Method for streaming object models that have a plurality of versioned states
    1.
    发明授权
    Method for streaming object models that have a plurality of versioned states 有权
    具有多个版本化状态的流对象模型的方法

    公开(公告)号:US06625663B1

    公开(公告)日:2003-09-23

    申请号:US09533343

    申请日:2000-03-23

    IPC分类号: G06F15163

    CPC分类号: G06F9/4493

    摘要: A computer-implemented method for streaming object models with a plurality of versioned states. The method comprises the steps of streaming out each versioned state for each nonderived classifier-level feature; streaming out each versioned state of each unowned object within the model; streaming out versioned state of each owned object not yet streamed out; streaming out the versioned state of each link not represented by a reference; and streaming out a version graph using normal XMI.

    摘要翻译: 一种用于流传输具有多个版本化状态的对象模型的计算机实现的方法。 该方法包括以下步骤:为每个非分类级别特征流出每个版本化状态; 流出模型中每个未知对象的每个版本化状态; 流出每个拥有对象的版本化状态尚未流式传输; 流出未由引用表示的每个链接的版本化状态; 并使用正常的XMI流出版本图。

    Method in a computing system for comparing XMI-based XML documents for identical contents
    2.
    发明授权
    Method in a computing system for comparing XMI-based XML documents for identical contents 有权
    用于比较基于XMI的XML文档用于相同内容的计算系统中的方法

    公开(公告)号:US06502112B1

    公开(公告)日:2002-12-31

    申请号:US09384690

    申请日:1999-08-27

    IPC分类号: G06F1727

    摘要: A method is provided by a computer system processing XMI-based XML documents. The method compares two such XMI-based XML documents for identical content. The method begins with the step of parsing each of the documents to create for each a semantic graph of the document's objects. Next, a list of names of properties for each of the objects having significant order is read. For each of the objects, and then for each object's property not listed as having significant order, values of said properties are sorted. Finally, the objects of the semantic graphs are compared.

    摘要翻译: 一种处理基于XMI的XML文档的计算机系统提供了一种方法。 该方法比较两个这样的基于XMI的XML文档,用于相同的内容。 该方法开始于解析每个文档以为每个文档对象的语义图形创建的步骤。 接下来,读取具有重要顺序的每个对象的属性的名称的列表。 对于每个对象,然后对于未列为具有重要顺序的每个对象的属性,将对所述属性的值进行排序。 最后,比较语义图的对象。

    Method for merging versions of a model in an object oriented repository
    3.
    发明授权
    Method for merging versions of a model in an object oriented repository 有权
    在面向对象的存储库中合并模型的版本的方法

    公开(公告)号:US06415299B1

    公开(公告)日:2002-07-02

    申请号:US09547642

    申请日:2000-04-12

    IPC分类号: G06F1730

    摘要: A computer-implemented method is provided for merging a source version into a target version of a model in an object oriented repository. The method comprises the steps of building a first list as a collection of versions that occur only in a history of the source version; and, building a second list as a collection of versions that occur only in a history of the target version. Next, a dual history is created as a union of the first and second lists. After this, either the first or the second list is identified as the non-preferred history wherein either the source or the target version is preferred. The first list is then added to the history of the target version. Any conflicts occurring in the dual history are next reported; and, such conflicts are resolved by adding the target version to a list of end versions on each conflicting attribute value and link occurring in the non-preferred history.

    摘要翻译: 提供了一种计算机实现的方法,用于将源版本合并到面向对象的存储库中的模型的目标版本。 该方法包括以下步骤:将第一列表构建为仅在源版本的历史中发生的版本的集合; 并将第二个列表构建为仅在目标版本的历史中发生的版本集合。 接下来,创建双历史作为第一和第二列表的联合。 之后,将第一或第二列表识别为非首选历史,其中源或目标版本优选。 然后将第一个列表添加到目标版本的历史记录中。 下一次报道双历史上发生的任何冲突; 并且通过将目标版本添加到在非优选历史记录中发生的每个冲突属性值和链接的最终版本列表来解决这些冲突。

    Method for merging repository updates with incomplete work in progress
    4.
    发明授权
    Method for merging repository updates with incomplete work in progress 有权
    将存储库更新与不完整工作进行合并的方法

    公开(公告)号:US06275787B1

    公开(公告)日:2001-08-14

    申请号:US09172569

    申请日:1998-10-14

    IPC分类号: G06F1300

    摘要: In a computing system executing a repository program, a computer-implemented method is disclosed for merging updates to models stored in the repository with a user's incomplete work-in-progress currently being executed by the computer system. Once a repository transaction is begun, for each package within a model that is checked out by the user, a check-in operation is performed. The model is extracted for use by a modeling tool and the repository transaction is aborted, thereby undoing the check-in but leaving the extracted model intact for use by the modeling tool. A report may be generated of any inconsistencies between the model and modeling rules imposed by the check-in operation of the repository.

