name) + '\'') # Get dataset. Raise exception if dataset is unknown dataset = datastore.get_dataset(identifier) if dataset is None: raise ValueError('unknown dataset \'' + identifier + '\'') # Make sure that position is a valid column index in the new dataset if position < 0 or position > len(dataset.columns): raise ValueError('invalid column index \'' + str(position) + '\'') # Get identifier for new column col_id = dataset.max_column_id() + 1 # Insert new column into schema schema = list(dataset.columns) new_column = MimirDatasetColumn(col_id, name, name) schema.insert(position, new_column) # Create a view for the modified schema col_list = [] for col in schema: if col.identifier == new_column.identifier: # Note: By no (April 2018) this requires Mimir to run with the # XNULL option. Otherwise, in some scenarios setting the all # values in the new column to NULL may cause an exception. col_list.append(" CAST('' AS int) AS " + col.name_in_rdb) else: col_list.append(col.name_in_rdb) sql = 'SELECT ' + ','.join( col_list) + ' FROM ' + dataset.table_name + ';' view_name, dependencies = mimir.createView(dataset.table_name, sql) # Store updated dataset information with new identifier ds = datastore.register_dataset(table_name=view_name, columns=schema, row_counter=dataset.row_counter, annotations=dataset.annotations) return VizualApiResult(ds)
def compute_empty_dataset(self, args, context): """Execute empty dataset command. Parameters ---------- args: vizier.viztrail.command.ModuleArguments User-provided command arguments context: vizier.engine.task.base.TaskContext Context in which a task is being executed Returns ------- vizier.engine.task.processor.ExecResult """ outputs = ModuleOutputs() default_columns = [("''", "unnamed_column")] ds_name = args.get_value(pckg.PARA_NAME).lower() if ds_name in context.datasets: raise ValueError('dataset \'' + ds_name + '\' exists') if not is_valid_name(ds_name): raise ValueError('invalid dataset name \'' + ds_name + '\'') try: source = "SELECT {};".format(", ".join( default_val + " AS " + col_name for default_val, col_name in default_columns)) view_name, dependencies = mimir.createView(dict(), source) columns = [ MimirDatasetColumn(identifier=col_id, name_in_dataset=col_defn[1]) for col_defn, col_id in zip(default_columns, range(len(default_columns))) ] ds = context.datastore.register_dataset(table_name=view_name, columns=columns, row_counter=1) provenance = ModuleProvenance( write={ ds_name: DatasetDescriptor(identifier=ds.identifier, columns=ds.columns, row_count=ds.row_count) }, read=dict( ) # Need to explicitly declare a lack of dependencies. ) outputs.stdout.append( TextOutput("Empty dataset '{}' created".format(ds_name))) except Exception as ex: provenance = ModuleProvenance() outputs.error(ex) return ExecResult(is_success=(len(outputs.stderr) == 0), outputs=outputs, provenance=provenance)
def filter_columns(self, identifier, columns, names, datastore): """Dataset projection operator. Returns a copy of the dataset with the given identifier that contains only those columns listed in columns. The list of names contains optional new names for the filtered columns. A value of None in names indicates that the name of the corresponding column is not changed. Raises ValueError if no dataset with given identifier exists or if any of the filter columns are unknown. Parameters ---------- identifier: string Unique dataset identifier columns: list(int) List of column identifier for columns in the result. names: list(string) Optional new names for filtered columns. datastore : vizier.datastore.fs.base.FileSystemDatastore Datastore to retireve and update datasets Returns ------- vizier.engine.packages.vizual.api.VizualApiResult """ # Get dataset. Raise exception if dataset is unknown dataset = datastore.get_dataset(identifier) if dataset is None: raise ValueError('unknown dataset \'' + identifier + '\'') # The schema of the new dataset only contains the columns in the given # list. A column might need to be renamed. schema = list() col_list = [] for i in range(len(columns)): col_idx = get_index_for_column(dataset, columns[i]) col = dataset.columns[col_idx] if not names[i] is None: schema.append( MimirDatasetColumn(identifier=col.identifier, name_in_dataset=names[i], name_in_rdb=col.name_in_rdb)) else: schema.append(col) col_list.append(col.name_in_rdb) sql = 'SELECT ' + ','.join( col_list) + ' FROM ' + dataset.table_name + ';' view_name, dependencies = mimir.createView(dataset.table_name, sql) # Store updated dataset information with new identifier ds = datastore.register_dataset(table_name=view_name, columns=schema, row_counter=dataset.row_counter, annotations=dataset.annotations.filter( columns=columns, rows=dataset.row_ids)) return VizualApiResult(ds)
