Re: newbie struggling with advanced filter

[brownh hartford-hwp com (Haines Brown)]

> Haines Brown wrote:
> > I have a worksheet that has data A2:N1178, and I want to filter it to
> > create a new worksheet that has select columns from the original as
> > well as select rows based on a filter.
> >
> > My first question is whether I necessarily must use an advanced
> > filter. The autofilter seems to have a maximum capacity of 150
> > rows, which is far too small for my purpose. Is this so?
>
> What version of Gnumeric are you using?  I'm running 1.6.1 and there 
> isn't this limitation.

gnumeric-1.4.3

> > This just gives me: "The Given Criteria are Invalid". So my question
> > is, why?
>
> I get that error if the column heading in the criteria field(s) don't 
> match columns in the data fields.  So I suggest you've check that you've 
> typed 'exp.year' correctly and it matches one of the other columns.

I didn't find the manual very helpful and so spent some hours playing
with gnumeric Advanced Filter in order to infer its rules.

The "List Range" I find must start with column A and Row 1 (the
labels), but need not extend all the way to the right or to the bottem
of the available worksheet data. I find I can use either absolute or
relative cell references ($A$7 or A7) in a simple filter operation.

The "Criteria Range" refers to two free rows of cells. The first
row must correspond to the columns in the List Range and it must
duplicate the worksheet's column labels exactly. The second row of 
cells holds a filter criterion for one or more of the columns
specified by List Range. I tried to use an alphabetic filter (for
example, <=Jon ), but it always returned "no matching records". I
found that I was able to impose multiple numeric filter criteria.
  
In short, my need to create a filtered spreadsheet seems to have been
resolved, and I thank you for that help, but unfortunately, when
entering data I had spelled out months rather than using their number
(sep rather than 9), and so if I can't filter alphabetically, I'll
have to convert the data in the worksheets. 

-- 
 
       Haines Brown
       KB1GRM