Summarize with dynamic variable involved

[gcam032]

Hi All,

I have some data stored int he following format:

Key	Timestamp	DynField
1234	2019-07-24 23:59:33	{"isA":false,"isRR":false,"isE":true,"isM":null,"isVS":0,"isBot":false}
3938	2019-07-24 12:04:32	{"isA":true,"isRR":false,"isE":false,"isM":null,"isVS":1000,"isBot":false}
3982	2019-07-24 01:33:13	{"isA":false,"isRR":true,"isE":false,"isM":null,"isVS":0,"isBot":true}

The combinations in the DynField can be true/false/null (except for isVS that is 0 or 1000).

My goal is to have the following output:

Date	count key	count key where isA = true	count key where isRR = true
2019-07-24	3	1	1

 and so on... with a column for each of the variables with the dynamic field.

Can anyone help on how to best use Dynamic fields when summarizing?

Thanks,

g

[#2]

consider the following option:

datatable(key:long, Timestamp:datetime, d:dynamic)
[
    1234, datetime(2019-07-24 23:59:33), dynamic({"isA":false,"isRR":false,"isE":true, "isM":null,"isVS":0,   "isBot":false}),
    3938, datetime(2019-07-24 12:04:32), dynamic({"isA":true, "isRR":false,"isE":false,"isM":null,"isVS":1000,"isBot":false}),
    3982, datetime(2019-07-24 01:33:13), dynamic({"isA":false,"isRR":true, "isE":false,"isM":null,"isVS":0,   "isBot":true })
]
| summarize count_key = dcount(key), 
            count_key_where_isA_is_true = dcountif(key, d.isA == true),
            count_key_where_isRR_is_true = dcountif(key, d.isRR == true)
         by Date = startofday(Timestamp)