A cross-platform collaborative loading method for CAD drawings

Abstract

This invention provides a cross-platform collaborative loading method for CAD drawings, relating to the field of data processing technology. The method includes: performing rasterization and layering processing on the original CAD file on a central platform to obtain a multi-level raster tile set; storing the raster tiles in a temporary storage unit using a pyramid-level index; obtaining the updated view state by real-time monitoring of user interaction events, comparing the updated view state with the original CAD file to obtain view parameters; extracting the target raster tile set; converting it to a neutral exchange format set; scheduling a container format conversion service based on format support characteristics, and distributing it compatiblely to M edge platforms; and performing synchronous rendering and replacement of cross-platform tile files based on M role permissions. This invention solves the technical problem of existing technologies lacking effective CAD file format compatibility processing, leading to cumbersome data conversion between platforms, and consequently reducing data accuracy and reliability.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and specifically to a cross-platform collaborative loading method for CAD drawings. Background Art

[0002] With the digital transformation of the construction, manufacturing, and engineering industries, the application of CAD (Computer-Aided Design) systems is becoming increasingly common. CAD systems are typically used to create accurate two-dimensional or three-dimensional graphic models. As project scales up and the need for team collaboration increases, the demand for collaborative operation of CAD files is growing, especially in large-scale projects spanning multiple locations and time zones, where real-time collaboration and data consistency become particularly important.

[0003] However, existing technologies typically require processing large amounts of CAD data, which is loaded directly into the system without optimization, resulting in slow system response and a poor user experience, especially in environments with limited network conditions or hardware performance. Furthermore, different platforms support CAD file formats differently, lacking a unified data format, which necessitates cumbersome data conversion between platforms. This can lead to information loss or errors during the data conversion process, reducing the accuracy and reliability of the data. Summary of the Invention

[0004] This application provides a cross-platform collaborative loading method for CAD drawings, which addresses the technical problem that existing technologies have different platforms supporting different CAD file formats and lack effective data format compatibility handling, resulting in cumbersome data conversion between platforms, which may lead to information loss or errors during the data conversion process, reducing the accuracy and reliability of the data.

[0005] In view of the above problems, this application provides a cross-platform collaborative loading method for CAD drawings.

[0006] This application provides a cross-platform collaborative loading method for CAD drawings. The method includes: performing rasterization and layering processing on the original CAD file on a central platform to obtain a multi-level raster tile set; storing the multi-level raster tile set in a temporary storage unit of the central platform using a pyramid hierarchical index; obtaining view parameters by comparing the updated view state with the original CAD file after obtaining the updated view state through real-time monitoring of user interaction events; extracting a target raster tile set from the temporary storage unit using the view parameters as an index; converting the target raster tile set into a neutral exchange format set on the central platform; scheduling a container format conversion service according to the format support characteristics of M edge platforms to compatiblely distribute the neutral exchange format set to the M edge platforms; and performing synchronous rendering and replacement of cross-platform tile files based on the M role permissions of the M edge platforms after receiving edge modification information.

[0007] The technical solution provided in this application has at least the following technical effects or advantages:

[0008] By rasterizing and layering the original CAD files and storing them in the central platform's temporary storage unit using a pyramid hierarchical index, large-scale CAD drawing data can be processed efficiently. This rasterization and layering process transforms complex drawing data into a tile format that is easy to store and access. The pyramid hierarchical index ensures that tile data at different resolutions can be retrieved quickly, thereby improving the system's access speed and responsiveness. By monitoring user interaction events in real time to obtain updated view states and generating view parameters based on this, the system can flexibly respond to users' dynamic interaction needs. After each user interaction, the system accurately obtains the view parameters required by the user by comparing the updated view state with the original CAD file, ensuring that the tile set that meets the requirements is extracted. This not only improves the user experience and avoids unnecessary resource waste, but also optimizes system performance, ensuring that tile extraction is dynamic and efficient. The extracted target raster tile set is converted into a neutral exchange format set, enabling cross-platform compatible transmission and exchange of data and avoiding data format incompatibility issues between different platforms. By scheduling container format conversion services, the neutral exchange format set is compatiblely distributed to multiple edge platforms, ensuring that different platforms can process and display data according to their format support characteristics, improving data processing efficiency and maximizing the processing capabilities of each platform. After receiving modification information from the edge platforms, the central platform replaces the cross-platform tile files according to the role permissions of each edge platform and triggers synchronous rendering on the edge platforms, ensuring data consistency and correctness. This mechanism ensures data synchronization between multiple platforms, keeping the tile data updates consistent across platforms and rendering the latest content in a timely manner, improving collaborative work efficiency and avoiding data delays and version inconsistencies.

[0009] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0010] Figure 1 A schematic diagram of a cross-platform collaborative loading method for CAD drawings provided in this application;

[0011] Figure 2 This is a flowchart illustrating the process of obtaining a multi-level grid tile set in a cross-platform collaborative loading method for CAD drawings provided in this application. Detailed Implementation

[0012] This application provides a cross-platform collaborative loading method for CAD drawings, which addresses the technical problem that existing technologies have different platforms supporting different CAD file formats and lack effective data format compatibility handling, resulting in cumbersome data conversion between platforms, which may lead to information loss or errors during the data conversion process, reducing the accuracy and reliability of the data.

