One of the great things about postgres is that it allows indexing into a json object.
I have a column of data formatted a little bit like this:
{"Items":
[
{"RetailPrice":6.1,"EffectivePrice":0,"Multiplier":1,"ItemId":"53636"},
{"RetailPrice":0.47,"EffectivePrice":0,"Multiplier":1,"ItemId":"53404"}
]
}
What I'd like to do is find the average RetailPrice of each row with these data.
Something like
select avg(json_extract_path_text(item_json, 'RetailPrice'))
but really I need to do this for each item in the items json object. So for this example, the value in the queried cell would be 3.285
How can I do this?
Could work like this:
WITH cte(tbl_id, json_items) AS (
SELECT 1
, '{"Items": [
{"RetailPrice":6.1,"EffectivePrice":0,"Multiplier":1,"ItemId":"53636"}
,{"RetailPrice":0.47,"EffectivePrice":0,"Multiplier":1,"ItemId":"53404"}]}'::json
)
SELECT tbl_id, round(avg((elem->>'RetailPrice')::numeric), 3) AS avg_retail_price
FROM cte c
, json_array_elements(c.json_items->'Items') elem
GROUP BY 1;
The CTE just substitutes for a table like:
CREATE TABLE tbl (
tbl_id serial PRIMARY KEY
, json_items json
);
json_array_elements() (Postgres 9.3+) to unnest the json array is instrumental.
I am using an implicit JOIN LATERAL here. Much like in this related example:
Query for element of array in JSON column
For an index to support this kind of query consider this related answer:
Index for finding an element in a JSON array
For details on how to best store EAV data:
Is there a name for this database structure?
Related
I have a table named "games" with 2 fields:
name (varchar)
data (json)
This is a sample row of data:
name
data
Test
{"players":["PlayerOne","PlayerTwo"],"topPlayers":["PlayerTen","PlayerThirteen"]}
Now I want to SELECT rows which have a "player" named "PlayerOne".
I've tried following SQL commands without success:
SELECT * FROM games WHERE data -> players = 'PlayerOne';
SELECT * FROM games WHERE data ->> players = 'PlayerOne';
The position of the array element won't be the same every time, so the direct reference you tried doesn't work - even after fixing your expression: data -> 'players' ->> 0 or data #>> '{players,0}'
Use the data type jsonb instead of json, and the jsonb containment operator #>:
SELECT *
FROM games
WHERE data #> '{"players":["PlayerOne"]}';
If you can't change the table definition, add a cast in the query:
...
WHERE data::jsonb #> '{"players":["PlayerOne"]}';
Either way, if the table is big, you want to support this with an index - an expression index in the latter case. See:
What's the proper index for querying structures in arrays in Postgres jsonb?
Postgres 9.4 jsonb array as table
I currently have a table sessions with a column actions(JSONB).
I am able to properly store an array from my frontend to my database using array_to_jsonb(array value).
The structure of my array (array value) looks like this:
var arrayStructure = [
{name: 'email client', check: false},
{name: 'send contract', check: false}
]
Unlike a lot of the questions I've seen on SO about modifying JSONB or accessing certain keys, I am only interested in converting the entire JSONB data back to an array for the frontend.
My temporary fix is to map through the array and use JSON.parse() in Javascript to restructure each object in the array. I cannot use JSON.parse() on the entire array, as it throws an error.
I'm looking for a query to bring back the array format to the JSONB data type stored. I've seen JSONB_SET, LATERAL, AND JSONB_ARRAY_ELEMENTS_TEXT. But not in a way that worked to bring back a proper array.
Starts As: JSONB in actions column in table named sessions
Should Result In: A query that brings back all rows, but with the actions column (JSONB) converted back to an array for the frontend:
select session_id, session_name, someFunction or lateral here(actions) from sessions
Screenshot of Sessions Table
I've tried queries like this:
SELECT
session_id,
actions::jsonb -> 'name' as name
FROM sessions;
And receive back null for name. I've tried ->> to access a deeper level, but that didn't work either.
This is half of the correct query result:
select session_id, jsonb_array_elements_text(actions)
from sessions
group by session_id;
Which results in this (only pay attention to results for session_id of 264):
query result
Now I have objects in their own rows as:
{"name": "some task", "check": "false}
When what I want for the actions column is:
[ {name: "some task", check: false}, {name: "other task", check: true} ]
So I need to further parse the JSON and group by session_id. I'm just struggling to build a sub-query that does that.
