An RPA process design system based on a VSCode plug-in

The RPA process design system based on the VS Code plugin solves the problems of interface clutter and collaboration in complex process design of existing RPA tools, realizes efficient process development and multi-person collaboration, and improves the system's adaptability and scalability.

CN120975738BActive Publication Date: 2026-04-17MYRON INTELLIGENT TECH (SHANGHAI) CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
MYRON INTELLIGENT TECH (SHANGHAI) CO LTD
Filing Date
2025-08-25
Publication Date
2026-04-17

AI Technical Summary

Technical Problem

Existing RPA tools suffer from problems such as cluttered graphical interfaces, difficulty in version control, conflicts in multi-person collaboration, and difficulty in embedding custom algorithms when designing complex processes, making it difficult to meet the needs of complex processes and collaboration.

Method used

The RPA workflow design system, based on the VS Code plugin, includes a text editing module, a graphical parameter configuration module, a project management module, and an execution adaptation module. Through algorithms such as format compliance verification, parameter validity verification, workflow difference comparison, and conflict detection, it achieves command completion, parameter configuration, version control, and script conversion.

Benefits of technology

It improves development efficiency and process accuracy, reduces maintenance costs, supports multi-person collaboration, enhances system adaptability and scalability, and enables non-professionals to participate in process automation design.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120975738B_ABST
    Figure CN120975738B_ABST
Patent Text Reader

Abstract

The application relates to an RPA process design system based on a VS Code plug-in and belongs to the technical field of process automation. The system comprises a text editing module, a graphical parameter configuration module, an engineering management module and an execution adaptation module. The text editing module loads preset instruction libraries and method signature libraries, receives structured instructions, performs format checking and parameter completion, analyzes and generates a memory instruction sequence list; the graphical parameter configuration module listens to the instruction selection state, generates a parameter form, verifies the validity of input parameters, generates instruction-parameter data and stores the data into a to-be-processed data pool; the engineering management module generates a standard directory structure, performs version control and conflict detection; and the execution adaptation module extracts data and converts the data into an executable script and performs verification. The system can efficiently realize the design, management and execution of an RPA process, and improves the convenience and accuracy of robot process automation design.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of process automation technology, specifically relating to an RPA process design system based on a VS Code plugin. Background Technology

[0002] Current mainstream RPA tools, such as UiPath and PowerAutomate, are mostly standalone desktop software that build processes through graphical drag-and-drop components, relying on Python or C# as their underlying language. While convenient for simple scenarios, these tools reveal shortcomings in complex processes: with numerous nodes, the graphical canvas becomes dense with connecting lines and overlapping nodes, leading to a chaotic structure and significantly increased maintenance costs; processes are stored in proprietary formats, making code-based viewing and editing impossible, version control difficult, and conflicts prone to occur during multi-user collaboration; the interface only supports preset component combinations, lacking text-based programming support, making it difficult to embed custom algorithms, and hindering debugging and refactoring; and they are disconnected from mainstream editors, with generated scripts often bound to specific languages, making adaptation to existing technology stacks difficult. These problems make traditional tools ill-suited for complex processes and collaboration needs, and the market urgently needs a solution that combines visual friendliness with code flexibility. Summary of the Invention

[0003] To address the aforementioned problems in the existing technology, this invention provides an RPA process design system based on the VS Code plugin;

[0004] The objective of this invention can be achieved through the following technical solutions:

[0005] An RPA workflow design system based on a VS Code plugin is characterized by comprising a text editing module, a graphical parameter configuration module, a project management module, and an execution adaptation module.

[0006] The text editing module loads a preset instruction library and method signature library, receives structured instructions input in namespace and method name format, compares the input content of the structured instructions with the templates in the instruction library through a format compliance verification algorithm, retrieves parameter options from the method signature library for completion based on the structured instruction prefix, and generates a memory instruction sequence list after parsing the structured instructions.