[0013] The technical solutions of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. It should be understood that the present invention is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention. It should also be noted that, for ease of description, only the parts related to the present invention are shown in the accompanying drawings, not all of them.

[0014] Examples, such as Figure 1 As shown, this application provides a cross-platform collaborative loading method for CAD drawings, the method comprising:

[0015] A100: The original CAD file is rasterized and layered on the central platform to obtain a multi-level raster tile set.

[0016] In one embodiment, the original CAD file is rasterized and layered on a central platform to obtain a multi-level raster tile set, such as... Figure 2 As shown, the method step A100 provided by the present invention includes:

[0017] A110: Based on the scale characteristics, the original CAD file is divided into multi-level sub-drawing sets according to spatial index, wherein the multi-level sub-drawing sets are bound to multi-level fixed scales.

[0018] A120: Preset reference resolution, reference raster size, and hierarchical parameter tuning rules.

[0019] A130: Starting from the reference resolution and reference raster scale, multi-level rasterization parameters are calculated according to the hierarchical parameter adjustment rules and multi-level fixed scales, wherein the multi-level rasterization parameters include multi-level resolution and multi-level raster scale.

[0020] A140: Call the Mapnik toolchain to perform batch rasterization processing of the multi-level sub-drawing sets according to the multi-level rasterization parameters, and output the multi-level raster tile set, wherein each raster tile is bound to a hierarchical affiliation, row and column coordinates, and tile spatial range.

[0021] Specifically, in this embodiment, the scale parameters (e.g., 1:100, 1:200, etc.) defined in the CAD file are first parsed, and then a spatial indexing algorithm is used to divide several CAD drawings in the original CAD file into layers, so that each layer corresponds to a specific display scale range. For example, the sub-drawing set at the 1:100 level only contains the geometric elements that need to be displayed at that scale, ensuring data separation at different detail levels during subsequent processing.

[0022] The reference resolution is defined as the pixel precision of the initial level (such as the maximum scale level), for example, each pixel represents an actual size of 0.05 mm; the reference grid scale is set as the basic pixel size of the tile (such as 512×512 pixels).

[0023] The hierarchical parameter tuning rules are based on the principle of proportional scaling. For example, with each additional level, the resolution decreases exponentially by a coefficient of 2, while the raster size increases accordingly. Based on this, starting from the reference resolution and reference raster scale, according to the hierarchical parameter tuning rules and multi-level fixed scales, multi-level rasterization parameters in sequence form from high-precision details to low-precision overviews are calculated. The multi-level rasterization parameters include multi-level resolution and multi-level raster scale.

[0024] The Mapnik toolchain is invoked to perform batch rasterization of the multi-level sub-drawing sets according to the multi-level resolution and multi-level raster scale mapping, and the multi-level raster tile set is output.

[0025] Meanwhile, during the image rasterization process, the hierarchical information (such as hierarchical number), row and column coordinates (based on the indexing rules of the tile matrix), and spatial range of each raster tile are automatically generated to ensure that the tiles can be accurately indexed and located in subsequent applications, and the obtained hierarchical information, row and column coordinates, and tile spatial range are bound to the corresponding raster tiles.

[0026] The resulting multi-level grid tile set is subsequently organized in a pyramid structure and supports a dynamic loading mechanism, which can match a number of corresponding tiles when the user zooms in on the view in the visualization interface.

[0027] A200: The multi-level raster tile set is stored in the temporary storage unit of the central platform using a pyramid hierarchical index.

[0028] In one embodiment, a pyramid hierarchical index is used to store the multi-level raster tile set to a temporary storage unit of the central platform. The method step A200 provided by this invention includes:

[0029] A210: Establish a directory structure based on the multi-level fixed scale to obtain a standardized storage path, wherein the tile storage path rule of the standardized storage path is level-row number-column number.

[0030] A220: Extract the multi-level tile spatial range set from the multi-level grid tile set to establish a multi-level hierarchy-precision mapping set as a spatial index database.

[0031] A230: After updating the multi-level raster tile set to the spatial index database according to the standardized storage path, the spatial index database is migrated to the temporary storage unit.

[0032] Specifically, in this embodiment, the root directory is the level number, and the storage address is generated according to the three-level path rule of "level / row number / column number". For example, the tile in the 3rd row and 5th column of the 2nd level is stored as Level2 / Row3 / Col5.webp. Each level in the directory structure corresponds to a preset scale range (e.g., Level1 corresponds to 1:100-1:200). The row and column numbers are generated based on the Cartesian coordinate system indexing rule of the tile matrix to ensure the physical spatial continuity of adjacent tiles.

[0033] From the metadata bound to the multi-level raster tile set, the layer affiliation, row and column coordinates, and spatial range coordinates of each tile are parsed (e.g., bottom left corner X=1200mm, Y=3500mm, top right corner X=1800mm, Y=4100mm). This data is organized into a spatial index table using a four-dimensional index (layer-Z, minimum X-Xmin, minimum Y-Ymin, maximum X-Xmax). Simultaneously, a layer-precision mapping table is established, recording the resolution of each level (e.g., Level_1 is 0.025mm / pixel) and the actual size range covered by the tile (e.g., a single tile represents a 500mm × 500mm area). The mapping table and the spatial index table are merged based on tile consistency to form a key-value pair database supporting range queries, serving as the spatial index database.

