How can I extract json data from snowflake - json

Question

I'm working with Snowflake, and have a batch of JSON data in a variant datatype. I'm going to paste just the first entry (I hope) out of megabytes of data.
I need to figure out the SQL to extract the data elements from it.
This query:
SELECT
src:value::string AS valuest,
src:value:evaluation_forms as evaluation_forms,
src:value:evaluation_forms:evaluations as evaluations,
src:value:evaluation_forms:evaluations:channel_meta as channel_meta,
src:value:evaluations:channel_meta:agent_first_name as agent_first_name
FROM stage.intuit_quality_raw_table
retrieves JSON data in valuest, JSON data in evaluation_forms, and nulls in evaluations, channel_meta, and agent_first_name.
I suspect that that is because of the '[' characters that occur in from of the evaluations and channel_meta entries, but I just don't know how to deal with that sort of data.
I aplogize the being so long. The first entry is something over 300 lines long, and I'm afraid to try to shorten it because I fear that I will delete something important. I've already added the ending batch of stuff, which I'm also afraid that I might have messed up.
Anyway, here it is:
{"value":
{"evaluation_forms":
[ {"evaluations":
[ {"channel_meta":
{"after_call_work_time": [],
"agent_first_name": ["KATRINA"],
"agent_hung_up": [],
"agent_last_name": ["COX"],
"agent_unique_id": ["LO_00130604"],
"agent_username": [],
"alternate_call_id": [],
"audio_file_name": ["b161aa50-cda9-42e2-ad16-117a9cf08b92.wav"],
"call_direction": ["INBOUND"],
"call_id": ["b161aa50-cda9-42e2-ad16-117a9cf08b92"],
"channel_type": "Call",
"customer_id": [],
"customer_segment": [],
"disposition": [],
"disposition_time": [],
"end_at": ["2023-01-24 12:00:00"],
"file_location": ["blank"],
"handle_time": ["1824"],
"hold_time": [],
"how_call_ended": ["Transferred"],
"how_call_originated": [],
"id": "3cb5a3e5-0ab8-4a38-9f6b-4ab3b2e24351",
"ivr_options": [],
"line": ["Intuit"],
"notes": [],
"queue": ["Product Support"],
"queue_time": [],
"start_at": ["2023-01-24 12:00:00"]
"subqueue": [ "Product Support - Desktop" ],
"supplemental_field_1": ["Expert Service Score: 3"],
"supplemental_field_10": [],
"supplemental_field_2": ["Derived TNPS Score: 0"],
"supplemental_field_3": [],
"supplemental_field_4": [],
"supplemental_field_5": [],
"supplemental_field_6": [],
"supplemental_field_7": [],
"supplemental_field_8": [],
"supplemental_field_9": [],
"talk_time": ["78"],
"team_id": [],
"times_held": ["2"],
"total_time": []
},
"evaluation_list": [
{"evaluation_meta":
{"ack_info":
{"ack_init_time":
null,
"ack_status": "NOT_INITIATED",
"ack_time": null,
"dispute_raised_at": null,
"dispute_raised_by_email": null,
"dispute_resolved_at": null,
"dispute_resolved_by_email": null
},
"agent_email": "lo_00130604#liveopsnation.com",
"agent_id": "63d77af042a6982e25008bf3",
"agent_name": "KATRINA COX",
"agent_status": "ACTIVE",
"created_at": "2023-02-06T02:16:53.637Z",
"evaluation_purpose": null,
"evaluation_type": "Manual",
