A batch capture method for structural elements of software interface

By using selectors in RPA robots to acquire structured data, the problems of the system software not providing APIs and the data structure not being obvious are solved. This enables efficient structured data acquisition and page-turning operations, generating intuitive data results.

CN114691265BActive Publication Date: 2024-11-15DATAGRAND TECH INC
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202210224918.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-03-07
Publication Date
2024-11-15
Estimated Expiration
2042-03-07

AI Technical Summary

Technical Problem

Existing technologies suffer from problems such as the lack of APIs, data pagination, and unclear data structures when acquiring structured data, resulting in low acquisition efficiency and a high risk of errors.

Method used

During the development process, three selectors are obtained, corresponding to the ancestor node, child node group, and target node group, respectively. During the execution process, the RPA robot uses these selectors to extract structured data and performs page-turning operations on different pages to generate a two-dimensional array of structured data.

Benefits of technology

It supports retrieving data from system software that does not provide an API, automatically flips through pages to retrieve paginated data, simplifies the understanding of data structures, improves retrieval efficiency, and generates intuitive data results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114691265B_ABST
    Figure CN114691265B_ABST
Patent Text Reader

Abstract

The present invention relates to a method for batch crawling of structured elements of a software interface, including a development process stage and an operation process stage, in which three selectors are obtained, and the three selectors correspond to an ancestor node, a child node group, and a target node in the child node group respectively; in the operation process stage, an RPA robot extracts structured data through three selectors, obtains a node group composed of target nodes, and the RPA robot obtains the attributes of the target node in the node group and generates a two-dimensional array of structured data according to the node group and the child node group. The present invention solves the problems encountered in the prior art when the RPA robot obtains structured data, that is, the system software does not provide an API, data paging, and the data structure is not obvious, simplifies the process of the RPA robot obtaining result data, and improves the efficiency of the RPA robot in obtaining result data.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of RPA robot process automation, and in particular to a method for batch crawling structured elements of a software interface. Background Art

[0002] Robotic Process Automation (RPA) is software that runs on personal computers or servers. Its main function is to simulate human control of various system software and automatically execute business processes. It is suitable for large-scale, highly repetitive business scenarios. One of the main functions of RPA robots is to obtain data on the system software interface, including structured data (structured data refers to data in a fixed format in a record file, such as tables, directories, etc.).

[0003] Currently, the main method for obtaining structured data is to call the Application Programming Interface (API) provided by the system software. This method has the following problems:

[0004] 1. It is invalid for system software that does not provide API;

[0005] 2. It is invalid for structured data displayed in pages;

[0006] 3. It is invalid for data with unclear structure.

[0007] In order to solve the above technical problems, the solution proposed in the prior art is to obtain each data in the structure separately, and then splice them together according to the discovered rules. However, this solution has the following problems:

[0008] 1. It is difficult to discover the patterns of structured data;

[0009] 2. When the amount of data is large, it takes time to obtain each data separately;

[0010] 3. Combining the acquired data according to specific rules is inefficient and prone to errors. Summary of the invention

[0011] In order to solve the problems encountered in the prior art when the RPA robot obtains structured data, such as the system software does not provide an API, data paging, and unclear data structure, simplify the process of the RPA robot obtaining result data, and improve the efficiency of the RPA robot in obtaining result data, the present invention provides a method for batch crawling structured elements of a software interface.

[0012] The technical objectives of the present invention are achieved through the following technical solutions:

[0013] A method for batch crawling structured elements of a software interface includes a development process stage and an operation process stage. In the development process stage, three selectors are obtained, and the three selectors correspond to an ancestor node, a child node group, and a target node in the child node group, respectively. In the operation process stage, an RPA robot extracts structured data through the three selectors to obtain a node group composed of target nodes. The RPA robot obtains the attributes of the target node in the node group and generates a two-dimensional array of structured data according to the node group and the child node group.

[0014] Furthermore, during the process operation phase, when the structured data is on different pages, the RPA robot turns the pages, obtains the structured data on each page, and stitches the obtained structured data together.

[0015] Furthermore, the selector saves the element attributes and hierarchical relationships in the target node, child node group, and ancestor node.

