Web-based automation collection and large model constraint generation manual generation system and method

Abstract

The application discloses a kind of operation manual generation system and method based on Web automation collection and large model constraint generation, it is related to software automation and artificial intelligence technical field.System includes system information input module, Web automation collection module, structured data processing module, large model constraint generation module, template rendering output module and version incremental update module.System receives the access information input by user, by script non-invasive type simulation login and traverses target system, recursive analysis menu tree, positioning operation element and automatically screenshot;It is encapsulated as the structured dataset of uniform format to collect data;Load large model with strong constraint prompt word to generate standard operation manual text;Finally, according to pre-set template, graphic text filling and rendering output are carried out.Incremental update is only carried out to change part when system version iteration is carried out.The application realizes the full-process non-human intervention automatic generation of operation manual, and has high consistency and strong expansibility.

Description

Technical Field

[0001] This invention belongs to the fields of software automation, web data acquisition, artificial intelligence natural language processing, and document automation generation technology, specifically relating to an operation manual generation system and method based on web automated acquisition and large model constraint generation. Background Technology

[0002] Currently, various back-end management systems, business systems, operation and maintenance platforms, and government systems are widely deployed in enterprises and institutions. Furthermore, with changing business needs, system functions are being iterated and updated more frequently. As essential technical documents for software delivery, system operation and maintenance, and daily operations, user manuals must maintain a high degree of consistency with the actual interface, functional modules, and operating procedures of the web system.

[0003] However, existing methods for generating and maintaining web system operation manuals generally rely on manual writing or simple document editing tools, which have the following serious technical shortcomings in practical applications: 1. The data collection efficiency is extremely low, and inconsistencies between the documents and the system's images and text are prone to occur.

[0004] Current technologies lack automated front-end parsing mechanisms for web systems. When writing manual documentation, it relies on manual login, manual page-by-page screenshots, and word-by-word writing of operation steps. Due to omissions and subjective negligence in manual operations, it is easy to miss hidden functional modules or incorrectly record complex operation paths, resulting in the generated manual content being seriously outdated or deviating from the actual system interface.

[0005] 2. It lacks non-intrusive structured data acquisition capabilities, making it impossible to provide a reliable data source for automated document generation.

[0006] Traditional data collection tools typically require the target system to expose specific API interfaces, modify front-end code, or directly read the underlying database. For many deployed, closed web systems that do not reserve data interfaces, existing technologies cannot automatically and recursively obtain key structured information such as menu trees, page routes, function buttons, and DOM structure without intruding on or modifying the target system. This results in a lack of standardized data input in the subsequent document generation process.

[0007] 3. When generating documents directly from a general large model, the output format is uncontrollable and the content logic is chaotic.

[0008] Large general models (LLMs) are based on probabilistic generation mechanisms. Without strict contextual structure constraints and fixed output rules, they are prone to producing "illusions" or arbitrary formatting logic, resulting in generated documents with problems such as missing chapters, non-standard step expressions, and inaccurate matching of text and images, which cannot directly meet the delivery requirements of standard documents.

[0009] 4. Version updates are delayed, and maintenance costs are extremely high.

[0010] Current technology lacks an automated mechanism for identifying and comparing changes to web system functionality. When a web system undergoes version iteration, it cannot automatically locate specific changed pages or elements, requiring manual verification, screenshotting, and document rewriting, which easily leads to delays in document updates and significant duplication of effort.

[0011] 5. Poor document formatting and lack of automated format standardization.

[0012] Existing document generation tools can only provide simple text filling or static template editing. They cannot automatically map dynamically collected structured data, multi-level menu numbers, interface screenshots, and generated text to standard typesetting styles. Typesetting, table of contents generation, and the combination of text and images still heavily rely on manual secondary processing. Summary of the Invention

[0013] The following provides a brief overview of one or more aspects to offer a basic understanding of them. This overview is not an exhaustive summary of all conceived aspects, nor is it intended to identify key or decisive elements of all aspects, nor to define the scope of any or all aspects. Its sole purpose is to present some concepts of one or more aspects in a simplified form to prepare for the more detailed descriptions that follow.

[0014] The purpose of this invention is to provide a system and method for generating operation manuals based on Web-based automated data collection and large model constraint generation. This system solves the technical problems of low efficiency, inconsistent graphics and text, non-standard format, delayed updates, and inability to automatically generate operation manuals for Web systems. It enables the generation of complete, standardized operation manuals that are completely consistent with the system with just one click by inputting the system URL, account, and password, without the need for manual involvement in writing, screenshotting, typesetting, and proofreading.

