Heterogeneous model multi-view data consistency interaction system based on xmi standard

Abstract

This invention discloses a heterogeneous model multi-view data consistency interaction system based on the XMI standard, including a model element parsing module for parsing model element data in XMI format model files; a graphics information parsing module for parsing graphics data structures in XMI format model files; a view mapping construction module for identifying and extracting extended structure fragments representing view mapping relationships from XMI format model files; and a conversion and standardization module for generating unified model representation data supported by the target platform. An import and verification module is used to import the unified model representation data into the target platform's data model management component, graphics rendering module, and view management module, respectively, and execute a consistency verification mechanism to generate consistency verification records and reports, and update the import status identifier. This invention achieves complete exchange and cross-platform restoration of model data between M-Design and CSM platforms.

Description

Technical Field

[0001] This invention relates to the field of model data exchange and consistency processing, and in particular to a heterogeneous model multi-view data consistency interaction system based on the XMI standard. Background Technology

[0002] In software modeling and system design, different modeling platforms typically use their own metamodel definitions, data storage structures, and graphical representation specifications, resulting in various heterogeneous model data formats. To achieve cross-platform model data exchange and sharing, the industry generally adopts XMI (XML Metadata Interchange) as a standardized data exchange format. This format encodes model elements, graphical information, and view mapping relationships into an XML structure, enabling the transfer of models between different platforms. However, due to differences in metaclass hierarchy, attribute definition methods, graphical rendering rules, and view organization patterns among modeling platforms, even under the XMI format, direct import and export cannot guarantee complete consistency in semantics, structure, and graphical representation of model data.

[0003] In existing technologies, common methods mainly focus on single-dimensional data transformation, such as only implementing metaclass type mapping for model elements, or only performing geometric and layout transformations on graphical data. These methods often neglect the synchronous processing of field mapping relationships between views in multi-view modeling scenarios, leading to problems such as lost view fields and invalid mapping rules after cross-platform import. Furthermore, in the consistency verification stage, existing solutions mostly remain at the file level or simple field level comparison, lacking a comprehensive verification mechanism covering the three dimensions of model element attributes, graphical features, and view mapping relationships. This makes it difficult to promptly detect complex differences such as semantic ambiguity, attribute mapping errors, and graphical distortion. In addition, some platforms fail to maintain the index binding between model elements, graphical data, and view mapping relationships during the import process, causing the imported model view to fail to accurately reflect the structural semantics of the source platform.

[0004] Existing technologies for implementing multi-view model data interaction on heterogeneous modeling platforms based on the XMI standard generally suffer from the following shortcomings: a lack of a unified model representation data construction mechanism, making it difficult to simultaneously cover model elements, graphical data, and view mapping relationships; a lack of multi-dimensional consistency verification strategies, making it impossible to comprehensively evaluate the correctness of cross-platform import results; and a lack of refined parsing and semantic association processing of view mapping relationships, leading to data inconsistencies or missing rules between views. These problems all affect the accurate transfer and usability of model data in scenarios such as multi-domain collaborative modeling, model reuse, and inter-platform model migration. Summary of the Invention

[0005] One objective of this invention is to propose a heterogeneous model multi-view data consistency interaction system based on the XMI standard. This invention describes in detail a method for unified conversion and multi-dimensional consistency verification of model element structure, graphic data structure and structured view mapping relationship set between multi-source heterogeneous modeling platforms through modules such as model element parsing, graphic information parsing, view mapping construction, transformation and standardization, import and verification. It has the advantages of strong cross-platform model semantic consistency, high graphic reproduction accuracy and complete multi-view mapping rules.

[0006] According to an embodiment of the present invention, a heterogeneous model multi-view data consistency interaction system based on the XMI standard includes:

[0007] The model element parsing module is used to parse model element data in XMI format model files;

[0008] The graphics information parsing module is used to parse the graphics data structure in the XMI format model file;

[0009] The view mapping construction module is used to identify and extract extended structure fragments representing view mapping relationships from XMI format model files, parse the metadata structure in the extended structure fragments, extract the mapping unit and its contained source view name, target view name and field mapping rules, generate field mapping paths, and semantically associate the field mapping paths with the model element structure to construct a set of structured view mapping relationships.

[0010] The transformation and standardization module is used to generate unified model representation data supported by the target platform based on metaclass mapping rules, attribute alignment rules, and graph transformation rule sets.

[0011] The import and verification module is used to import the unified model representation data into the data model management component, graphics rendering module and view management module of the target platform, respectively, and perform consistency verification, including model element consistency verification, graphics feature consistency verification and view mapping consistency verification, generate consistency verification records and consistency verification reports, and update the import status flag of the unified model representation data.

[0012] Optionally, modules can be integrated using the following methods:

[0013] S1. Receive the XMI format model file exported from the source platform;

[0014] S2. Parse the model element data and generate a structural representation of the model elements;

[0015] S3. Parse the graphical data structure and generate the graphical data structure;

[0016] S4. Extract the mapping pairs between the source view and the target view, and construct a set of structured view mapping relationships based on the mapping rule type and field attribute correspondence.

[0017] S5. Based on the metaclass mapping rules, attribute alignment rules and graphics transformation rule set, perform type conversion, attribute mapping and graphics format standardization processing on the model element structure representation, stereotype label information and graphics structure to generate unified model expression data supported by the target platform.

[0018] S6. Import the unified model representation data into the data model management component, graphics rendering module, and view management module of the target platform respectively, and perform consistency verification in three dimensions: model elements, graphics data, and structured view mapping. Generate consistency verification records and consistency verification reports, and update the import status flag of the unified model representation data.

[0019] Optionally, both the source platform and the target platform can be heterogeneous modeling platforms that conform to the XMI standard, and they can serve as each other's source and target in different application scenarios to support bidirectional conversion and consistent interaction of model data.

[0020] Optionally, the XMI format model file includes model element data, graphical data structure, and view mapping relationship data.

[0021] Optionally, the bidirectional communication and interoperability refers to the complete bidirectional conversion and interaction of multi-view model data (including model element data, graphic data structure and view mapping relationship data) conforming to the XMI standard between the M-Design platform and the CSM platform.

