Re: psql has some accessibility issues on Windows - Mailing list pgsql-bugs

From Thomas Munro
Subject Re: psql has some accessibility issues on Windows
Date
Msg-id CA+hUKGLZVtWkzKz=duBG-8ctuNOdbxWtyBdtk8xH1BJNKFwrZg@mail.gmail.com
Whole thread Raw
In response to psql has some accessibility issues on Windows  (Jelte Fennema <Jelte.Fennema@microsoft.com>)
List pgsql-bugs
On Fri, Jul 17, 2020 at 10:39 PM Jelte Fennema
<Jelte.Fennema@microsoft.com> wrote:
> Datetimes are not being read by the screen reader as datetimes. So instead of saying "9th of March 2020, 10 hours 28
minutesand 46 seconds" it will instead say "2020 dash zero three dash zero nine 10 colon 28 colon 46"
 

I'll leave the table-display topics to Pavel (see some of his work at
[1] and [2], which I'm guessing might help here), but about the
timestamps:

I'm surprised it can't understand 2020-07-20 10:56:54.992092+12.
That's an extremely widely used date format, ISO 8601 date + time,
with the common modification of replacing "T" (between date part and
time part) with a space.  That's not formally allowed by ISO 8601 but
RFC3339 says it's OK ("5.6. Internet  Date/Time Format"), and a lot of
other software uses it.  You can also ask PostgreSQL to show dates in
a couple of other non-standard formats (try SET datestyle = postgres
for a historical format), but IMHO, RFC3339 should be high on the list
of formats understood by software trying to recognise timestamps.

[1] http://okbob.blogspot.com/2014/10/styles-for-unicode-borders-are-merged.html
[2] https://github.com/okbob/pspg



pgsql-bugs by date:

Previous
From: Tobias Völk
Date:
Subject: Improvement for query planner? (no, not about count(*) again ;-))
Next
From: Amit Kapila
Date:
Subject: Re: Buffers from parallel workers not accumulated to upper nodes with gather merge