[0034] The information of each grid tile in the multi-level grid tile set is added to the spatial index database according to a standardized storage path. This process involves storing the storage path and related spatial information of each tile into the database so that it can be retrieved and managed efficiently in the future.

[0035] After the database update is complete, the spatial index database is migrated to a temporary storage unit. A temporary storage unit is a temporary data storage space, typically used for efficiently storing and processing intermediate data, avoiding load on the main storage system. In some cases, the temporary storage unit may be a high-speed storage device (such as SSD or RAM) or a dedicated storage server for processing and analysis. The temporary storage unit provides a fast-access data source for subsequent data operations. Storing the spatial index database in temporary storage ensures efficient reading and processing in subsequent operations, avoiding the need to access slower storage locations each time.

[0036] A300: After obtaining the updated view status by listening to user interaction events in real time, the view parameters are obtained by comparing the updated view status with the original CAD file.

[0037] In one embodiment, after obtaining the updated view state by real-time monitoring of user interaction events, the view parameters are obtained by comparing the updated view state with the original CAD file. The method step A300 provided by this invention includes:

[0038] A310: Select a dynamic listening mechanism based on the device type of the central platform to capture user interaction events in order to extract the updated view state, wherein the updated view state includes viewport boundary coordinates and real-time zoom level.

[0039] A320: The real-time hierarchy is obtained by mapping the real-time zoom level to the multi-level fixed scale for hierarchical correction.

[0040] A330: After normalizing the viewport boundary coordinates to global boundary coordinates, the global boundary coordinates are projected onto the original CAD file for range validity comparison, and the valid viewport coordinates are output. The real-time level and the valid viewport coordinates constitute the view parameters.

[0041] Specifically, in this embodiment, an appropriate listening mechanism is selected based on the device type of the central platform. For example, desktop devices typically use a mouse and keyboard as input devices, so an event-based listening mechanism can be selected, such as listening to mouse wheel and drag events. Mobile devices commonly use touch screens as input devices, so a touch event listening mechanism can be selected, such as touch zoom, swipe, and other gesture operations.

[0042] When users change the view's display through interactions (such as mouse wheel zooming, touch gesture zooming, dragging and panning), the monitoring mechanism captures these operations in real time. For example, when a user changes the map's zoom level using a zoom gesture (such as a pinch gesture or scroll wheel), or changes the map's display area through dragging, the current viewport boundary coordinates are extracted based on the user's panning operation and zoom level. These coordinates define the visible area of ​​the current view. Simultaneously, the current real-time zoom level is calculated; for example, there may be multiple preset zoom levels (such as 1:1000, 1:500, etc.), and the user's current zoom state is recorded to determine the appropriate tile precision for loading. By capturing these interaction events and extracting relevant state information, the view state is updated in real time. The updated view state includes the viewport boundary coordinates and the real-time zoom level, providing necessary information for subsequent image or tile loading.

[0043] Based on the user's current real-time zoom level, it is mapped to a preset multi-level fixed scale. For example, if the user is viewing a drawing at a 1:500 scale, this real-time zoom level needs to be mapped to the corresponding scale level. Level correction determines the actual layer loaded based on the current scale and the user's zoom operation. This correction process involves refining or simplifying the displayed content of the layers. For instance, when the user zooms from 1:500 to 1:1000, the level of detail in the drawing automatically decreases, and the loaded content becomes more simplified. This correction method ensures that the system dynamically adjusts the loaded layer content according to the current zoom level, thus avoiding incomplete or inaccurate display of the drawing content due to excessive or insufficient zoom. After scale mapping and level correction, the resulting real-time level is the actual drawing display level for the current view. This level determines which precision of raster tiles the system should load, ensuring that the drawing's precision and display detail remain consistent across different zoom levels.

[0044] The viewport boundary coordinates acquired in real time are usually local coordinates relative to the current view. In order to compare them in the global coordinate system, the local viewport boundary coordinates must first be converted into global coordinates. This process maps the viewport boundary coordinates to a global coordinate system (such as the actual geographic coordinate system of a map or drawing), so that the viewport's extent can be effectively compared with the spatial extent of the original CAD file. The normalization process includes converting the starting point (such as the top left corner) and ending point (such as the bottom right corner) of the local coordinates into global coordinates. For example, the top left corner of the original CAD file corresponds to (xmin, ymax) in the global coordinate system, while the bottom right corner corresponds to (xmax, ymin).

[0045] The normalized global boundary coordinates are mapped to the spatial range of the original CAD file. All drawing content in the original CAD file has a predefined spatial range, such as the actual geographic coordinate range or the design boundary of the drawing. Through the projection formula, the global coordinates are converted into coordinates consistent with the coordinate system of the original CAD file, ensuring that the coordinates are aligned with the boundary of the CAD file after the coordinate conversion.

[0046] By comparing the global boundary coordinates with the spatial range of the original CAD file, it checks whether the viewport area selected by the user exceeds the valid range of the drawing. If the viewport boundary is within the valid range, the view is considered valid, and the valid viewport coordinates are output. The areas corresponding to these valid coordinates are used to load the drawing content or tile data.