[0007] The graphical parameter configuration module monitors the command selection status in the text editing module, queries the corresponding parameter template based on the detected structured command, and generates a form on the side of the interface for input parameter items in the detected structured command in the corresponding parameter template; receives the input parameter items, performs format and logic checks on the input parameter items through a parameter validity verification algorithm, converts them into a data exchange format, binds them with the corresponding memory command sequence to generate command-parameter data, and stores them in the data pool to be processed;

[0008] The project management module generates a standard directory structure including process files, dependency libraries, and configuration files based on project creation; during version control, the process files before and after modification are compared based on the process difference intelligent comparison algorithm, and the changed instruction lines are marked; when users collaborate, the process conflict detection mechanism determines the correlation between the received remote modification instructions and local instructions, and generates conflict instruction prompts.

[0009] The execution adaptation module extracts the instruction-parameter data from the data pool to be processed, converts it into an executable script based on the syntax rules of the target language, and performs syntax verification on the executable script to generate a legal script file.

[0010] Specifically, when comparing the structured instructions with the instruction library template, the format compliance verification algorithm splits the input content and template according to the namespace and method name hierarchy, splits the namespace into three-level nodes, and the method name as the leaf node; calculates the matching degree level by level, compares the character consistency of each level node, the matching degree of a single node is the ratio of the number of identical characters to the total number of characters, and the total matching degree is the weighted average of the matching degrees of each level node.

[0011] Specifically, before retrieving the parameter options from the method signature library, the parameter derivation algorithm first parses the namespace and method name in the structured instruction prefix, locates the corresponding class in the method signature library through the namespace, and then finds the method signature under the class based on the method name; extracts the parameter list from the method signature as completion options, and displays the completion options in the order of parameter definition in the method signature library.

[0012] Specifically, when the parameter validity verification algorithm checks the format of the input parameter item, it identifies the data type of the parameter item through the parameter template; and calls the format verification rules corresponding to the data type. The format verification rules include: validating illegal characters for string type, validating pure numbers within the valid range for integer type, and validating corresponding values ​​or abbreviations for boolean type.

[0013] Specifically, when performing logical checks, the parameter validity verification algorithm initiates a scanning mechanism to iterate through all other input parameter items in the data pool to be processed. During the scanning process, logical association rules between parameters are established based on the inherent relationships and business logic between the parameters. These logical association rules are stored in a structured form. After the logical association rules are established, the current parameter item is compared with each of the logical association rules. Based on the logical reasoning engine, the dependency relationships and value range constraints between each parameter are analyzed to generate a logical conflict judgment.

[0014] Specifically, when the process difference intelligent comparison algorithm compares the process files, it first parses the process files before and after modification into a sequence of instructions including instruction ID, namespace, method name, and parameter set; it then locates the added instruction, deleted instruction, and modified instruction by matching the instruction ID; for the modified instruction, it compares each field of the namespace, method name, and parameter set for each parameter value, records the specific differences, and generates a difference report which is stored in the version log of the project management module.

[0015] Specifically, when the process conflict detection mechanism performs correlation judgment, it extracts the instruction identifier and parameter association relationship between the remote modification instruction and the local instruction; if the two instruction identifiers are the same and the version numbers are consistent, or the parameter association relationship points to the same operation object and the modification content is different, a conflict instruction prompt is generated, along with a comparison table of modification details between the remote and local instructions.

[0016] Specifically, when the execution adaptation module converts the target language's syntax rules into the executable script, it first loads the syntax mapping rule library corresponding to the target language; it then traverses the memory instruction sequence, converting the namespace of each instruction into an import statement in the target language, converting the method name into a function call, and converting the parameter items into literals in the target language according to their data types.

[0017] Specifically, when the execution adaptation module performs syntax verification on the executable script, it calls the syntax checker of the target language. The syntax checker outputs results including the error location and error type. The execution adaptation module maps the error location back to the corresponding instruction index in the memory instruction sequence, highlights it in yellow in the text editing module, and displays the relevant information in the error panel.

[0018] Specifically, during the comparison process, the format compliance verification algorithm records the types of differences between the input content and the instruction library template: namespace errors include missing nodes and incorrect node names; method name errors include misspellings and case mismatches; missing format symbols include missing namespace separators and parentheses after the method name.

[0019] Specifically, when the execution adaptation module detects a script syntax error, it initiates an error tracing algorithm to extract the line number and code fragment from the syntax error. It then uses the mapping table of the execution adaptation module to find the associated memory instruction index. Based on the memory instruction index, it retrieves the namespace and method name of the original structured instruction from the memory instruction sequence, queries the instruction location table of the text editing module, and locates the input line number of the instruction in the text editing area. Through the instruction-parameter binding relationship, it finds the parameter input item ID associated with the instruction in the graphical parameter configuration module and retrieves the position record of the parameter form. Finally, it generates a tracing path according to the hierarchy of script error line, memory instruction index, original instruction line number, and parameter input item ID.

