Re: Subsecond decoded-time again

[Lars Ingebrigtsen]

Paul Eggert <address@hidden> writes:

> On 9/29/19 6:01 AM, Lars Ingebrigtsen wrote:
>> I thought I remembered that we settled on giving a special parameter to
>> `decode-time' to make it return the sub-second bits, too?
>
> Yes, it's the optional new FORM parameter of decode-time. E.g,,
> (decode-time '(1566009571321 . 1000) t t) yields ((31321 . 1000) 39 2
> 17 8 2019 6 nil 0). This example is taken from the manual.

Ah, I was confused by this bit in the doc string:

----
The optional FORM specifies the form of the SEC member.  If ‘integer’,
SEC is an integer; if t, SEC uses the same resolution as TIME.  An
omitted or nil FORM is currently treated like ‘integer’, but this may
change in future Emacs versions.
----

I think "if t, SEC uses the same resolution as TIME" could possibly be
expanded upon.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no