EPICS Controls Argonne National Laboratory

Experimental Physics and
Industrial Control System

1994  1995  1996  1997  1998  <19992000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024  Index 1994  1995  1996  1997  1998  <19992000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
<== Date ==> <== Thread ==>

Subject: Re: Y2K
From: "William L. Cruise" <[email protected]>
To: Andrew Johnson <[email protected]>
Cc: [email protected]
Date: Tue, 27 Jul 1999 10:07:46 -1000
Andrew is definitely right about this being an application
issue, and probably more right about not changing how the
basic tsStampToText works.  

However, what is really needed in our case is a direct 
conversion from timestamp to numerical values.  I did
see the TS_TEXT_MONDDYYYY option, but converting to the
month name and then to the month number would be too
unwieldy and inefficient.  So, we'll be writing our own
conversion routine, and for sure it will be Y2K compliant!

All I wanted to do was give a heads up to others who
may have used this function.  It seems that the EPICS
underlying time is OK -- it's just how you fetch it that
might cause a problem.

Bill

Andrew Johnson wrote:
> 
> Bill Cruise wrote:
> >
> > Well, I got quite a surprise when our system display showed
> > 1 Jan 1900!  Turns out that tsStampToText() is used to break
> > out the timestamp (which itself is OK), and it only has an
> > option for a "YY" type format.  Unfortunately this function
> > call is at the heart of our system.  We are running R3.13.beta11,
> > in case there are version differences in this area.
> 
> tsStampToText actually supports two output formats:
>     TS_TEXT_MONDDYYYY  Mon dd, yyyy hh:mm:ss.nano-secs
>     TS_TEXT_MMDDYY     mm/dd/yy hh:mm:ss.nano-secs
> Within the EPICS base source code the MMDDYY format is only ever requested
> in test code or logging output from the sequencer.  Changing the
> TS_TEXT_MMDDYY format to generate a 4-digit year would probably break more
> code than it fixed due to the longer string generated, which the
> application might not be able to handle properly.
> 
> The corresponding routime tsTextToStamp converts any 2-digit year < 90 to
> a timestamp in the year 20yy (Note to EPICS maintainers in ~2080: you may
> want to look at this) and handles 4-digit years correctly.
> 
> IMHO this is an application issue, not something that needs to be changed
> in EPICS base.
> 
> - Andrew
> --
> I brake for tail-gaters

-- 

William L. Cruise                             [email protected]
Electronics System Supervisor                   FAX:    808 885-7288
Canada-France-Hawaii Telescope                  Voice:  808 885-3122
P.O. Box 1597                     http://www.cfht.hawaii.edu/~cruise
Kamuela, HI   96743


References:
Y2K Bill Cruise
Re: Y2K Andrew Johnson

Navigate by Date:
Prev: Xy566 Waveform Garrett D. Rinehart
Next: CapFast Summer Special Dr. Chong Lee
Index: 1994  1995  1996  1997  1998  <19992000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
Navigate by Thread:
Prev: Re: Y2K Andrew Johnson
Next: Xy566 Waveform Garrett D. Rinehart
Index: 1994  1995  1996  1997  1998  <19992000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
ANJ, 10 Aug 2010 Valid HTML 4.01! · Home · News · About · Base · Modules · Extensions · Distributions · Download ·
· Search · EPICS V4 · IRMIS · Talk · Bugs · Documents · Links · Licensing ·