[0020] Specifically, the execution adaptation module initiates the process based on the process dynamic priority scheduling algorithm; constructs a state space, which includes the number of remaining instructions that have not yet been executed, the intensity of resource demand, and the urgency of the deadline; based on the constructed state space, the process dynamic priority scheduling algorithm continuously optimizes the action strategy: monitors the state parameters of the process execution environment, including CPU utilization, memory usage, and task queue length, and generates the current state space in combination with the state space; based on the current state space, the process dynamic priority scheduling algorithm continuously iteratively updates the reward feedback value table to evaluate the expected benefits of different actions, dynamically adjusts the CPU time slice allocation ratio of each process, and constructs a state-action-reward triplet model.

[0021] The beneficial effects of this invention are as follows:

[0022] In terms of development efficiency, the text editing module's format validation and parameter completion functions reduce input errors and repetitive operations; the graphical parameter configuration module enables visual parameter configuration, and combined with validity verification algorithms, improves setting efficiency; the project management module's standardized management and difference comparison functions reduce version management costs. Regarding process accuracy, the format compliance verification algorithm accurately identifies instruction errors, parameter validity verification ensures parameter correctness from both format and logic perspectives, and the execution adaptation module's syntax validation and error tracing functions further enhance process accuracy. In team collaboration, the process conflict detection mechanism avoids collaboration conflicts, and version control and difference comparison functions clearly record modification paths, ensuring smooth and efficient collaboration. The system is highly adaptable and scalable; the execution adaptation module supports multi-language script conversion, the dynamic priority scheduling algorithm improves adaptability to complex environments, and the modular design facilitates functional expansion. Furthermore, the system lowers the barrier to entry for robotic process automation design, enabling non-professionals to participate in development and promoting the popularization of robotic process automation technology. Attached Figure Description

[0023] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0024] Figure 1 This is a structural diagram of an RPA workflow design system based on a VS Code plugin according to the present invention; Detailed Implementation

[0025] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0026] Please see Figure 1 An RPA process design system based on a VS Code plugin, characterized in that it includes a text editing module, a graphical parameter configuration module, a project management module, and an execution adaptation module;

[0027] The text editing module loads a preset instruction library and method signature library, receives structured instructions input in namespace and method name format, compares the input content of the structured instructions with the templates in the instruction library through a format compliance verification algorithm, retrieves parameter options from the method signature library to complete the instructions based on the structured instruction prefix, and generates a memory instruction sequence list after parsing the structured instructions.

[0028] The graphical parameter configuration module monitors the command selection status in the text editing module, queries the corresponding parameter template based on the detected structured command, and generates a form on the side of the interface for input parameter items in the detected structured command in the corresponding parameter template; receives the input parameter items, performs format and logic checks on the input parameter items through a parameter validity verification algorithm, converts them into a data exchange format, binds them with the corresponding memory command sequence to generate command-parameter data, and stores them in the data pool to be processed;

[0029] The project management module generates a standard directory structure including process files, dependency libraries, and configuration files based on project creation; during version control, the process files before and after modification are compared based on the process difference intelligent comparison algorithm, and the changed instruction lines are marked; when users collaborate, the process conflict detection mechanism determines the correlation between the received remote modification instructions and local instructions, and generates conflict instruction prompts.

[0030] The execution adaptation module extracts the instruction-parameter data from the data pool to be processed, converts it into an executable script based on the syntax rules of the target language, and performs syntax verification on the executable script to generate a legal script file.

[0031] Specifically, when comparing the structured instructions with the instruction library template, the format compliance verification algorithm splits the input content and template according to the namespace and method name hierarchy, splits the namespace into three-level nodes, and the method name as the leaf node; calculates the matching degree level by level, compares the character consistency of each level node, the matching degree of a single node is the ratio of the number of identical characters to the total number of characters, and the total matching degree is the weighted average of the matching degrees of each level node.

