Extract nested values as columns Google BigQuery? - sql

I have a table with nested values, like the following:
I'd like to grab the values, with keys as columns without multiple cross joins.
i.e.
SELECT
owner_id,
owner_type,
domain,
metafields.value AS name,
metafields.value AS image,
metafields.value AS location,
metafields.value AS draw
FROM
example_table
Obviously, the above won't work for this, but the following output would be desired:
In the actual table there are hundreds of metafields per owner_id, and hundreds of owner_ids, and owner_types. Multiple joins to other tables for owner_types is fine, but for the same owner type, I don't want to have to join multiple times.
Basically, I need to be able to select the key to which the column corresponds, and display the relevant value for that column. Without, having to display every metafield available.
Any way of doing this?

Consider below approach
select * except(id) from (
select t.* except(metafields),
to_json_string(t) id, key, value
from your_table t, unnest(metafields) kv
)
pivot (min(value) for key in ('name', 'image', 'location', 'draw'))
if applied to sample data in your question - output is

You can use the subqueries and SAFE_offset statement and get a value from an array at a specific location.
Also, you need to use STRING_AGG, which returns a value (either STRING or BYTES) obtained by concatenating non-null values.
With the information you shared, you can use the query below.
With this code, you will get all the columns separated by a comma:
WITH sequences AS
(
SELECT 1 as ID,"product" AS owner_type,"beta.com" AS domain,["name","image","lcation","draw"] AS metalfields_key, ["big","pic.png","utha","1"] AS metalfields_value
),
Val as(
SELECT distinct id, owner_type,domain, value FROM sequences, sequences.metalfields_value as value, sequences.metalfields_key
), text as(
SELECT
id, owner_type, domain,
STRING_AGG(value ORDER BY value) AS Text
FROM Val
GROUP BY owner_type, domain, id
)
In this code, you will get each element that is separated by a comma and return them by columns.
SELECT DISTINCT t1.id, t1.owner_type,domain,
split(t1.text, ',')[SAFE_offset(1)] as name,
split(t1.text, ',')[SAFE_offset(2)] as image,
split(t1.text, ',')[SAFE_offset(3)] as location,
split(t1.text, ',')[SAFE_offset(0)] as draw
from text as t1
You can see the result.

Related

Flatten out a Record field

Assume we have a BigQuery table like the following:
Where "Col1" is a Record field. What would be the most efficient way to "flatten out" the table so that part of the Record field becomes columns:
Consider below option
select * from (
select key, name, value
from your_table, unnest(col1)
)
pivot (any_value(value) for name in ('A','B','C','D'))
if applied to sample data as in your question
the output is

Selecting Columns which are not null in Athena(Metabase)

I have a table of 1000+ columns in Athena(Metabase), and I want to know how can I extract only those columns which are not null for a certain group of ID.
Typically, this would need an UNPIVOTING of your columns to rows and then check where not null and then back to PIVOT.
From the documentation, Athena may do it simpler.
As documented here
SELECT filter(ARRAY [-1, NULL, 10, NULL], q -> q IS NOT NULL)
Which returns:
[-1,10]
Unfortunately, since there is no ability to be dynamic until we get to an array, this looks like:
WITH dataset AS (
SELECT
ID,
ARRAY[field1, field2, field3, .....] AS fields
FROM
Your1000ColumnTable
)
SELECT ID, SELECT filter(fields, q -> q IS NOT NULL)
FROM dataset
If you need to access the column names from the array, use a mapping to field names when creating the array as seen here

