aminghadersohi commented on code in PR #39922: URL: https://github.com/apache/superset/pull/39922#discussion_r3284881259
########## superset/mcp_service/chart/plugins/big_number.py: ########## @@ -0,0 +1,220 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. + +"""Big number chart type plugin.""" + +from __future__ import annotations + +from typing import Any + +from superset.mcp_service.chart.chart_utils import ( + _big_number_chart_what, + _summarize_filters, + is_column_truly_temporal, + map_big_number_config, +) +from superset.mcp_service.chart.plugin import BaseChartPlugin +from superset.mcp_service.chart.schemas import BigNumberChartConfig, ColumnRef +from superset.mcp_service.chart.validation.dataset_validator import DatasetValidator +from superset.mcp_service.common.error_schemas import ChartGenerationError + + +class BigNumberChartPlugin(BaseChartPlugin): + """Plugin for big_number chart type.""" + + chart_type = "big_number" + display_name = "Big Number" + native_viz_types = { + "big_number": "Big Number with Trendline", + "big_number_total": "Big Number", + } + + def pre_validate( + self, + config: dict[str, Any], + ) -> ChartGenerationError | None: + if "metric" not in config: + return ChartGenerationError( + error_type="missing_metric", + message="Big Number chart missing required field: metric", + details=( + "Big Number charts require a 'metric' field " + "specifying the value to display" + ), + suggestions=[ + "Add 'metric' with name and aggregate: " + "{'name': 'revenue', 'aggregate': 'SUM'}", + "The aggregate function is required (SUM, COUNT, AVG, MIN, MAX)", + "Example: {'chart_type': 'big_number', " + "'metric': {'name': 'sales', 'aggregate': 'SUM'}}", + ], + error_code="MISSING_BIG_NUMBER_METRIC", + ) + + metric = config.get("metric", {}) + if not isinstance(metric, dict): + return ChartGenerationError( + error_type="invalid_metric_type", + message="Big Number metric must be a dict with 'name' and 'aggregate'", + details=( + f"The 'metric' field must be an object, got {type(metric).__name__}" + ), + suggestions=[ + "Use a dict: {'name': 'col', 'aggregate': 'SUM'}", + "Valid aggregates: SUM, COUNT, AVG, MIN, MAX", + ], + error_code="INVALID_BIG_NUMBER_METRIC_TYPE", + ) + if not metric.get("aggregate") and not metric.get("saved_metric"): + return ChartGenerationError( + error_type="missing_metric_aggregate", + message=( + "Big Number metric must include an aggregate function " + "or reference a saved metric" + ), + details=( + "The metric must have an 'aggregate' field or 'saved_metric': true" + ), + suggestions=[ + "Add 'aggregate': {'name': 'col', 'aggregate': 'SUM'}", + "Or use a saved metric: {'name': 'metric', 'saved_metric': true}", + "Valid aggregates: SUM, COUNT, AVG, MIN, MAX", + ], + error_code="MISSING_BIG_NUMBER_AGGREGATE", + ) + + show_trendline = config.get("show_trendline", False) + temporal_column = config.get("temporal_column") + if show_trendline and not temporal_column: + return ChartGenerationError( + error_type="missing_temporal_column", + message="Trendline requires a temporal column", + details=( + "When 'show_trendline' is True, " + "a 'temporal_column' must be specified" + ), + suggestions=[ + "Add 'temporal_column': 'date_column_name'", + "Or set 'show_trendline': false for number only", + "Use get_dataset_info to find temporal columns", + ], + error_code="MISSING_TEMPORAL_COLUMN", + ) + + return None + + def extract_column_refs(self, config: Any) -> list[ColumnRef]: + if not isinstance(config, BigNumberChartConfig): + return [] + refs: list[ColumnRef] = [config.metric] + # temporal_column is a str field, not a ColumnRef — validate it exists + if config.temporal_column: + refs.append(ColumnRef(name=config.temporal_column)) + if config.filters: + for f in config.filters: + refs.append(ColumnRef(name=f.column)) + return refs + + def to_form_data( + self, config: Any, dataset_id: int | str | None = None + ) -> dict[str, Any]: + return map_big_number_config(config) + + def post_map_validate( + self, + config: Any, + form_data: dict[str, Any], + dataset_id: int | str | None = None, + ) -> ChartGenerationError | None: + """Verify the trendline temporal column is a real temporal SQL type. + + This check was previously baked into map_config_to_form_data() in + chart_utils.py as a special case. Moving it here keeps the dispatcher + clean and makes the constraint explicit and discoverable. + """ + if not isinstance(config, BigNumberChartConfig): + return None + if not (config.show_trendline and config.temporal_column): + return None + + if not is_column_truly_temporal(config.temporal_column, dataset_id): + return ChartGenerationError( + error_type="non_temporal_trendline_column", + message=( + f"Big Number trendline requires a temporal SQL column; " + f"'{config.temporal_column}' is not temporal." + ), + details=( + f"Column '{config.temporal_column}' does not have a temporal " + f"SQL type (DATE, DATETIME, TIMESTAMP). The trendline requires " + f"a true temporal column for DATE_TRUNC to work." + ), + suggestions=[ + "Use get_dataset_info