[0015] The technical solution of this invention is as follows: This invention discloses an operation manual generation system based on Web automated data acquisition and large model constraint generation, comprising: a system information input module, a Web automated data acquisition module, a structured data processing module, a large model constraint generation module, and a template rendering output module, wherein each module is sequentially connected in communication; wherein: The system information input module is used to receive and verify the information input by the user to form standardized startup parameters; The Web automated data collection module is used to simulate logging into the target system non-intrusively through scripts based on the standardized startup parameters, recursively parse the menu tree, traverse page routes and function buttons, automatically locate page operation elements and complete interface screenshots, and bind screenshots to corresponding function nodes. The structured data processing module is used to encapsulate the collected menu, page, button and screenshot information into a structured dataset in a unified format; The large model constraint generation module is used to load strong constraint prompts containing preset chapter structures and output rules, and generate operation manual text that conforms to the specifications using the structured dataset as context. The template rendering output module is used to map the operation manual text and interface screenshots to fields, fill in images and text, and unify the format according to a preset template, and then render and output the operation manual document.

[0016] According to an embodiment of the operation manual generation system based on Web-based automated data collection and large model constraint generation of the present invention, the system further includes: The version incremental update module is used to re-collect and compare the differences between the old and new data during system iteration, and perform partial updates on the changed parts.

[0017] According to an embodiment of the operation manual generation system based on Web-based automated data collection and large model constraint generation of the present invention, the information input by the user includes: system name, access URL, login account and password; the script includes: Python script.

[0018] According to an embodiment of the operation manual generation system based on Web-based automated data acquisition and large model constraint generation of the present invention, the Web-based automated data acquisition module includes: The simulated login submodule is used to submit account and password using automated script tools to complete session persistence and credential acquisition. The menu tree parsing submodule is used to recursively extract multi-level menus through DOM parsing and construct the functional hierarchy structure of the target system. The page traversal and element positioning submodule is used to automatically enter each page, locate input boxes, buttons, drop-down lists and form operation elements, and record the name, position and function of the elements; The automatic screenshot submodule is used to take screenshots of each page and key operation areas, and bind the screenshot path to the corresponding function node.

[0019] According to an embodiment of the operation manual generation system based on Web-based automated data collection and large model constraint generation of the present invention, the structured data processing module encapsulates the collected information into JSON format data containing the following fields: system name, version, menu level, page URL, element name, operation type, screenshot path, and business description.

[0020] According to an embodiment of the operation manual generation system based on Web-based automated data collection and large model constraint generation of the present invention, the large model constraint generation module has built-in strong constraint prompts. The text content generated by the forced constraints includes: system overview, login method, function operation steps, permission instructions, common problems and troubleshooting; and each function operation step is required to include: operation location, execution action, expected result and precautions.

[0021] According to an embodiment of the operation manual generation system based on Web-based automated data collection and large model constraint generation of the present invention, the template rendering output module has a built-in preset Word template, which includes preset chapters, fonts, paragraphs, numbering, chart placeholders and table of contents fields; during rendering, text backfilling, image insertion, page numbering and table of contents updating are automatically performed.

[0022] According to an embodiment of the operation manual generation system based on Web-based automated data collection and large model constraint generation of the present invention, the operation manual document format output by the template rendering output module includes an editable Word document and an archived PDF document, and simultaneously generates a version change record page.

[0023] According to an embodiment of the operation manual generation system based on Web automated data collection and large model constraint generation of the present invention, the version incremental update module identifies the added, modified or deleted function points by comparing the hash values ​​of menu structure, page routing and element features, and only regenerates and replaces the text and screenshots corresponding to the changed function points.

[0024] According to an embodiment of the operation manual generation system based on Web-based automated data acquisition and large model constraint generation of the present invention, the system has an automatic configuration function, automatically identifies new Web systems, and completes the entire process of data acquisition, generation, and rendering adaptation. When a new Web system is connected, the system information input module directly receives the name, access URL, login account and password of the new Web system. The Web automated acquisition module, structured data processing module, large model constraint generation module and template rendering output module automatically complete the entire process of acquisition, generation and rendering of the new Web system.