[0032] Specifically, before retrieving the parameter options from the method signature library, the parameter derivation algorithm first parses the namespace and method name in the structured instruction prefix, locates the corresponding class in the method signature library through the namespace, and then finds the method signature under the class based on the method name; extracts the parameter list from the method signature as completion options, and displays the completion options in the order of parameter definition in the method signature library.

[0033] Specifically, when the parameter validity verification algorithm checks the format of the input parameter item, it identifies the data type of the parameter item through the parameter template; and calls the format verification rules corresponding to the data type. The format verification rules include: validating illegal characters for string type, validating pure numbers within the valid range for integer type, and validating corresponding values ​​or abbreviations for boolean type.

[0034] Specifically, when performing logical checks, the parameter validity verification algorithm initiates a scanning mechanism to iterate through all other input parameter items in the data pool to be processed. During the scanning process, logical association rules between parameters are established based on the inherent relationships and business logic between the parameters. These logical association rules are stored in a structured form. After the logical association rules are established, the current parameter item is compared with each of the logical association rules. Based on the logical reasoning engine, the dependency relationships and value range constraints between each parameter are analyzed to generate a logical conflict judgment.

[0035] When the input parameter is submitted to the parameter configuration panel, the algorithm first calls the parameter template bound to the cross-system adaptation and process creation module to extract the preset data type of the parameter to achieve accurate identification. Then, it calls the format validation rules corresponding to the data type to perform checks. String types are scanned character by character to check for illegal characters defined in the parameter template; integer types are verified to be pure numbers within the specified valid range; and boolean types are checked to see if they are preset legal values ​​or their abbreviations. If they do not conform to the corresponding rules, they are marked as format abnormal. During the logic check, the algorithm automatically starts a scanning mechanism to comprehensively traverse other input parameter items in the pending data pool of the dynamic parameter configuration module, collecting relevant information. Based on the inherent relationships and business logic between parameters, the algorithm constructs structured logical association rules in key-value pairs stored in the rule base through a rule engine. These rules include dependency relationships between parameters and value range constraints. After the rules are established, the value of the current parameter item is compared with the logical association rules in the rule base item by item. The logical reasoning engine analyzes the dependency relationships and value range constraints between each parameter. If conflicts exist, a logical conflict judgment is generated that clearly involves the parameter item and its specific content. Through the coordinated execution of format checks and logical checks, the algorithm comprehensively verifies the validity of the parameter items and feeds back the verification results to the dynamic parameter configuration module in real time. If an anomaly is found, the parameter configuration results are prevented from being synchronized to the .rpa file until the parameter item is corrected to meet the requirements.

[0036] Specifically, when the process difference intelligent comparison algorithm compares the process files, it first parses the process files before and after modification into a sequence of instructions including instruction ID, namespace, method name, and parameter set; it then locates the added instruction, deleted instruction, and modified instruction by matching the instruction ID; for the modified instruction, it compares each field of the namespace, method name, and parameter set for each parameter value, records the specific differences, and generates a difference report which is stored in the version log of the project management module.

[0037] Specifically, when the process conflict detection mechanism performs correlation judgment, it extracts the instruction identifier and parameter association relationship between the remote modification instruction and the local instruction; if the two instruction identifiers are the same and the version numbers are consistent, or the parameter association relationship points to the same operation object and the modification content is different, a conflict instruction prompt is generated, along with a comparison table of modification details between the remote and local instructions.

[0038] Specifically, when the execution adaptation module converts the target language's syntax rules into the executable script, it first loads the syntax mapping rule library corresponding to the target language; it then traverses the memory instruction sequence, converting the namespace of each instruction into an import statement in the target language, converting the method name into a function call, and converting the parameter items into literals in the target language according to their data types.

[0039] Specifically, when the execution adaptation module performs syntax verification on the executable script, it calls the syntax checker of the target language. The syntax checker outputs results including the error location and error type. The execution adaptation module maps the error location back to the corresponding instruction index in the memory instruction sequence, highlights it in yellow in the text editing module, and displays the relevant information in the error panel.

[0040] Specifically, during the comparison process, the format compliance verification algorithm records the types of differences between the input content and the instruction library template: namespace errors include missing nodes and incorrect node names; method name errors include misspellings and case mismatches; missing format symbols include missing namespace separators and parentheses after the method name.

