[osis-core] Grain regex

Todd Tillinghast osis-core@bibletechnologieswg.org
Wed, 14 Aug 2002 13:24:48 -0600


Is that [ and ] for str and ch?

Todd

> -----Original Message-----
> From: owner-osis-core@bibletechnologieswg.org [mailto:owner-osis-
> core@bibletechnologieswg.org] On Behalf Of Patrick Durusau
> Sent: Wednesday, August 14, 2002 12:41 PM
> To: osis-core@bibletechnologieswg.org
> Subject: Re: [osis-core] Grain regex
> 
> Todd,
> 
> For consistency I would prefer [ and ].
> 
> BTW, my laptop has died (on/off switch problem I suspect) and has been
> shipped to the local source for repairs. (died about 11 AM today) As
you
> probably suspect, was using it for the pretty documentation with
XMLSpy.
> Switching to validation with Xerces but will not have pretty
> documentation abilities until  the laptop returns. May need to call on
> you for generating the docs.
> 
> Any other outstanding issues that will require changes that need to be
> validated? (I don't know that Steve's earlier reply resolved any other
> issues with the current candidate for release. If not, post separate
> subject line messages).
> 
> Patrick
> 
> Todd Tillinghast wrote:
> 
> >So would both ch and str be followed by [ and ] or just str?
> >
> >Todd
> >
> >>-----Original Message-----
> >>From: owner-osis-core@bibletechnologieswg.org [mailto:owner-osis-
> >>core@bibletechnologieswg.org] On Behalf Of Steve DeRose
> >>Sent: Wednesday, August 14, 2002 6:03 AM
> >>To: osis-core@bibletechnologieswg.org
> >>Subject: RE: [osis-core] Grain regex
> >>
> >>At 02:56 PM -0600 08/13/02, Todd Tillinghast wrote:
> >>
> >>>
> >>>Isn't the string everything after the : until the end of the
> >>>
> >reference
> >
> >>>OR until the - in the case of the first value in a range reference.
> >>>
> >>Could -- but that makes the syntax non-extensible;  and undelimited
> >>strings make me nervous...
> >>
> >>>How would we know when the number ends for the ch:number case?
Same
> >>>rules.  Neither case requires [ ] to know when the start and end
are.
> >>>
> >>>Todd
> >>>
> >>True; at the moment. But it seems cleaner to me to throw in the ()
or
> >>[], because
> >>
> >>a) it makes the three parts (work/id/grain) parallel
> >>
> >>b) it allows for later adding grains with >1 parameter, or adding
> >>length or other parameters to existing grains
> >>
> >>c) it makes it clearer what the scope is in range references, so the
> >>'-' is less prone to look like it's part of the grain syntax
> >>
> >>d) it makes the scope of the grain parameter visually obvious
> >>
> >>
> >>--
> >>
> >>Steve DeRose -- http://www.stg.brown.edu/~sjd
> >>Chair, Bible Technologies Group -- http://www.bibletechnologies.net
> >>Email: sderose@speakeasy.net
> >>Backup email: sjd@stg.brown.edu
> >>
> 
> --
> Patrick Durusau
> Director of Research and Development
> Society of Biblical Literature
> pdurusau@emory.edu
> 
>