[0025] This invention also discloses a method for generating operation manuals based on web-based automated data collection and large model constraint generation, comprising the following steps: Step S1: Receive and verify the information input by the user to form standardized startup parameters; Step S2: Based on the standardized startup parameters, the system simulates login to the target system non-intrusively using a script, recursively parses the menu tree, traverses page routes and function buttons, automatically locates page operation elements and completes interface screenshots, and binds the screenshots to the corresponding function nodes. Step S3: Encapsulate the collected menu, page, button and screenshot information into a structured dataset in a unified format; Step S4: Load strongly constrained prompts containing preset chapter structures and output rules, and generate compliant operation manual text using the structured dataset as context; Step S5: Map the operation manual text and the interface screenshots to fields, fill in images and text, and unify the format according to the preset template, and then render and output the operation manual document.

[0026] According to an embodiment of the operation manual generation method based on Web-based automated data collection and large model constraint generation of the present invention, the method further includes: Step S6: During system iteration, re-collect and compare the differences between the old and new data, and perform a partial update on the changed parts.

[0027] The present invention also discloses a computer system comprising: Memory, used to store computer programs; A processor is used to execute the computer program to implement the operation manual generation method based on Web-based automated data acquisition and large model constraint generation as described above.

[0028] The present invention also discloses a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the operation manual generation method based on Web-based automated data acquisition and large model constraint generation as described above.

[0029] The present invention also discloses a computer program product, including a computer program / instruction, which, when executed by a processor, implements the operation manual generation method based on Web-based automated data collection and large model constraint generation as described above.

[0030] Compared with the prior art, the present invention has the following advantages: 1. This invention, through the establishment of a web-automated data collection module, uses standardized startup parameters to automatically traverse the target system via scripts. It recursively parses the menu tree, traverses page routes and function buttons, automatically locates page operation elements, and takes screenshots, binding the screenshots to corresponding function nodes. This feature achieves full and accurate acquisition of interface information and operation paths, avoiding omissions and deviations caused by manual screenshotting and recording. This significantly improves data collection efficiency and ensures that the generated operation manual achieves 100% consistency between the generated manual and the actual interface and functions of the target system.

[0031] 2. This invention employs a non-intrusive data collection module via a web automation module, acquiring data from the target system solely by simulating user browser behavior. A structured data processing module encapsulates the collected menu, page, button, and screenshot information into a structured dataset with a unified field format, including system name, version, menu level, page URL, element name, operation type, screenshot path, and business description. This feature allows the system to universally access the target system without requiring open interfaces, code modifications, or database configuration changes. Therefore, it achieves non-intrusive data collection from any closed or deployed web system, providing a standardized, accurate, and reliable structured contextual data source for large model generation.

[0032] 3. This invention loads strongly constrained prompts containing preset chapter structures and output rules through a large model constraint generation module. The text content generated under these constraints includes a system overview, login method, functional operation steps, permission instructions, frequently asked questions, and troubleshooting. Furthermore, each functional operation step must include the operation location, the action to be performed, the expected result, and precautions. This feature, through prompt engineering, strictly limits the generation boundaries and output structure of the large model, overcoming the problems of random generation and "illusion" effects. This ensures that the output content of the large model is logically clear, the steps are expressed in a standardized manner, and perfectly conforms to standard document requirements.

[0033] 4. This invention, through its incremental update module, identifies added, modified, or deleted functionalities by comparing the hash values ​​of menu structure, page routing, and element features. It only regenerates and replaces the text and screenshots corresponding to the changed functionalities. This feature changes the traditional method of manual full-scale verification and complete rewriting, achieving refined localized location and automatic replacement of differences. Therefore, it possesses the technical effect of efficiently identifying system version iteration changes, achieving lightweight local updates, and significantly reducing the subsequent maintenance and update costs of documents.

[0034] 5. This invention utilizes a template rendering output module with a built-in preset Word template (including preset chapters, fonts, paragraphs, numbering, chart placeholders, and table of contents fields). During rendering, it automatically maps fields, fills in text and images, inserts pictures, arranges page numbers, updates the table of contents, and unifies the formatting of the user manual text and interface screenshots. It outputs an editable Word document and an archived PDF document, generating a version change record page. This feature achieves automatic stitching of text, screenshots, and layout styles without manual intervention, thus achieving standardized user manual layout, highly unified formatting, and fully automated output delivery. Attached Figure Description

[0035] The above-described features and advantages of the present invention will be better understood after reading the following detailed description of embodiments of the present disclosure in conjunction with the accompanying drawings. In the drawings, components are not necessarily drawn to scale, and components having similar related properties or features may have the same or similar reference numerals.

