How to apply multiple filters to a group by query

[vikramsinghchandel]

How can i apply multiple filters to a group by query following does not work filters = (pydruid.utils.filters.Dimension("unit")=='000721') & (pydruid.utils.filters.Dimension("val") > 0) query1 = query.groupby( datasource=dataSource, granularity='minute',

intervals='2016-08-01/p12w',

        #intervals='2016-08-01/pt24h',
        intervals='2016-08-11/2016-08-15',
        dimensions=["GPN"],
        filter=filters,
        aggregations={"val_sum": ag.doublesum("val"),"val_count": ag.count("val")},
        post_aggregations={"Avg": (pag.Field("val_sum") / pag.Field("val_count"))},
        context={"timeout": 600000}#,
       # limit_spec={
            #"type": "default",
           # "limit": 50,
          #  "columns" : ["sensor_val_sum","sensor_val_count"]
       # }
    )

Gives following error Traceback (most recent call last): File "Test_Query_2Filters.py", line 41, in context={"timeout": 600000}#, File "/usr/local/lib/python2.7/dist-packages/pydruid/client.py", line 191, in groupby query = self.query_builder.groupby(kwargs) File "/usr/local/lib/python2.7/dist-packages/pydruid/query.py", line 316, in groupby return self.build_query(query_type, args) File "/usr/local/lib/python2.7/dist-packages/pydruid/query.py", line 250, in build_query query_dict[key] = Filter.build_filter(val) File "/usr/local/lib/python2.7/dist-packages/pydruid/utils/filters.py", line 90, in build_filter filter['fields'] = [Filter.build_filter(f) for f in filter['fields']] File "/usr/local/lib/python2.7/dist-packages/pydruid/utils/filters.py", line 87, in build_filter filter = filter_obj.filter['filter'] AttributeError: 'bool' object has no attribute 'filter'

[var23rav]

Import the Filter class

from pydruid.utils.filters import Filter
filter_unit_obj = Filter(dimension='unit', value='000721')

val > 0, these kind work can be handled by make use of javascript functionality

filter_val_obj = Filter(
                    type= 'javascript',
                    dimension= 'val',
                    function= "function(x){ return x > 0}"
                )

Now join them like which can be substitue to the query.groupby function.

query_filter_obj = Filter(
                     type= 'and',
                     fields= [ filter_unit_obj, filter_val_obj]
                    )

[DzakirinMD]

Hope they can add this into filter documentation. Really helpful