On Tue, 21 May 2013 10:41:39 -0500, Paul Gilmartin wrote:

>On Tue, 21 May 2013 10:06:12 -0500, Bill Godfrey wrote:
>
>>On Tue, 21 May 2013 08:24:41 -0400, Charles Mills wrote:
>>>
>>>Also, thanks for your STCK(E) converter. 
>>>http://www.longpelaexpertise.com.au/toolsTOD.php Have used that from time to 
>>>time.
>>>
>> The STCK converter is not handling leap seconds. If you give it STCK values 
>> from various years listed in chapter 4 of Principles of Operation, the 
>> returned time for many of them is off by approximately (not exactly) the 
>> number of leap seconds that were not taken into account.
>> 
>Isn't that for compatibility with the STCKCONV macro?
>
Perhaps, but the results are still sometimes a second off from STCKCONV. Not a 
big deal, but it might matter to some people.
>>Also, if you give it the value 7d91048bca000000 for 1/1/1970 (not in the 
>>manual) it says it is before Jan 1 1970. That's not a leap seconds problem, 
>>but it is incorrect.
>> 
>Sounds like a POSIX limitation.
>
I don't see how POSIX would have any affect. 1-1-1970 at 00:00:00 is a valid 
date in POSIX. The site does say "This tool can only convert TOD values after 
Jan 1, 1970" and it does give a correct result for Jan 2 1970, so they really 
mean "after". STCKCONV converts the value I gave above to Jan 1 1970 at 
00:00:00 as expected.

Bill

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to