[0047] By combining real-time hierarchy and effective viewport coordinates, view parameters are generated. These view parameters are used for subsequent drawing or tile loading operations. Based on these view parameters, tile or drawing data corresponding to the user's view area is loaded, ensuring that the loading and display of drawings at different levels and zoom levels can meet the user's needs.

[0048] In one embodiment, by mapping the real-time scaling level to the multi-level fixed scale for hierarchical correction, the real-time hierarchy is obtained. The method step A320 provided by this invention includes:

[0049] A321: Extract the multi-level scaling level set of the multi-level sub-drawing set.

[0050] A322: Based on the data distribution characteristics of the multi-level scaling set, a multi-level initial scaling interval is constructed through clustering.

[0051] A323: Perform connection compensation on the multi-level initial scaling intervals to obtain multi-level standard scaling intervals.

[0052] A324: Map the real-time zoom level to the multi-level standard zoom range to locate the target zoom range.

[0053] A325: After mapping and binding the multi-level standard scaling range to the multi-level fixed scale, locate the real-time level in the directory structure according to the correspondence between the target scaling range and the multi-level fixed scale.

[0054] Specifically, in this embodiment, the multi-level sub-drawing sets have been divided according to different scales. Each sub-drawing set is bound to a specific scaling level. Each scaling level represents the fineness or display range of the drawing. All the extracted scaling levels are organized to form a multi-level scaling level set. This set contains all available scaling levels of the drawing and is sorted according to the fineness and display range of the scale, for example, sorted by scale from smallest to largest.

[0055] Analyze the data distribution characteristics of a multi-level zoom level set to identify the correlation between zoom levels. For example, some zoom levels may have small variations, while others may have large differences. For instance, the data distribution characteristics can be analyzed based on factors such as the interval between scales and the content density of the drawing. The analysis of data distribution characteristics will focus on finding clustered areas, that is, areas where the changes between zoom levels are relatively stable or continuous.

[0056] Clustering algorithms, such as K-means and hierarchical clustering, can group multi-scale scaling sets together, clustering similar scaling levels. Clustering algorithms can help identify different scaling intervals where scaling levels have similar characteristics or patterns of change. For example, if the scaling levels between 1:1000 and 1:5000 vary little, while there is a significant difference between 1:100 and 1:2000, then clustering algorithms can place 1:1000 and 1:5000 in the same interval, while dividing other scales into different intervals.

[0057] Based on the clustering results, a multi-level initial scaling interval is constructed. Each interval contains a set of continuous and similar scaling levels, which allows for a smoother transition to different scaling levels during user operation without excessive jumps.

[0058] Multi-level initial zoom intervals may present transition issues between different intervals. For example, the scale difference between some intervals may be significant, resulting in an abrupt zooming experience when switching from one interval to another. Analyzing the boundaries of each interval and checking for excessively large zoom level differences, especially in the transition between different intervals, helps to prevent users from experiencing excessive jumps when zooming by adjusting the boundaries of the zoom intervals. Transition compensation can be achieved by setting transition zoom levels to make the transition between different intervals smoother. For example, if the scale difference between two intervals is too large, compensation can be achieved by adding an intermediate transition interval to reduce the difference.

[0059] The compensated zoom range forms the final multi-level standard zoom range. Each standard zoom range has a clear boundary, ensuring that users can smoothly transition during the zooming process and will not experience obvious discomfort when switching from one range to another, providing a more accurate and smooth user experience for zooming operations.

[0060] The real-time zoom level is mapped to a multi-level standard zoom range to determine which specific target zoom range the level belongs to. The target zoom range refers to the zoom range corresponding to the drawing area currently viewed by the user. This is achieved by matching the real-time zoom level with the boundary of the standard zoom range. After the target zoom range is determined, the corresponding layer and tile can be further located based on this range to ensure that the zoom level during user operation matches the drawing content.

[0061] Based on the defined multi-level standard scaling intervals, each interval is mapped and bound to the actual multi-level fixed scale. This means that each standard scaling interval will be mapped to the corresponding scale level, such as 1:1000, 1:500, etc. This binding relationship ensures that each scaling interval corresponds to multiple scales of different precision, which can more accurately locate and display the content of the drawing.

[0062] Based on the correspondence between the target zoom range and the multi-level fixed scale, the appropriate scale for displaying the drawing is determined. Once the target scale level is determined, the drawing data at that level can be located in the predefined directory structure. The directory structure contains file paths organized according to scale and level, with each path representing the drawing content at a zoom level. Through the above steps, the corresponding real-time level is accurately located in the directory structure based on the correspondence between the target zoom range and the multi-level fixed scale. The real-time level indicates the specific details and display precision of the drawing content.

[0063] A400: Using the view parameters as an index, extract the target raster tile set from the temporary storage unit.

[0064] In one embodiment, using the view parameters as an index, the method step A400 of the present invention extracts the target raster tile set from the temporary storage unit, and includes:

[0065] A410: Combine the real-time hierarchy and effective viewport coordinates into a structured query key.

[0066] A420: Based on the structured query key, drive the spatial index database to perform a spatial range intersection query to filter and obtain a discrete raster tile set covering the effective viewport coordinates.

[0067] A430: Spatially stitch the discrete grid tile set to obtain the target grid tile set.