I need to SELECT a group of columns not null

Basically, i have a table that have a series of columns named:
ATTRIBUTE10, ATTRIBUTE11, ATTRIBUTE12 ... ATTRIBUTE50
I want a query that gives me all the columns from ATTRIBUTE10 to ATTRIBUTE50 not null
As others have commented we aren't exactly sure of your requirements, but if you want a list the UNPIVOT can do that...
SELECT attribute , value
FROM
(SELECT * from YourFile) p
UNPIVOT
(value FOR attribute IN
(attribute1, attribute2, attribute3, etc.)
)AS unpvt
May be you can use where condition for all columns Or use between operator as below.
For All Columns
where ATTRIBUTE10 is not null and ATTRIBUTE11 is not null ...... and ATTRIBUTE50 is not null
By using between operator
where ATTRIBUTE10 between ATTRIBUTE11 and ATTRIBUTE50
One way to approach the problem is to unfold your table-with-a-zillion-like-named-attributes into one in which you've got one attribute per row, with appropriate foreign keys back to the original table. So something like:
CREATE TABLE ATTR_TABLE AS
SELECT ID_ATTR, ID_TABLE_WITH_ATTRS, ATTR
FROM (SELECT ((ID_TABLE_WITH_ATTRS-1)*100)+1 AS ID_ATTR, ID_TABLE_WITH_ATTRS, ATTRIBUTE10 AS ATTR FROM TABLE_WITH_ATTRS UNION ALL
SELECT ((ID_TABLE_WITH_ATTRS-1)*100)+2, ID_TABLE_WITH_ATTRS, ATTRIBUTE11 FROM TABLE_WITH_ATTRS UNION ALL
SELECT ((ID_TABLE_WITH_ATTRS-1)*100)+3, ID_TABLE_WITH_ATTRS, ATTRIBUTE12 FROM TABLE_WITH_ATTRS);
This only unfolds ATTRIBUTE10, ATTRIBUTE11, and ATTRIBUTE12, but you should be able to get the idea - the rest of the attributes just requires a little cut-n-paste on your part.
You can then query this table to find your non-NULL attributes as
SELECT *
FROM ATTR_TABLE
WHERE ATTR IS NOT NULL
ORDER BY ID_ATTR
Hopefully the difficulty you're encountering in dealing with this table-with-a-zillion-repeated-fields teaches you a hard lesson about exactly why tables with repeated fields or groups of fields are a Bad Idea.
dbfiddle here

How to efficiently select records matching substring in another table using BigQuery?

