Collaborative control methods, devices, equipment, and storage media for bottom input boxes

By overlaying a native input box layer at the bottom of the webpage view and listening to changes in the DOM tree, the pop-up interface is automatically identified and hidden, solving the problem of coordinated control between the bottom input box and the pop-up window. This achieves centralized management and automated operation, reduces development and maintenance costs, and improves the user experience.

CN123114348APending Publication Date: 2026-11-17CHINA PING AN PROPERTY INSURANCE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611328418.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-28
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In existing technologies, the collaborative control of the bottom input box and pop-up in hybrid architecture mobile applications suffers from problems such as scattered logic, reliance on manual operation and easy omissions. Furthermore, full WebView-based and full native solutions lead to a surge in development costs and a loss of business flexibility.

Method used

By overlaying a native input box layer at the bottom of the webpage view and listening to the addition event of child nodes in the document object model to obtain multi-dimensional feature information, the pop-up interface is automatically identified and the native input box is hidden, thus realizing automatic collaborative control between the pop-up and the input box.

Benefits of technology

It achieves centralized management of pop-up recognition logic, avoiding the need for developers to manually add notification logic, reducing maintenance costs, and improving the consistency of user experience and system performance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123114348A_ABST
    Figure CN123114348A_ABST
Patent Text Reader

Abstract

This invention belongs to the field of data processing technology and discloses a collaborative control method, device, equipment, and storage medium for a bottom input box. The method includes: overlaying a native input box layer on the bottom area of ​​a webpage view while the webpage view is displayed; listening for child node addition events in the document object model of the webpage view; when a child node addition event is detected, acquiring multi-dimensional feature information of the added node; when the values ​​of the multi-dimensional feature information meet preset conditions, determining the added node as a pop-up interface; and when the added node is a pop-up interface, controlling the native input box to be hidden. This invention can be applied to business management systems in fintech, healthcare, and other fields, solving the inherent defects of existing manual notification solutions, such as "dispersed logic, reliance on manual intervention, and easy omissions," while fully WebView-based and fully native solutions suffer from increased development costs and loss of business flexibility.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of data processing technology and is applied in the fields of financial technology and healthcare. In particular, it relates to a collaborative control method, device, equipment and storage medium for a bottom input box. Background Technology

[0002] In hybrid architecture mobile applications, to balance cross-platform development efficiency and native user experience, developers typically use a WebView to carry H5 page content, while overlaying native UI components (such as a bottom input box) on top of the WebView to build the application interface. This architecture is particularly common in insurance and healthcare applications, where users use the bottom input box to consult intelligent customer service for insurance product information or describe their symptoms to an online doctor.

[0003] However, during the operation of an H5 page, the appearance of pop-ups (such as product detail pop-ups, department selection pop-ups, etc.) can create visual conflicts and interactive interference with the native input box at the bottom. To solve this problem, the following three mainstream solutions exist in the current technology: The first approach is a manual notification-based native input box collaboration solution. This solution uses a native input box component at the bottom of the H5 page. When a pop-up window appears on the H5 page, the H5 business code actively calls the native interface to notify the native layer to hide the input box. This solution has minimal intrusion into the H5 business code, but it has significant drawbacks: the notification logic for different pop-up components (such as the middle dialog and the bottom pop-up) is inconsistent; some pop-ups notify the native layer to display a mask, while others notify the native layer to hide the input box, resulting in fragmented system logic and maintenance difficulties.

[0004] Taking insurance service applications as an example, when a user inquires about insurance products on the intelligent customer service page, clicking the "Product Details" button triggers a product details pop-up (dialog type). Developers need to manually add code to the native notification layer to hide the input box in the pop-up's display logic. However, when the user clicks the "Quick Reply" button at the bottom of the page, a bottom pop-up layer (popup type) is triggered. Its notification logic is completely different from that of the dialog pop-up; the former displays a mask in the native notification layer, while the latter hides the input box. Because there are many types of pop-ups on insurance product pages (product comparison pop-ups, coverage pop-ups, insurance confirmation pop-ups, rate calculation pop-ups, etc.), and different pop-ups are written by different developers at different times, the notification logic is difficult to manage uniformly, and developers are very likely to miss or confuse the notification code.

