[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Brian T's queries

• Subject: Brian T's queries
• From: Nick Spadaccini <nick@xxxxxxxxxxxxx>
• Date: Tue, 26 Sep 2000 03:26:59 +0100 (BST)

Brian Toby mailed these questions, but I think their answers are relevant
to all on this mail list so I have responded here.

> 1) Is the following valid? 
> 
> 	_data_name1
> ; this is a multiline data value
> this is the second line
> ;	_data_name2 'data item for name2'
> 
> I had thought that the terminating semicolon had to be the only
> character on a line, but rereading the 1991 Acta paper leads me to
> believe the above is valid.

Yes it is valid, certainly in STAR and there has never been (to my
recollection) a requirement in CIF that the delimiting token be
<terminate><semi_colon><terminate>. Of course one could argue that it is
good practice to start _data_name2 on a new line, but it is not a
requirement of the format.

> 2) I want to confirm that the following is not valid:
> 
> 	_data_item _data_value
> 
> My reading is that if a data value begins with _ it must be quoted. So
> it would have to be written:
> 
> 	_data_item '_data_value'

That is correct. The productions for <NON_QUOTED_1_STRING> and
<NON_QUOTED_2_STRING> explicitly exclude leading _ (underscores), hence
such data values must be quoted in some way.

> 3) What if a data item begins with loop_ or data_?
> 
> 	_data_item data_value
> 
> I did not see a rule requiring them to be quoted.

These are excluded also, but by reference to the notes (in blue
background) accompanying the definitions of <NON_QUOTED_1_STRING> and
<NON_QUOTED_2_STRING>. Why, are they not part of the productions? Because
they would become unnecessarily complicated and long. I have adopted the
practice used in many language BNFs and Syntax diagrams, for instance
formally specify what an identifier is, and then add a note saying that
the identifier string cannot be the following reserved words, etc.

It simplifies things considerably.

Hope this helps.

cheers

Nick

--------------------------------
Dr Nick Spadaccini
Department of Computer Science              voice: +(61 8) 9380 3452
University of Western Australia               fax: +(61 8) 9380 1089
Nedlands, Perth,  WA  6907                 email: nick@cs.uwa.edu.au
AUSTRALIA                        web: http://www.cs.uwa.edu.au/~nick

Reply to: [list | sender only]

• Prev by Date: RE: The CIF BNF
• Next by Date: RE: The CIF BNF
• Prev by thread: Re: Powder CIF Proposals
• Next by thread: CIF BNF Part 2
• Index(es):
  • Date
  • Thread