[0016] Further, the development process phase includes the following steps:

[0017] Step 1: Manually select two interface elements of the same control type in the structured data;

[0018] Step 2: Analyze the hierarchy and position relationship of the two interface elements with the help of the analysis module, find the common ancestor node of the two interface elements, and use the selector of the ancestor node as the first selector;

[0019] Step 3: Get all child node groups of the ancestor node, and use the selector for searching child node groups from the ancestor node as the second selector;

[0020] Step 4: Use the selector of the node with the same control type in the child node group as the third selector; if the two interface elements in step 1 are at the child node group level, the third selector is empty.

[0021] Furthermore, during the process run phase, the RPA robot extracts structured data through three selectors, including the following steps:

[0022] Step 1: The RPA robot finds the ancestor node based on the first selector;

[0023] Step 2: The RPA robot finds the child node group based on the second selector;

[0024] Step 3: Based on the second selector, the RPA robot searches for the corresponding target node under each child node of the child node group and forms a target node group.

[0025] Step 4: The RPA robot obtains the specified attributes of the target node and forms a two-dimensional array of structured data according to the target node group and the child node group.

[0026] Furthermore, the analysis module is a syntax analysis algorithm.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] 1. The method for batch crawling of structured elements of software interfaces of the present invention supports system software that does not provide an API for extracting structured data;

[0029] 2. Use the RPA robot to automatically turn pages to obtain structured data in paginated form;

[0030] 3. The method of the present invention supports the extraction of data with unclear structure;

[0031] 4. The method of the present invention is simple to operate and does not require understanding of the structural rules between data;

[0032] 5. The generated results are three element selectors, which intuitively show the result relationship of the data, making it easy to understand and modify;

[0033] 6. The method of the present invention greatly improves the efficiency of acquiring structured data. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic diagram of the element nodes of the system application interface in the present invention.

[0035] Figure 2 This is the organizational chart in Example 1.

[0036] Figure 3 This is the organizational chart in Example 2.

[0037] Figure 4 This is the organizational chart in Example 3. DETAILED DESCRIPTION

[0038] The technical solution of the present invention is further described below in conjunction with specific embodiments:

[0039] A method for batch crawling of structural elements of software interfaces, such as Figure 1-4 As shown in the figure, it includes the development process stage and the running process stage. A selector is a mode for selecting elements, which is stored and transmitted in the form of text, such as a CSS selector. The function of a selector is to locate interface elements. The selector stores some attributes of the target interface elements and the hierarchical relationship between these elements. According to the attributes in the selector, the RPA robot can locate one or more interface elements that meet the selector conditions.

[0040] During the development process, two interface elements in the structured data are captured and three selectors are obtained through analysis:

[0041] Step 1: Manually select two interface elements (referred to as node A and node B) in the structured data. If the control types (the control type is referred to as control type C) of the two interface elements are different, a prompt is given to reselect.

[0042] Step 2: Analyze the hierarchical and positional relationship of the two data through the analysis module. The analysis module uses a general syntax analysis algorithm to find the common ancestor node of the two elements (referred to as node G), and use the selector of node G as the first selector. The function of the syntax analysis algorithm is to analyze the two input selector texts and generate the output three selector texts according to specific rules. Through the analysis and processing of the text, the analysis and processing of the selection conditions are indirectly realized.

[0043] Step 3: Get all child node groups of node G (referred to as node group F), and use the selector for searching node group F from node G as the second selector;

[0044] Step 4: Use the selector that searches for a node of control type C from a node in node group F as the third selector; if node A and node B are at the level of node group F, that is, node A and node B directly belong to node G, the third selector is empty.

[0045] During the run process phase, the RPA robot extracts structured data based on three selectors:

[0046] Step 1: The RPA robot finds node G based on the first selector.

[0047] Step 2: The RPA robot finds node group F based on the second selector.

[0048] Step 3: Based on the second selector, the RPA robot starts from each node in node group F, searches for all target nodes, and forms a target node group (node ​​group C).

[0049] Step 4: The RPA robot obtains the specified attributes of the target node and forms a two-dimensional array of structured data according to node group C and node group F.

[0050] Step 5: If you need to turn the page, the RPA robot automatically clicks the page-turning button to turn the page. The new page obtains structured data from the first step and splices the previously obtained data together.