[0036] Figure 1 This is a schematic diagram of an embodiment of the operation manual generation system based on Web-based automated data collection and large model constraint generation of the present invention.

[0037] Figure 2 This is a flowchart of an embodiment of the operation manual generation method based on Web-based automated data collection and large model constraint generation of the present invention. Detailed Implementation

[0038] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. It should be noted that the aspects described below with reference to the accompanying drawings and specific embodiments are merely exemplary and should not be construed as limiting the scope of protection of the present invention in any way.

[0039] Figure 1 This describes the principle of an embodiment of the operation manual generation system based on Web-based automated data collection and large model constraint generation, as per the present invention. See also... Figure 1 The system in this embodiment includes the following modules: system information input module, Web automated acquisition module, structured data processing module, large model constraint generation module, and template rendering output module, and each module is connected to the system in sequence.

[0040] The system information input module is used to receive and verify the information input by the user, and form standardized startup parameters.

[0041] For example, user input includes: system name, access URL, login account, and password. Validation methods include, for example, automatically validating the URL format and the validity of the account and password. After successful validation, the above information is structured and encapsulated (e.g., encapsulated into a JSON-formatted configuration file, dictionary object, or command-line startup parameters) to form standardized startup parameters containing system configuration information. These parameters are then passed to the web automation data collection module as the entry point for the automation script.

[0042] The Web automation data collection module is used to standardize startup parameters and simulate login to the target system non-intrusively through scripts. It recursively parses the menu tree, traverses page routes and function buttons, automatically locates page operation elements and completes interface screenshots, and binds the screenshots to the corresponding function nodes.

[0043] Scripts, for example, are Python scripts.

[0044] The web-automated data collection module further includes: The simulated login submodule is used to submit account passwords using automated script tools (such as the Selenium or Requests library in Python) to complete session persistence and credential acquisition; The menu tree parsing submodule is used to recursively extract multi-level menus through DOM parsing and construct the functional hierarchy structure of the target system. Specifically, it identifies front-end node tags with multi-level hierarchical relationships through DOM parsing, recursively extracts the node text and route URLs of the first-level to multi-level menus, and then uses the extracted menu tree as the main skeleton of the system's functional hierarchy to guide the depth-first navigation of subsequent page traversal submodules, and maps the menu hierarchy to the multi-level chapter directory in the user manual. The page traversal and element positioning submodule is used to automatically enter each page, locate input boxes, buttons, drop-down lists and form operation elements, and record the name, position and function of the elements; The automatic screenshot submodule is used to take screenshots of each page and key operation areas, and bind the screenshot path to the corresponding function node.

[0045] The structured data processing module is used to encapsulate the collected menu, page, button, and screenshot information into a structured dataset in a unified format. The structured dataset is JSON format data containing the following fields: system name, version, menu level, page URL, element name, operation type, screenshot path, and business description.

[0046] The large model constraint generation module is used to load strong constraint prompts containing preset chapter structures and output rules, and generate standardized operation manual text using a structured dataset as the context.

[0047] The large model constraint generation module has built-in strong constraint prompts. The text content generated by the forced constraints includes: system overview, login method, function operation steps, permission instructions, common problems and troubleshooting; and each function operation step must include: operation location, execution action, expected result and precautions.

[0048] For example, a template for strongly constrained prompts may include: a character setting area, a context data injection area, a chapter structure constraint area, and a step format strongly constrained area. Its structure and processing logic are as follows: [Character Setting Area]: "You are a professional software system documentation expert, responsible for generating standard operating manuals based on the input structured data."

Context Data Injection Area

[0049] [Chapter Structure Constraints]: This section strictly limits the highest level heading of the output Markdown / text, and mandates that the output content must contain only the following six first-level chapters in sequence: 1. System Overview, 2. Login Method, 3. Function Operation Steps, 4. Permission Explanation, 5. Frequently Asked Questions, 6. Troubleshooting.