"evaluator_email": "info#ara-gs.com",
"evaluator_id":"63cb37e42db1035ae08285d1",
"evaluator_name": "RAMON CARPIO JR�",
"evaluator_status": "ACTIVE",
"id": "63e06221ea5e4c7202572252",
"modified_at": "2023-02-06T02:17:35.229Z",
"status": "EDITED"
},
"response": {
"comment": "",
"sections": [
{
"id": "3fe2faeb-e8bc-4a16-9fd1-49166a3eed4d",
"questions": [
{
"comment": "",
"fail": false,
"id": "4c057d99-aae7-4432-8634-6c41e4cde734",
"score_obtained": 1,
"value": "Yes"
},
{"comment": "",
"fail": false,
"id": "f23b5e63-b24c-49f7-aa46-eb7499cdf89d",
"score_obtained": 1,
"value": "Yes"
},
{"comment": "",
"fail": false,
"id": "4798443d-eaf3-46e7-bc9f-1f03ccee871a",
"score_obtained": 1,
"value": "Yes"
}],
"score_obtained": 3
},
{"id": "f5a83974-3be3-4f69-a9de-49866213b383",
"questions": [ {
"comment": "",
"fail": false,
"id": "52ec152b-fce8-464d-a8cf-b4ae82bfb5d9",
"score_obtained": 0,
"value": "N/A"
},
{"comment": "",
"fail": false,
"id": "627ce8fd-2d29-44d9-8c3d-49dfdab9f7b2",
"score_obtained": 0,
"value": "N/A"
},
{ "comment": "",
"fail": false,
"id": "a6efd5fe-42e1-4411-b197-ee715b579281",
"score_obtained": 0,
"value": "N/A"
},
{"comment": "",
"fail":false,
"id": "aea7d958-c27e-44ab-a32f-c2388b8c6aab",
"score_obtained": 1,
"value": "Yes"
},
{"comment": "",
"fail": false,
"id": "860ac0b2-c325-443d-8f91-63879d09bac8",
"score_obtained": 0,
"value": "N/A"
} ],
"score_obtained": 1
},
{"id": "6a5fa573-8dc5-4dbb-b6cd-2e01445c3b4c",
"questions": [
{"comment": "",
"fail": false,
"id": "22e2cc11-c886-4b56-8de6-e97b773de4d0",
"score_obtained": 1,
"value": "Yes"
},
{"comment": "Agent transferred call incorrectly. Another agent advised her of it then call disconnected. ",
"fail":true,
"id": "39b2ab0d-7372-408e-b8e2-886be10239de",
"score_obtained": 0,
"value": "No"}
],
"score_obtained": 1
},
{"id": "d1deed99-e8aa-49f1-8c32-079887fd6838",
"questions": [
{"comment": "",
"fail": false,
"id": "b951c53c-52b9-4a08-b3c6-70298643db22",
"score_obtained": 1,
"value": "Yes"
},
{"comment": "",
"fail": false,
"id": "2321f675-01f9-4975-b7e9-d57feb865d0e",
"score_obtained": 1,
"value": "Yes"
},
{"comment": "",
"fail": false,
"id": "6b0eb24c-f31c-41f0-85d3-56805d42284d",
"score_obtained": 1,
"value": "Yes"
},
{"comment": "",
"fail":false,
"id": "7e5e7a2e-60c0-4416-814e-fe19c9ab2361",
"score_obtained": 1,
"value": "Yes"
},
{
"comment": "",
"fail": false,
"id": "19f6ec0a-4c4c-49b6-962e-d029a0326d27",
"score_obtained": 1,
"value": "No"
} ],
"score_obtained": 5
},
{"id": "6c41a906-3d69-469d-9391-c814a6124ec1",
"questions": [
{"comment": "",
"fail": false,
"id": "d72fca9d-323d-4238-980e-c0aec746e1fc",
"score_obtained": 1,
"value": "Yes"},
{
"comment": "",
"fail": false,
"id": "d1af8d8e-6a97-4111-bad7-a64695f2c3f7",
"score_obtained": 1,
"value": "Yes"
},
{
"comment": "",
"fail":true,
"id": "c1251cb6-c3fe-4f69-bbe5-9510f150d18a",
"score_obtained": 0,
"value": "No"
}],
"score_obtained": 2
},
{
"id": "ca1c275b-1cd0-49c9-a4d7-9fd0544b4bbf",
"questions": [
{
"comment": "",
"fail": true,