[0041] Specifically, when the execution adaptation module detects a script syntax error, it initiates an error tracing algorithm to extract the line number and code fragment from the syntax error. It then uses the mapping table of the execution adaptation module to find the associated memory instruction index. Based on the memory instruction index, it retrieves the namespace and method name of the original structured instruction from the memory instruction sequence, queries the instruction location table of the text editing module, and locates the input line number of the instruction in the text editing area. Through the instruction-parameter binding relationship, it finds the parameter input item ID associated with the instruction in the graphical parameter configuration module and retrieves the position record of the parameter form. Finally, it generates a tracing path according to the hierarchy of script error line, memory instruction index, original instruction line number, and parameter input item ID.

[0042] Specifically, the execution adaptation module initiates the process based on the process dynamic priority scheduling algorithm; constructs a state space, which includes the number of remaining instructions that have not yet been executed, the intensity of resource demand, and the urgency of the deadline; based on the constructed state space, the process dynamic priority scheduling algorithm continuously optimizes the action strategy: monitors the state parameters of the process execution environment, including CPU utilization, memory usage, and task queue length, and generates the current state space in combination with the state space; based on the current state space, the process dynamic priority scheduling algorithm continuously iteratively updates the reward feedback value table to evaluate the expected benefits of different actions, dynamically adjusts the CPU time slice allocation ratio of each process, and constructs a state-action-reward triplet model.

[0043] In this embodiment, a specific implementation of an RPA workflow design system based on a VS Code plugin is as follows:

[0044] An e-commerce platform processes over 5,000 orders daily. Manual processing involves several steps: order data retrieval (exporting to Excel from the platform backend), information verification (checking the validity of shipping addresses and contact information), inventory updates (deducting the corresponding product quantities from the ERP system), and logistics order generation (creating waybills via a courier API). Each order takes an average of 5 minutes to process, and customer complaints due to manual data entry errors reach 3%. A system designed based on the aforementioned RPA process can build a fully automated processing chain, solving both efficiency and accuracy issues.

[0045] Implementation process of each module of the system

[0046] Text editing module application

[0047] Command library loading: After the system starts, the text editing module automatically loads the e-commerce command library, where the "order" namespace contains three levels of nodes (first level "e-commerce business", second level "order management", and third level "data operation"), and associated method names include "get" (query), "verify" (verification), etc.; the signature of "order.get" in the method signature library is "order.get(startTime:str,endTime:str,status:enum{paid / unpaid / canceled})".

[0048] Command Input and Validation: When operators enter "order.get" in the text editing area, the format compliance validation algorithm breaks down the input content in real time.

[0049] The namespace "order" was compared with the template's three-level nodes. The character matching degrees of the "e-commerce business", "order management" and "data operation" nodes were 100%, 100% and 92% respectively ("order" and "data operation" have abbreviation mapping). The weighted average (weight ratio 3:3:4) reached 96.8%, and the format was judged to be compliant.

[0050] The method name "get" matches the leaf node exactly, triggering command completion. A list of parameter options pops up below the editing area, arranged in the order of "startTime", "endTime", and "status", with each option labeled with its type (e.g., "startTime: date string, e.g., 2024-01-01").

[0051] Memory instruction generation: After the operator selects the parameter options, the system automatically completes "order.get (startTime=?,endTime=?,status=?)", and after parsing, generates a list of memory instruction sequences (numbered INS-001), which includes metadata such as instruction type (query), priority (medium), and parameter placeholders.

[0052] Graphical parameter configuration module application

[0053] Parameter panel trigger: When the operator positions the cursor on the "order.get" command line, a parameter configuration panel automatically pops up on the right. The top of the panel displays the associated command ID (INS-001), and the bottom shows the form in the "Basic Parameters" tab.

[0054] The “startTime” input box has a calendar selector next to it, the “endTime” input box is bound to a linked validation (default is greater than startTime), and “status” is a drop-down selection box (options “paid”, “unpaid”, and “canceled”).

[0055] Parameter input and verification:

[0056] After the operators select "startTime=2024-01-01", "endTime=2024-01-31", and "status=paid", the parameter validity verification algorithm will be activated.

[0057] Format check: A string scan was performed on “startTime”, and “2024-01-01” was detected. It contains a “-” separator and conforms to the “YYYY-MM-DD” format, with no illegal characters (such as “*” or “&”). “status” is “paid”, which matches the enumerated value exactly, so the format check passed.