Steps to Create Set Up:
create table fakeSessions (
session_id serial primary key,
name varchar(20),
list jsonb
)
insert into fakeSessions(name, list)
VALUES(
'running',
'["{\"name\":\"inquired\",\"check\":false}", "{\"name\":\"sent online guide\",\"check\":false}", "{\"name\":\"booked!\",\"check\":false}"]'
)
insert into fakeSessions(name, list)
VALUES(
'snowboarding',
'["{\"name\":\"rental\",\"check\":false}", "{\"name\":\"booked ski passes\",\"check\":false}", "{\"name\":\"survey\",\"check\":false}"]'
)
The closest query I've created:
with exports as (
select jsonb_array_elements_text(actions)::jsonb as doc from sessions
)
select array_agg(doc) from
exports, sessions
group by session_id;
Get the text values, and then apply an aggregate function to those returned rows. Just can't get the select array_agg(doc) to work as expected. Most likely because I need a different function in that place.
Does this help?
demo:db<>fiddle
SELECT
jsonb_agg(elem)
FROM
sessions, jsonb_array_elements(actions) as elem
jsonb_array_elements() expands the jsonb array into one row each jsonb element
jsonb_agg() aggregates these jsonb elements into one big array.
I was able to reach the answer by building off of your query! Thank you so much. The query that got the expected outcome was this:
with exports as (
select session_id, jsonb_array_elements_text(actions)::jsonb as doc from sessions
)
select session_id, jsonb_agg(doc) from exports
group by session_id;
I wasn't using the jsonb_agg() function once I got the elements. The difference for getting the exact format was just using jsonb_array_elements_text::jsonb
I have a DB with a jsonb column where each row essentially holds an array of name value pairs. Example for a single jsonb value:
[
{"name":"foo", "value":"bar"},
{"name":"biz", "value":"baz"},
{"name":"beep", "value":"boop"}
]
How would I query for rows that contain a partial value? I.e., find rows with the JSON object key value ilike '%ba%'?
I know that I can use SELECT * FROM tbl WHERE jsoncol #> '[{"value":"bar"}]' to find rows where the JSON is that specific value, but how would I query for rows containing a pattern?
There are no built in jsonb operators nor any indexes supporting this kind of filter directly (yet).
I suggest an EXISTS semi-join:
SELECT t.*
FROM tbl t
WHERE EXISTS (
SELECT FROM jsonb_array_elements(t.jsoncol) elem
WHERE elem->>'value' LIKE '%ba%'
);
It avoids redundant evaluations and the final DISTINCT step you would need to get distinct rows with a plain CROSS JOIN.
If this still isn't fast enough, a way more sophisticated specialized solution for the given type of query would be to extract a concatenated string of unique values (with a delimiter that won't interfere with your search patterns) per row in an IMMUTABLE function, build a trigram GIN index on the functional expression and use the same expression in your queries.
Related:
Search for nested values in jsonb array with greater operator
Find rows containing a key in a JSONB array of records
Create Postgres JSONB Index on Array Sub-Object
Aside, if your jsonb values really look like the example, you could trim a lot of noise and just store:
[
{"foo":"bar"},
{"biz":"baz"},
{"beep":"boop"}
]
You can use the function jsonb_array_elements() in a lateral join and use its result value in the WHERE clause:
select distinct t.*
from my_table t
cross join jsonb_array_elements(jsoncol)
where value->>'value' like '%ba%'
Please, read How to query jsonb arrays with IN operator for notes about distinct and performance.
I have a jsonb field in a PostgreSQL table which was supposed to contain a dictionary like data aka {} but few of its entries got an array due to source data issues.
I want to weed out those entries. One of the ways is to perform following query -
select json_field from data_table where cast(json_field as text) like '[%]'
But this requires converting each jsonb field into text. With data_table having order of 200 million entries, this looks like bit of an overkill.
I investigated pg_typeof but it returns jsonb which doesn't help differentiate between a dictionary and an array.
Is there a more efficient way to achieve the above?
How about using the json_typeof function?
select json_field from data_table where json_typeof(json_field) = 'array'
In my table mytable I have a json field called data and I inserted json with a lot of keys & values.
I know that it's possible to select individual fields like so:
SELECT data->'mykey' as mykey from mytable
But how can I get an overview of all of the json keys on a certain depth? I would have expected something like
SELECT data->* from mytable
but that doesn't work. Is there something similar?
You can use the json_object_keys() function to get all the top-level keys of a json value:
SELECT keys.*
FROM mytable, json_object_keys(mytable.data) AS keys (mykey);
If you want to search at a deeper level, then first extract that deeper level from the json value using the #> operator:
SELECT keys.*
FROM mytable, json_object_keys(mytable.data #> '{level1, level2}') AS keys (mykey);
Note that the function returns a set of text, so you should invoke the function as a row source.
If you are using the jsonb data type, then use the jsonb_object_keys() function.