[0051] The following is explained in conjunction with specific embodiments:

[0052] Taking a company as an example, employees are regarded as elements, employee names and genders are regarded as element attributes, and the company's organizational structure corresponds to the hierarchical relationship of the elements.

[0053] Example 1

[0054] Male A is a member of the development team of the R&D department, and Male B is a member of the testing team of the R&D department;

[0055] Analyze the similarities between male A and male B and generate three selectors, which correspond to the R&D department, the team leader under the R&D department, and the male team members managed by each team leader.

[0056] Selector 1 - Select the R&D department;

[0057] Selector 2 - selects all group leaders;

[0058] Selector 3 - selects all male group members;

[0059] The RPA robot extracts structured data through three selectors and obtains a table in which the rows and columns correspond to the R&D department team leader and the male team members managed by the R&D department team leader.

[0060] Example 2

[0061] Male C is the development team leader of the R&D department, and female D is the test team leader of the R&D department. Analyze the common points between male C and female D to generate three selectors, which correspond to the R&D department, the team leader under the R&D department, and the team members managed by each team leader.

[0062] Selector 1 - Select the R&D department;

[0063] Selector 2 - selects all group leaders;

[0064] Selector 3 - selects the group leader itself, and selector 3 is empty;

[0065] The RPA robot extracts structured data through three selectors. For example, a single table can be obtained, including all the team leaders of the R&D department.

[0066] Example 3

[0067] Male E is the leader of the development team of the R&D department, and male F is a member of the development team of the R&D department. By analyzing the similarities between male E and male F, it is found that male E and male F have different positions relative to the public node R&D department and are at different levels. The generation of the selector fails and needs to be re-selected.

[0068] This embodiment is only a further explanation of the present invention and is not a limitation of the present invention. After reading this specification, those skilled in the art can make non-creative modifications to this embodiment as needed, but as long as it is within the scope of the claims of the present invention, it will be protected by the patent law.

Claims

1. A method for batch crawling of structural elements of software interfaces. It is characterized in that It includes a development process stage and an operation process stage. In the development process stage, three selectors are obtained. The three selectors correspond to the ancestor node, the child node group, and the target node in the child node group, respectively. The selectors save the element attributes and hierarchical relationships in the target node, the child node group, and the ancestor node. In the operation process stage, the RPA robot extracts structured data through the three selectors to obtain a node group composed of target nodes. The RPA robot obtains the attributes of the target node in the node group and generates a two-dimensional array of structured data according to the node group and the child node group. The development process phase includes the following steps: Step 1: Manually select two interface elements of the same control type in the structured data; Step 2: Analyze the hierarchy and position relationship of the two interface elements with the help of the analysis module, find the common ancestor node of the two interface elements, and use the selector of the ancestor node as the first selector; Step 3: Get all child node groups of the ancestor node, and use the selector for searching child node groups from the ancestor node as the second selector; Step 4: Use the selector of the node with the same control type in the child node group as the third selector; if the two interface elements in step 1 are at the child node group level, the third selector is empty; During the process run phase, the RPA robot extracts structured data through three selectors, including the following steps: Step 1: The RPA robot finds the ancestor node based on the first selector; Step 2: The RPA robot finds the child node group based on the second selector; Step 3: Based on the third selector, the RPA robot searches for the corresponding target node under each child node of the child node group and forms a target node group. Step 4: The RPA robot obtains the specified attributes of the target node and forms a two-dimensional array of structured data according to the target node group and the child node group.

2. A method for batch capturing structural elements of a software interface according to claim 1, It is characterized in that During the process operation phase, when the structured data is on different pages, the RPA robot turns the pages, obtains the structured data on each page, and stitches the obtained structured data together.

3. A method for batch capturing structural elements of a software interface according to claim 1, It is characterized in that The analysis module is a syntax analysis algorithm.

Citation Information

Patent Citations

  • Method and device for determining similar elements, computer equipment and storage medium

    CN111460259A

  • Systems and methods of activity target selection for robotic process automation

    CN113015956A

  • Universal interface element automatic detection system

    CN113176922A

  • Graphical element search technique selection, fuzzy logic selection of anchors and targets, and / or hierarchical graphical element identification for robotic process automation

    CN113811850A

  • Data model for machine data for semantic search

    US20140074887A1