[0068] Specifically, in this embodiment, the real-time level represents the current zoom level and the precision of the drawing display. The effective viewport coordinates are the boundary coordinates of the view area that are finally determined after validity comparison. The real-time level and the effective viewport coordinates are combined to form a structured query key. This query key contains the spatial range (boundary coordinates) of the viewport and the display level information of the drawing. It can be used to locate the corresponding tile data.

[0069] A spatial extent intersection query is performed in the spatial index database using a structured query key. The goal is to identify raster tiles that overlap with the user's current view area based on the viewport coordinates in the query key, retrieve all tiles that meet the criteria, and ensure that the spatial extent of these tiles intersects with the valid viewport coordinates. The query results contain multiple raster tiles that cover different portions of the viewport coordinate range. All discrete raster tiles are filtered out; these tiles may represent different areas of the drawing, distributed in multiple locations, but they all cover the stated valid viewport coordinates. Each tile contains a specific spatial coordinate range, along with its corresponding layer and scale.

[0070] The selected discrete raster tile set is spatially stitched together. This stitching operation combines the spatial extent and coordinates of each tile to form a continuous image. This involves merging multiple tiles according to their spatial positions, resulting in a seamlessly connected final image. The resulting target raster tile set contains complete view content, displaying the drawing data at the user's current location. These tile sets correspond to the real-time layer, ensuring the image accuracy and detail seen by the user meets scaling requirements. The generated target raster tile set is loaded and displayed to the user. Based on the user's scaling operations, different tile sets are dynamically loaded and stitched together to ensure accurate display of the drawing at different scaling levels.

[0071] A500: The target raster tile set is converted into a neutral exchange format set on the central platform.

[0072] Specifically, the Neutral Exchange Format set is a standardized set of data formats designed to ensure high compatibility of data exchange between different platforms. These formats are typically universal, such as JSON, XML, CSV, or other cross-platform supported standard formats. The use of Neutral Exchange Format ensures that no format incompatibility issues occur when data is transmitted between different platforms.

[0073] To achieve cross-platform compatibility, the target raster tile set is converted into a neutral exchange format set. The conversion process includes: converting image or raster data into JSON or XML format for easy transmission and exchange; and standardizing the geographic information and attributes of the raster tiles to ensure data consistency and integrity. The converted data is stored or transmitted in a neutral exchange format set to ensure subsequent operations can be performed smoothly on different platforms.

[0074] A600: Based on the format support characteristics of the M edge platforms, schedule container format conversion services to compatiblely distribute the neutral exchange format set to the M edge platforms.

[0075] In one embodiment, based on the format support characteristics of the M edge platforms, a container format conversion service is scheduled to distribute the neutral exchange format set compatiblely to the M edge platforms. Step A600 of the method provided by this invention includes:

[0076] A610: Collect M format support lists through interaction with the M edge platforms.

[0077] A620: With the goal of minimizing tasks, the duplicate requirements of the M format support lists are merged to obtain the format support features, wherein the format support features include H format conversion tasks.

[0078] A630: After matching container instances according to the H format conversion tasks, and combining the conversion characteristics of the neutral exchange format set, drive the container instances to convert and distribute the neutral exchange format set to the M edge platforms.

[0079] Specifically, in this embodiment, edge platforms are typically computing nodes in a distributed system. They can perform data processing or transformation locally, avoiding sending all data to a central server. Each edge platform may support different file formats and data types. Therefore, it is necessary to interact with each edge platform to collect its supported file formats and obtain a list of M supported formats. Each platform may support multiple file formats (such as PDF, JPEG, PNG, TIFF, CAD, etc.). These file format lists can be obtained through API calls, query interfaces, or by directly reading the platform's configuration file.

[0080] Duplicate requirements refer to situations where multiple edge platforms may support the same format or perform the same format conversion task. For example, multiple platforms may support converting from PDF to PNG or from JPEG to TIFF. To reduce unnecessary duplicate tasks, the M format support lists are merged to identify these duplicate format requirements. The core objective of the merging process is to find the format conversion tasks that are commonly supported on all platforms and unify them to avoid repeatedly performing the same tasks in subsequent processing.

[0081] The goal of task minimization is to reduce the number of format conversion tasks that need to be performed. For example, if multiple edge platforms support the same format conversion task, these tasks can be merged and executed in a centralized location, thereby reducing the consumption of processing resources. The merged format support features include H format conversion tasks, each representing the need to convert from one format to another.

[0082] In distributed systems, containerization technology is used to package and run applications. Each container instance typically has the ability to handle specific tasks. In this step, based on H format conversion tasks, a suitable container instance is selected. Each format conversion task requires specific processing resources and environment. By comparing the task requirements and the capabilities of the container instance (e.g., whether it supports relevant format conversion tools), the most suitable container instance is selected to handle these tasks.

[0083] A neutral exchange format set refers to a standardized set of intermediate formats that are cross-platform and cross-system compatible, enabling efficient conversion between different formats, such as JSON, XML, CSV, or other standardized data exchange formats. Conversion characteristics refer to how these formats are converted between different systems. For example, some formats need to be converted to a neutral format (such as JSON) before being converted to the target format (such as PDF). These conversion characteristics help determine how the conversion should be performed within the container instance and ensure that data can be accurately converted from one format to another.