[0058] Logical check: Scan the data pool to be processed (currently containing the pre-command INS-000: "system.login(platform='e-commerce backend')"). Establish a rule base based on business logic: "endTime must be later than startTime" and "query time span does not exceed 90 days". The algorithm compares the input value with the rules; "2024-01-31-2024-01-01=30 days" meets the constraints, generating a "logically conflict-free" result.

[0059] Data storage: The parameter configuration result is converted into JSON format ({"startTime":"2024-01-01","endTime":"2024-01-31","status":"paid"}), and then stored in the pending data pool after being bound to INS-001. The records in the pool display "Command ID: INS-001, Parameter Status: Confirmed".

[0060] Project Management Module Application

[0061] Catalog Generation: After clicking "New Project," the project management module generates a standard catalog.

[0062] / orderProcess (root directory)

[0063] / src / orderProcess.rpa (process file)

[0064] / dependencies / (dependencies, automatically download requests==2.31.0, pandas==2.1.4)

[0065] / config / environment.json (Configuration file, containing e-commerce backend API address and ERP system port)

[0066] Version control and collaboration:

[0067] Operations staff A modified "inventory.reduce(sku='EC2024001',quantity=5)" to "quantity=10" and submitted the changes. The intelligent comparison algorithm for process differences analyzed the orderProcess.rpa files before and after the modification.

[0068] The instruction sequence was analyzed, and the modification instruction was located through instruction ID=INS-005. A field-by-field comparison revealed that the "quantity" value changed from 5 to 10, while other fields (SKU, namespace) remained unchanged. A difference report was generated: "INS-005: inventory.reduce parameter quantity changed from 5→10, modified by: A, time: 2024-02-01 10:15".

[0069] Operations personnel B simultaneously modified the command "quantity=8" and retrieved the remote version, triggering the process conflict detection mechanism:

[0070] Extract the identifier (ID=INS-005) and parameter association (both pointing to the inventory deduction of sku=EC2024001) of the remote instruction (quantity=10) and the local instruction (quantity=8), and determine that they are in conflict.

[0071] The command line in the editing area is marked with a double red underline, and a conflict panel pops up: the left side shows the remote modification "quantity=10 (A, 10:15)", and the right side shows the local modification "quantity=8 (B, 10:18)". It provides the options of "adopt remote", "keep local" and "manual input". After selecting "manual input 12" for B, the system automatically records the merge result to the version log.

[0072] Execute the adaptation module application

[0073] Script conversion: The adaptation module extracts the instruction-parameter data of INS-001 from the data pool to be processed, and loads the Python syntax mapping rule library.

[0074] The namespace "order" is mapped to "import ecommerce.order as order".

[0075] The method name "get" is mapped to "order.query()".

[0076] The parameter "startTime" is converted to "start_time='2024-01-01'" (automatically adding underscores to conform to Python naming conventions).

[0077] Generate an executable script:

[0078] import ecommerce.order as order

[0079] order.query(start_time='2024-01-01',

[0080] end_time='2024-01-31',status='paid')

[0081] Syntax validation and error handling:

[0082] If an operator mistakenly enters "endTime='2024-01-31'" as the parameter, the Python syntax checker will output the error: "Line 2, syntax error: keyword argument 'endTime' is not defined, it should be 'end_time'".

[0083] The execution adapter module locates line 15 in the text editing area through the mapping table (script line number → memory instruction index → ​​text editing line number). This line is highlighted in yellow, and the error panel on the right displays: "Associated instruction INS-001, parameter 'endTime' is not named according to the specification, it is recommended to change it to 'end_time'". Clicking the prompt will automatically correct it.

[0084] System performance

[0085] After system deployment, the order processing workflow will be executed automatically at 9:00 AM daily.

[0086] Efficiency: The processing time for a single order has been reduced from 5 minutes to 10 seconds, and the average daily processing volume has increased to 10,000 orders (including automatic operation at night).

[0087] Accuracy: Issues such as address verification errors (e.g., insufficient number of digits in the mobile phone number) and overselling of inventory have been completely eliminated, and the customer complaint rate has dropped from 3% to 0.1%.

