Hello Eike,
Eike Rathke wrote:
Hi Leonard,
I just noticed the following formula in one of the issues:
"=IF(AND((ISERROR(SEARCH("Pass",$F8))),
(ISERROR(SEARCH("Warn",$F8))),
(ISERROR(SEARCH("Closed",$F8)))),0,1)"
IIRC we already mentioned the to be implemented IFERROR() function in
this context somewhere.
Well, I will let you simplify this formula using IFERROR(). However, I
sincerely hope that implementing my proposal will take less time than
simplifying the above formula using IFERROR(). ;-)
And take my hint, there might be a remote chance that you will fail.
Lets assume now that SEARCH() will return 0 if the string is NOT found.
Then this whole formula ends up as:
"=IF(OR(SEARCH("Pass",$F8);
SEARCH("Warn",$F8);
SEARCH("Closed",$F8)); 1; 0)"
We got rid of 3 functions and a lot of additional brackets. Now, this is
a simplification. The only way to go even further, is to implement
arrays within SEARCH(), but that's another story.
I'll be back with more comments when I have some more time available.
Sincerely,
Leonard
And this only because SEARCH() returns an ERROR if the string is not found.
[http://www.openoffice.org/issues/show_bug.cgi?id=91146]
That issue has a different topic.
How many examples should I provide to get the developers implement an
alternative Search function that returns 0 IF the string is not found.
I don't think it is a matter of the amount of examples. Usually no issue
is. Actually the process would be:
1. Propose the new function to the OASIS formula subcommittee, using the
comments form at OASIS. You know where it is.
2. Wait until the proposal is discussed and accepted.
3. Find someone who will implement it.
But, once IFERROR() was implemented, a different SEARCH() would only be
necessary to distinguish between not found and propagated error. So for
the overall enhancement it may be better to skip steps #1 and #2 and
start with #3 to implement IFERROR(), as that may be useful in all
context where an expression may return an error.
The explanation that this is different than Excel is strange considering
issue 5658, which is definitely a more severe inconsistency than this
one [http://www.openoffice.org/issues/show_bug.cgi?id=5658].
Nice to throw in an unrelated but one of the most controversially
discussed issues with the most useless comments we have :-/ This really
does not help.
Eike
Issue 5658 is the single most severe issue regarding interoperability
with MS Excel. Anything else can be circumvented, because people will
know that something went wrong when they see some ERROR popping up. And
correcting an error is possible once you spot it. But issue 5658 is
different, because the user will never suspect that something went wrong.
Please accept my expertise, I have seen many things.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