[0005] Similar issues exist in healthcare applications. When a user clicks the "Select Department" button on an online consultation page, a department selection pop-up appears (dialog type); clicking the "Upload Report" button opens a report upload pop-up (popup type); and clicking the "Medical Records" button opens a medical record viewing pop-up. These three pop-ups have different notification mechanisms, requiring developers to write separate logic for each pop-up to interact with the native input fields. This results in code scattered across different modules, leading to extremely high maintenance costs.

[0006] The second solution is a fully WebView-based pop-up solution. This solution makes all pop-ups independent WebView pages, completely isolating them from the main page at the rendering layer. While this avoids direct conflict between pop-ups and input fields, it severely disrupts the existing H5 business ecosystem. All pop-ups need to be redeveloped using new H5 pages, and a complex cross-WebView communication mechanism needs to be established between the pop-ups and the main page. In insurance service applications, this means that dozens of pop-ups, such as product details pop-ups, coverage pop-ups, and insurance confirmation pop-ups, need to be redeveloped. In healthcare applications, pop-ups such as department selection pop-ups, report upload pop-ups, and medical record viewing pop-ups also need to be completely rewritten, multiplying development and maintenance costs.

[0007] The third option is a fully native approach. This approach implements all UI components on the page (including pop-ups and input fields) using native code to achieve the ultimate user experience. However, this approach requires independent development for iOS, Android, and HarmonyOS platforms, resulting in a huge investment of manpower. Furthermore, native applications rely on app store review mechanisms, making hot updates impossible and leading to long cycles for fixing urgent issues. This shortcoming is particularly prominent in rapidly iterating business scenarios such as insurance and healthcare, where product terms updates or consultation process adjustments all require app store approval, preventing timely delivery to users.

[0008] In summary, existing manual notification solutions have inherent drawbacks such as "dispersed logic, reliance on manual intervention, and susceptibility to omissions," while fully WebView-based and fully native solutions bring new problems such as a sharp increase in development costs and a loss of business flexibility. Summary of the Invention

[0009] This invention provides a collaborative control method, device, equipment, and storage medium for a bottom input box, which can solve the inherent defects of existing manual notification solutions, such as "dispersed logic, reliance on manual intervention, and easy omissions," as well as the technical problems of increased development costs and loss of business flexibility in fully WebView-based and fully native solutions.

[0010] In a first aspect, the present invention provides a collaborative control method for a bottom input box, comprising: When the webpage view is displayed, a native input box layer is overlaid in the bottom area of ​​the webpage view. The native input box layer is displayed by default and contains native input boxes. Listen for new child node events in the document object model of the webpage view; When a new child node event is detected, the feature information of the new node in multiple dimensions is obtained; When the values ​​of the feature information in the multiple dimensions meet the preset conditions, the newly added node is determined to be a pop-up interface; When the newly added node is the pop-up interface, control the native input box to be hidden.

[0011] Secondly, the present invention provides a collaborative control device for a bottom input box, comprising: An overlay module is used to overlay a native input box layer on the bottom area of ​​a webpage view when the webpage view is in display mode. The native input box layer is in display mode by default and contains native input boxes. The event listener module is used to listen for new child node events in the document object model of the webpage view; The information acquisition module is used to acquire multi-dimensional feature information of the newly added node when the newly added child node event is detected. The node determination module is used to determine the newly added node as a pop-up interface when the values ​​of the feature information of the multiple dimensions meet the preset conditions. The control module is used to control the native input box to be hidden when the newly added node is the pop-up interface.

[0012] Thirdly, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described collaborative control method for the bottom input box.

[0013] Fourthly, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the steps of the aforementioned collaborative control method for the bottom input box.

[0014] Understandably, this invention, through its architectural design of "overlaying a native input box layer on top of the webpage view," replaces the approach of developing separate native components for different underlying operating systems found in the "fully native solution." Since the native input box layer is overlaid on the webpage view, the H5 business code does not need to concern itself with the differences in the underlying operating system. A single set of H5 code can run normally on various underlying operating systems and in various external deployment scenarios, fundamentally solving the problems of drastically increased development costs and loss of business flexibility inherent in the fully native solution.