def create_dataset( self, columns: List[DatasetColumn], rows: List[DatasetRow], properties: Dict[str, Any] = None, human_readable_name: str = "Untitled Dataset", backend_options: Optional[List[Tuple[str, str]]] = None, dependencies: Optional[List[str]] = None) -> MimirDatasetHandle: """Create a new dataset in the datastore. Expects at least the list of columns and the rows for the dataset. Parameters ---------- columns: list(vizier.datastore.dataset.DatasetColumn) List of columns. It is expected that each column has a unique identifier. rows: list(vizier.datastore.dataset.DatasetRow) List of dataset rows. properties: dict(string, any), optional Annotations for dataset components Returns ------- vizier.datastore.dataset.DatasetDescriptor """ # Get unique identifier for new dataset properties = {} if properties is None else properties backend_options = [] if backend_options is None else backend_options dependencies = [] if dependencies is None else dependencies identifier = 'DS_' + get_unique_identifier() columns = [ col if isinstance(col, MimirDatasetColumn) else MimirDatasetColumn( identifier=col.identifier, name_in_dataset=col.name, data_type=col.data_type) for col in columns ] table_name, schema = mimir.loadDataInline( schema=[{ "name": base.sanitize_column_name(col.name), "type": col.data_type } for col in columns], rows=[row.values for row in rows], result_name=identifier, human_readable_name=human_readable_name, dependencies=dependencies, properties=properties) # Insert the new dataset metadata information into the datastore return MimirDatasetHandle.from_mimir_result(table_name=table_name, schema=schema, properties=properties, name=human_readable_name)
def empty_dataset( self, datastore: Datastore, filestore: Filestore, initial_columns: List[Tuple[str, str]] = [("''", "unnamed_column")] ) -> VizualApiResult: """Create (or load) a new dataset from a given file or Uri. It is guaranteed that either the file identifier or the url are not None but one of them will be None. The user name and password may only be given if an url is given. The resources refer to any resoures (e.g., file identifier) that have been generated by a previous execution of the respective task. This allows to associate an identifier with a downloaded file to avoid future downloads (unless the reload flag is True). Parameters ---------- datastore : vizier.datastore.fs.base.FileSystemDatastore Datastore to retireve and update datasets filestore: vizier.filestore.Filestore Filestore to retrieve uploaded datasets Returns ------- vizier.engine.packages.vizual.api.VizualApiResult """ assert (isinstance(datastore, MimirDatastore)) ds = datastore.create_dataset( columns=[ MimirDatasetColumn(identifier=id, name_in_dataset=col, data_type="varchar") for id, (default, col) in enumerate(initial_columns) ], rows=[ DatasetRow( identifier=str(id), values=[default for default, col in initial_columns]) for id in range(1, 2) ], human_readable_name="Empty Table", ) return VizualApiResult(dataset=ds)
def insert_column(self, identifier: str, position: int, name: str, datastore: Datastore) -> VizualApiResult: """Insert column with given name at given position in dataset. Raises ValueError if no dataset with given identifier exists, if the specified column position is outside of the current schema bounds, or if the column name is invalid. Parameters ---------- identifier: string Unique dataset identifier position: int Index position at which the column will be inserted name: string, optional New column name datastore : vizier.datastore.fs.base.FileSystemDatastore Datastore to retireve and update datasets Returns ------- vizier.engine.packages.vizual.api.VizualApiResult """ # Raise ValueError if given colum name is invalid if not is_valid_name(name): raise ValueError('invalid column name \'' + str(name) + '\'') # Get dataset. Raise exception if dataset is unknown dataset = datastore.get_dataset(identifier) if dataset is None: raise ValueError('unknown dataset \'' + identifier + '\'') # Make sure that position is a valid column index in the new dataset if position < 0 or position > len(dataset.columns): raise ValueError('invalid column index \'' + str(position) + '\'') # Get identifier for new column col_id = dataset.max_column_id() + 1 # Insert new column into schema schema = list(dataset.columns) new_column = MimirDatasetColumn(col_id, name, name) schema.insert(position, new_column) command = {"id": "insertColumn", "name": name, "position": position} response = mimir.vizualScript(dataset.identifier, command) return VizualApiResult.from_mimir(response)