    摘要翻译: 在执行存储库程序的计算系统中,公开了一种计算机实现的方法,用于将存储在存储库中的模型的更新合并到当前由计算机系统执行的用户的不完整工作。 一旦存储库事务开始,对于由用户检出的模型中的每个包,执行签入操作。 该模型被提取以供建模工具使用,并且存储库事务被中止,从而撤销该检入,但是将提取的模型完整地保留以供建模工具使用。 可能会生成模型与存储库的登录操作强加的建模规则之间的任何不一致的报告。

    Method for controlling access to a multiplicity of objects using a customizable object-oriented access control hook
    5.
    发明授权
    Method for controlling access to a multiplicity of objects using a customizable object-oriented access control hook 有权
    使用可定制的面向对象的访问控制钩来控制对多个对象的访问的方法

    公开(公告)号:US06766457B1

    公开(公告)日:2004-07-20

    申请号:US09456768

    申请日:1999-12-07

    IPC分类号: G06F1214

    CPC分类号: G06F21/6218 Y10S707/99939

    摘要: A computer-implemented object-oriented method for controlling access to a multiplicity of objects is disclosed. The method includes creating specific access control object types, each including a pre-check method for implementing a pre-defined access control policy. Each one of the multiplicity of objects to be controlled is then associated with one of the access control objects. Next, upon an attempt to invoke a feature of any one of the multiplicity of objects, a determination is made if one of the multiplicity of objects is linked to an access control object, and if yes; the pre-check method for the access control object associated with the one of the multiplicity of objects is performed to determine whether to grant access.

    摘要翻译: 公开了一种用于控制对多个对象的访问的计算机实现的面向对象的方法。 该方法包括创建特定的访问控制对象类型,每个类型包括用于实现预定义的访问控制策略的预检查方法。 然后将要控制的多个对象中的每一个与访问控制对象之一相关联。 接下来,在尝试调用多个对象中的任何一个的特征时,确定多个对象中的一个是否被链接到访问控制对象,如果是; 执行与多个对象中的一个对象相关联的访问控制对象的预检查方法,以确定是否允许访问。

    Method for converting a UML rendering of an RSM-based metamodel to a UML rendering of MOF-based metamodel
    6.
    发明授权
    Method for converting a UML rendering of an RSM-based metamodel to a UML rendering of MOF-based metamodel 失效
    将基于RSM的元模型的UML呈现转换为基于MOF的元模型的UML呈现的方法

    公开(公告)号:US06684386B1

    公开(公告)日:2004-01-27

    申请号:US09592747

    申请日:2000-06-13

    IPC分类号: G06F944

    CPC分类号: G06F9/4492

    摘要: A computer-implemented method for converting a UML rendering of an RSM-based metamodel to a UML rendering of a MOF-based metamodel. The method includes the steps of removing inheritance from classes defined within a Repository Services Model (“RSM”) and removing each explicit “construct” operation from each class in the RSM-based metamodel Next, each use in the RSM-based metamodel of an RSM type is changed to use a non-RSM type. After this, each element of the RSM-based metamodel is converted to a corresponding MOF-based element and a determination is made as to whether or not the RSM naming service is used in the RSM-based metamodel, and if so “name” attributes are added that would have been inherited from the RSM classes. A > stereotype is added to the UML rendering of the MOF-based metamodel; and the MOF properties are set on the UML rendering.

    摘要翻译: 一种用于将基于RSM的元模型的UML呈现转换为基于MOF的元模型的UML呈现的计算机实现的方法。 该方法包括从存储库服务模型(“RSM”)中定义的类中删除继承的步骤,并从基于RSM的元模型中的每个类中删除每个显式“构造”操作。接下来,每个使用基于RSM的元模型 RSM类型更改为使用非RSM类型。 之后,将基于RSM的元模型的每个元素转换为相应的基于MOF的元素,并确定在基于RSM的元模型中是否使用RSM命名服务,如果是“名称”属性 被添加,这将被继承自RSM类。 将基于MOF的元模型的UML渲染中添加一个“元模型”构造型; 并且在UML渲染上设置MOF属性。

    Method for versioning a UML model in a repository in accordance with an updated XML representation of the UML model
    7.
    发明授权
    Method for versioning a UML model in a repository in accordance with an updated XML representation of the UML model 有权
    根据UML模型的更新的XML表示形式在存储库中对UML模型进行版本化的方法