[0022] During the forward import process, the XMI format model files exported from the M-Design platform are processed by the above modules and converted into unified model expression data that conforms to the semantics and rendering requirements of the CSM platform before being imported into CSM.

[0023] During the reverse import process, the XMI format model file exported from the CSM platform is processed by the above modules and converted into unified model expression data that conforms to the semantics and rendering requirements of the M-Design platform before being imported into M-Design.

[0024] During the bidirectional data exchange process, the consistency of multi-dimensional data in the logical view, physical view, and requirement view is maintained. Through structured mapping and consistency verification mechanisms, it is ensured that the cross-view element relationships and graphic layout styles are not lost or distorted during the import and export process.

[0025] Optionally, step S2 includes the following specific steps:

[0026] S21. Based on the model element data in the XMI format model file, identify and extract the unique identifier information of each model element to distinguish different element instances;

[0027] S22. Extract the metaclass type information of the model elements and construct a type mapping relationship consistent with the metamodel definition;

[0028] S23. Analyze the prefiguration label information attached to the model elements and identify the semantic role of the prefiguration label information in the specific modeling context;

[0029] S24. Extract the set of attributes contained in the model elements and construct the key-value mapping relationship between attribute names and attribute values;

[0030] S25. Identify the relationships between model elements, including relationship type, relationship direction, relationship endpoints and multiplicity information, and construct a structured element relationship graph;

[0031] S26. Based on the extracted unique identifier information, metaclass type information, stereotype label information, attribute set and association relationship, a preliminary model element structure representation is generated.

[0032] Optionally, the graphic data structure extracts the graphic identifier, graphic type, graphic size, position coordinates, color attribute and trajectory path corresponding to the model element to generate the graphic data structure.

[0033] Optionally, step S4 includes the following specific steps:

[0034] S41. Identify and extract extended structure fragments used to represent view mapping relationships from the XMI format model file, and parse the metadata structure in the extended structure fragments;

[0035] S42. Extract mapping units from the metadata structure. Each mapping unit contains source view name and target view name fields.

[0036] S43. Parse the mapping rules defined within each mapping unit and extract the mapping rule type, source field name, and target field name;

[0037] S44. Based on the extracted mapping units and mapping rules, construct a set of structured view mapping relationships and establish field mapping paths between source view elements and target view elements;

[0038] S45. Organize each view pair and its corresponding field mapping rules into key-value mapping items according to the mapping pair structure;

[0039] S46. Semantically associate the structured view mapping relationship set with the model element structure to identify the attribute or identifier synchronization rules of model elements across multiple views.

[0040] Optionally, step S5 includes the following specific steps:

[0041] S51. Based on the metaclass mapping rules, perform type conversion operations on the metaclass types in the model element structure representation to generate the metaclass layer structure expression supported by the target modeling platform.

[0042] The metaclass mapping rules refer to a set of rules used to guide the semantic alignment, structural transformation, and attribute mapping of metaclass types of source platform model elements to target platform metaclass types. This set of rules includes the following three core parts:

[0043] Metaclass mapping table: Defines a one-to-one correspondence between semantically identical or functionally equivalent metaclass types in the source platform and the target platform, used to implement direct mapping.

[0044] Equivalent substitution mechanism: When the metaclass type of the source platform does not have the exact same semantic definition or structural form in the target platform, the equivalent metaclass type with the closest semantic function, attribute set and association relationship in the target platform is selected by the preset substitution rules to replace it, so as to ensure the equivalence of structure and function, and to add mapping annotations when necessary to retain the original semantic information.

[0045] Semantic preservation strategy: When performing type conversion and replacement, semantic tag mapping, attribute value conversion, and relational constraint reuse are used to ensure that the converted model in the target platform can fully retain the business meaning and design intent of the source platform for the naming, scope, attribute semantics and association semantics of the metaclass.

[0046] S52. Based on the attribute alignment rules, convert the source attribute key-value pairs in the attribute set of the model element into the target attribute set, and complete the attribute name mapping, data type conversion and default value completion.

[0047] The attribute alignment rule refers to the process of performing item-by-item comparison and renaming of the attribute set defined in the source platform element during the model element structure conversion process, according to the preset attribute mapping table.

[0048] The attribute mapping table explicitly lists the one-to-one mapping relationship between the attribute names of the source platform and the corresponding attribute names of the target platform. When there are semantically similar but differently named attributes in the target platform, the attribute alias table is used for conversion. When there are extended attributes in the source platform that are not supported by the target platform, attribute filtering is performed, marked as "unmappingable attributes", and removed from the target structure. When a required attribute of the target platform is missing in the source platform, the default attribute template is called to assign an initial value to ensure the integrity and compliance of the attribute structure.

[0049] S53. Based on the semantic role of the stereotype tags, perform mapping processing on the stereotype tag information and output transformed tags that conform to the stereotype definition of the target modeling platform;

[0050] S54. Based on the target platform type, graphic semantic tags and style features, select suitable geometric mapping rules, color normalization rules and layout adjustment rules from the graphic conversion rule set, and perform rule matching and conversion operations in sequence to generate a graphic data structure that meets the semantic requirements of the target platform.

[0051] The graphics transformation rule set refers to a set of rules used to transform the graphics data structure in the source platform into the graphics semantics and rendering format supported by the target platform. It includes geometric mapping rules, color normalization rules and layout adjustment rules. Through rule matching and transformation operations, graphics data structures that meet the requirements of the target platform's rendering engine are generated.

[0052] S55. Based on the conversion results of S54, according to the target platform's graphics file format, encoding specifications and rendering engine requirements, complete the formatted encoding, data structure rearrangement and storage mapping of graphics geometric features, color features and layout features, and output the final graphics data structure that can be directly loaded and rendered.

