Re: [oiic-formation-discuss] Open formula group, and getting ahead ofourselves?

"Dave Pawson" <>
wrote on 06/22/2008 04:31:32 AM:

> Just had a quick look at the latest open formula group output.

> 

> For many of the functions I see such as:

> Syntax: MIDB( Text T ; BytePosition Start ; ByteLength Length )

> 

> The pseudotypes ByteLength and BytePosition are Integers, but their

> exact meanings and values are not further defined by this

> specification.

> 

> I.e. for (some part) of the spec a function is defined and each term

> of the function is refined. Not exactly a formal grammar, but nodding

> in that direction.

> 

> If there's anyone on this group also on the formula group, how

> might we suggest they consider testability for their syntax statements?

> 

You can send comments on the OpenFormula draft to
the ODF public comment list.  I know that David Wheeler, chair of
that subcommittee, monitors that list.

An overall to-do item for ODF 1.2 is to express on
syntax formally in EBNF form.  This should be testable, though it
would require some custom coding.  For example, a SAX parser that
catches all spreadsheet formula attributes and then passes the formulas
to a syntax checker, perhaps based on lex/yacc to ensure it matches the
EBNF productions.

-Rob