[0050] [Strong Constraints on Step Format]: Under Section 3. Functional Operation Steps, you are required to break down the operation nodes into specific sub-steps based on the menu hierarchy and element information in the JSON dataset; and in the strong constraint Prompt, set a few-sample example or format validation regular expression to force each specific operation step to be output in the following fixed four-element structure:

Operation Location

Action to be performed

Notes

[0051] The template rendering output module is used to map the operation manual text and interface screenshots to preset templates, fill in the text and images, and unify the format, and then render and output the operation manual document.

[0052] The template rendering output module has a built-in preset Word template, which includes preset chapters, fonts, paragraphs, numbering, chart placeholders, and table of contents fields; during rendering, it automatically performs text backfilling, image insertion, page numbering, and table of contents updates.

[0053] For example, the Word template pre-defines text and image placeholders corresponding to system chapters and data structures. The specific field mapping and rendering logic is as follows: Template placeholder design: The template includes pre-defined global attribute placeholders and combined placeholders for specific operation steps; Image and text binding and field mapping: The rendering engine extracts the standard text output by the large model constraint generation module and the screenshot paths recorded in the structured dataset; through functional nodes, it associates and maps the specific step text of the large model generation with the screenshot files saved locally or on the server by the automatic screenshot submodule; Adaptive image rendering: When inserting a screenshot into the image placeholder in a Word template, the rendering engine automatically reads the original pixel aspect ratio of the image and automatically scales the image size according to the page margins and paragraph width of the Word template to ensure that the image is centered and does not exceed the document layout boundaries.

[0054] The template rendering output module outputs operation manual documents in editable Word documents and archived PDF documents, and simultaneously generates a version change record page. When the incremental update module identifies changes in the system, this sub-module automatically extracts the incremental update log (including change time, name of the changed function node, change type such as add / modify / delete, and comparison of hash values ​​before and after the change), and automatically inserts and renders a version change record table page after the first page of the Word template, realizing automated archiving of the version evolution history.

[0055] Preferably, the system in this embodiment further includes: a version incremental update module, used to re-collect and compare the differences between the old and new data during system iteration, and perform local updates on the changed parts.

[0056] The incremental update module identifies added, modified, or deleted function points by comparing the hash values ​​of menu structure, page routing, and element features, and only regenerates and replaces the text and screenshots corresponding to the changed function points.

[0057] For example, the specific calculation mechanism for hash values ​​is as follows: Menu and route hash calculation: The collected menu tree node names, hierarchical relationships and page URLs are concatenated into a menu string in a fixed order, and a menu structure hash value is generated using a preset hash algorithm (such as SHA-256 or MD5); Page element feature hash calculation: For each functional page, dynamically changing random DOM attributes (such as dynamically generated style class names or random IDs) are filtered out, and only the core attribute strings of key operation elements (including element name, type such as button / input box, positioning XPath or core Text) are extracted, concatenated and calculated to generate element feature hash value; Node-level hash mapping table construction: Each functional node (such as "User Management - Add User") is bound to its corresponding menu hash value and element feature hash value to form the hash mapping dictionary table (Hash Map) of the current version.

[0058] For identifying newly added, modified, or deleted function points, a key-value comparison is performed between the newly generated hash mapping dictionary table and the previously archived hash mapping dictionary table. New node identification: If a certain function node ID does not exist in the old dictionary table, then the node is determined to be a new function node; Deleted node identification: If a certain function node ID in the old dictionary table no longer exists in the new dictionary table, then the node is determined to be a deleted function node; Modified Node Identification: If the same function node ID exists in the old and new dictionary tables, but their corresponding menu hash values ​​or element feature hash values ​​are inconsistent, then the node is determined to be a modified function node.

[0059] After identifying the changed functionalities, the system does not need to perform a full rewrite, but instead performs partial data reconstruction and document node replacement: Local large model text generation: Only extract the structured dataset corresponding to the function nodes that are determined to be "new" or "modified", send it to the large model constraint generation module, and regenerate the operation step text corresponding to the node with strong constraint prompt words; Partial screenshot update: The automatic screenshot submodule is only retried for "Add" or "Modify" function nodes to update the corresponding local screenshot files; Precise replacement of Word document bookmarks / nodes: During the initial rendering of a Word document, the template rendering output module embeds hidden bookmarks or custom XML node tags with function node IDs at the titles and paragraphs of each functional chapter; during incremental updates, the rendering engine directly locates the bookmark position. 1. For "modified nodes", directly clear the original text and images within the bookmark range, and then insert the newly generated text and new screenshots back into the bookmark. 2. For "deleted nodes", directly delete the entire chapter / paragraph corresponding to the bookmark; 3. For "New Node", locate the end of the parent menu chapter bookmark and append the newly generated chapter content and screenshot.