[0053] S56. Maintain the index binding relationship between the model element structure, the graphical data structure and the view mapping relationship set to form an intermediate model representation dataset with consistent structure, clear semantics and graphical standardization, and output unified model expression data;

[0054] Optionally, the target platform rendering engine requirements include: the graphics data must use file formats supported by the target platform (XML, JSON, SVG) and conform to its encoding specifications; the graphics objects must define node, connection, and grouping structures according to the hierarchical organization rules of the target platform; coordinates and dimensions must be described using the coordinate system reference and units specified by the target platform; color values ​​must conform to the color model (RGB, RGBA) and color accuracy requirements used by the platform; layout information must meet the parsing rules of the platform's layout engine, including alignment, spacing, anchor point positions, etc.; and upper limits are set on file size, number of graphics nodes, and rendering complexity to ensure that the graphics data can be loaded without loss, parsed correctly, and displayed efficiently in the rendering engine.

[0055] Optionally, step S6 includes the following specific steps:

[0056] S61. The unified model expression data includes a standardized model element structure, a graphical data structure, and a set of structured view mapping relationships, and maintains the index binding relationship between the three.

[0057] S62. Import the model element structure into the data model management component of the target platform, create model element instances according to the metaclass layer structure specification defined by the target platform, and complete the metaclass structure registration and attribute set initialization.

[0058] S63. Import the graphic data structure into the graphics rendering module of the target platform, generate and deploy graphic objects according to graphic type, coordinates, size, color and trajectory path, build a graphic view supported by the target platform and maintain consistency with the graphic data structure.

[0059] S64. Import the structured view mapping relationship set into the view management module of the target platform, register the view pair names and field mapping rules, and establish an attribute mapping index table between views within the target platform.

[0060] S65. Perform a consistency check task, load the model data before transformation and the imported model data after transformation, construct comparison datasets in the dimensions of model elements, graphical data, and structured view mapping, and perform consistency checks respectively:

[0061] At the model element level, unique identifiers, metaclass types, stereotype tags and attribute sets are compared one by one, and anomalies such as missing attributes, type mismatch, name mapping errors and default value filling errors are recorded.

[0062] In the dimension of graphic data, the graphic type, size, coordinate position, color parameters and trajectory path are compared according to the graphic node number to record anomalies such as geometric deformation, coordinate offset, color distortion and trajectory loss.

[0063] In the structured view mapping dimension, verify the consistency of view names, field mapping rules and semantic correspondence, and record anomalies such as field omissions, rule errors and semantic ambiguities;

[0064] S66. Compile and classify the comparison results of each dimension to form a consistency verification record list, record the comparison status, difference type and field number, and determine whether the difference constitutes an inconsistency item based on the preset threshold. For differences exceeding the threshold, add the location, difference description and possible cause explanation.

[0065] S67. Summarize inconsistencies to generate a consistency verification report. The report includes field names, dimensions, difference types, example values, and remediation suggestions. Update the import status flag of the unified model representation data based on the verification results. Mark "verification passed" if there are no inconsistencies, and mark "verification failed" if there are inconsistencies. Attach a list of differences and suggested processing paths to complete the cross-platform import and graphical reproduction process of multi-view model data.

[0066] The beneficial effects of this invention are:

[0067] This invention introduces a model element parsing, graphic information parsing, and view mapping construction mechanism based on the XMI standard. This enables comprehensive parsing and semantic association of the model element structure, graphic data structure, and structured view mapping relationship set in the XMI format model file exported from the source platform. This effectively solves the problems of easy loss of model semantics and incomplete attribute mapping in the cross-platform transmission process in the prior art.

[0068] In the conversion and standardization stage, this invention applies metaclass mapping rules, attribute alignment rules, and graphic transformation rule sets to metaclass types, attribute sets, and graphic structures respectively, ensuring the structural consistency, attribute integrity, and graphic reproduction accuracy of the unified model representation data on different target platforms, and avoiding display errors and layout offsets caused by format incompatibility or differences in rendering mechanisms between different modeling tools.

[0069] In the import and verification phase, this invention imports unified model representation data into the data model management component, graphics rendering module, and view management module of the target platform, respectively. It performs multi-level verification in three dimensions: model element consistency, graphics feature consistency, and view mapping consistency, generates a consistency verification report, and updates the import status identifier. This achieves highly reliable transmission and traceable verification of cross-platform multi-view model data, and improves the accuracy and controllability of the model interaction process. Attached Figure Description

[0070] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0071] Figure 1 This is a flowchart of the method for a heterogeneous model multi-view data consistency interaction system based on the XMI standard proposed in this invention.

[0072] Figure 2 This is a system flowchart of the heterogeneous model multi-view data consistency interaction system based on the XMI standard proposed in this invention;

[0073] Figure 3 This is a flowchart of the consistency verification process for the heterogeneous model multi-view data consistency interaction system based on the XMI standard proposed in this invention.

[0074] Figure 4 This is a flowchart illustrating the bidirectional interoperability between importing the CSM model into M-Design and importing the M-Design model into CSM, as proposed in this invention. Detailed Implementation

[0075] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0076] refer to Figure 1-4 A heterogeneous model multi-view data consistency interaction system based on the XMI standard, including:

[0077] The model element parsing module is used to parse model element data in XMI format model files;

[0078] The graphics information parsing module is used to parse the graphics data structure in the XMI format model file;

[0079] The view mapping construction module is used to identify and extract extended structure fragments representing view mapping relationships from XMI format model files, parse the metadata structure in the extended structure fragments, extract the mapping unit and its contained source view name, target view name and field mapping rules, generate field mapping paths, and semantically associate the field mapping paths with the model element structure to construct a set of structured view mapping relationships.

[0080] The transformation and standardization module is used to generate unified model representation data supported by the target platform based on metaclass mapping rules, attribute alignment rules, and graph transformation rule sets.

[0081] The import and verification module is used to import the unified model representation data into the data model management component, graphics rendering module and view management module of the target platform, respectively, and perform consistency verification, including model element consistency verification, graphics feature consistency verification and view mapping consistency verification, generate consistency verification records and consistency verification reports, and update the import status flag of the unified model representation data.