I have a table of several million strings that I want to match against a table of about twenty thousand strings like this:
#standardSQL
SELECT record.* FROM `record`
JOIN `fragment` ON record.name
LIKE CONCAT('%', fragment.name, '%')
Unfortunately this is taking an awful long time.
Considering that the fragment table is only 20k records, can I load it into a JavaScript array using a UDF and match it that way? I'm trying to figure out how to this right now but perhaps there's already some magic I could do here to make this faster. I tried a CROSS JOIN and got resource exceeded fairly quickly. I've also tried using EXISTS but I can't reference the record.name inside that subquery's WHERE without getting an error.
Example using Public Data
This seems to reflect about the same amount of data ...
#standardSQL
WITH record AS (
SELECT LOWER(text) AS name
FROM `bigquery-public-data.hacker_news.comments`
), fragment AS (
SELECT LOWER(name) AS name, COUNT(*)
FROM `bigquery-public-data.usa_names.usa_1910_current`
GROUP BY name
)
SELECT record.* FROM `record`
JOIN `fragment` ON record.name
LIKE CONCAT('%', fragment.name, '%')
Below is for BigQuery Standard SQL
#standardSQL
WITH record AS (
SELECT LOWER(text) AS name
FROM `bigquery-public-data.hacker_news.comments`
), fragment AS (
SELECT DISTINCT LOWER(name) AS name
FROM `bigquery-public-data.usa_names.usa_1910_current`
), temp_record AS (
SELECT record, TO_JSON_STRING(record) id, name, item
FROM record, UNNEST(REGEXP_EXTRACT_ALL(name, r'\w+')) item
), temp_fragment AS (
SELECT name, item FROM fragment, UNNEST(REGEXP_EXTRACT_ALL(name, r'\w+')) item
)
SELECT AS VALUE ANY_VALUE(record) FROM (
SELECT ANY_VALUE(record) record, id, r.name name, f.name fragment_name
FROM temp_record r
JOIN temp_fragment f
USING(item)
GROUP BY id, name, fragment_name
)
WHERE name LIKE CONCAT('%', fragment_name, '%')
GROUP BY id
above was completed in 375 seconds, while original query is still running at 2740 seconds and keep running, so I will not even wait for it to complete
Mikhail's answer appears to be faster - but lets have one that doesn't need to SPLIT nor separate the text into words.
First, compute a regular expression with all the words to be searched:
#standardSQL
WITH record AS (
SELECT text AS name
FROM `bigquery-public-data.hacker_news.comments`
), fragment AS (
SELECT name AS name, COUNT(*)
FROM `bigquery-public-data.usa_names.usa_1910_current`
GROUP BY name
)
SELECT FORMAT('(%s)',STRING_AGG(name,'|'))
FROM fragment
Now you can take that resulting string, and use it in a REGEX ignoring case:
#standardSQL
WITH record AS (
SELECT text AS name
FROM `bigquery-public-data.hacker_news.comments`
), largestring AS (
SELECT '(?i)(mary|margaret|helen|more_names|more_names|more_names|josniel|khaiden|sergi)'
)
SELECT record.* FROM `record`
WHERE REGEXP_CONTAINS(record.name, (SELECT * FROM largestring))
(~510 seconds)
As eluded to in my question, I worked on a version using a JavaScript UDF which solves this albeit in a slower way than the answer I accepted. For completeness, I'm posting it here because perhaps someone (like myself in the future) may find it useful.
CREATE TEMPORARY FUNCTION CONTAINS_ANY(str STRING, fragments ARRAY<STRING>)
RETURNS STRING
LANGUAGE js AS """
for (var i in fragments) {
if (str.indexOf(fragments[i]) >= 0) {
return fragments[i];
}
}
return null;
""";
WITH record AS (
SELECT text AS name
FROM `bigquery-public-data.hacker_news.comments`
WHERE text IS NOT NULL
), fragment AS (
SELECT name AS name, COUNT(*)
FROM `bigquery-public-data.usa_names.usa_1910_current`
WHERE name IS NOT NULL
GROUP BY name
), fragment_array AS (
SELECT ARRAY_AGG(name) AS names, COUNT(*) AS count
FROM fragment
GROUP BY LENGTH(name)
), records_with_fragments AS (
SELECT record.name,
CONTAINS_ANY(record.name, fragment_array.names)
AS fragment_name
FROM record INNER JOIN fragment_array
ON CONTAINS_ANY(name, fragment_array.names) IS NOT NULL
)
SELECT * EXCEPT(rownum) FROM (
SELECT record.name,
records_with_fragments.fragment_name,
ROW_NUMBER() OVER (PARTITION BY record.name) AS rownum
FROM record
INNER JOIN records_with_fragments
ON records_with_fragments.name = record.name
AND records_with_fragments.fragment_name IS NOT NULL
) WHERE rownum = 1
The idea is that the list of fragments is relatively small enough that it can be processed in an array, similar to Felipe's answer using regular expressions. The first thing I do is create a fragment_array table which is grouped by the fragment lengths ... a cheap way of preventing an over-sized array which I found can cause UDF timeouts.
Next I create a table called records_with_fragments that joins those arrays to the original records, finding only those which contain a matching fragment using the JavaScript UDF CONTAINS_ANY(). This will result in a table containing some duplicates since one record may match multiple fragments.
The final SELECT then pulls in the original record table, joins to records_with_fragments to determine which fragment matched, and also uses the ROW_NUMBER() function to prevent duplicates, e.g. only showing the first row of each record as uniquely identified by its name.
Now, the reason I do the join in the final query is because in my actual data there are more fields I want besides just the string being matched. Earlier on in my actual data I create a table of DISTINCT strings which then later need to be re-joined.
Voila! Not the most elegant but it gets the job done.

Postgres Query of an Array using LIKE

I am querying a database in Postgres using psql. I have used the following query to search a field called tags that has an array of text as it's data type:
select count(*) from planet_osm_ways where 'highway' = ANY(tags);
I now need to create a query that searches the tags fields for any word starting with the letter 'A'. I tried the following:
select count(*) from planet_osm_ways where 'A%' LIKE ANY(tags);
This gives me a syntax error. Any suggestions on how to use LIKE with an array of text?
Use the unnest() function to convert array to set of rows:
SELECT count(distinct id)
FROM (
SELECT id, unnest(tags) tag
FROM planet_osm_ways) x
WHERE tag LIKE 'A%'
The count(dictinct id) should count unique entries from planet_osm_ways table, just replace id with your primary key's name.
That being said, you should really think about storing tags in a separate table, with many-to-one relationship with planet_osm_ways, or create a separate table for tags that will have many-to-many relationship with planet_osm_ways. The way you store tags now makes it impossible to use indexes while searching for tags, which means that each search performs a full table scan.
Here is another way to do it within the WHERE clause:
SELECT COUNT(*)
FROM planet_osm_ways
WHERE (
0 < (
SELECT COUNT(*)
FROM unnest(planet_osm_ways) AS planet_osm_way
WHERE planet_osm_way LIKE 'A%'
)
);