Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!cs.utexas.edu!wuarchive!decwrl!ucbvax!ucdavis!csusac!unify!nico From: nico@unify.uucp (Nico Nierenberg) Newsgroups: comp.databases Subject: Re: variable record length Message-ID: Date: 2 Feb 90 17:20:39 GMT References: <410@ssc.UUCP> <1990Jan25.000347.7138@oracle.com> <738@dgis.dtic.dla.mil> <742@dgis.dtic.dla.mil> <7388@tank.uchicago.edu> Reply-To: nico@unify.UUCP (Nico Nierenberg) Distribution: usa Organization: Unify Corporation, Sacramento, CA, USA Lines: 36 In article <7388@tank.uchicago.edu> monty@delphi.UUCP (Monty Mullig) writes: >let's define some terms. > >to my mind, a variable length field is one in which no fixed maximum >or minimum is specified by the user. its declaration might be >something like: > > fieldname varlen; I don't really think that is important whether or not a maximum length must be declared. After all the user could simply declare this length to be the maximum acceptable field length for this data type. This would seem to be purely a semantic issue. If Oracle has a maximum size it could take the above syntax and convert it to a declaration of a field of the maximum length. > >now, the system might place some restrictions on the size of the >field, such as it must be between 0 and 65,534 bytes long, but the >user doesn't specify a limit. perhaps a better name for this kind of >field would be an "arbitrary length field". is this kind of field >available with oracle ? > Strangely enough I think that this is a bigger issue since it is the physical restriction on the capability of the data type. If the system restricts it to 65,534 bytes and the user needs a 2MB image then there is a problem. By the way Unify doesn't require the size to be specified in advance just in case anyone thinks I'm biased by our design decisions. -- --------------------------------------------------------------------- Nicolas Nierenberg "No matter where you go, Unify Corp. there you are." nico@unify