[0082] This invention proposes a heterogeneous model multi-view data consistency interaction system based on the XMI standard. Through a model element parsing module, a graphics information parsing module, and a view mapping construction module, it parses the model element data, graphics data structure, and view mapping relationships in the XMI format model file, respectively, and generates a set of structured view mapping relationships semantically associated with the model element structure. Combined with a conversion and standardization module, based on metaclass mapping rules, attribute alignment rules, and graphics conversion rule sets, it converts the source platform model data into unified model representation data supported by the target platform. Finally, through an import and verification module, the unified model representation data is imported into various processing components of the target platform, and consistency verification is performed in three dimensions: model elements, graphics features, and view mappings, generating consistency verification records and reports, thereby achieving accurate conversion and consistent interaction of model data between different modeling platforms.

[0083] In this embodiment, the modules are interconnected using the following method:

[0084] S1. Receive the XMI format model file exported from the source platform;

[0085] S2. Parse the model element data and generate a structural representation of the model elements;

[0086] S3. Parse the graphical data structure and generate the graphical data structure;

[0087] S4. Extract the mapping pairs between the source view and the target view, and construct a set of structured view mapping relationships based on the mapping rule type and field attribute correspondence.

[0088] S5. Based on the metaclass mapping rules, attribute alignment rules and graphics transformation rule set, perform type conversion, attribute mapping and graphics format standardization processing on the model element structure representation, stereotype label information and graphics structure to generate unified model expression data supported by the target platform.

[0089] S6. Import the unified model representation data into the data model management component, graphics rendering module, and view management module of the target platform respectively, and perform consistency verification in three dimensions: model elements, graphics data, and structured view mapping. Generate consistency verification records and consistency verification reports, and update the import status flag of the unified model representation data.

[0090] This invention provides a heterogeneous model multi-view data consistency interaction method based on the XMI standard. It receives an XMI format model file exported from a source platform, sequentially parses the model element data and graphical data structure, extracts and constructs a set of structured view mapping relationships between the source and target views, and combines metaclass mapping rules, attribute alignment rules, and graphical transformation rule sets to perform type conversion, attribute mapping, and graphical format standardization on the model element structure, stereotype labels, and graphical structure. This generates unified model representation data that can be directly used by the target platform, which is then imported into the target platform's data model management component, graphical rendering module, and view management module. Consistency checks are performed across the three dimensions of model elements, graphical data, and view mappings, generating consistency verification results and updating the data import status, thus achieving accurate conversion and consistent interaction of cross-platform model data.

[0091] In this embodiment, both the source platform and the target platform can be heterogeneous modeling platforms that conform to the XMI standard, and they can serve as each other's source and target in different application scenarios to support bidirectional conversion and consistent interaction of model data.

[0092] Source platform: refers to the heterogeneous modeling platform used to initiate model data export operations and act as a data provider during multi-view model data interaction. The source platform of this invention can be a modeling platform conforming to the XMI standard, such as CSM or M-Design, and can interchange roles with the target platform depending on the data exchange direction in different application scenarios.

[0093] Target platform: refers to the heterogeneous modeling platform used in the multi-view model data interaction process to receive and import model data files output by the source platform, and to complete parsing, mapping, rendering, and consistency verification operations under the method of this invention. The target platform of this invention can be a modeling platform conforming to the XMI standard, such as CSM or M-Design, and can interchange roles with the source platform according to the data exchange direction in different application scenarios.

[0094] In this embodiment, the XMI format model file contains model element data, graphical data structure, and view mapping relationship data.

[0095] In this embodiment, step S2 includes the following specific steps:

[0096] S21. Based on the model element data in the XMI format model file, identify and extract the unique identifier information of each model element to distinguish different element instances;

[0097] S22. Extract the metaclass type information of the model elements and construct a type mapping relationship consistent with the metamodel definition;

[0098] S23. Analyze the prefiguration label information attached to the model elements and identify the semantic role of the prefiguration label information in the specific modeling context;

[0099] S24. Extract the set of attributes contained in the model elements and construct the key-value mapping relationship between attribute names and attribute values;

[0100] S25. Identify the relationships between model elements, including relationship type, relationship direction, relationship endpoints and multiplicity information, and construct a structured element relationship graph;

[0101] S26. Based on the extracted unique identifier information, metaclass type information, stereotype label information, attribute set and association relationship, a preliminary model element structure representation is generated.

[0102] This invention parses the model element data in the XMI format model file, extracting unique identifiers, metaclass types, stereotype labels, attribute sets, and relationships between elements in sequence. It constructs type mapping relationships and attribute key-value mapping relationships consistent with the metamodel definition, and generates a structured element relationship graph. Finally, it forms a preliminary model element structure representation containing unique identifiers, type information, semantic labels, attribute data, and relationship structures, laying the foundation for subsequent cross-platform model conversion and consistent interaction.

[0103] In this embodiment, the graphic data structure extracts the graphic identifier, graphic type, graphic size, position coordinates, color attribute and trajectory path corresponding to the model element to generate the graphic data structure.

[0104] In this embodiment, the bidirectional interoperability refers to the fact that both the source and target platforms are heterogeneous modeling platforms conforming to the XMI standard, and they can serve as each other's source and target in different application scenarios, achieving complete exchange of model element data, graphical data structures, and view mapping relationship data, while maintaining consistency of multi-view data during the import and export process. Taking the M-Design platform and the CSM platform as examples, this invention demonstrates bidirectional interoperability including two modes: a forward import process and a reverse import process.

[0105] Forward import process: The model element parsing module receives and parses the XMI format model file exported from the M-Design platform, generating a model element structure representation; the graphics information parsing module parses the graphics data structure; the view mapping construction module extracts view mapping relationships and constructs a set of structured view mapping relationships; the conversion and standardization module converts the above structure into unified model expression data that conforms to the semantics and rendering requirements of the CSM platform according to metaclass mapping rules, attribute alignment rules, and graphics conversion rule sets; finally, the import and verification module imports the unified model expression data into the data model management component, graphics rendering module, and view management module of the CSM platform, and performs consistency verification.

[0106] The reverse import process is as follows: The model element parsing module receives and parses the XMI format model file exported from the CSM platform, generating a model element structure representation; the graphics information parsing module parses the graphics data structure; the view mapping construction module extracts the view mapping relationships and constructs a set of structured view mapping relationships; the transformation and standardization module converts the above structure into unified model expression data that conforms to the semantics and rendering requirements of the M-Design platform, based on metaclass mapping rules, attribute alignment rules, and graphics transformation rule sets; finally, the import and verification module imports the unified model expression data into the data model management component, graphics rendering module, and view management module of the M-Design platform, and performs consistency verification.

