Re: SQL/JSON path issues/questions

Alexander Korotkov <a.korotkov@postgrespro.ru>

From: Alexander Korotkov <a.korotkov@postgrespro.ru>
To: Liudmila Mantrova <l.mantrova@postgrespro.ru>
Cc: Kyotaro Horiguchi <horikyota.ntt@gmail.com>, Thom Brown <thom@linux.com>, PostgreSQL-development <pgsql-hackers@postgresql.org>
Date: 2019-07-03T20:59:01Z
Lists: pgsql-hackers
Hi!

On Wed, Jul 3, 2019 at 5:27 PM Liudmila Mantrova
<l.mantrova@postgrespro.ru> wrote:
>
> I have rechecked the standard and I agree that we should use "filter
> expression" whenever possible.
> "A filter expression must be enclosed in parentheses..." looks like an
> oversight, so I fixed it. As for what's actually enclosed, I believe we
> can still use the word "condition" here as it's easy to understand and
> is already used in our docs, e.g. in description of the WHERE clause
> that serves a similar purpose.
> The new version of the patch fixes the terminology, tweaks the examples,
> and provides some grammar and style fixes in the jsonpath-related chapters.


It looks good to me.  But this sentence looks a bit too complicated.

"It can be followed by one or more accessor operators to define the
JSON element on a lower nesting level by which to filter the result."

Could we phrase this as following?

"In order to filter the result by values lying on lower nesting level,
@ operator can be followed by one or more accessor operators."

------
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company



Commits

  1. Fixes for jsonpath filter expression elements table in docs

  2. Assorted fixes for jsonpath documentation

  3. Fix description for $varname jsonpath variable

  4. Improve documentation for jsonpath like_regex predicate

  5. Support 'q' flag in jsonpath 'like_regex' predicate