    公开(公告)号:US06330569B1

    公开(公告)日:2001-12-11

    申请号:US09345291

    申请日:1999-06-30

    IPC分类号: G06F1730

    摘要: In a computer system executing a repository program and having a memory, a method is disclosed for versioning a UML model in the repository in accordance with an updated XML representation of the UML model. The method includes the steps of identifying differences between UML objects in the UML model and XML objects in the XML file. Where there is a difference between one of the UML objects and one of the XML objects, it is reserved as a ghost object. The ghost object is updated to match each corresponding one of the XML objects. Objects whose target references have been changed are reserved and updated. Finally, the UML objects are updated in the repository that refer to ghost objects updated in the preceding step. The method includes making changes to the ghost object Attributes, References and owned elements to match Attributes, References and owned elements, respectively, found in the XML object.

    摘要翻译: 在执行存储库程序并具有存储器的计算机系统中,公开了根据UML模型的更新的XML表示来在存储库中版本化UML模型的方法。 该方法包括识别UML模型中的UML对象与XML文件中的XML对象之间的差异的步骤。 在其中一个UML对象与其中一个XML对象之间存在差异的地方,它被保留为一个ghost对象。 更新ghost对象以匹配每个对应的一个XML对象。 其目标引用被更改的对象被保留和更新。 最后,UML对象在存储库中被更新,它引用了上一步中更新的ghost对象。 该方法包括对ghost对象Attributes,References和所有的元素进行更改,以分别匹配XML对象中的Attributes,References和拥有的元素。

    Object-oriented apparatus and method in a computer system for a
programming language compiler to select optimal pointer kinds for code
generation
    8.
    发明授权
    Object-oriented apparatus and method in a computer system for a programming language compiler to select optimal pointer kinds for code generation 失效
    用于编程语言编译器的计算机系统中的面向对象的装置和方法,用于为代码生成选择最佳指针种类

    公开(公告)号:US6129460A

    公开(公告)日:2000-10-10

    申请号:US992825

    申请日:1997-12-17

    IPC分类号: G06F9/45

    CPC分类号: G06F8/437 G06F8/443

    摘要: The invention disclosed is an object-oriented apparatus and method in a computer system for a programming language compiler to select optimal pointer kinds for code generation. The invention offers important advantages over current compiler practice. In Particular, the invention provides improved performance for compiled programs through generation of more optimal selection of pointer kinds and variable allocations. Also, the invention provides a general framework for optimizing pointer usage, as opposed to leaving such optimization to special cases.

    摘要翻译: 所公开的发明是用于编程语言编译器的计算机系统中的面向对象的装置和方法,用于为代码生成选择最佳指针种类。 本发明提供了比现有编译器实践更重要的优点。 特别地,本发明通过产生更优选的指针种类和可变分配来提供编译程序的改进性能。 此外,本发明提供了用于优化指针使用的一般框架,而不是将这种优化留给特殊情况。

    Method for tracking changes in source locations in a compiler
    9.
    发明授权
    Method for tracking changes in source locations in a compiler 失效
    跟踪编译器中源位置变化的方法

    公开(公告)号:US5978586A

    公开(公告)日:1999-11-02

    申请号:US978706

    申请日:1997-11-26

    IPC分类号: G06F9/45

    CPC分类号: G06F8/42 G06F8/443

    摘要: The disclosed method creates a source object for each source language element parsed by the compiler and associates each such created source object with objects created from source language elements to produce source-mapped objects. Next, the source-mapped objects are transformed into corresponding machine-instruction objects and a binary file of the machine-instruction objects is created by: i) adjusting a source data object to a source object associated with a machine instruction object, thereby identifying a lowest level of inlining where the call history of the source object differs from the call history of the previous source object to which the source data object was adjusted; ii) if the source data object identifies no differences from the previous source object, then continuing at step i) for the next machine instruction; and, iii) for each level of the call history beginning with the level returned in the preceding step and continuing through the last level, putting information obtained from the source data object in a source table stored in the binary file.

    摘要翻译: 所公开的方法为编译器解析的每个源语言元素创建源对象,并将每个这样创建的源对象与从源语言元素创建的对象相关联以产生源映射对象。 接下来,源映射对象被转换成相应的机器指令对象,并且机器指令对象的二进制文件通过以下方式创建:i)将源数据对象调整为与机器指令对象相关联的源对象,从而识别 源对象的呼叫历史与调整源数据对象的先前源对象的呼叫历史不同的内联最低级别; ii)如果源数据对象识别出与先前源对象没有差异,则在步骤i)继续下一个机器指令; 以及iii)针对从前一步骤中返回的级别开始并继续通过最后一级的呼叫历史的每个级别,将从源数据对象获得的信息放入存储在二进制文件中的源表中。