Experimental Physics and Industrial Control System
Hi Michael,
On 2011-10-18 Michael Davidsaver wrote:
> On 10/18/2011 11:30 AM, Andrew Johnson wrote:
> > The number of elements of a long string (char array) must always include
> > the terminating nil character, thus an empty string must have an element
> > count of one, not zero. If you don't include the terminator in your
> > element count your string might get old garbage appended to it.
>
> Ok, so this covers the case of long strings. What about other types
> (eg. DOUBLE)? Isn't this an issue of general correctness/proper error
> reporting?
I believe there are a number of places in the dbAccess code that can't handle
zero-length arrays. Passing a zero element count sometimes means just process
the attributes (alarm, timestamp etc.), don't copy any data elements at all; a
second call is used to access the data elements in those cases.
I'm not saying this is how I'd /like/ it to work, but it is how I believe some
parts of the code currently operate.
- Andrew
--
Optimization is the process of taking something that works and
replacing it with something that almost works, but costs less.
-- Roger Needham
- References:
- aSub and zero length arrays Michael Davidsaver
- Re: aSub and zero length arrays Andrew Johnson
- Re: aSub and zero length arrays Michael Davidsaver
- Navigate by Date:
- Prev:
RE: How the EPICS IOC determines the link type {CA_LINK/DB_LINK/CONSTANT} Abadie Lana
- Next:
Build EPICS base-3.14.12.1-T222 in Linux problem. Chen Xue
- Index:
1994
1995
1996
1997
1998
1999
2000
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: aSub and zero length arrays Michael Davidsaver
- Next:
Build EPICS base-3.14.12.1-T222 in Linux problem. Chen Xue
- Index:
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
<2011>
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024