[0088] Collaboration: Supports 5 operations staff to maintain processes simultaneously, reducing the average time for conflict resolution from 15 minutes to 6 minutes, and version rollback response time to <30 seconds.

[0089] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. An RPA workflow design system based on a VS Code plugin, characterized in that, It includes a text editing module, a graphical parameter configuration module, a project management module, and an execution adaptation module; The text editing module loads a preset instruction library and method signature library, receives structured instructions input in namespace and method name format, and compares the input content of the structured instructions with the templates in the instruction library through a format compliance verification algorithm. The process is as follows: the input content and template are split according to the hierarchy of namespace and method name, the namespace is split into three-level nodes, and the method name is used as a leaf node. The matching degree is calculated level by level, and the character consistency of each node is compared. The matching degree of a single node is the ratio of the number of identical characters to the total number of characters, and the total matching degree is the weighted average of the matching degrees of each level of nodes. According to the structured instruction prefix, the parameter options are retrieved from the method signature library to complete the instruction. The structured instruction is parsed to generate a list of memory instruction sequences. The graphical parameter configuration module monitors the command selection status in the text editing module, queries the corresponding parameter template based on the detected structured command, and generates a form on the side of the interface for input parameter items in the detected structured command in the corresponding parameter template; receives the input parameter items, performs format and logic checks on the input parameter items through a parameter validity verification algorithm, converts them into a data exchange format, binds them with the corresponding memory command sequence to generate command-parameter data, and stores them in the data pool to be processed; The format and logic checks are performed by identifying the data type of the parameter item through the parameter template and calling the format validation rules corresponding to the data type to perform format checks. The scanning mechanism is initiated to traverse other input parameter items in the data pool to be processed. Based on the inherent relationship between parameters and business logic, structured logical association rules are established. The current parameter item is compared with the logical association rules item by item. The logical reasoning engine analyzes the dependency relationship and value range constraints between parameters and generates logical conflict judgments. The project management module generates a standard directory structure based on project creation, including process files, dependency libraries, and configuration files. During version control, a process difference intelligent comparison algorithm is used to compare the process files before and after modification. The process is as follows: First, the process files before and after modification are parsed into a sequence of instructions including instruction ID, namespace, method name, and parameter set; the instruction ID is used to match and locate new instructions, deletion instructions, and modification instructions; for the modification instructions, each field is compared with the namespace, method name, and each parameter value in the parameter set, the specific differences are recorded, and a difference report is generated; the changed instruction lines are marked; when users collaborate, the process conflict detection mechanism performs correlation judgment between the received remote modification instructions and local instructions, and generates conflict instruction prompts. The execution adaptation module extracts the instruction-parameter data from the data pool to be processed, converts it into an executable script based on the syntax rules of the target language, and performs syntax verification on the executable script to generate a legal script file. When the execution adaptation module detects a script syntax error, it initiates an error tracing algorithm: extracting the line number and code fragment from the syntax error, finding the associated memory instruction index through a mapping table; obtaining the namespace and method name of the original structured instruction from the memory instruction sequence based on the memory instruction index, querying the instruction location table to locate the input line number of the instruction in the text editing area; finding the parameter input item ID associated with the instruction in the graphical parameter configuration module through the instruction-parameter binding relationship, and retrieving the position record of the parameter form; and generating a tracing path according to the hierarchy of script error line, memory instruction index, original instruction line number, and parameter input item ID.

2. The system according to claim 1, characterized in that, When comparing the structured instructions with the instruction library template, the format compliance verification algorithm splits the input content and template according to the hierarchy of namespace and method name, splits the namespace into three-level nodes, and the method name as the leaf node; The matching degree is calculated level by level, and the character consistency of each node is compared. The matching degree of a single node is the ratio of the number of identical characters to the total number of characters, and the total matching degree is the weighted average of the matching degrees of each level of nodes.

3. The system according to claim 1, characterized in that, When retrieving parameter options from the method signature library for instruction completion, a parameter deduction algorithm is executed: first, the namespace and method name in the structured instruction prefix are parsed; the corresponding class is located in the method signature library through the namespace; then, the method signature under the class is found based on the method name; the parameter list in the method signature is extracted as completion options, and the completion options are displayed in the order in which the parameters are defined in the method signature library.