[0107] Through the two import modes described above, this invention enables full, lossless, and semantically aligned bidirectional exchange of model data (including graphics) between M-Design and CSM, covering various view types such as logical view, physical view, and requirement view, and ensuring the integrity and consistency of cross-view element relationships during the import and export process.

[0108] In this embodiment, step S4 includes the following specific steps:

[0109] S41. Identify and extract extended structure fragments used to represent view mapping relationships from the XMI format model file, and parse the metadata structure in the extended structure fragments;

[0110] S42. Extract mapping units from the metadata structure. Each mapping unit contains source view name and target view name fields.

[0111] S43. Parse the mapping rules defined within each mapping unit and extract the mapping rule type, source field name, and target field name;

[0112] S44. Based on the extracted mapping units and mapping rules, construct a set of structured view mapping relationships and establish field mapping paths between source view elements and target view elements;

[0113] S45. Organize each view pair and its corresponding field mapping rules into key-value mapping items according to the mapping pair structure;

[0114] S46. Semantically associate the structured view mapping relationship set with the model element structure to identify the attribute or identifier synchronization rules of model elements across multiple views.

[0115] This invention parses the extended structural fragments in an XMI format model file that represent view mapping relationships, extracts mapping units containing source and target view names, and parses their internal mapping rules to obtain the mapping rule type and source and target field names. Based on this, a structured view mapping relationship set is constructed, establishing field mapping paths between source and target view elements. The view mapping set is then organized into key-value mapping items according to the mapping pair structure, and further semantically associated with the model element structure, thereby clarifying the attribute or identifier synchronization rules of model elements across multiple views.

[0116] In this embodiment, the consistency verification mechanism refers to the process of establishing comparison datasets in the dimensions of model elements, graphics data, and structured view mapping for the landing results of unified model expression data on the target platform during the cross-platform import of model data, and performing the entire process of item-by-item comparison and anomaly recording.

[0117] Specifically, it includes:

[0118] Model element dimension: Compare unique identifiers, metaclass types, stereotype labels and attribute sets to detect missing attributes, type mismatches, name mapping errors and default value filling errors;

[0119] Graphical data dimensions: Compare graphic type, size, coordinate position, color parameters and trajectory path to detect geometric deformation, coordinate offset, color distortion and trajectory loss;

[0120] Structured view mapping dimension: Validate the mapping rules and semantic correspondence of view names and fields, and detect missing fields, rule errors and semantic ambiguity.

[0121] All discrepancies are categorized and statistically analyzed by dimension to generate a consistency verification record list. Based on preset thresholds, it is determined whether they constitute inconsistencies. Finally, a consistency verification report is output, which includes the discrepancy type, field number, example value, and repair suggestions, and is used to update the import status identifier of the unified model representation data.

[0122] The consistency verification mechanism runs through the entire process of forward and reverse imports, and is an important guarantee for bidirectional mutual import and communication. It can ensure that multi-dimensional data such as logical view, physical view, and demand view do not suffer structural damage or semantic loss during cross-platform transmission.

[0123] In this embodiment, the process of constructing the structured view mapping relationship set is as follows:

[0124] The parsed mapping units are grouped according to the combination of source view name and target view name. Within each group, mapping rules are organized, and the mapping rule type, source field name, and target field name are extracted. Each mapping rule is combined with a unique identifier or attribute reference in the model element structure to generate a mapping path from the source field to the target field. Based on this, using the view pair as the key, the corresponding field mapping path and its associated rule entries are used as values ​​to organize key-value mapping items and summarize them to form a set of structured view mapping relationships. At the same time, the index binding relationship with the model element structure is recorded to identify the attribute or identifier synchronization rules between multiple views.

[0125] In this embodiment, step S5 includes the following specific steps:

[0126] S51. Based on the metaclass mapping rules, perform type conversion operations on the metaclass types in the model element structure representation to generate the metaclass layer structure expression supported by the target modeling platform.

[0127] The metaclass mapping rules refer to a set of rules used to guide the semantic alignment and type conversion of model element metaclass types in the source platform to metaclass types in the target platform; the rule set includes a metaclass correspondence table, attribute inheritance handling strategy, equivalent substitution mechanism for incompatible metaclasses, and semantic preservation strategy;

[0128] S52. Based on the attribute alignment rules, convert the source attribute key-value pairs in the attribute set of the model element into the target attribute set, and complete the attribute name mapping, data type conversion and default value completion.

[0129] The attribute alignment rule refers to the process of performing item-by-item comparison and renaming of the attribute set defined in the source platform element during the model element structure conversion process, according to the preset attribute mapping table.

[0130] The attribute mapping table explicitly lists the one-to-one mapping relationship between the attribute names of the source platform and the corresponding attribute names of the target platform. When there are semantically similar but differently named attributes in the target platform, the attribute alias table is used for conversion. When there are extended attributes in the source platform that are not supported by the target platform, attribute filtering is performed, marked as "unmappingable attributes", and removed from the target structure. When a required attribute of the target platform is missing in the source platform, the default attribute template is called to assign an initial value to ensure the integrity and compliance of the attribute structure.

[0131] S53. Based on the semantic role of the stereotype tags, perform mapping processing on the stereotype tag information and output transformed tags that conform to the stereotype definition of the target modeling platform;

[0132] S54. Based on the target platform type, graphic semantic tags and style features, select suitable geometric mapping rules, color normalization rules and layout adjustment rules from the graphic conversion rule set, and perform rule matching and conversion operations in sequence to generate a graphic data structure that meets the semantic requirements of the target platform.

[0133] S55. Based on the conversion results of S54, according to the target platform's graphics file format, encoding specifications and rendering engine requirements, complete the formatted encoding, data structure rearrangement and storage mapping of graphics geometric features, color features and layout features, and output the final graphics data structure that can be directly loaded and rendered.