[0060] After the update is complete, the system automatically generates a change summary that includes the name of the changed node, the change type (added / modified / deleted), the change time, and the hash values ​​before and after the change. This summary is automatically written to the "Version Change Record Page" on the first page of the document and triggers the regeneration of the PDF document.

[0061] Furthermore, preferably, the system in this embodiment has an automatic configuration function, which automatically identifies newly added Web systems and completes the entire process of collection, generation, and rendering adaptation: when a newly added Web system is accessed, the system information input module directly receives the name, access URL, login account, and password of the newly added Web system, and the Web automated collection module, structured data processing module, large model constraint generation module, and template rendering output module automatically complete the entire process of collection, generation, and rendering adaptation for the newly added Web system.

[0062] Figure 2 A flowchart illustrating an embodiment of the operation manual generation method based on Web-based automated data collection and large model constraint generation according to the present invention is shown. (Reference) Figure 2 The method in this embodiment includes the following steps.

[0063] Step S1: Receive and verify the information input by the user to form standardized startup parameters.

[0064] For example, user input includes: system name, access URL, login account, and password. Validation methods include, for example, automatically validating the URL format and the validity of the account and password. After successful validation, the above information is structured and encapsulated (e.g., encapsulated into a JSON-formatted configuration file, dictionary object, or command-line startup parameters) to form standardized startup parameters containing system configuration information. These parameters are then passed to the web automation data collection module as the entry point for the automation script.

[0065] Step S2: Based on standardized startup parameters, the script non-intrusively simulates login to the target system, recursively parses the menu tree, traverses page routes and function buttons, automatically locates page operation elements and completes interface screenshots, and binds the screenshots to the corresponding function nodes.

[0066] Scripts, for example, are Python scripts.

[0067] Step S2 further includes: Simulated login: Use automated script tools (such as the Selenium or Requests library in Python) to submit account and password, and complete session persistence and credential acquisition; Menu tree parsing: Multi-level menus are recursively extracted through DOM parsing to construct the functional hierarchy of the target system. Specifically, front-end node tags with multi-level hierarchical relationships are identified through DOM parsing, and the node text and route URLs of the first-level to multi-level menus are recursively extracted. The extracted menu tree is then used as the main skeleton of the system's functional hierarchy to guide the depth-first navigation of sub-modules in subsequent page traversals, and the menu hierarchy is mapped to the multi-level chapter directory in the user manual. Page traversal and element location: Automatically enter each page, locate input boxes, buttons, dropdown lists and form operation elements, and record the name, position and function of the elements; Automatic screenshot: Takes screenshots of each page and key operation areas, and binds the screenshot path to the corresponding function node.

[0068] Step S3: Encapsulate the collected menu, page, button, and screenshot information into a structured dataset in a unified format.

[0069] The structured dataset is JSON formatted data containing the following fields: system name, version, menu level, page URL, element name, operation type, screenshot path, and business description.

[0070] Step S4: Load strongly constrained prompts containing preset chapter structures and output rules, and generate compliant operation manual text using the structured dataset as context.

[0071] The text content generated by the mandatory constraints includes: system overview, login method, function operation steps, permission description, common problems and troubleshooting; and each function operation step must include: operation location, execution action, expected result and precautions.

[0072] Step S5: Map the operation manual text and the interface screenshots to fields, fill in images and text, and unify the format according to the preset template, then render and output the operation manual document.

[0073] The built-in preset Word templates include preset chapters, fonts, paragraphs, numbering, chart placeholders, and table of contents fields; during rendering, text backfilling, image insertion, page numbering, and table of contents updates are performed automatically.

[0074] Preferably, the method of this embodiment further includes: Step S6: During system iteration, re-collect and compare the differences between the old and new data, and perform a partial update on the changed parts.

[0075] By comparing the hash values ​​of menu structure, page routing, and element features, the system identifies newly added, modified, or deleted function points, and only regenerates and replaces the text and screenshots corresponding to the changed function points.

