Ray Denenberg, Library of Congress wrote:
> From: "Adam Dickmeiss" <[log in to unmask]>
>> I don't think we'll be implementing it.. Regular expressions suit us
>> fine at this point. AFAIK.
> 
> The reason for the proposal in the first place was to support subfield
> queries like:
> marc.533$7=/substring="1:4" 1992
> to search (in this example) on year of publication.
> (All other hypothetical uses of substring are speculative at this point.)
> 
> How would you accomplish this query with a regular expression?
Usually when dealing with character positions.. that is dealt with 
during indexing. We then search for 'date of publication'..

In real life also MARC records are "polluted" .. There are "many" kinds 
of dates and in many cases people did enter them.. Hence a _real_ 
programming language / preprocessing is needed anyway.. During indexing :-)

/ Adam

> 
> --Ray
> 
>