[0134] S56. Maintain the index binding relationship between the model element structure, the graphical data structure and the view mapping relationship set to form an intermediate model representation dataset with consistent structure, clear semantics and graphical standardization, and output unified model expression data;

[0135] This invention converts the metaclass types in the model element structure representation into a metaclass layer structure expression supported by the target modeling platform based on metaclass mapping rules. It combines attribute alignment rules to complete the name mapping, data type conversion, and default value completion of source attributes to target attributes, and performs semantic mapping on stereotype tags. At the same time, it selects and executes geometric mapping, color normalization, and layout adjustment operations based on the graphics transformation rule set, and completes the formatted encoding and structural rearrangement of the graphics data structure according to the target platform format and rendering engine requirements. Finally, it maintains the index binding of the model element structure, graphics data structure, and view mapping relationship set, generating unified model expression data with consistent structure, clear semantics, and standardized graphics.

[0136] In this embodiment, the model element structure conversion process refers to the entire process in which the system exports the XMI format model file from the source platform during the model data interaction process, performs unified structure and semantic mapping through the intermediate conversion module, and then imports it into the target platform. This process is used to achieve model data interoperability and consistent graphical representation between heterogeneous modeling platforms.

[0137] In this embodiment, step S6 includes the following specific steps:

[0138] S61. The unified model expression data includes a standardized model element structure, a graphical data structure, and a set of structured view mapping relationships, and maintains the index binding relationship between the three.

[0139] S62. Import the model element structure into the data model management component of the target platform, create model element instances according to the metaclass layer structure specification defined by the target platform, and complete the metaclass structure registration and attribute set initialization.

[0140] S63. Import the graphic data structure into the graphics rendering module of the target platform, generate and deploy graphic objects according to graphic type, coordinates, size, color and trajectory path, build a graphic view supported by the target platform and maintain consistency with the graphic data structure.

[0141] S64. Import the structured view mapping relationship set into the view management module of the target platform, register the view pair names and field mapping rules, and establish an attribute mapping index table between views within the target platform.

[0142] S65. Perform a consistency check task, load the model data before transformation and the imported model data after transformation, construct comparison datasets in the dimensions of model elements, graphical data, and structured view mapping, and perform consistency checks respectively:

[0143] At the model element level, unique identifiers, metaclass types, stereotype tags and attribute sets are compared one by one, and anomalies such as missing attributes, type mismatch, name mapping errors and default value filling errors are recorded.

[0144] In the dimension of graphic data, the graphic type, size, coordinate position, color parameters and trajectory path are compared according to the graphic node number to record anomalies such as geometric deformation, coordinate offset, color distortion and trajectory loss.

[0145] In the structured view mapping dimension, verify the consistency of view names, field mapping rules and semantic correspondence, and record anomalies such as field omissions, rule errors and semantic ambiguities;

[0146] S66. Compile and classify the comparison results of each dimension to form a consistency verification record list, record the comparison status, difference type and field number, and determine whether the difference constitutes an inconsistency item based on the preset threshold. For differences exceeding the threshold, add the location, difference description and possible cause explanation.

[0147] S67. Summarize inconsistencies to generate a consistency verification report. The report includes field names, dimensions, difference types, example values, and remediation suggestions. Update the import status flag of the unified model representation data based on the verification results. Mark "verification passed" if there are no inconsistencies, and mark "verification failed" if there are inconsistencies. Attach a list of differences and suggested processing paths to complete the cross-platform import and graphical reproduction process of multi-view model data.

[0148] After generating unified model representation data, this invention imports the standardized model element structure, graphical data structure, and structured view mapping relationship set into the data model management component, graphics rendering module, and view management module of the target platform, respectively. It completes the construction of model element instances, generation of graphical objects, and registration of view mapping indexes. It performs consistency verification in three dimensions: model elements, graphical data, and structured view mapping, counts and determines differences, generates a consistency verification report containing a list of differences and repair suggestions, and finally updates the import status of unified model representation data, realizing cross-platform import and graphical reproduction of multi-view model data.

[0149] In this embodiment, the preset threshold refers to the judgment benchmark value used in the consistency verification process to determine whether the comparison results of the three dimensions of model elements, graphic data, and structured view mapping constitute an "inconsistency item," specifically including:

[0150] The threshold for model element dimensions is determined by the field matching ratio. When the matching ratio is lower than the preset percentage (usually set between 95% and 100%), it is determined to be an inconsistency.

[0151] Dimension thresholds for graphical data include the following categories:

[0152] Coordinate offset tolerance: When the coordinate position offset distance of the same graphic node exceeds the set pixel value range (1 to 5 pixels), it is judged as inconsistent;

[0153] Size difference tolerance: When the size change of the graphic exceeds the preset percentage range (2% to 5%), it is judged as inconsistent;

[0154] Color difference tolerance value: When the color difference value of the graphic exceeds the set color difference tolerance range (1 to 3 color difference units), it is judged as inconsistent.

[0155] The structured view mapping dimension threshold is determined by the matching ratio of field mapping rules. When the matching ratio is lower than the preset percentage (set between 98% and 100%), it is determined to be an inconsistency.

[0156] The above thresholds can be set through experimental statistics or business rules according to the accuracy requirements of different projects and the actual data characteristics, and can be adjusted in the system configuration.

[0157] In this embodiment, the metaclass hierarchical structure specification refers to a set of structural specifications in the target modeling platform used to define the inheritance hierarchy, structural constraints, and type organization forms between the metaclasses to which model elements belong. The structural specifications include metaclass inheritance system, attribute inheritance rules, stereotype adaptation requirements, and semantic preservation constraints, which are used to guide the construction and registration of model element instances in the target platform, ensuring that the semantics of model elements are not lost, types do not conflict, and structures are consistent after cross-platform conversion.