[0084] When a container instance is matched and assigned to a format conversion task, the container instance initiates the corresponding conversion process based on the conversion characteristics of the neutral exchange format set. Through the computing power of the container instance, the conversion process is distributed to M edge platforms as needed. Each platform processes the received data according to its own support list. For example, if a platform supports PDF format, it will receive and store the converted PDF file; if other platforms support PNG format, it will receive and store PNG format files.

[0085] In one embodiment, after matching container instances according to the H format conversion tasks, and combining the conversion characteristics of the neutral exchange format set, the container instances are driven to convert and distribute the neutral exchange format set to the M edge platforms. The method step A630 provided by this invention includes:

[0086] A631: Based on the H format conversion tasks, container instances are matched in the container cluster to obtain E format conversion containers, wherein the container cluster is pre-registered on the central platform.

[0087] A632: Based on the neutral exchange format set, predict the time consumption of the H format conversion tasks to obtain the conversion time of the H formats.

[0088] A633: Based on the real-time load of the container and the time taken for H format conversions, dynamically allocate the H format conversion tasks to the E format conversion containers for compatibility conversion, and output H compatible raster tile sets.

[0089] A634: Based on the attribution relationship between the H format conversion tasks and the M format support lists, distribute the H compatible raster tile sets to the M edge platforms.

[0090] Specifically, in this embodiment, the container cluster is a container pool pre-registered on the central platform, containing multiple container instances. These instances can be dynamically scheduled and allocated according to task requirements. Each container instance has different functions and computing capabilities to handle different task types. The central platform manages these container clusters and coordinates task allocation and container instance selection.

[0091] In a container cluster, a suitable container instance is selected based on the requirements of each format conversion task. For example, some tasks need to process large-scale files, while other tasks may only involve simple conversions. Based on the computational requirements and complexity of the task, a container instance that can handle it efficiently is selected. The E format conversion containers are a set of container instances selected from the container cluster based on the requirements of the H format conversion tasks. These container instances are responsible for handling specific format conversion tasks and ensuring that the tasks can be completed efficiently.

[0092] Based on the input and output formats of each format conversion task, the usage of the neutral exchange format set during the conversion process can be determined. Different conversion tasks involve different neutral format conversion paths, affecting the conversion time. A time-consuming prediction model is established. This model predicts the conversion time of each task based on factors such as task type, input and output formats, data volume, and the processing capacity of the target container instance. The H format conversion times refer to the predicted execution time of each task. These time-consuming predictions will provide a basis for subsequent task scheduling, helping the system to better allocate tasks and avoid container instances being overloaded or idle.

[0093] Dynamically monitor the real-time load of each container instance to ensure that tasks are allocated reasonably based on the current resource status of the container instance (such as CPU, memory usage, disk space, etc.). Based on the load of the container instance and the predicted task execution time, dynamically distribute tasks to container instances with appropriate loads. For example, assign tasks with longer execution times to container instances with idle resources and lower loads, and assign tasks with lighter execution times to container instances with higher resource utilization, ensuring load balance across all container instances and preventing individual container instances from becoming overloaded.

[0094] Each container instance executes the format conversion task assigned to it and performs compatibility conversions according to the task requirements. For example, if an input file is in PDF format and the target format is PNG, the container instance will convert the PDF file to PNG format according to the task requirements. During the conversion process, the container instance outputs H compatible raster tile sets, which represent the converted image or data blocks and are outputs compatible with the target format.

[0095] The M format support lists represent the sets of formats supported by different edge platforms. During the task processing, the input and output formats of each format conversion task are matched against the platform's supported formats. Based on these matchups, it is determined which platform the converted data (i.e., the compatible raster tile set) will be distributed to. According to the match between each task and the format support list, the corresponding compatible raster tile set is distributed to the appropriate edge platform. If a platform supports the target format (e.g., a platform supports PNG format), the converted PNG format tile set will be distributed to that platform. Each edge platform, after receiving the tile set, stores, displays, or processes it further, depending on the platform's capabilities.

[0096] A700: After receiving the edge modification information, the central platform performs cross-platform tile file synchronous rendering and replacement according to the M role permissions of the M edge platforms.

[0097] In one embodiment, after receiving edge modification information, the central platform performs cross-platform tile file synchronous rendering and replacement based on the M role permissions of the M edge platforms. The method step A700 provided by this invention includes:

[0098] A710: The central platform receives edge modification information consisting of N edge modification compressed packages, wherein the edge modification compressed package includes a modified tile ID, a modified tile file, and a platform ID, and N≤M.

[0099] A720: Invoke the conflict priority rules of the M edge platforms.

[0100] A730: Based on ID consistency, the edge modification information is modified to detect conflicts according to N modified tile IDs, and U edge modification compressed packages are selected from the edge modification information according to N platform IDs and conflict priority rules.

[0101] A740: Extract U modified tile files from the U edge-modified compressed package.

[0102] A750: After temporarily writing the U modified tile files into the neutral exchange format set based on the U modified tile IDs, cross-platform tile file overwriting and replacement is performed.