[0076] Furthermore, this invention also discloses a computer system, including a memory and a processor. The memory is used to store computer programs; the processor is used to execute the computer programs to achieve, for example... Figure 2 The steps of an embodiment of the operation manual generation method based on Web-based automated data collection and large model constraint generation are shown.

[0077] Furthermore, this invention also discloses a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the following... Figure 2 The steps of an embodiment of the operation manual generation method based on Web-based automated data collection and large model constraint generation are shown.

[0078] Furthermore, this invention also discloses a computer program product, including a computer program / instructions, which, when executed by a processor, implements as follows: Figure 2 The steps of an embodiment of the operation manual generation method based on Web-based automated data collection and large model constraint generation are shown.

[0079] Although the methods described above are illustrated and depicted as a series of actions for the sake of simplicity, it should be understood and appreciated that these methods are not limited by the order of the actions, as some actions may occur in a different order and / or concurrently with other actions from the illustrations and descriptions herein or not illustrated and described herein but which may be understood by those skilled in the art, according to one or more embodiments.

[0080] Those skilled in the art will further appreciate that the various illustrative logic blocks, modules, circuits, and algorithm steps described in conjunction with the embodiments disclosed herein can be implemented as electronic hardware, computer software, or a combination of both. To clearly illustrate this interchangeability between hardware and software, the various illustrative components, blocks, modules, circuits, and steps are described above in a generalized manner in terms of their functionality. Whether such functionality is implemented as hardware or software depends on the specific application and the design constraints imposed on the overall system. Those skilled in the art may implement the described functionality in different ways for each specific application, but such implementation decisions should not be construed as departing from the scope of the invention.

[0081] The various illustrative logic blocks, modules, and circuits described in conjunction with the embodiments disclosed herein can be implemented or performed using a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. The general-purpose processor may be a microprocessor, but in alternatives, it may be any conventional processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors cooperating with a DSP core, or any other such configuration.

[0082] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of both. The software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to a processor such that the processor can read and write information to / from the storage medium. In an alternative, the storage medium may be integrated into the processor. The processor and storage medium may reside in an ASIC. The ASIC may reside in a user terminal. In an alternative, the processor and storage medium may reside as discrete components in the user terminal.

[0083] In one or more exemplary embodiments, the described functionality may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software as a computer program product, the functionality may be stored or transmitted as one or more instructions or code on or through a computer-readable medium. A computer-readable medium includes both computer storage media and communication media, encompassing any medium that facilitates the transfer of a computer program from one location to another. A storage medium may be any available medium accessible to a computer. By way of example and not limitation, such a computer-readable medium may include RAM, ROM, EEPROM, CD-ROM or other optical disc storage, disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and is accessible to a computer. Any connection is also legitimately referred to as a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of a medium. As used in this article, disk and disc include compact discs (CDs), laser discs, optical discs, digital multi-purpose discs (DVDs), floppy disks, and Blu-ray discs. Disks typically reproduce data magnetically, while discs reproduce data optically using lasers. Combinations of these should also be included within the scope of computer-readable media.

[0084] The prior description of this disclosure is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to this disclosure will be apparent to those skilled in the art, and the general principles defined herein may be applied to other variations without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not intended to be limited to the examples and designs described herein, but should be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A user manual generation system based on Web-based automated data collection and large model constraint generation, characterized in that, include: The system comprises an information input module, a web-based automated data acquisition module, a structured data processing module, a large model constraint generation module, and a template rendering output module, all of which are interconnected sequentially. Among these modules: The system information input module is used to receive and verify the information input by the user to form standardized startup parameters; The Web automated data collection module is used to simulate logging into the target system non-intrusively through scripts based on the standardized startup parameters, recursively parse the menu tree, traverse page routes and function buttons, automatically locate page operation elements and complete interface screenshots, and bind screenshots to corresponding function nodes. The structured data processing module is used to encapsulate the collected menu, page, button and screenshot information into a structured dataset in a unified format; The large model constraint generation module is used to load strong constraint prompts containing preset chapter structures and output rules, and generate operation manual text that conforms to the specifications using the structured dataset as context. The template rendering output module is used to map the operation manual text and interface screenshots to fields, fill in images and text, and unify the format according to a preset template, and then render and output the operation manual document.

2. The operation manual generation system based on Web-based automated data acquisition and large model constraint generation according to claim 1, characterized in that, The system also includes: The version incremental update module is used to re-collect and compare the differences between the old and new data during system iteration, and perform partial updates on the changed parts.