"id": "4f3268ba-4e8d-46b8-a50a-69ea9de88a6a",
"score_obtained": 0,
"value": "No"
},
{
"comment": "",
"fail": true,
"id": "cb82d886-8e2c-4d4c-a8f8-b5c769bf289a",
"score_obtained": 0,
"value": "No"
},
{
"comment": "",
"fail": true,
"id": "ea9aa819-6905-4f15-a702-3713bbb033a9",
"score_obtained": 0,
"value": "No"
}
],
"score_obtained": 0
},
{
"id": "5743b6e8-ac6f-4d34-8561-f7e5191d5f37",
"questions": [
{
"comment": "",
"fail": false,
"id": "af04685e-5c73-47be-8282-493960cce3e0",
"score_obtained": 1,
"value": "Yes"
},
{
"comment": "",
"fail": false,
"id": "67b8ea9f-d481-42c2-9343-e586875e0929",
"score_obtained": 1,
"value": "Yes"
},
{
"comment": "",
"fail":false,
"id": "9fc25686-a63c-4db4-b2aa-dcf8666d2147",
"score_obtained": 1,
"value": "Yes"
},
{
"comment": "",
"fail": false,
"id": "410616c0-0d41-4322-b6a3-df11c3c28963",
"score_obtained": 1,
"value": "Yes"
},
{
"comment": "",
"fail": false,
"id": "575c978e-365d-4b83-842e-acfe41de09ae",
"score_obtained": 1,
"value": "Yes"
} ],
"score_obtained": 5
},
"id": "bced1392-24b2-404f-91be-9df6d68e0c8d",
"questions": [
{
"comment": "",
"fail": false,
"id": "e2250777-8bba-4fee-b9b4-1535b42be14c",
"score_obtained": 0,
"value": "No"
},
{
"comment": "",
"fail": false,
"id": "533281a0-63db-4af6-b41e-7b967a456332",
"score_obtained": 0,
"value": "No"
},
{
"comment": "",
"fail": false,
"id": "c9cf8f64-427b-4927-8ffb-02454e6daa43",
"score_obtained": 0,
"value": "No"
},
{
"comment": "",
"fail": false,
"id": "7c4df8ba-0d89-4703-b4fe-3e6f88c3fa76",
"score_obtained": 0,
"value": "No"
},
"comment": "",
"fail": false,
"id": "0bf1e679-a300-475e-9fb8-613317d42570",
"score_obtained": 0,
"value": "No"
},
{
"comment": "Advise agent get the link with forms update
dates to provide clients. No transfer needed to be completed. ",
" fail": false,
"id": "52ba5be6-7def-469a-a31f-7c74c86730ac",
"score_obtained": 0,
"value": ""
}
],
"score_obtained": 0
}
]},
"scores_obtained": {
"auto_fail": false,
"final_score": 17,
"grade_assigned": "Pass",
"percent_score": 77.27,
"total_bonus": 0,
"total_penalties": 0,
"total_points": 22
}
}
]
}
]}]}}
Thanks,

Answer 1

You can use FLATTEN function to parse nested arrays in JSON.
refer to
https://docs.snowflake.com/en/user-guide/querying-semistructured#using-the-flatten-function-to-parse-nested-arrays

Answer 2

To simplify, I have taken a small part of your JSON data and tested it.
CREATE OR REPLACE TABLE xyz
(
src variant
)
AS
SELECT PARSE_JSON(column1) AS src
FROM VALUES
('{"value":
{"evaluation_forms":
[ {"evaluations":
[ {"channel_meta":
{"after_call_work_time": [],
"agent_first_name": ["KATRINA"],
"agent_hung_up": [],
"agent_last_name": ["COX"]
}
} ]
} ]
}
} '
) v;
SELECT
src:value AS valuest,
src:value:evaluation_forms as evaluation_forms,
v.value:evaluations as evaluations,
ve.value:channel_meta as channel_meta,
Fn.value as agent_first_name
FROM
xyz
, LATERAL FLATTEN(INPUT => SRC:value:evaluation_forms) v
, LATERAL FLATTEN(INPUT => v.value:evaluations) ve
, LATERAL FLATTEN(INPUT => ve.value:channel_meta:agent_first_name) fn
;