[0103] Specifically, in this embodiment, the edge modification compressed package is a collection of modification information packaged and sent by each edge platform. It contains modifications to the tile data, including the modified tile ID, the modified tile file, and the platform ID. The modified tile ID is a unique ID that identifies the tile, including the tile's layer, row number, and column number. This information allows for precise location of each tile. The modified tile file contains the modified content of the tile, which may be a change in the tile data itself or an update to certain image, map, or other files. The platform ID is the edge platform ID that identifies the platform that generated the modification, helping to track which platforms submitted the modification and facilitating conflict resolution and data source management.

[0104] N modified compressed packages generated by N (N≤M) edge platforms will be transmitted to the central platform. The size and content of these modified compressed packages can vary depending on the amount of tile data being modified. The number N refers to the number of edge platforms participating in the modification submission, which can be at most equal to the total number of edge platforms M.

[0105] In a distributed system, different edge platforms may modify the same tile simultaneously. For example, tiles with the same row and column number may be updated on different platforms. In this case, data conflicts will occur, meaning that the content of the same tile has been modified on multiple platforms. In this situation, conflict priority rules are needed to determine which modifications should be kept and which should be discarded.

[0106] Conflict priority rules are a predefined set of rules used to determine how to resolve conflicts when multiple edge platforms submit changes. For example, time priority means the latest change has higher priority, so the change with the latest timestamp will be selected; platform priority means that changes from certain platforms are given priority, such as setting specific edge platforms as having higher priority or being more trusted platforms; platform load priority means that if a platform is heavily loaded, changes from other platforms will be given higher priority. The central platform handles conflicts according to these conflict priority rules. If changes to the same tile come from multiple platforms, the central platform will decide which changes to retain according to the preset rules.

[0107] Each modified tile ID is compared to ensure that modifications to the same tile ID on different platforms do not conflict. If the same tile is modified on multiple platforms—for example, if the tile's image content, data, or attributes are updated simultaneously on different platforms—a conflict is considered to have occurred, and further rules must be used to determine the final modification. If a conflict is detected, a conflict priority rule is applied to decide which modification to retain. Finally, U valid modification archives are selected from N edge-modified archives. These archives contain the final modified content after conflict resolution.

[0108] From the selected U edge modification compressed packages, extract the actual U modification tile files. These tile files contain images, data, or other formatted files that represent the updated content of the tiles.

[0109] The extracted U modified tile files are temporarily written to a neutral exchange format set. With the support of the neutral exchange format set, tile files can be overwritten and replaced across multiple platforms. Specifically, on each target platform, the extracted tile files overwrite the original tile files. Overwriting and replacing tile files can ensure that the tile content of different platforms remains consistent, resolving conflicts caused by multiple platforms modifying the same tile.

[0110] In one embodiment, the method steps provided by the present invention further include:

[0111] A760: Receives modification and retention instructions sent by the user.

[0112] A770: Based on the modification retention instruction, the U modified tile files are overwritten into the spatial index database.

[0113] Specifically, a modification retention instruction is a user-sent command indicating whether to retain certain modifications. Instructions include retaining specified modifications, deleting specified modifications, etc. During data processing, users may request to retain certain modifications while ignoring others. The instruction is parsed to identify the tile IDs and modification content involved, ensuring that each modification instruction is accurately applied to the corresponding tile file.

[0114] Based on the user's modification retention instructions, modified tile files are selectively overwritten into the spatial index database. If the user's instructions require certain modifications to be retained, these modified tile files are officially written to the database; if the user's instructions require certain modifications to be deleted or not retained, these modifications are skipped or the original data is restored. This step ensures that users have precise control over modifications to the tile data, avoids unnecessary modifications being applied, and ensures that the database stores the final, confirmed tile data.

[0115] In summary, any of the methods or steps described above can be stored as computer instructions or programs in various types of computer memory, and the computer instructions or programs can be recognized by various types of computer processors to implement any of the above methods or steps.

[0116] Based on the above specific embodiments of the present invention, any improvements and modifications made to the present invention by those skilled in the art without departing from the principle of the present invention shall fall within the patent protection scope of the present invention.

Claims

1. A cross-platform collaborative loading method for CAD drawings, characterized in that, The method includes: The original CAD file is rasterized and layered on the central platform to obtain a multi-level raster tile set. The multi-level raster tile set is stored in the temporary storage unit of the central platform using a pyramid hierarchical index; After obtaining the updated view status by listening to user interaction events in real time, the view parameters are obtained by comparing the updated view status with the original CAD file; Using the view parameters as an index, extract the target raster tile set from the temporary storage unit; The target raster tile set is converted into a neutral exchange format set on the central platform. Based on the format support characteristics of the M edge platforms, schedule the container format conversion service to compatiblely distribute the neutral exchange format set to the M edge platforms; After receiving edge modification information, the central platform performs cross-platform tile file synchronous rendering and replacement based on the M role permissions of the M edge platforms. The method includes: The central platform receives edge modification information consisting of N edge modification compressed packages, wherein each edge modification compressed package includes a modified tile ID, a modified tile file, and a platform ID, and N≤M; Invoke the conflict priority rules of the M edge platforms; Based on ID consistency, modification conflict detection is performed on the edge modification information according to N modified tile IDs, and U edge modification compressed packages are selected from the edge modification information according to N platform IDs and conflict priority rules; Extract U modified tile files from the U edge-modified compressed package; After temporarily writing the U modified tile files into the neutral exchange format set based on the U modified tile IDs, cross-platform tile file overwriting and replacement are performed.