def compute(self, command_id, arguments, context): """Compute results for commands in the Mimir package using the set of user-provided arguments and the current database state. Parameters ---------- command_id: string Unique identifier for a command in a package declaration arguments: vizier.viztrail.command.ModuleArguments User-provided command arguments context: vizier.engine.task.base.TaskContext Context in which a task is being executed Returns ------- vizier.engine.task.processor.ExecResult """ outputs = ModuleOutputs() # Get dataset. Raise exception if dataset is unknown. ds_name = arguments.get_value(pckg.PARA_DATASET).lower() dataset = context.get_dataset(ds_name) mimir_table_name = dataset.identifier # Keep track of the name of the input dataset for the provenance # information. input_ds_name = ds_name if command_id == cmd.MIMIR_GEOCODE: geocoder = arguments.get_value(cmd.PARA_GEOCODER) # Add columns for LATITUDE and LONGITUDE column_counter = dataset.max_column_id() + 1 cname_lat = dataset.get_unique_name('LATITUDE') cname_lon = dataset.get_unique_name('LONGITUDE') dataset.columns.append( MimirDatasetColumn( identifier=column_counter, name_in_dataset=cname_lat, data_type=DATATYPE_REAL ) ) dataset.columns.append( MimirDatasetColumn( identifier=column_counter + 1, name_in_dataset=cname_lon, data_type=DATATYPE_REAL ) ) house = arguments.get_value(cmd.PARA_HOUSE_NUMBER, raise_error=False, default_value=None) street = arguments.get_value(cmd.PARA_STREET, raise_error=False, default_value=None) city = arguments.get_value(cmd.PARA_CITY, raise_error=False, default_value=None) state = arguments.get_value(cmd.PARA_STATE, raise_error=False, default_value=None) params = { 'houseColumn': dataset.column_by_id(house).name_in_rdb if house is not None and house != '' else None, 'streetColumn': dataset.column_by_id(street).name_in_rdb if street is not None and street != '' else None, 'cityColumn': dataset.column_by_id(city).name_in_rdb if city is not None and city != '' else None, 'stateColumn': dataset.column_by_id(state).name_in_rdb if state is not None and state != '' else None, 'geocoder': geocoder#, #'latitudeColumn': Option[String], #'longitudeColumn': Option[String], #'cacheCode': Option[String] } elif command_id == cmd.MIMIR_KEY_REPAIR: column = dataset.column_by_id(arguments.get_value(pckg.PARA_COLUMN)) params = { "key" : column.name_in_rdb } elif command_id == cmd.MIMIR_MISSING_KEY: column = dataset.column_by_id(arguments.get_value(pckg.PARA_COLUMN)) params = column.name_in_rdb # Set MISSING ONLY to FALSE to ensure that all rows are returned #params += ['MISSING_ONLY(FALSE)'] # Need to run this lens twice in order to generate row ids for # any potential new tuple elif command_id == cmd.MIMIR_MISSING_VALUE: params = list() for col in arguments.get_value(cmd.PARA_COLUMNS, default_value=[]): f_col = dataset.column_by_id(col.get_value(pckg.PARA_COLUMN)) param = f_col.name_in_rdb col_constraint = col.get_value( cmd.PARA_COLUMNS_CONSTRAINT, raise_error=False ) if col_constraint == '': col_constraint = None #if not col_constraint is None: # param = param + ' ' + str(col_constraint).replace("'", "\'\'").replace("OR", ") OR (") #param = '\'(' + param + ')\'' params.append(param) elif command_id == cmd.MIMIR_PICKER: # Compute the input columns inputs = [] for col in arguments.get_value(cmd.PARA_SCHEMA): c_col = col.get_value(cmd.PARA_PICKFROM) column = dataset.column_by_id(c_col) inputs.append(column.name_in_rdb) # Compute the output column output = arguments.get_value(cmd.PARA_PICKAS, default_value = inputs[0]) if output == "": output = inputs[0] else: output = dataset.get_unique_name(output.strip().upper()) # Compute the final parameter list params = { "inputs" : inputs, "output" : output } elif command_id == cmd.MIMIR_TYPE_INFERENCE: params = [str(arguments.get_value(cmd.PARA_PERCENT_CONFORM))] elif