4. The system according to claim 1, characterized in that, When the parameter validity verification algorithm checks the format of the input parameter item, it identifies the data type of the parameter item through the parameter template; Invoke the format validation rules corresponding to the data type. The format validation rules include: validating illegal characters for string types, validating pure numbers within the valid range for integer types, and validating corresponding values ​​or abbreviations for boolean types.

5. The system according to claim 1, characterized in that, When performing logical checks, the parameter validity verification algorithm will activate a scanning mechanism to iterate through all other input parameter items in the data pool to be processed. During the scanning process, based on the inherent relationship between parameters and business logic, logical association rules between parameters are established, and the logical association rules are stored in a structured form. After establishing the logical association rules, the current parameter item is compared with the logical association rules item by item. Based on the logical reasoning engine, the dependency relationship and value range constraints between the parameters are analyzed to generate a logical conflict judgment.

6. The system according to claim 1, characterized in that, When the process difference intelligent comparison algorithm compares the process files, it first parses the process files before and after modification into a sequence of instructions including instruction ID, namespace, method name, and parameter set; it then locates the added instruction, deleted instruction, and modified instruction by matching the instruction ID; for the modified instruction, it compares each field of the namespace, method name, and parameter set for each parameter value, records the specific differences, and generates a difference report which is stored in the version log of the project management module.

7. The system according to claim 1, characterized in that, When the process conflict detection mechanism performs correlation judgment, it extracts the instruction identifier and parameter association relationship between the remote modification instruction and the local instruction. If the two instruction identifiers are the same and the version numbers are consistent, or the parameter association relationship points to the same operation object but the modification content is different, a conflict instruction prompt is generated, along with a comparison table of modification details between the remote and local instructions.

8. The system according to claim 1, characterized in that, When the execution adaptation module converts the target language's syntax rules into the executable script, it first loads the syntax mapping rule library corresponding to the target language; it then traverses the memory instruction sequence, converting the namespace of each instruction into an import statement in the target language, converting the method name into a function call, and converting the parameter items into literals in the target language according to their data types.

9. The system according to claim 1, characterized in that, When the execution adaptation module performs syntax verification on the executable script, it calls the syntax checker of the target language. The syntax checker outputs results including the error location and error type. The execution adaptation module maps the error location back to the corresponding instruction index in the memory instruction sequence, highlights it in yellow in the text editing module, and displays the relevant information in the error panel.

10. The system according to claim 1, characterized in that, During the comparison process, the format compliance verification algorithm records the types of differences between the input content and the instruction library template: namespace errors include missing nodes and incorrect node names; method name errors include misspellings and case mismatches; missing format symbols include missing namespace separators and parentheses after the method name.

11. The system according to claim 1, characterized in that, When the execution adaptation module detects a script syntax error, it initiates an error tracing algorithm to extract the line number and code snippet from the syntax error. It then uses the mapping table of the execution adaptation module to find the associated memory instruction index. Based on the memory instruction index, it retrieves the namespace and method name of the original structured instruction from the memory instruction sequence, queries the instruction location table of the text editing module, and locates the input line number of the instruction in the text editing area. Through the instruction-parameter binding relationship, it finds the parameter input item ID associated with the instruction in the graphical parameter configuration module and retrieves the position record of the parameter form. Finally, it generates a tracing path according to the hierarchy of script error line, memory instruction index, original instruction line number, and parameter input item ID.

12. The system according to claim 1, characterized in that, The execution adaptation module initiates the process based on the process dynamic priority scheduling algorithm; it constructs a state space, which includes the number of remaining instructions that have not yet been executed, the intensity of resource demand, and the urgency of the deadline; based on the constructed state space, the process dynamic priority scheduling algorithm continuously optimizes the action strategy: it monitors the state parameters of the process execution environment, including CPU utilization, memory usage, and task queue length, and generates the current state space in combination with the state space; based on the current state space, the process dynamic priority scheduling algorithm continuously iteratively updates the reward feedback value table to evaluate the expected benefits of different actions, dynamically adjusts the CPU time slice allocation ratio of each process, and constructs a state-action-reward triplet model.

Citation Information

Patent Citations

  • Robot operation system based on RPA editor process control

    CN114625447A

  • Item-rule code generation method and apparatus combining RPA and ai, and electronic device

    WO2023226129A1