[0158] Example 1:

[0159] To verify the feasibility and effectiveness of this invention in practical engineering applications, it was deployed in the cross-platform model collaboration environment of a large engineering design firm. This firm has long used two heterogeneous modeling platforms, M-Design and CSM, to complete system design and structural analysis. Due to differences in model data structures, graphics rendering methods, and multi-view mapping rules, traditional import / export processes often result in problems such as lost model elements, misaligned graphic layouts, and incorrect attribute mappings. For example, during import, logical views generated by the CSM platform cannot be fully rendered in M-Design, and some stereotype tags in M-Design lose their semantic relationships after rendering on the CSM platform. These problems directly affect the efficiency of cross-team collaboration, leading to repeated modeling and manual verification, thus extending the project cycle by an average of more than 15%.

[0160] After deployment, the system first inputs the XMI format model file exported from the source platform into the model element parsing module and the graphics information parsing module to extract the standardized model element structure and graphics data structure. Then, the view mapping construction module generates the mapping relationship between multiple views. Subsequently, the conversion and standardization module, based on metaclass mapping rules, attribute alignment rules, and graphics conversion rule sets, transforms the model element types, attribute sets, and graphics layout formats of different platforms into unified model expression data. Finally, this data is loaded into the model management component, graphics rendering module, and view management module of the target platform through the import and verification module, and a consistency verification mechanism is executed to ensure semantic and structural consistency across the three dimensions of model elements, graphics features, and multi-view mapping. The entire process is transparent to the user; designers only need to perform an export and import operation once on the source platform to complete a complete bidirectional conversion from M-Design to CSM or from CSM to M-Design.

[0161] In the verification phase, the testing team selected 12,458 model elements, 8,632 graphical objects, and 1,254 view mapping pairs as a sample set for bidirectional import testing. Import time, the number of consistency verification anomalies, and the pass rate were statistically analyzed. Test results showed that whether importing from M-Design to CSM or vice versa, the import time remained within 20 seconds, and the consistency verification pass rate was above 99.9%. Anomalies were mainly concentrated in a very small number of color parameter precision differences and missing mappings in a few view fields, without affecting the overall semantics and rendering effect of the model. Compared with traditional manual conversion, this invention significantly reduces the workload of manual comparison and repair, shortening the model interaction process from an average of 40 minutes to less than 20 seconds, and avoiding structural errors caused by repeated modeling. It also greatly improves cross-platform collaboration efficiency, as shown in Table 1.

[0162] Table 1. XMI bidirectional import performance and consistency verification results.

[0163] Import direction Total number of model elements Total number of graphic objects Number of view mapping pairs Import time (seconds) Number of consistency check anomalies Consistency check pass rate M-Design→CSM 12458 8632 1254 18.7 7 99.94% CSM→M-Design 12458 8632 1254 19.2 9 99.92%

[0164] In the bidirectional import test between M-Design and CSM, the total number of model elements, the total number of graphic objects, the number of view mapping pairs, the import time, the number of consistency check anomalies, and the consistency check pass rate were statistically analyzed.

[0165] The results in Table 1 above demonstrate that the present invention is not only significantly superior to traditional manual methods in terms of data import efficiency, but also has extremely high accuracy in maintaining multi-dimensional consistency, effectively solving the problems of structural distortion and semantic loss in the process of multi-view data interaction between heterogeneous modeling platforms.

[0166] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A heterogeneous model multi-view data consistency interaction system based on the XMI standard, characterized in that, include: The model element parsing module is used to parse model element data in XMI format model files; The graphics information parsing module is used to parse the graphics data structure in the XMI format model file; The view mapping building module is used to identify and extract extended structure fragments representing view mapping relationships from XMI format model files, and parse the metadata structure in the extended structure fragments; The transformation and standardization module is used to generate unified model representation data supported by the target platform based on metaclass mapping rules, attribute alignment rules, and graph transformation rule sets. The import and verification module is used to import the unified model representation data into the data model management component, graphics rendering module and view management module of the target platform, respectively, and to perform a consistency verification mechanism, including model element consistency verification, graphics feature consistency verification and view mapping consistency verification, generate consistency verification records and consistency verification reports, and update the import status identifier of the unified model representation data.

2. The heterogeneous model multi-view data consistency interaction system based on the XMI standard according to claim 1, characterized in that, The modules are connected in the following way: S1. Receive the XMI format model file exported from the source platform; S2. Parse the model element data and generate a structural representation of the model elements; S3. Parse the graphical data structure and generate the graphical data structure; S4. Extract the mapping pairs between the source view and the target view, and construct a set of structured view mapping relationships based on the mapping rule type and field attribute correspondence. S5. Based on the metaclass mapping rules, attribute alignment rules and graphics transformation rule set, perform type conversion, attribute mapping and graphics format standardization processing on the model element structure representation, stereotype label information and graphics structure to generate unified model expression data supported by the target platform. S6. Import the unified model representation data into the data model management component, graphics rendering module, and view management module of the target platform respectively, and perform consistency verification in three dimensions: model elements, graphics data, and structured view mapping. Generate consistency verification records and consistency verification reports, and update the import status flag of the unified model representation data.

3. The heterogeneous model multi-view data consistency interaction system based on the XMI standard according to claim 2, characterized in that, Both the source platform and the target platform can be heterogeneous modeling platforms that conform to the XMI standard, and they can serve as each other's source and target in different application scenarios to support bidirectional data exchange and consistent interaction.

4. The heterogeneous model multi-view data consistency interaction system based on the XMI standard according to claim 3, characterized in that, The bidirectional communication and interoperability refers to the complete bidirectional conversion and interaction of multi-view model data conforming to the XMI standard between the M-Design platform and the CSM platform. During the forward import process, the XMI format model files exported from the M-Design platform are processed by the above modules and converted into unified model expression data that conforms to the semantics and rendering requirements of the CSM platform before being imported into CSM. During the reverse import process, the XMI format model file exported from the CSM platform is processed by the above modules and converted into unified model expression data that conforms to the semantics and rendering requirements of the M-Design platform before being imported into M-Design. During the bidirectional data exchange process, the consistency of multi-dimensional data in the logical view, physical view, and requirement view is maintained. Through structured mapping and consistency verification mechanisms, it is ensured that the cross-view element relationships and graphic layout styles are not lost or distorted during the import and export process.

5. The heterogeneous model multi-view data consistency interaction system based on the XMI standard according to claim 2, characterized in that, The XMI format model file contains model element data, graphical data structure, and view mapping relationship data.