3. The operation manual generation system based on Web-based automated data collection and large model constraint generation according to claim 1, characterized in that, The information entered by the user includes: system name, access URL, login account and password; the script includes: Python script.

4. The operation manual generation system based on Web-based automated data acquisition and large model constraint generation according to claim 1, characterized in that, The Web automated data collection module includes: The simulated login submodule is used to submit account and password using automated script tools to complete session persistence and credential acquisition. The menu tree parsing submodule is used to recursively extract multi-level menus through DOM parsing and construct the functional hierarchy structure of the target system. The page traversal and element positioning submodule is used to automatically enter each page, locate input boxes, buttons, drop-down lists and form operation elements, and record the name, position and function of the elements; The automatic screenshot submodule is used to take screenshots of each page and key operation areas, and bind the screenshot path to the corresponding function node.

5. The operation manual generation system based on Web-based automated data acquisition and large model constraint generation according to claim 1, characterized in that, The structured data processing module encapsulates the collected information into JSON format data containing the following fields: system name, version, menu level, page URL, element name, operation type, screenshot path, and business description.

6. The operation manual generation system based on Web-based automated data acquisition and large model constraint generation according to claim 1, characterized in that, The large model constraint generation module has built-in strong constraint prompts. The text content generated by the forced constraints includes: system overview, login method, function operation steps, permission instructions, common problems and troubleshooting; and each function operation step must include: operation location, execution action, expected result and precautions.

7. The operation manual generation system based on Web-based automated data collection and large model constraint generation according to claim 1, characterized in that, The template rendering output module has a built-in preset Word template, which includes preset chapters, fonts, paragraphs, numbering, chart placeholders, and table of contents fields; during rendering, it automatically performs text backfilling, image insertion, page numbering, and table of contents updates.

8. The operation manual generation system based on Web-based automated data acquisition and large model constraint generation according to claim 1 or 7, characterized in that, The template rendering output module outputs operation manual documents in editable Word documents and archived PDF documents, and simultaneously generates a version change record page.

9. The operation manual generation system based on Web-based automated data acquisition and large model constraint generation according to claim 2, characterized in that, The incremental update module identifies added, modified, or deleted function points by comparing the hash values ​​of menu structure, page routing, and element features, and only regenerates and replaces the text and screenshots corresponding to the changed function points.

10. The operation manual generation system based on Web-based automated data acquisition and large model constraint generation according to claim 1, characterized in that, The system has an automatic configuration function, automatically identifying new Web systems and completing the entire process of data collection, generation, and rendering adaptation. When a new Web system is connected, the system information input module directly receives the name, access URL, login account and password of the new Web system. The Web automated acquisition module, structured data processing module, large model constraint generation module and template rendering output module automatically complete the entire process of acquisition, generation and rendering of the new Web system.

11. A method for generating operation manuals based on Web-based automated data collection and large model constraint generation, characterized in that, Includes the following steps: Step S1: Receive and verify the information input by the user to form standardized startup parameters; Step S2: Based on the standardized startup parameters, the system simulates login to the target system non-intrusively using a script, recursively parses the menu tree, traverses page routes and function buttons, automatically locates page operation elements and completes interface screenshots, and binds the screenshots to the corresponding function nodes. Step S3: Encapsulate the collected menu, page, button and screenshot information into a structured dataset in a unified format; Step S4: Load strongly constrained prompts containing preset chapter structures and output rules, and generate compliant operation manual text using the structured dataset as context; Step S5: Map the operation manual text and the interface screenshots to fields, fill in images and text, and unify the format according to the preset template, and then render and output the operation manual document.

12. The method for generating operation manuals based on Web-based automated data collection and large model constraint generation according to claim 11, characterized in that, The method also includes: Step S6: During system iteration, re-collect and compare the differences between the old and new data, and perform a partial update on the changed parts.

13. A computer system, characterized in that, include: Memory, used to store computer programs; A processor is configured to execute the computer program to implement the operation manual generation method based on Web-based automated data acquisition and large model constraint generation as described in claim 11 or 12.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the operation manual generation method based on Web-based automated data collection and large model constraint generation as described in claim 11 or 12.

15. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instruction is executed by the processor, it implements the operation manual generation method based on Web-based automated data collection and large model constraint generation as described in claim 11 or 12.

Images