Function to return per-column counts?

On 09/28/2017 04:34 PM, Seamus Abshere wrote:

hey,

Does anybody have a function lying around (preferably pl/pgsql) that
takes a table name and returns coverage counts?

What is "coverage count"?

cheers

--
Tomas Vondra http://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

On Thu, Sep 28, 2017 at 12:15 PM, Tomas Vondra <tomas.vondra@2ndquadrant.com

wrote:

On 09/28/2017 04:34 PM, Seamus Abshere wrote:

hey,

Does anybody have a function lying around (preferably pl/pgsql) that
takes a table name and returns coverage counts?

What is "coverage count"?

​I'm guessing it's what is described here:
https://www.red-gate.com/blog/sql-cover​

IIUC, this is "code coverage" for things kept in your RDMS system, such as
triggers, procedures, and other "code" items which are implicitly part of
your application code.

cheers

--
Tomas Vondra http://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

--
I just child proofed my house.
But the kids still manage to get in.

Maranatha! <><
John McKown

Does anybody have a function lying around (preferably pl/pgsql) that
takes a table name and returns coverage counts?

What is "coverage count"?

Ah, I should have explained better. I meant how much of a column is
null.

Basically you have to

0. count how many total records in a table
1. discover the column names in a table
2. for each column name, count how many nulls and subtract from total
count

If nobody has one written, I'll write one and blog it.

Thanks!
Seamus

PS. In a similar vein, we published
http://blog.faraday.io/how-to-do-histograms-in-postgresql/ which gives
plpsql so you can do:

SELECT * FROM histogram($table_name_or_subquery, $column_name)

--
Seamus Abshere, SCEA
https://www.faraday.io
https://github.com/seamusabshere
https://linkedin.com/in/seamusabshere

--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

On Thu, Sep 28, 2017 at 3:31 PM, Seamus Abshere <seamus@abshere.net> wrote:

Does anybody have a function lying around (preferably pl/pgsql) that
takes a table name and returns coverage counts?

What is "coverage count"?

Ah, I should have explained better. I meant how much of a column is
null.

Basically you have to

0. count how many total records in a table
1. discover the column names in a table
2. for each column name, count how many nulls and subtract from total
count

If nobody has one written, I'll write one and blog it.

Thanks!
Seamus

PS. In a similar vein, we published
http://blog.faraday.io/how-to-do-histograms-in-postgresql/ which gives
plpsql so you can do:

SELECT * FROM histogram($table_name_or_subquery, $column_name)

--
Seamus Abshere, SCEA
https://www.faraday.io
https://github.com/seamusabshere
https://linkedin.com/in/seamusabshere

--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

I can't really do the full query for you, but the following should be able
to give you a head start:

SELECT c.relname AS table,
a.attname AS column,
a.attnum AS colnum,
s.stanullfrac as pct_null,
s.stadistinct
FROM pg_class c
JOIN pg_attribute a ON a.attrelid = c.oid
JOIN pg_statistic s ON (s.starelid = c.oid AND s.staattnum = a.attnum)
WHERE c.relname = 'your_table_name'
AND a.attnum > 0
ORDER BY 3

--
*Melvin Davidson*
I reserve the right to fantasize. Whether or not you
wish to share my fantasy is entirely up to you.