[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [office-formula] Section 4 Types: TBD: Should "blank" also be a possible value result?
Hi David, On Tuesday, 2006-11-21 17:47:34 -0500, David A. Wheeler wrote: > > > >An "empty cell" return value for example is needed > > > > in [HV]LOOKUP results, where for an empty cell the expression > > > > ISBLANK(LOOKUP(...)) must yield TRUE (OOo doesn't implement that yet). > > > > > > I'm don't think that [HV]LOOKUP actually return empty cell in many implementations. > > > Should we spec it anyway, or leave that implementation-defined this go around? > > > I'm thinking of leaving that implementation-defined, I don't see lots of > > > consensus on this point. > > > > People actually use it in their documents originating from "that other" > > application, we do have bug reports because we don't suppport it at the > > moment.. at least in the [HV]LOOKUP cases it really makes sense, any > > other functions? > > Oh, sorry, I was unclear. I think we MUST support [HV]LOOKUP. > The only question is, should we require anything if the lookup fails? > Ideally, we'd return an Error, but I believe that many don't do so; > the obvious solution is to leave allow an implementation-defined result > if the lookup fails (for this version of the spec). Ah, no, this is a misunderstanding. If the lookup fails it is always an error, an Empty can only be returned for a successful lookup. The nature of the [HV]LOOKUP functions is to lookup a value in a given area/vector and from the same row/column respectively an associated vector return a corresponding value, which may be an empty cell. Eike -- Automatic string conversions considered dangerous. They are the GOTO statements of spreadsheets. --Robert Weir on the OpenDocument formula subcommittee's list.
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]