6. The heterogeneous model multi-view data consistency interaction system based on the XMI standard according to claim 2, characterized in that, S2 includes the following specific steps: S21. Based on the model element data in the XMI format model file, identify and extract the unique identifier information of each model element to distinguish different element instances; S22. Extract the metaclass type information of the model elements and construct a type mapping relationship consistent with the metamodel definition; S23. Analyze the prefiguration label information attached to the model elements and identify the semantic role of the prefiguration label information in the specific modeling context; S24. Extract the set of attributes contained in the model elements and construct the key-value mapping relationship between attribute names and attribute values; S25. Identify the relationships between model elements, including relationship type, relationship direction, relationship endpoints and multiplicity information, and construct a structured element relationship graph; S26. Based on the extracted unique identifier information, metaclass type information, stereotype label information, attribute set and association relationship, a preliminary model element structure representation is generated.

7. The heterogeneous model multi-view data consistency interaction system based on the XMI standard according to claim 2, characterized in that, The graphic data structure extracts the graphic identifier, graphic type, graphic size, position coordinates, color attributes, and trajectory path corresponding to the model elements to generate the graphic data structure.

8. The heterogeneous model multi-view data consistency interaction system based on the XMI standard according to claim 2, characterized in that, S4 includes the following specific steps: S41. Identify and extract extended structure fragments used to represent view mapping relationships from the XMI format model file, and parse the metadata structure in the extended structure fragments; S42. Extract mapping units from the metadata structure. Each mapping unit contains source view name and target view name fields. S43. Parse the mapping rules defined within each mapping unit and extract the mapping rule type, source field name, and target field name; S44. Based on the extracted mapping units and mapping rules, construct a set of structured view mapping relationships and establish field mapping paths between source view elements and target view elements; S45. Generate a set of structured view mapping relationships, and organize each view pair and its corresponding field mapping rules into key-value mapping items according to the mapping pair structure; S46. Semantically associate the structured view mapping relationship set with the model element structure to identify the attribute or identifier synchronization rules of model elements across multiple views.

9. The heterogeneous model multi-view data consistency interaction system based on the XMI standard according to claim 2, characterized in that, S5 includes the following specific steps: S51. Based on the metaclass mapping rules, perform type conversion operations on the metaclass types in the model element structure representation to generate the metaclass layer structure expression supported by the target modeling platform. The metaclass mapping rules refer to a set of rules used to guide the semantic alignment and type conversion of model element metaclass types in the source platform to metaclass types in the target platform; the rule set includes a metaclass correspondence table, attribute inheritance handling strategy, equivalent substitution mechanism for incompatible metaclasses, and semantic preservation strategy; S52. Based on the attribute alignment rules, convert the source attribute key-value pairs in the attribute set of the model element into the target attribute set, and complete the attribute name mapping, data type conversion and default value completion. The attribute alignment rule refers to the process of performing item-by-item comparison and renaming of the attribute set defined in the source platform element during the model element structure conversion process, according to the preset attribute mapping table. S53. Based on the semantic role of the stereotype tags, perform mapping processing on the stereotype tag information and output transformed tags that conform to the stereotype definition of the target modeling platform; S54. Based on the target platform type, graphic semantic tags and style features, select suitable geometric mapping rules, color normalization rules and layout adjustment rules from the graphic conversion rule set, and perform rule matching and conversion operations in sequence to generate a graphic data structure that meets the semantic requirements of the target platform. S55. Based on the aforementioned graphic data structure, in accordance with the target platform's graphic file format, encoding specifications, and rendering engine requirements, complete the formatted encoding, data structure rearrangement, and storage mapping of the graphic geometric features, color features, and layout features, and output the final graphic data structure that can be directly loaded and rendered. S56. Maintain the index binding relationship between the model element structure, the graphical data structure and the view mapping relationship set to form an intermediate model representation dataset with consistent structure, clear semantics and graphical standardization, and output unified model expression data.

10. The heterogeneous model multi-view data consistency interaction system based on the XMI standard according to claim 2, characterized in that, S6 includes the following specific steps: S61. The unified model expression data includes a standardized model element structure, a graphical data structure, and a set of structured view mapping relationships, and maintains the index binding relationship between the three. S62. Import the model element structure into the data model management component of the target platform, create model element instances according to the metaclass layer structure specification defined by the target platform, and complete the metaclass structure registration and attribute set initialization. S63. Import the graphic data structure into the graphics rendering module of the target platform, generate and deploy graphic objects according to graphic type, coordinates, size, color and trajectory path, build a graphic view supported by the target platform and maintain consistency with the graphic data structure. S64. Import the structured view mapping relationship set into the view management module of the target platform, register the view pair names and field mapping rules, and establish an attribute mapping index table between views within the target platform. S65. Perform a consistency check task, load the model data before transformation and the imported model data after transformation, construct comparison datasets in the dimensions of model elements, graphical data, and structured view mapping, and perform consistency checks respectively: At the model element level, unique identifiers, metaclass types, stereotype tags and attribute sets are compared one by one to record missing attributes, type mismatches, name mapping errors and default value filling errors. In the dimension of graphic data, the graphic type, size, coordinate position, color parameters and trajectory path are compared according to the graphic node number to record geometric deformation, coordinate offset, color distortion and trajectory loss; In the structured view mapping dimension, verify the consistency of view with name, field mapping rules and semantic correspondence, and record field omissions, rule errors and semantic ambiguities; S66. Compile and classify the comparison results of each dimension to form a consistency verification record list, record the comparison status, difference type and field number, and determine whether the difference constitutes an inconsistency item based on the preset threshold. For differences exceeding the threshold, add the location, difference description and possible cause explanation. S67. Summarize inconsistencies to generate a consistency verification report. The report includes field names, dimensions, difference types, example values, and remediation suggestions. Based on the verification results, update the import status flag of the unified model representation data. If there are no inconsistencies, mark it as "verification passed"; if there are inconsistencies, mark it as "verification failed" and attach a list of differences and suggested processing paths to complete the cross-platform import and graphical reproduction process of multi-view model data.

Images