Issue 4547: Sections 1.8, 2.8, 3.1, 3.2, Editorial (cad-ftf) Source: NIST (Mr. David Flater, dflater@nist.gov) Nature: Uncategorized Issue Severity: Summary: Sections 1.8, 2.8, 3.1, 3.2, Editorial: Compliance points / levels and optional interfaces are spread around the document in a way that would hamper formal claims of conformance. Resolution: see above Revised Text: Changes to Section 2.1.2.1, CadServer Attributes Delete the sentence "The compliance point (compliance) identifies the level of compliance for this CadServer." Update the following descriptions: is_interactive - TRUE indicates support for the UserInterface compliance point defined in Section 1.7. is_persistent_id - TRUE indicates support for the PersistentIdentifiers compliance point defined in Section 1.7. [no change to is_accurate] is_parametric - TRUE indicates support for the Parametrics compliance point defined in Section 1.7.. Add this one: is_extended_geometry - TRUE indicates support for the GeometryExtensions compliance point defined in Section 1.7. Changes to Section 2.1.3.1, CadSystem Operations Replace the CadSystem::is_interactive operation with get_gui - Returns a reference to the CadUserInterface. An implementation that does not support the UserInterface compliance point shall throw GuiUnsupported when this operation is invoked. Changes to Section 2.2.2.2, Model Operations To the descriptions of the operations new_persistent_ids, persistent_ids_status, and persistent_ids_to_entities, append the following: "An implementation that does not support the PersistentIdentifiers compliance point shall throw PidUnsupported when this operation is invoked." Changes to Section 2.3.2.2, Entity Operations In the description of get_persistent_id, rephrase the last sentence as "An implementation that does not support the PersistentIdentifiers compliance point shall throw PidUnsupported when this operation is invoked." Changes to Section 2.3.6, Exceptions and Struct Add new exception: GuiUnsupported - Exception thrown when the client invokes an operation that requires the UserInterface compliance point on an implementation that does not support it. Rephrase existing exception: PidUnsupported - Exception thrown when the client invokes an operation that requires the PersistentIdentifiers compliance point on an implementation that does not support it. Changes to Section 2.6.2, DesignFeature Interface Append the following to the description of get_parameter_set: "If the implementation does not support the Parametrics compliance point (i.e., does not allow parametric regeneration of CAD entities), get_parameter_set shall return a zero-length sequence." Changes to Section 3.1, Summary of optional versus mandatory interfaces, and Section 3.2, Proposed compliance points These sections are now completely redundant and should be deleted. Changes to IDL / UML CadFoundation.idl: add exception GuiUnsupported. CadConnection.idl: replace boolean is_interactive (out CadUserInterface gui) with CadUserInterface get_gui () raises (GuiUnsupported); CadConnection.idl: add is_extended_geometry to OptionsStruct. CadConnection.idl: delete compliance from NativeCadAttr. CadMain.idl: add CadFoundation::PidUnsupported to the raises lists of new_persistent_ids, persistent_ids_status, and persistent_ids_to_entities. Actions taken: August 20, 2001: received issue July 5, 2002: closed issue Discussion: It was agreed to reorganize the Compliance points as follows: Identifier Description Base All interfaces in all modules except CadGeometryExtens. GeometryExtensions Base, plus all interfaces in CadGeometryExtens. PersistentIdentifiers Base, plus support for persistent identifiers as described in Section 2.2.2.2, Model Operations, and Section 2.3.2.2, Entity Operations. UserInterface Base, plus support for a user interface as described in Section 2.1.3.1, CadSystem Operations. Parametrics Base, plus support for parametric regeneration of CAD entities as described in Section 2.6.2, DesignFeature Interface. If the UserInterface compliance point is not supported, the CadUserInterface interface becomes inaccessible (PdmSystem::get_gui throws GuiUnsupported) and its implementation becomes moot. Similarly, if the Parametrics compliance point is not supported, the Parameter interface becomes inaccessible (DesignFeature::get_parameter_set returns an empty sequence). Otherwise, all interfaces called out by Base are mandatory. Support for persistent identifiers, a user interface, or parametric regeneration of CAD entities are indicated by declaring support for the respective compliance points. Additionally, implementations may vary in their support for accuracy, presentation properties, and session-level identification of CAD entities. However, these variations are exposed through specified behaviors of the mandatory interfaces. End of Annotations:===== From: David Flater MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Message-ID: <15233.23657.990795.407660@mouse.msid.cme.nist.gov> Date: Mon, 20 Aug 2001 14:52:25 -0400 (EDT) To: cad-ftf@omg.org, issues@omg.org Cc: jcad_services@yahoogroups.com Subject: CAD FTF issues Re: Telecon Thursday In-Reply-To: <5.1.0.14.2.20010820125611.01f5b260@popserve.grc.nasa.gov> References: <5.1.0.14.2.20010820125611.01f5b260@popserve.grc.nasa.gov> X-Mailer: VM 6.72 under 21.1 (patch 10) "Capitol Reef" XEmacs Lucid X-Public-Key: Available at http://www.mel.nist.gov/msidstaff/flater/flater.htm X-Keyserver: wwwkeys.pgp.net Content-Type: text/plain; charset=us-ascii X-UIDL: 6@Zd9[cj!!nSi!!)!od9 Russell W. Claus writes: > Enclosed please find an unofficial list of the CAD Services issues. The > official list is at: > > http://cgi.omg.org/issues/cad-ftf.html A couple of my eval team issues were deferred to the FTF but don't appear in this list. If they have been somehow addressed since Boston, I cannot tell because there is no document on the OMG server more recent than mfg/2001-07-10. [*] Sections 1.8, 2.8, 3.1, 3.2, Editorial: Compliance points / levels and optional interfaces are spread around the document in a way that would hamper formal claims of conformance. I could potentially submit revised text for both of these issues. Regarding the first, an example of how I would approach it should be appearing soon in the latest PDM draft at mfg/01-08-02 or mfg/01-08-03, or can be found now in the mfg-ppe mailing list archive. The second one would ideally just require sitting down and documenting the process, which I should be able to figure out. (Less ideally, it could uncover problems that need to be addressed.) [*] Although I am getting email through jcad_services@yahoogroups.com, the group web page tells me I am not a member and therefore not able to check if there are more recent files stored there. I have requested (another) subscription, pending moderator approval. Regards, -- David Flater, National Institute of Standards and Technology, U.S.A.