2. The cross-platform collaborative loading method for CAD drawings as described in claim 1, characterized in that, The original CAD file is rasterized and layered on a central platform to obtain a multi-level raster tile set. The method includes: Based on the scale characteristics, the original CAD file is divided into multi-level sub-drawing sets according to spatial index, wherein the multi-level sub-drawing sets are bound to multi-level fixed scales; Preset the baseline resolution, baseline raster size, and hierarchical parameter tuning rules; Starting from the reference resolution and reference raster scale, multi-level rasterization parameters are calculated according to the hierarchical parameter adjustment rules and multi-level fixed scales. The multi-level rasterization parameters include multi-level resolution and multi-level raster scale. The Mapnik toolchain is invoked to perform batch rasterization of the multi-level sub-drawing sets according to the multi-level rasterization parameters, and the multi-level raster tile set is output, wherein each raster tile is bound to a hierarchical affiliation, row and column coordinates, and tile spatial range.

3. The cross-platform collaborative loading method for CAD drawings as described in claim 2, characterized in that, The method of storing the multi-level raster tile set to the temporary storage unit of the central platform using a pyramid hierarchical index includes: A directory structure is established based on the multi-level fixed scale to obtain a standardized storage path, wherein the tile storage path rule of the standardized storage path is level-row number-column number; Extract the multi-level tile spatial range set from the multi-level grid tile set to establish a multi-level hierarchy-precision mapping set as a spatial index database; After updating the multi-level raster tile set to the spatial index database according to the standardized storage path, the spatial index database is migrated to the temporary storage unit.

4. The cross-platform collaborative loading method for CAD drawings as described in claim 1, characterized in that, Based on the format support characteristics of the M edge platforms, a container format conversion service is scheduled to distribute the neutral exchange format set compatiblely to the M edge platforms. The method includes: M format support lists are collected by interacting with the M edge platforms; With the goal of minimizing tasks, duplicate requirements are merged from the M format support lists to obtain the format support features, wherein the format support features include H format conversion tasks; After matching container instances based on the H format conversion tasks, and combining the conversion characteristics of the neutral exchange format set, the container instances are driven to convert and distribute the neutral exchange format set to the M edge platforms.

5. The cross-platform collaborative loading method for CAD drawings as described in claim 4, characterized in that, After matching container instances based on the H format conversion tasks, and combining the conversion characteristics of the neutral exchange format set, the container instances are driven to convert and distribute the neutral exchange format set to the M edge platforms. The method includes: Based on the H format conversion tasks, container instances are matched in the container cluster to obtain E format conversion containers, wherein the container cluster is pre-registered on the central platform; Based on the neutral exchange format set, the time consumption of the H format conversion tasks is predicted to obtain the conversion time of the H formats. Based on the real-time load of the container and the time taken for H format conversions, the H format conversion tasks are dynamically allocated to the E format conversion containers for compatibility conversion, and H compatible raster tile sets are output. Based on the attribution relationship between the H format conversion tasks and the M format support lists, the H compatible raster tile sets are distributed to the M edge platforms.

6. The cross-platform collaborative loading method for CAD drawings as described in claim 3, characterized in that, After obtaining the updated view state by real-time monitoring of user interaction events, the view parameters are obtained by comparing the updated view state with the original CAD file. The method includes: A dynamic listening mechanism is selected based on the device type of the central platform to capture user interaction events in order to extract the updated view state, wherein the updated view state includes viewport boundary coordinates and real-time zoom level. The real-time hierarchy is obtained by mapping the real-time zoom level to the multi-level fixed scale for hierarchical correction. After normalizing the viewport boundary coordinates to global boundary coordinates, the global boundary coordinates are projected onto the original CAD file for range validity comparison, and the valid viewport coordinates are output. The real-time level and the valid viewport coordinates constitute the view parameters.

7. The cross-platform collaborative loading method for CAD drawings as described in claim 6, characterized in that, The method for obtaining a real-time hierarchy by mapping the real-time zoom level to the multi-level fixed scale for hierarchy correction includes: Extract the multi-level scaling level set of the multi-level sub-drawing set; Based on the data distribution characteristics of the multi-level scaling level set, a multi-level initial scaling interval is constructed through clustering. The multi-level initial scaling intervals are connected and compensated to obtain multi-level standard scaling intervals; Map the real-time zoom level to the multi-level standard zoom range to locate the target zoom range; After mapping and binding the multi-level standard scaling range to the multi-level fixed scale, the real-time level is located in the directory structure according to the correspondence between the target scaling range and the multi-level fixed scale.

8. The cross-platform collaborative loading method for CAD drawings as described in claim 6, characterized in that, Using the view parameters as an index, the method for retrieving the target raster tile set from the temporary storage unit includes: The real-time hierarchy and effective viewport coordinates are combined into a structured query key; Based on the structured query key, the spatial index database is driven to perform a spatial range intersection query to filter and obtain a discrete raster tile set covering the effective viewport coordinates; The discrete grid tile set is spatially stitched together to obtain the target grid tile set.

9. The cross-platform collaborative loading method for CAD drawings as described in claim 3, characterized in that, The method further includes: Receive modification retention instructions sent by the user; According to the modification retention instruction, the U modified tile files are overwritten into the spatial index database.

Images