superset详解(九)--数据查询

数据的查询和展示是superset的核心功能，前端用D3.js来渲染各种图标，后端用pandas来处理各种数据。
superset的数据查询功能也很简单，如果懂sql的话，只要在sueprset的explore页面配置一下左侧的图标类型，查询条件，例如，过滤字段，聚合查询/非聚合查询的字段，查询时间等，然后点页面上方的查询按钮，页面右侧就会按你选择的图表和查询条件来展示查询结果。今天我们来分析一下superset是如何查询数据并且把数据展示出来的。

superset查询数据的原理：

1.将前端的配置信息form_data传给explore_json函数
2.根据所选的图表类型，找到对应的图表类
3.根据过滤条件生成sql查询语句
4.根据数据库的连接条件找到对应的数据库engine,创建engine
5.使用pandas的read_sql函数获取查询结果，并生成一个dataframe

具体看一下代码：

1. superset/views/core.py Superset类的explore_json

@log_this
    @has_access_api
    @expose('/explore_json///', methods=['GET', 'POST'])
    @expose('/explore_json/', methods=['GET', 'POST'])
    def explore_json(self, datasource_type=None, datasource_id=None):
        try:
            csv = request.args.get('csv') == 'true'             
            query = request.args.get('query') == 'true'
            xlsx=request.args.get('xlsx') == 'true'
            force = request.args.get('force') == 'true'
            form_data = self.get_form_data()[0]
            datasource_id, datasource_type = self.datasource_info(
                datasource_id, datasource_type, form_data)
        except Exception as e:
            logging.exception(e)
            return json_error_response(
                utils.error_msg_from_exception(e),
                stacktrace=traceback.format_exc())
        return self.generate_json(datasource_type=datasource_type,
                                  datasource_id=datasource_id,
                                  form_data=form_data,
                                  csv=csv,
                                  query=query,
                                  force=force,xlsx=xlsx)

2. superset/viz.py BaseViz的get_df

    def get_df(self, query_obj=None):
        """Returns a pandas dataframe based on the query object"""
        if not query_obj:
            query_obj = self.query_obj()
        if not query_obj:
            return None

        self.error_msg = ''
        self.results = None

        timestamp_format = None
        if self.datasource.type == 'table':
            dttm_col = self.datasource.get_col(query_obj['granularity'])
            if dttm_col:
                timestamp_format = dttm_col.python_date_format

        # The datasource here can be different backend but the interface is common
        self.results = self.datasource.query(query_obj, self.viz_type)   # Sqltable.query方法
        self.query = self.results.query
        self.status = self.results.status
        self.error_message = self.results.error_message
        pass

sqltable.query:

    def query(self, query_obj, viz_type=None, special_sql=None):
        def format_time_grain(index, time_grain_sqla=None):
            if index.name == '__timestamp':
                index = index.apply(lambda x: time_grain_convert(x, time_grain_sqla))
            return index

        qry_start_dttm = datetime.now()
        status = QueryStatus.SUCCESS
        error_message = None
        df = None

        if special_sql:
            sql = special_sql
        else:
            time_grain_sqla = query_obj["extras"].get('time_grain_sqla',None)
            sql = self.get_query_str(query_obj)

        try:
            df = self.database.get_df(sql, self.schema)            # database的get_df方法
        except Exception as e:
            status = QueryStatus.FAILED
            logging.exception(e)
            error_message = (
                self.database.db_engine_spec.extract_error_message(e))
       pass

database.get_df:

    def get_df(self, sql, schema):
        sql = sql.strip().strip(';')
        eng = self.get_sqla_engine(schema=schema)         # 创建engine
        df = pd.read_sql(sql, eng)           # 获取查询结果

        def needs_conversion(df_series):
            if df_series.empty:
                return False
            if isinstance(df_series[0], (list, dict)):
                return True
            return False

        for k, v in df.dtypes.items():
            if v.type == numpy.object_ and needs_conversion(df[k]):
                df[k] = df[k].apply(utils.json_dumps_w_dates)
        return df