[0015] Meanwhile, because the native input box layer is separated from the H5 business code, the H5 business logic can obtain the interactive experience of the native input box without modification, thus solving the problem of significant intrusion into H5 business logic. Furthermore, the native input box layer is displayed by default, ensuring that users can immediately see and use the bottom input box after entering the page.

[0016] This invention enables the system to automatically identify all types of pop-up interfaces by uniformly monitoring changes in the DOM tree. This achieves centralized management of pop-up recognition logic, with all pop-up recognition completed through the same monitoring and judgment mechanism. There is no longer a need to write different notification code for different types of pop-ups, fundamentally eliminating the problem of "dispersed logic".

[0017] This invention automatically detects the appearance of pop-up windows by listening to the addition of child nodes in the document object model. It then uses a multi-dimensional feature scoring algorithm to intelligently identify the new node, eliminating the need for developers to manually add any notification logic to the pop-up code. When a new node is determined to be a pop-up, the system automatically hides the native input box. The entire process is automated and does not rely on active calls from H5 business code, fundamentally eliminating the problems of "manual dependence and easy omissions."

[0018] Upon automatically recognizing the pop-up interface, this invention automatically controls the native input box to be hidden, ensuring the pop-up interface is fully displayed without being obscured by the native input box. This directly solves the problem of "input boxes failing to hide in time when pop-ups appear, causing interface overlap and abnormal interaction." Furthermore, in this invention, the hiding action is automatically triggered by the system after pop-up detection, avoiding the manual triggering of input box hiding by H5 business code in existing technologies. This automates the process, replacing manual notification. In addition, the overlay architecture of the native input box layer ensures that the "hide" operation only applies to the native layer, without affecting the rendering of the webpage view content, guaranteeing a consistent user experience. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating a collaborative control method for a bottom input box in one embodiment of the present invention.

[0021] Figure 2 yes Figure 1A flowchart of step S130.

[0022] Figure 3a and Figure 3b This is a schematic diagram of the display and hidden states of the native input box in one embodiment of the present invention.

[0023] Figure 4 This is another flowchart illustrating the collaborative control method for the bottom input box in one embodiment of the present invention.

[0024] Figure 5 This is another flowchart illustrating the collaborative control method for the bottom input box in one embodiment of the present invention.

[0025] Figure 6 yes Figure 5 A flowchart of step S172.

[0026] Figure 7 yes Figure 5 This is another flowchart illustrating the collaborative control method for the bottom input box in one embodiment of the present invention.

[0027] Figure 8 This is a schematic diagram of the collaborative control device for the bottom input box in one embodiment of the present invention.

[0028] Figure 9 This is a schematic diagram of the structure of a computer device according to an embodiment of the present invention.

[0029] Figure 10 This is another structural schematic diagram of a computer device according to one embodiment of the present invention. Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] Figure 1 A flowchart of the collaborative control method for the bottom input box provided in an embodiment of the present invention is shown below. Figure 1 As shown, the collaborative control method for the bottom input box provided in this embodiment of the invention includes the following steps.

[0032] Step S110: In the display state of the webpage view, overlay the native input box layer on the bottom area of ​​the webpage view; The native input box layer is displayed by default, and it contains native input boxes. In one possible implementation, a native input box layer is overlaid on the bottom area of ​​the webpage view while it is in the display state. Here, the webpage view refers to the view component (such as WebView) used to load and render webpage content on an electronic device, and the native input box layer refers to a layer composed of input box components provided by the operating system's native framework (such as UIKit in iOS or the View system in Android). The native input box layer is displayed by default and contains native input boxes.

[0033] Specifically, in a hybrid architecture application, the web view occupies the entire screen or a large portion of it, displaying the H5 page content. The native input box layer is overlaid on top of the web view and located at the bottom of the screen. The native input box layer and the web view have a different hierarchy in the vertical direction (Z-axis): the native input box layer is higher than the web view, ensuring that the native input box layer is visually in front of the web view, allowing the user to see and interact with it.

