DateAndTime class>>#fromSeconds: and #fromMilliseconds: have problems
Locked
DateAndTime class>>#fromSeconds: and #fromMilliseconds: have problems
 
Administrator
|
This test was run in a 32-bit 9.2 image, but I suspect the defect is ancient. DateAndTime fromSeconds: 0 1901-01-01T00:00:00-08:00 DateAndTime fromSeconds: -1 1900-12-31T23:59:59-08:00DateAndTime fromSeconds: -12 1900-12-31T23:59:48-08:00 DateAndTime fromSeconds: -123 1900-12-31T23:57:57-08:00 DateAndTime fromSeconds: -1234 1900-12-31T23:39:26-08:00 DateAndTime fromSeconds: -12345 1900-12-31T20:34:15-08:00 DateAndTime fromSeconds: -123456 1900-12-30T13:42:24-08:00 DateAndTime fromSeconds: -1234567 1900-12-17T17:03:53-08:00 DateAndTime fromSeconds: -12345678 1900-08-11T02:38:42-08:00 DateAndTime fromSeconds: -123456789 1897-02-01T02:26:51-08:00 DateAndTime fromSeconds: -1234567890 1861-11-17T00:28:30-07:52:58 DateAndTime fromSeconds: -12345678901 1509-10-13T04:44:59-07:52:58 DateAndTime fromMilliseconds: 0 1901-01-01T00:00:00-08:00 DateAndTime fromMilliseconds: -1 1900-12-31T23:59:59.999-08:00 DateAndTime fromMilliseconds: -12 1900-12-31T23:59:59.988-08:00 DateAndTime fromMilliseconds: -123 1900-12-31T23:59:59.877-08:00 DateAndTime fromMilliseconds: -1234 1900-12-31T23:59:58.766-08:00 DateAndTime fromMilliseconds: -12345 1900-12-31T23:59:47.655-08:00 DateAndTime fromMilliseconds: -123456 1900-12-31T23:57:56.544-08:00 DateAndTime fromMilliseconds: -1234567 1900-12-31T23:39:25.433-08:00 DateAndTime fromMilliseconds: -12345678 1900-12-31T20:34:14.322-08:00 DateAndTime fromMilliseconds: -123456789 1900-12-30T13:42:23.-7-8-9-08:00 DateAndTime fromMilliseconds: -1234567890 1900-12-17T17:03:52.-8-9-08:00 DateAndTime fromMilliseconds: -12345678901 1900-08-11T02:38:41.-90-1-08:00 DateAndTime fromMilliseconds: -123456789012 1897-02-01T02:26:50.0-1-2-08:00 DateAndTime fromMilliseconds: -1234567890123 1861-11-17T00:28:29.-1-2-3-07:52:58 DateAndTime fromMilliseconds: -12345678901234 1509-10-13T04:44:58.-2-3-4-07:52:58 You received this message because you are subscribed to the Google Groups "VA Smalltalk" group. To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email]. To view this discussion on the web visit https://groups.google.com/d/msgid/va-smalltalk/953df8ba-3425-468f-b446-b8a663f453b2%40googlegroups.com. |
Re: DateAndTime class>>#fromSeconds: and #fromMilliseconds: have problems
|
Administrator
|
DateAndTime class>>#fromSeconds:offsetSeconds: does appear to work correctly based on all the numbers I tried below. On Wednesday, April 29, 2020 at 10:27:32 AM UTC-7, Richard Sargent wrote:
You received this message because you are subscribed to the Google Groups "VA Smalltalk" group. To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email]. To view this discussion on the web visit https://groups.google.com/d/msgid/va-smalltalk/1c8d72ac-6b90-4fdf-a0f7-be09e01c8a4c%40googlegroups.com. |
Re: DateAndTime class>>#fromSeconds: and #fromMilliseconds: have problems
|
Administrator
|
Spoke too soon! DateAndTime fromSeconds: -2545887519165/1000 offsetSeconds: -25200 yields 1820-04-28T10:01:20.-1-6-5-07:00 On Wednesday, April 29, 2020 at 10:39:39 AM UTC-7, Richard Sargent wrote:
You received this message because you are subscribed to the Google Groups "VA Smalltalk" group. To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email]. To view this discussion on the web visit https://groups.google.com/d/msgid/va-smalltalk/b699d33a-3d97-43b5-9b56-ab669630ba89%40googlegroups.com. |
Re: DateAndTime class>>#fromSeconds: and #fromMilliseconds: have problems
|
Administrator
|
Aha! It looks like DateAndTime>>#printMilliseconds:on: fails
to recognize "negative time" from negative milliseconds when printing
the millisecond portion of the time. Changing DateAndTime>>#printOn: to add %_ to the time format correctly prints 1820-04-28T10:01:20.835. Doing so also makes the subsequent (and defective) statement self printMilliseconds: self asMilliseconds on: aStream. unnecessary. It turns out that once upon a time, we didn't have standard time. We used local mean time (LMT). If you use a date far enough in the past, you get LMT time zone entries with a time zone offset that is directly related to the named place's longitude. In the case of Los Angeles, this is -07:52:58. Weird, but what can you do? On Wednesday, April 29, 2020 at 11:27:44 AM UTC-7, Richard Sargent wrote:
You received this message because you are subscribed to the Google Groups "VA Smalltalk" group. To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email]. To view this discussion on the web visit https://groups.google.com/d/msgid/va-smalltalk/776da0e5-75bb-40e4-9da1-905efd62dd93%40googlegroups.com. |
Re: DateAndTime class>>#fromSeconds: and #fromMilliseconds: have problems
|
It would be nice if just adding '%_' to the time format would work, but unfortunately it isn't quite that simple:
-- - %_ doesn't supply a leading period (.), so you instead need to add '.%_' to get the proper separator - %_ doesn't work correctly (in this context) for a time with zero milliseconds - it will print a single 0 in this case So I think something like this will work better: | tfmt | tfmt := (self asMilliseconds rem: 1000) ~= 0 ifTrue: [ tfmt := 'T%H:%M:%S.%_' ] "$NON-NLS$" ifFalse: [ tfmt := 'T%H:%M:%S' ]. "$NON-NLS$" Locale current lcTime copy dFmt: '%Y-%m-%d'; "$NON-NLS$" tFmt: tfmt; printDate: self date on: aStream; printTime: self time on: aStream. " self printMilliseconds: self asMilliseconds on: aStream." DateAndTime fromMilliseconds: -123456789 1900-12-30T13:42:23.211-05:00 DateAndTime year:2020 day: 101 hour: 11 minute: 34 second:17 2020-04-10T11:34:17-04:00 John
You received this message because you are subscribed to the Google Groups "VA Smalltalk" group. To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email]. To view this discussion on the web visit https://groups.google.com/d/msgid/va-smalltalk/41574481-961a-45bd-bb12-439f47ef73e6%40googlegroups.com. |
Re: DateAndTime class>>#fromSeconds: and #fromMilliseconds: have problems
|
Administrator
|
On Mon, May 18, 2020 at 10:51 AM John O'Keefe <[hidden email]> wrote:
Good point, John. I agree with your proposal. The results look good, like what one would expect.
You received this message because you are subscribed to the Google Groups "VA Smalltalk" group. To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email]. To view this discussion on the web visit https://groups.google.com/d/msgid/va-smalltalk/CAGNapEMoq2e%3DjYgg-TYPGufz8L3MpKmPduZmo9hK%2BmCJmWUnDQ%40mail.gmail.com. |
«
Return to Instantiations
|
1 view|%1 views
| Free forum by Nabble | Edit this page |