IPP Mail Archive: FW: IPP> OPS - 3 Options for Filters on Ge

IPP Mail Archive: FW: IPP> OPS - 3 Options for Filters on Ge

FW: IPP> OPS - 3 Options for Filters on Get-Job-Attributes

From: McDonald, Ira (imcdonald@sharplabs.com)
Date: Fri Aug 25 2000 - 16:04:44 EDT

  • Next message: Manros, Carl-Uno B: "IPP> ADM - IPP Phone Conference - 000830"

    -----Original Message-----
    From: McDonald, Ira
    Sent: Friday, August 25, 2000 1:01 PM
    To: 'Michael Sweet'; McDonald, Ira
    Subject: RE: IPP> OPS - 3 Options for Filters on Get-Job-Attributes

    Hi Michael,

    Thanks for your quick reply. I agree that #3 is the best choice.

    The encoding safety of Tom Hastings' (Option #1) pretty simple
    (ANDs and ORs) method using a new type of Group in the operation
    SHOULD be fine. But Bob Herriot and I are somewhat concerned that
    existing implementations MAY not be resilient to finding a new
    (and unexpected) Group in a received Get-xxx operation request.

    And there's a nice simplicity in taking LDAPv3 filter syntax
    and using it verbatim.

    A modest effort would be required to specify the canonical string
    encoding of all IPP attribute syntaxes (not difficult - mostly
    done already where they're used in the SLP/LDAP Printer Schema).

    Cheers,
    - Ira McDonald

    -----Original Message-----
    From: Michael Sweet [mailto:mike@easysw.com]
    Sent: Friday, August 25, 2000 11:17 AM
    To: McDonald, Ira
    Subject: Re: IPP> OPS - 3 Options for Filters on Get-Job-Attributes

    "McDonald, Ira" wrote:
    > ...
    > Option 3 - new 'xxx-filter' operation attribute
    > - use LDAP and SLP common string expression syntax (RFC 2254)
    > - perfect compatibility with directory-enabled applications
    > - full set of operators (approximate equivalence, presence,
    > substring, etc.)
    > - requires string encoding of all values (not IPP native)

    Of the three choices, #3 would seem to be the best one to use in the
    long run. I see some complications with the implementation of #1,
    as it requires more complicated request generation and processing
    code and is not as flexible as #3.

    #2 uses collections, and my opinions on collections should be well
    known...

    -- 
    ______________________________________________________________________
    Michael Sweet, Easy Software Products                  mike@easysw.com
    Printing Software for UNIX                       http://www.easysw.com
    



    This archive was generated by hypermail 2b29 : Fri Aug 25 2000 - 16:13:52 EDT