psql - add SHOW_ALL_RESULTS option

Поиск
Список
Период
Сортировка
От Fabien COELHO
Тема psql - add SHOW_ALL_RESULTS option
Дата
Msg-id alpine.DEB.2.21.1904132231510.8961@lancre
обсуждение исходный текст
Ответы RE: psql - add SHOW_ALL_RESULTS option  ("Iwata, Aya" <iwata.aya@jp.fujitsu.com>)
Список pgsql-hackers
Hello devs,

The attached patch implements a new SHOW_ALL_RESULTS option for psql, 
which shows all results of a combined query (\;) instead of only the last 
one.

This solves a frustration that intermediate results were hidden from view 
for no good reason that I could think of.

For that, call PQsendQuery instead of (mostly not documented) PQexec, and 
rework how results are processed afterwards.

Timing is moved to ProcessQueryResults to keep the last result handling 
out of the measured time. I think it would not be a big deal to include 
it, but this is the previous behavior.

In passing, refactor a little and add comments. Make function names about 
results plural or singular consistently with the fact the it processes one 
or several results. Change "PrintQueryResult" to "HandleQueryResult" 
because it was not always printing something. Also add a HandleCopyResult 
function, which makes the patch a little bigger by moving things around 
but clarifies the code.

Code in "common.c" is actually a little shorter than the previous version. 
From my point of view the code is clearer than before because there is 
only one loop over results, not an implicit one within PQexec and another 
one afterwards to handle copy.

Add a few tests for the new feature.

IMHO this new setting should be on by default: few people know about \; so 
it would not change anything for most, and I do not see why those who use 
it would not be interested by the results of all the queries they asked 
for.

-- 
Fabien.
Вложения

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Tom Lane
Дата:
Сообщение: Re: Commit message / hash in commitfest page.
Следующее
От: Tomas Vondra
Дата:
Сообщение: Re: Should the docs have a warning about pg_stat_reset()?