to find columns with temporal SQL types", + "Set 'show_trendline': false to use any column as the metric", + "If the column contains dates stored as integers, " + "consider casting it in a virtual dataset", + ], + error_code="NON_TEMPORAL_TRENDLINE_COLUMN", + ) + + return None + + def generate_name(self, config: Any, dataset_name: str | None = None) -> str: + what = _big_number_chart_what(config) + context = _summarize_filters(getattr(config, "filters", None)) + return self._with_context(what, context) + + def resolve_viz_type(self, config: Any) -> str: + show_trendline = getattr(config, "show_trendline", False) + temporal_column = getattr(config, "temporal_column", None) + if show_trendline and temporal_column: + return "big_number" + return "big_number_total" + + def normalize_column_refs(self, config: Any, dataset_context: Any) -> Any: + config_dict = config.model_dump() + + if config_dict.get("metric") and not config_dict["metric"].get("saved_metric"): + config_dict["metric"]["name"] = DatasetValidator._get_canonical_column_name( + config_dict["metric"]["name"], dataset_context + ) Review Comment: The normalization is already in place. Both `big_number.py` and `pie.py` branch on `saved_metric` in `normalize_column_refs`: when `saved_metric=True`, the name is resolved through `DatasetValidator._get_canonical_metric_name()`, which performs a case-insensitive lookup exclusively against `available_metrics` and returns the canonical dataset name. Non-saved metrics go through `_get_canonical_column_name` instead. The described bug doesn't exist in the current code. ########## superset/mcp_service/chart/plugins/pie.py: ########## @@ -0,0 +1,128 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. + +"""Pie chart type plugin.""" + +from __future__ import annotations + +from typing import Any + +from superset.mcp_service.chart.chart_utils import ( + _pie_chart_what, + _summarize_filters, + map_pie_config, +) +from superset.mcp_service.chart.plugin import BaseChartPlugin +from superset.mcp_service.chart.schemas import ColumnRef, PieChartConfig +from superset.mcp_service.chart.validation.dataset_validator import DatasetValidator +from superset.mcp_service.common.error_schemas import ChartGenerationError + + +class PieChartPlugin(BaseChartPlugin): + """Plugin for pie chart type.""" + + chart_type = "pie" + display_name = "Pie / Donut Chart" + native_viz_types = { + "pie": "Pie Chart", + } + + def pre_validate( + self, + config: dict[str, Any], + ) -> ChartGenerationError | None: + missing_fields = [] + + if "dimension" not in config: + missing_fields.append("'dimension' (category column for slices)") + if "metric" not in config: + missing_fields.append("'metric' (value metric for slice sizes)") + + if missing_fields: + return ChartGenerationError( + error_type="missing_pie_fields", + message=( + f"Pie chart missing required fields: {', '.join(missing_fields)}" + ), + details=( + "Pie charts require a dimension (categories) and a metric (values)" + ), + suggestions=[ + "Add 'dimension' field: {'name': 'category_column'}", + "Add 'metric' field: {'name': 'value_column', 'aggregate': 'SUM'}", + "Example: {'chart_type': 'pie', 'dimension': {'name': 'product'}, " + "'metric': {'name': 'revenue', 'aggregate': 'SUM'}}", + ], + error_code="MISSING_PIE_FIELDS", + ) + + return None + + def extract_column_refs(self, config: Any) -> list[ColumnRef]: + if not isinstance(config, PieChartConfig): + return [] + refs: list[ColumnRef] = [config.dimension, config.metric] + if config.filters: + for f in config.filters: + refs.append(ColumnRef(name=f.column)) + return refs + + def to_form_data( + self, config: Any, dataset_id: int | str | None = None + ) -> dict[str, Any]: + return map_pie_config(config) + + def generate_name(self, config: Any, dataset_name: str | None = None) -> str: + what = _pie_chart_what(config) + context = _summarize_filters(config.filters) + return self._with_context(what, context) + + def resolve_viz_type(self, config: Any) -> str: + return "pie" + + def normalize_column_refs(self, config: Any, dataset_context: Any) -> Any: + config_dict = config.model_dump() + + if config_dict.get("dimension"): + config_dict["dimension"]["name"] = ( + DatasetValidator._get_canonical_column_name( + config_dict["dimension"]["name"], dataset_context + ) + ) + if config_dict.get("metric") and not config_dict["metric"].get("saved_metric"): + config_dict["metric"]["name"] = DatasetValidator._get_canonical_column_name( + config_dict["metric"]["name"], dataset_context Review Comment: The normalization is already in place. Both `big_number.py` and `pie.py` branch on `saved_metric` in `normalize_column_refs`: when `saved_metric=True`, the name is resolved through `DatasetValidator._get_canonical_metric_name()`, which performs a case-insensitive lookup exclusively against `available_metrics` and returns the canonical dataset name. Non-saved metrics go through `_get_canonical_column_name` instead. The described bug doesn't exist in the current code. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