[0034] The native input box layer is in a visible state by default, meaning that the native input box layer is visible after the application starts or the page finishes loading, and users can see the native input box at the bottom of the screen.

[0035] Optionally, the native input box layer can be overlaid on the web page view with a transparent background, allowing users to browse the content of the web page view while the native input box is visible.

[0036] Optionally, the bottom position of the native input box layer can be aligned with the bottom of the electronic device's screen or with the top edge of the system navigation bar on the screen. The specific alignment method can be determined based on the screen configuration of the electronic device and the interface design of the application.

[0037] As a concrete example, in the property insurance sector, when a user opens the insurance product consultation page of the Ping An Good Life App, this page is an H5 page displayed in a WebView. Simultaneously with the WebView display, the system overlays a native input box layer at the bottom of the WebView; this native input box layer contains a native input box.

[0038] As another concrete example, in the healthcare field, a user opens the online consultation page of the Ping An Health App. This page is an H5 page displayed in a WebView. Simultaneously with the WebView display, the system overlays a native input box layer at the bottom of the WebView; this native input box layer contains a native input box.

[0039] Step S120: Listen for the child node addition event of the document object model of the web page view; One possible implementation involves listening for new child node addition events in the webpage view's Document Object Model (DOM).

[0040] The DOM is a structured representation of an H5 page loaded by a web view. All elements in an H5 page (such as pop-ups, dialog boxes, buttons, input boxes, navigation bars, etc.) exist in the DOM tree as nodes. The child node addition event refers to the event triggered when a node is added to the DOM tree as a child node of another node.

[0041] Specifically, during the execution of an H5 page loaded in a webpage view, dynamic changes in page content (such as pop-up windows appearing, new content loading, and interface updates triggered by user interaction) will cause changes in the DOM tree structure. When a new child node is added to the DOM tree of the H5 page, the system captures this event through a listening mechanism.

[0042] Optionally, a MutationObserver can be used to monitor changes in the DOM tree. During the initialization of the H5 page when the webpage view loads, a MutationObserver instance is created, and its monitoring options are configured. These options can include monitoring changes to the list of child nodes. When a child node is added to the DOM tree, the MutationObserver can capture the change and trigger the corresponding callback function. This embodiment of the invention does not limit the handling methods in environments where MutationObserver is unavailable; alternative solutions such as polling mechanisms can be used to detect changes in the DOM tree.

[0043] Optionally, events for adding child nodes include, but are not limited to: events triggered by adding nodes to the DOM tree using JavaScript methods such as appendChild, insertBefore, and replaceChild, or events triggered by inserting new elements into the DOM tree using property modifications such as innerHTML and outerHTML.

[0044] Optionally, changes in DOM tree attributes can also be monitored, such as changes in the style, class, display, and visibility attributes of element nodes. When an element already existing in the DOM tree changes from a hidden state to a visible state due to attribute changes, although the element is not considered a "new child node," it may still need to be recognized as a pop-up interface. Therefore, in this embodiment of the invention, step S120 can also monitor changes in DOM tree attributes and treat attribute changes as events equivalent to the addition of a child node.

[0045] As a concrete example in the property insurance sector, when a user clicks the "Product Details" button on the insurance product consultation page of the Ping An Good Life App, the JavaScript code on the H5 page responds to the user's click by dynamically creating a pop-up node in the DOM tree and adding it as a child node of the body node. The system captures the addition event of this child node through a listening mechanism.

[0046] As another concrete example in the healthcare field, when a user clicks the "Select Department" button on the online consultation page of the Ping An Health App, the JavaScript code on the H5 page responds to the user's click by dynamically creating a department selection pop-up node in the DOM tree and adding it as a child node of the page container to the DOM tree. The system captures the addition event of this child node through a listening mechanism.

[0047] Step S130: When a child node addition event is detected, obtain the feature information of the added node in multiple dimensions. Specifically, when the listening mechanism in step S120 captures a child node addition event, the system obtains a reference to the added node and extracts multi-dimensional feature information of the added node. Feature information refers to multi-dimensional quantitative or qualitative indicators that can reflect the possibility of the added node as a pop-up interface.

