IPP Mail Archive: RE: IPP> last call comment on Override doc

IPP Mail Archive: RE: IPP> last call comment on Override doc

RE: IPP> last call comment on Override document

From: Herriot, Robert (Robert.Herriot@pahv.xerox.com)
Date: Wed Nov 29 2000 - 22:21:49 EST

  • Next message: Manros, Carl-Uno B: "IPP> RES - Summary of original Resource object with sub-typing for com parison - A further alternative"

    You asked me to provide examples of why I wanted to change four attributes
    in the document-overrides and page-overrides from rangeOfInteger to 1setOf
    rangeOfInteger. The four attributes are input-documents, output-documents,
    document-copies and pages.

    The mapping from JDF to IPP without the proposed changes is still easy. It
    just may be more verbose than if the proposed changes were in effect.

    For example, suppose a job needs a special media on 20 pages where none are
    consecutive.

    In JDF the override would consist of the specified media and the value of
    the pages attribute would be the 20 integers, expressed as 20 ranges.

    In IPP (without the change), the override would consist of 20 separate
    collection values. Each collection value would contain the specified media
    and the value of the pages attribute would be one of the 20 integers,
    expressed as a range. In IPP with the proposed change, the override would be
    like the one in JDF.

    One could argue that IPP doesn't need this change because this case is
    extremely rare. But there may be another reason -- consistency. In IPP we
    have been inconsistent about how we represent the abstract notion of a set
    of integers. We represent this concept in any of the following ways:
        a) 1setOf integer
        b) rangeOfInteger
        c) 1setOf rangeOfInteger
        d) 1setOf(integer | rangeOfInteger)

    Each of these solutions is really a representation of a set of integers
    optimized for or limited to the expected values. Solution d) is the most
    compact. Solution c) is the next most compact and is simpler than d). So
    that is why c) is a good proposal for all sets of integers.

    Bob Herriot

    > -----Original Message-----
    > From: don@lexmark.com [mailto:don@lexmark.com]
    > Sent: Thursday, November 16, 2000 6:38 AM
    > To: Robert.Herriot@pahv.xerox.com
    > Subject: Re: IPP> last call comment on Override document
    >
    >
    > Bob:
    >
    > Some examples of these might be help for everyone to
    > understand the limitations
    > and the difficulties in mapping that would result if these
    > changes were not
    > made.
    >
    > Thanks!
    >
    > **********************************************
    > * Don Wright don@lexmark.com *
    > * Chair, Printer Working Group *
    > * Chair, IEEE MSC *
    > * *
    > * Director, Strategic & Technical Alliances *
    > * Lexmark International *
    > * 740 New Circle Rd *
    > * Lexington, Ky 40550 *
    > * 859-232-4808 (phone) 859-232-6740 (fax) *
    > **********************************************
    >
    >
    >
    > Robert.Herriot%pahv.xerox.com@interlock.lexmark.com on
    > 11/15/2000 09:09:40 PM
    >
    > To: ipp%pwg.org@interlock.lexmark.com
    > cc: (bcc: Don Wright/Lex/Lexmark)
    > Subject: IPP> last call comment on Override document
    >
    >
    >
    > After doing a JDF to IPP mapping, I realized that four
    > attributes in the
    > override spec are too specific. For the four attributes
    > below, the spec says
    > that the syntax is rangeOfInteger but in JDF the same
    > attributes have a
    > syntax which is the equivalent of 1setOf rangeOfInteger. So
    > I suggest that
    > we change these four attributes to have the syntax "1setOf
    > rangeOfInteger(1:MAX))". Without this change the mapping
    > between JDF and IPP
    > is more difficult.
    >
    > input-documents (rangeOfInteger (1:MAX))
    > output-documents (rangeOfInteger (1:MAX))
    > document-copies (rangeOfInteger (1:MAX))
    > pages (1setOf rangeOfInteger(1:MAX))
    >
    > Bob Herriot
    >
    >
    >
    >



    This archive was generated by hypermail 2b29 : Wed Nov 29 2000 - 22:24:21 EST