command_id == cmd.MIMIR_SHAPE_DETECTOR: dseModel = arguments.get_value(cmd.PARA_MODEL_NAME) params = [] if not dseModel is None: params = [str(dseModel)] elif command_id == cmd.MIMIR_COMMENT: commentsParams = [] for idx, comment in enumerate(arguments.get_value(cmd.PARA_COMMENTS)): commentParam = {} # If target is defined, it is the column that we're trying to annotate # If unset (or empty), it means we're annotating the row. column_id = comment.get_value(cmd.PARA_EXPRESSION, None) if column_id is not None: column = dataset.column_by_id(column_id) commentParam['target'] = column.name_in_rdb # The comment commentParam['comment'] = comment.get_value(cmd.PARA_COMMENT) # If rowid is defined, it is the row that we're trying to annotate. # If unset (or empty), it means that we're annotating all rows rowid = comment.get_value(cmd.PARA_ROWID, None) if (rowid is not None) and (rowid != ""): # If rowid begins with '=', it's a formula if rowid[0] == '=': commentParam['condition'] = rowid[1:] else: commentParam['rows'] = [ int(rowid) ] #TODO: handle result columns commentsParams.append(commentParam) params = {'comments' : commentsParams} elif command_id == cmd.MIMIR_PIVOT: column = dataset.column_by_id(arguments.get_value(pckg.PARA_COLUMN)) params = { "target" : column.name_in_rdb, "keys" : [], "values" : [] } for col_arg in arguments.get_value(cmd.PARA_VALUES): col = dataset.column_by_id(col_arg.get_value(cmd.PARA_VALUE)) params["values"].append(col.name_in_rdb) for col_arg in arguments.get_value(cmd.PARA_KEYS, default_value=[]): col = dataset.column_by_id(col_arg.get_value(cmd.PARA_KEY)) params["keys"].append(col.name_in_rdb) if len(params["values"]) < 1: raise ValueError("Need at least one value column") # store_as_dataset = arguments.get_value(cmd.PARA_RESULT_DATASET) elif command_id == cmd.MIMIR_SHRED: params = { "keepOriginalColumns" : arguments.get_value(cmd.PARA_KEEP_ORIGINAL) } shreds = [] global_input_col = dataset.column_by_id(arguments.get_value(cmd.PARA_COLUMN_NAME)) for (idx, shred) in enumerate(arguments.get_value(cmd.PARA_COLUMNS)): output_col = shred.get_value(cmd.PARA_OUTPUT_COLUMN) if output_col is None: output_col = "{}_{}".format(global_input_col,idx) config = {} shred_type = shred.get_value(cmd.PARA_TYPE) expression = shred.get_value(cmd.PARA_EXPRESSION) group = shred.get_value(cmd.PARA_INDEX) if shred_type == "pattern": config["regexp"] = expression config["group"] = int(group) elif shred_type == "field": config["separator"] = expression config["field"] = int(group) elif shred_type == "explode": config["separator"] = expression elif shred_type == "pass": pass elif shred_type == "substring": range_parts = re.match("([0-9]+)(([+\\-])([0-9]+))?", expression) # print(range_parts) # Mimir expects ranges to be given from start (inclusive) to end (exclusive) # in a zero-based numbering scheme. # Vizier expects input ranges to be given in a one-based numbering scheme. # Convert to this format if range_parts is None: raise ValueError("Substring requires a range of the form '10', '10-11', or '10+1', but got '{}'".format(expression)) config["start"] = int(range_parts.group(1))-1 # Convert 1-based numbering to 0-based if range_parts.group(2) is None: config["end"] = config["start"] + 1 # if only one character, split one character elif range_parts.group(3) == "+": config["end"] = config["start"] + int(range_parts.group(4)) # start + length elif range_parts.group(3) == "-": config["end"] = int(range_parts.group(4)) # Explicit end, 1-based -> 0-based and exclusive cancel out else: raise ValueError("Invalid expression '{}' in substring shredder".format(expression)) # print("Shredding {} <- {} -- {}".format(output_col,config["start"],config["end"])) else: raise ValueError("Invalid Shredding Type '{}'".format(shred_type)) shreds.append({ **config, "op" : shred_type, "input" : global_input_col.name_in_rdb, "output" : output_col, }) params["shreds"] = shreds # store_as_dataset = arguments.get_value(cmd.PARA_RESULT_DATASET) else: raise ValueError("Unknown Mimir lens '{}'".format(command_id)) # Create Mimir lens mimir_lens_response = mimir.createLens( mimir_table_name, params, command_id, arguments.get_value(cmd.PARA_MATERIALIZE_INPUT, default_value=True), human_readable_name = ds_name.upper() ) lens_name = mimir_lens_response['name'] lens_schema = mimir_lens_response['schema'] lens_properties = mimir_lens_response['properties'] ds = MimirDatasetHandle.from_mimir_result(lens_name, lens_schema, lens_properties, ds_name) if command_id in LENSES_THAT_SHOULD_NOT_DISPLAY_TABLES: print_dataset_schema(outputs, ds_name, ds.columns) else: from vizier.api.webservice import server ds_output = server.api.datasets.get_dataset( project_id=context.project_id, dataset_id=ds.identifier, offset=0, limit=10 ) outputs.stdout.append(DatasetOutput(ds_output)) # Return task result return ExecResult( outputs=outputs, provenance=ModuleProvenance( read={input_ds_name: dataset.identifier}, write={ds_name: DatasetDescriptor( identifier=ds.identifier, name=ds_name, columns=ds.columns )} ) )
def rename_column(self, identifier, column_id, name, datastore): """Rename column in a given dataset. Raises ValueError if no dataset with given identifier exists, if the specified column is unknown, or if the given column name is invalid. Parameters ---------- identifier: string Unique dataset identifier column_id: int Unique column identifier name: string New column name datastore : vizier.datastore.fs.base.FileSystemDatastore Datastore to retireve and update datasets Returns ------- vizier.engine.packages.vizual.api.VizualApiResult """ # Raise ValueError if given colum name is invalid if not is_valid_name(name): raise ValueError('invalid column name \'' + str(name) + '\'') # Get dataset. Raise exception if dataset is unknown dataset = datastore.get_dataset(identifier) if dataset is None: raise ValueError('unknown dataset \'' + identifier + '\'') # Get the specified column that is to be renamed and set the column name # to the new name columns = list() schema = list(dataset.columns) colIndex = get_index_for_column(dataset, column_id) col = schema[colIndex] # No need to do anything if the name hasn't changed if col.name.lower() != name.lower(): sql = 'SELECT * FROM ' + dataset.table_name mimirSchema = mimir.getSchema(sql) # Create list of dataset columns colSql = '' idx = 0 for col in mimirSchema: col_id = len(columns) name_in_dataset = sanitize_column_name(col['name'].upper()) name_in_rdb = sanitize_column_name(col['name'].upper()) col = MimirDatasetColumn(identifier=col_id, name_in_dataset=name_in_dataset, name_in_rdb=name_in_rdb) if idx == 0: colSql = name_in_dataset + ' AS ' + name_in_rdb elif idx == colIndex: colSql = colSql + ', ' + name_in_dataset + ' AS ' + name col.name = name col.name_in_rdb = name else: colSql = colSql + ', ' + name_in_dataset + ' AS ' + name_in_rdb columns.append(col) idx = idx + 1 # Create view for loaded dataset sql = 'SELECT ' + colSql + ' FROM {{input}};' view_name, dependencies = mimir.createView(dataset.table_name, sql) # There are no changes to the underlying database. We only need to # change the column information in the dataset schema. # Store updated dataset to get new identifier ds = datastore.register_dataset(table_name=view_name, columns=columns, row_counter=dataset.row_counter, annotations=dataset.annotations) return VizualApiResult(ds) else: return VizualApiResult(dataset)
def execute_query(self, args, context): """Execute a SQL query in the given context. Parameters ---------- args: vizier.viztrail.command.ModuleArguments User-provided command arguments context: vizier.engine.task.base.TaskContext Context in which a task is being executed Returns ------- vizier.engine.task.processor.ExecResult """ # Get SQL source code that is in this cell and the global # variables source = args.get_value(cmd.PARA_SQL_SOURCE) if not source.endswith(';'): source = source + ';' ds_name = args.get_value(cmd.PARA_OUTPUT_DATASET, raise_error=False) # Get mapping of datasets in the context to their respective table # name in the Mimir backend mimir_table_names = dict() for ds_name_o in context.datasets: dataset_id = context.datasets[ds_name_o] dataset = context.datastore.get_dataset(dataset_id) if dataset is None: raise ValueError('unknown dataset \'' + ds_name_o + '\'') mimir_table_names[ds_name_o] = dataset.table_name # Module outputs outputs = ModuleOutputs() try: # Create the view from the SQL source view_name, dependencies = mimir.createView(mimir_table_names, source) sql = 'SELECT * FROM ' + view_name mimirSchema = mimir.getSchema(sql) columns = list() for col in mimirSchema: col_id = len(columns) name_in_dataset = col['name'] col = MimirDatasetColumn(identifier=col_id, name_in_dataset=name_in_dataset) columns.append(col) row_count = mimir.countRows(view_name) provenance = None if ds_name is None or ds_name == '': ds_name = "TEMPORARY_RESULT" ds = context.datastore.register_dataset(table_name=view_name, columns=columns, row_counter=row_count) ds_output = server.api.datasets.get_dataset( project_id=context.project_id, dataset_id=ds.identifier, offset=0, limit=10) ds_output['name'] = ds_name dependencies = dict((dep_name.lower(), context.datasets.get(dep_name.lower(), None)) for dep_name in dependencies) # print("---- SQL DATASETS ----\n{}\n{}".format(context.datasets, dependencies)) outputs.stdout.append(DatasetOutput(ds_output)) provenance = ModuleProvenance(write={ ds_name: DatasetDescriptor(identifier=ds.identifier, columns=ds.columns, row_count=ds.row_count) }, read=dependencies) except Exception as ex: provenance = ModuleProvenance() outputs.error(ex) # Return execution result return ExecResult(is_success=(len(outputs.stderr) == 0), outputs=outputs, provenance=provenance)
def create_dataset(self, columns, rows, human_readable_name=None, annotations=None, backend_options=[], dependencies=[]): """Create a new dataset in the datastore. Expects at least the list of columns and the rows for the dataset. Parameters ---------- columns: list(vizier.datastore.dataset.DatasetColumn) List of columns. It is expected that each column has a unique identifier. rows: list(vizier.datastore.dataset.DatasetRow) List of dataset rows. annotations: vizier.datastore.annotation.dataset.DatasetMetadata, optional Annotations for dataset components Returns ------- vizier.datastore.dataset.DatasetDescriptor """ # Get unique identifier for new dataset identifier = 'DS_' + get_unique_identifier() # Write rows to temporary file in CSV format tmp_file = os.path.abspath(self.base_path + identifier) # Create a list of columns that contain the user-vizible column name and # the name in the database db_columns = list() colSql = '' for col in map(base.sanitize_column_name, columns): db_columns.append( MimirDatasetColumn(identifier=col.identifier, name_in_dataset=col.name, name_in_rdb=col.name)) if colSql == '': colSql = col.name + ' AS ' + col.name else: colSql = colSql + ', ' + col.name + ' AS ' + col.name # Create CSV file for load with open(tmp_file, 'w') as f_out: writer = csv.writer(f_out, quoting=csv.QUOTE_MINIMAL) writer.writerow([col.name_in_rdb for col in db_columns]) for row in rows: record = helper.encode_values(row.values) writer.writerow(record) # Load CSV file using Mimirs loadCSV method. table_name = mimir.loadDataSource( tmp_file, True, True, human_readable_name=human_readable_name, backend_options=backend_options, dependencies=dependencies) os.remove(tmp_file) sql = 'SELECT ' + colSql + ' FROM {{input}};' view_name, dependencies = mimir.createView(table_name, sql) # Get number of rows in the view that was created in the backend row_count = mimir.countRows(view_name) # Insert the new dataset metadata information into the datastore return self.register_dataset(table_name=view_name, columns=db_columns, row_counter=row_count, annotations=annotations)
def load_dataset(self, f_handle=None, url=None, detect_headers=True, infer_types=True, load_format='csv', options=[], human_readable_name=None): """Create a new dataset from a given file or url. Expects that either the file handle or the url are not None. Raises ValueError if both are None or not None. Parameters ---------- f_handle : vizier.filestore.base.FileHandle, optional handle for an uploaded file on the associated file server. url: string, optional, optional Url for the file source detect_headers: bool, optional Detect column names in loaded file if True infer_types: bool, optional Infer column types for loaded dataset if True load_format: string, optional Format identifier options: list, optional Additional options for Mimirs load command human_readable_name: string, optional Optional human readable name for the resulting table Returns ------- vizier.datastore.mimir.dataset.MimirDatasetHandle """ if f_handle is None and url is None: raise ValueError('no load source given') elif not f_handle is None and not url is None: raise ValueError('too many load sources given') elif url is None: # os.path.abspath((r'%s' % os.getcwd().replace('\\','/') ) + '/' + f_handle.filepath) abspath = f_handle.filepath elif not url is None: abspath = url # Load dataset into Mimir init_load_name = mimir.loadDataSource(abspath, infer_types, detect_headers, load_format, human_readable_name, options) # Retrieve schema information for the created dataset sql = 'SELECT * FROM ' + init_load_name mimirSchema = mimir.getSchema(sql) # Create list of dataset columns columns = list() colSql = '' for col in mimirSchema: col_id = len(columns) name_in_dataset = base.sanitize_column_name(col['name'].upper()) name_in_rdb = base.sanitize_column_name(col['name'].upper()) col = MimirDatasetColumn(identifier=col_id, name_in_dataset=name_in_dataset, name_in_rdb=name_in_rdb) if colSql == '': colSql = name_in_dataset + ' AS ' + name_in_rdb else: colSql = colSql + ', ' + name_in_dataset + ' AS ' + name_in_rdb columns.append(col) # Create view for loaded dataset sql = 'SELECT ' + colSql + ' FROM {{input}};' view_name, dependencies = mimir.createView(init_load_name, sql) # TODO: this is a hack to speed up this step a bit. # we get the first row id and the count and take a range; # this is fragile and should be made better # # NOTE: This does not work because ROW_ID appears to be a string. # Thus, sorting not necessarily returns the smallest integer value # first. # row_count = mimir.countRows(view_name) return self.register_dataset(table_name=view_name, columns=columns, row_counter=row_count)
def compute(self, command_id, arguments, context): """Compute results for commands in the Mimir package using the set of user-provided arguments and the current database state. Parameters ---------- command_id: string Unique identifier for a command in a package declaration arguments: vizier.viztrail.command.ModuleArguments User-provided command arguments context: vizier.engine.task.base.TaskContext Context in which a task is being executed Returns ------- vizier.engine.task.processor.ExecResult """ outputs = ModuleOutputs() store_as_dataset = None update_rows = False lens_annotations = [] # Get dataset. Raise exception if dataset is unknown. ds_name = arguments.get_value(pckg.PARA_DATASET).lower() dataset = context.get_dataset(ds_name) mimir_table_name = dataset.table_name # Keep track of the name of the input dataset for the provenance # information. input_ds_name = ds_name if command_id == cmd.MIMIR_DOMAIN: column = dataset.column_by_id(arguments.get_value( pckg.PARA_COLUMN)) params = [column.name_in_rdb] elif command_id == cmd.MIMIR_GEOCODE: geocoder = arguments.get_value(cmd.PARA_GEOCODER) params = ['GEOCODER(' + geocoder + ')'] add_column_parameter(params, 'HOUSE_NUMBER', dataset, arguments, cmd.PARA_HOUSE_NUMBER) add_column_parameter(params, 'STREET', dataset, arguments, cmd.PARA_STREET) add_column_parameter(params, 'CITY', dataset, arguments, cmd.PARA_CITY) add_column_parameter(params, 'STATE', dataset, arguments, cmd.PARA_STATE) # Add columns for LATITUDE and LONGITUDE column_counter = dataset.max_column_id() + 1 cname_lat = dataset.get_unique_name('LATITUDE') cname_lon = dataset.get_unique_name('LONGITUDE') dataset.columns.append( MimirDatasetColumn(identifier=column_counter, name_in_dataset=cname_lat, data_type=DATATYPE_REAL)) dataset.columns.append( MimirDatasetColumn(identifier=column_counter + 1, name_in_dataset=cname_lon, data_type=DATATYPE_REAL)) params.append('RESULT_COLUMNS(' + cname_lat + ',' + cname_lon + ')') elif command_id == cmd.MIMIR_KEY_REPAIR: column = dataset.column_by_id(arguments.get_value( pckg.PARA_COLUMN)) params = [column.name_in_rdb] update_rows = True elif command_id == cmd.MIMIR_MISSING_KEY: column = dataset.column_by_id(arguments.get_value( pckg.PARA_COLUMN)) params = [column.name_in_rdb] # Set MISSING ONLY to FALSE to ensure that all rows are returned params += ['MISSING_ONLY(FALSE)'] # Need to run this lens twice in order to generate row ids for # any potential new tuple mimir_lens_response = mimir.createLens( dataset.table_name, params, command_id, arguments.get_value(cmd.PARA_MATERIALIZE_INPUT, default_value=True)) (mimir_table_name, lens_annotations) = (mimir_lens_response.lensName(), mimir_lens_response.annotations()) params = [ROW_ID, 'MISSING_ONLY(FALSE)'] update_rows = True elif command_id == cmd.MIMIR_MISSING_VALUE: params = list() for col in arguments.get_value(cmd.PARA_COLUMNS, default_value=[]): f_col = dataset.column_by_id(col.get_value(pckg.PARA_COLUMN)) param = f_col.name_in_rdb col_constraint = col.get_value(cmd.PARA_COLUMNS_CONSTRAINT, raise_error=False) if col_constraint == '': col_constraint = None if not col_constraint is None: param = param + ' ' + str(col_constraint).replace( "'", "\'\'").replace("OR", ") OR (") param = '\'(' + param + ')\'' params.append(param) elif command_id == cmd.MIMIR_PICKER: pick_from = list() column_names = list() for col in arguments.get_value(cmd.PARA_SCHEMA): c_col = col.get_value(cmd.PARA_PICKFROM) column = dataset.column_by_id(c_col) pick_from.append(column.name_in_rdb) column_names.append(column.name.upper().replace(' ', '_')) # Add result column to dataset schema pick_as = arguments.get_value(cmd.PARA_PICKAS, default_value='PICK_ONE_' + '_'.join(column_names)) pick_as = dataset.get_unique_name(pick_as.strip().upper()) dataset.columns.append( MimirDatasetColumn(identifier=dataset.max_column_id() + 1, name_in_dataset=pick_as)) params = ['PICK_FROM(' + ','.join(pick_from) + ')'] params.append('PICK_AS(' + pick_as + ')') elif command_id == cmd.MIMIR_SCHEMA_MATCHING: store_as_dataset = arguments.get_value(cmd.PARA_RESULT_DATASET) if store_as_dataset in context.datasets: raise ValueError('dataset \'' + store_as_dataset + '\' exists') if not is_valid_name(store_as_dataset): raise ValueError('invalid dataset name \'' + store_as_dataset + '\'') column_names = list() params = ['\'' + ROW_ID + ' int\''] for col in arguments.get_value(cmd.PARA_SCHEMA): c_name = col.get_value(pckg.PARA_COLUMN) c_type = col.get_value(cmd.PARA_TYPE) params.append('\'' + c_name + ' ' + c_type + '\'') column_names.append(c_name) elif command_id == cmd.MIMIR_TYPE_INFERENCE: params = [str(arguments.get_value(cmd.PARA_PERCENT_CONFORM))] elif command_id == cmd.MIMIR_SHAPE_DETECTOR: dseModel = arguments.get_value(cmd.PARA_MODEL_NAME) params = [] if not dseModel is None: params = [str(dseModel)] elif command_id == cmd.MIMIR_COMMENT: params = [] for comment in arguments.get_value(cmd.PARA_COMMENTS): c_expr = comment.get_value(cmd.PARA_EXPRESSION) c_cmnt = comment.get_value(cmd.PARA_COMMENT) c_rowid = comment.get_value(cmd.PARA_ROWID) if c_rowid is None: params.append('COMMENT(' + c_expr + ', \'' + c_cmnt + '\') ') else: params.append('COMMENT(' + c_expr + ', \'' + c_cmnt + '\', \'' + c_rowid + '\') ') result_cols = [] for col in arguments.get_value(cmd.PARA_RESULT_COLUMNS): c_name = col.get_value(pckg.PARA_COLUMN) result_cols.append(c_name) if len(result_cols) > 0: params.append('RESULT_COLUMNS(' + ','.join(result_cols) + ')') else: raise ValueError('unknown Mimir lens \'' + str(lens) + '\'') # Create Mimir lens if command_id in [ cmd.MIMIR_SCHEMA_MATCHING, cmd.MIMIR_TYPE_INFERENCE, cmd.MIMIR_SHAPE_DETECTOR ]: lens_name = mimir.createAdaptiveSchema(mimir_table_name, params, command_id.upper()) else: mimir_lens_response = mimir.createLens( mimir_table_name, params, command_id.upper(), arguments.get_value(cmd.PARA_MATERIALIZE_INPUT, default_value=True), human_readable_name=ds_name.upper()) (lens_name, lens_annotations) = (mimir_lens_response['lensName'], mimir_lens_response['annotations']) # Create a view including missing row ids for the result of a # MISSING KEY lens if command_id == cmd.MIMIR_MISSING_KEY: lens_name, row_counter = create_missing_key_view( dataset, lens_name, column) dataset.row_counter = row_counter # Create datastore entry for lens. if not store_as_dataset is None: columns = list() for c_name in column_names: col_id = len(columns) columns.append( MimirDatasetColumn(identifier=col_id, name_in_dataset=c_name)) ds = context.datastore.register_dataset( table_name=lens_name, columns=columns, annotations=dataset.annotations) ds_name = store_as_dataset else: ds = context.datastore.register_dataset( table_name=lens_name, columns=dataset.columns, annotations=dataset.annotations) # Add dataset schema and returned annotations to output if command_id in [ cmd.MIMIR_SCHEMA_MATCHING, cmd.MIMIR_TYPE_INFERENCE, cmd.MIMIR_SHAPE_DETECTOR ]: print_dataset_schema(outputs, ds_name, ds.columns) else: ds_output = server.api.datasets.get_dataset( project_id=context.project_id, dataset_id=ds.identifier, offset=0, limit=10) outputs.stdout.append(DatasetOutput(ds_output)) print_lens_annotations(outputs, lens_annotations) dsd = DatasetDescriptor(identifier=ds.identifier, columns=ds.columns, row_count=ds.row_count) result_resources = dict() result_resources[base.RESOURCE_DATASET] = ds.identifier # Return task result return ExecResult(outputs=outputs, provenance=ModuleProvenance( read={input_ds_name: dataset.identifier}, write={ds_name: dsd}, resources=result_resources))