Home > mailing lists

Thread: Use or not record count on examples

Use or not record count on examples

From

Marcos Pegoraro

Date:

23 November, 18:29:47

There is no rule of when the number of records at the end of the lists should be shown or not

Sometimes we show that line "(4 rows)", but sometimes not.

Should we have a standard for it ? Should we add them all or remove them all ?

I think they are useless. Or maybe shown when no rows are returned.

datatype.sgml

a | char_length
------+-------------
ok | 2

name | current_mood
------+--------------
Moe | happy
(1 row)

func.sgml

a | b
---+-----
1 | foo
2 |

Or maybe when not all records are shown, like in planstats.sgml

index | values | nulls | frequency | base_frequency
-------+----------+-------+-----------+----------------
0 | {0, 0} | {f,f} | 0.01 | 0.0001
1 | {1, 1} | {f,f} | 0.01 | 0.0001
...
49 | {49, 49} | {f,f} | 0.01 | 0.0001
50 | {50, 50} | {f,f} | 0.01 | 0.0001
...
97 | {97, 97} | {f,f} | 0.01 | 0.0001
98 | {98, 98} | {f,f} | 0.01 | 0.0001
99 | {99, 99} | {f,f} | 0.01 | 0.0001
(100 rows)

regards

Marcos

Re: Use or not record count on examples

From

David Rowley

Date:

25 November, 03:54:38

On Sun, 24 Nov 2024 at 01:30, Marcos Pegoraro <marcos@f10.com.br> wrote:
> There is no rule of when the number of records at the end of the lists should be shown or not
> Sometimes we show that line "(4 rows)", but sometimes not.
> Should we have a standard for it ? Should we add them all or remove them all ?

It looks like we have a few ways to try to trim down the query result
output to save space on the page. In [1] we do:

> SELECT format('|%10s|', 'foo');
> Result: |       foo|

I'm ok with the documentation being terse like this. If we were to
standardise on something, that would likely force those examples to be
much more verbose than they need to be. I don't think that verbosity
would add any additional value.

David

[1] https://www.postgresql.org/docs/current/functions-string.html

Re: Use or not record count on examples

From

Marcos Pegoraro

Date:

25 November, 19:08:58

Em dom., 24 de nov. de 2024 às 18:54, David Rowley <dgrowleyml@gmail.com> escreveu:

> SELECT format('|%10s|', 'foo');

This example you said returns one value, so I think it is ok.

I'm talking about lists with multiple fields or multiple lines. In [1] we have both modes.

I don't think that "(2 rows)" is useful

select jsonb_path_query(:'json', 'strict $.**.HR');
jsonb_path_query
------------------
73
135

select jsonb_path_query(:'json', 'lax $.track.segments[*].location ?(@[*] > 15)');
jsonb_path_query
------------------
47.763
47.706
(2 rows)

[1] https://www.postgresql.org/docs/current/functions-json.html#FUNCTIONS-SQLJSON-STRICT-AND-LAX-MODES

Re: Use or not record count on examples

From

Marcos Pegoraro

Date:

29 November, 19:13:30

There are lots of strange things related to this (... rows)

First, (0 rows) are completely useless. If no one record is shown, why do we need that list ?

We could change (0 rows), like this one on dblink.sgml

SELECT * FROM dblink_get_notify();
notify_name | be_pid | extra
-------------+--------+-------
(0 rows)

We could change it to

SELECT * FROM dblink_get_notify();<returnvalue>no rows returned</returnvalue>

Current version does not have these 19 rows as DOC says.

func.sgml

SELECT * FROM pg_ls_dir('.') WITH ORDINALITY AS t(ls,n);

...

(19 rows)

Show all settings counts 196 , but that is not true for current version.

show.sgml

SHOW ALL;

Why do we need to show number of lines of explain ? I think all them could be removed

explain.sgml

EXPLAIN SELECT * FROM foo WHERE i = 4;

QUERY PLAN
--------------------------------------------------------------
Index Scan using fi on foo (cost=0.00..5.98 rows=1 width=4)
Index Cond: (i = 4)
(2 rows)

Please let me know if you agree with some of these changes and I can create a patch with that

regards

Marcos