[0048] Optionally, a newly added node refers to the node added to the DOM tree in this child node addition event. If a single child node addition event involves multiple newly added nodes, feature information can be extracted for each newly added node separately.

[0049] Optionally, when obtaining multi-dimensional feature information of a newly added node, you can call the DOM API (such as getComputedStyle, getBoundingClientRect, getAttribute, etc.) to obtain the node's style information, position information, attribute information, etc.

[0050] As a specific example, in the property insurance field, on the insurance product consultation page of Ping An Good Life App, when a user clicks the "Product Details" button, the H5 page dynamically creates a pop-up node. After the system captures the new event of this child node through the listening mechanism, it obtains multiple dimensions of feature information of the newly added node, such as the list of class names, level value (i.e., z-index), size information, position information, background color, etc.

[0051] As another concrete example, in the healthcare field, on the online consultation page of the Ping An Health App, when a user clicks the "Select Department" button, the H5 page dynamically creates a department selection pop-up node. After the system captures the new event of this child node through the listening mechanism, it obtains multiple dimensions of feature information of the newly added node, such as the list of class names, z-index, size information, position information, background color, etc.

[0052] In some embodiments of the present invention, such as Figure 2 As shown, step S130 includes the following steps.

[0053] Step S131: Merge multiple child node addition events detected within the preset time window into a single child node addition event for processing. Specifically, during the execution of an H5 page loaded in a webpage view, dynamic changes in page content may trigger multiple DOM tree changes within a very short period of time. For example, the appearance of a pop-up interface may be accompanied by the sequential addition of multiple DOM nodes (such as adding the overlay node first, then the pop-up content node, then the close button node, etc.). If the pop-up interface judgment process is triggered immediately every time a child node is added, it will lead to the repeated execution of judgment logic in a short period of time, causing unnecessary consumption of computing resources and even causing page lag.

[0054] To address this, this invention introduces a merging mechanism: when a new child node event is detected, the judgment process of the pop-up interface is not executed immediately. Instead, the event is temporarily stored, and after the preset time window ends, the multiple new child node events detected within the window are merged into a single event for unified processing.

[0055] Optionally, the preset time window can be set according to the actual application scenario and device performance. In one possible implementation, the preset time window can be 16 milliseconds (corresponding to the duration of one frame at a 60Hz refresh rate). In another possible implementation, the preset time window can also be 10 milliseconds, 20 milliseconds, 30 milliseconds, or 50 milliseconds. The longer the preset time window, the more obvious the merging effect, but the delay in judgment and response also increases accordingly; the shorter the preset time window, the more timely the response, but the merging effect may be weakened. The specific value can be determined according to the balance requirements of real-time performance and performance in the actual application scenario.

[0056] Optionally, the specific implementation of the merging process may include: when a new child node event is detected, a timer is started, with a duration of a preset time window. During the timer's operation, if another new child node event is detected, this new event is also added to the current window's temporary queue, and the timer is reset (i.e., the debouncing mechanism). When the timer expires, all new child node events in the temporary queue are merged into a single event, triggering the subsequent decision-making process.

[0057] Optionally, in addition to merging events for newly added child nodes, attribute change events can also be included in the merging process along with events for newly added child nodes. For example, when an attribute change in the DOM tree (such as a change in style, class, display, or visibility attributes) is detected, this event can also be temporarily stored in a queue and merged with the events for newly added child nodes into a single event for unified processing.

[0058] Optionally, if the current environment does not support a MutationObserver, a polling mechanism can be used as an alternative. Specifically, a snapshot scan of the DOM tree can be performed every preset time interval (e.g., 100 milliseconds), and all changes between two scans can be merged into a single event for processing. The polling frequency can be dynamically adjusted according to device performance.

[0059] As a concrete example in the property insurance sector, on the insurance product consultation page of the Ping An Good Life App, after a user clicks the "Product Details" button, the H5 page sequentially adds a mask layer node, a pop-up container node, a pop-up title node, and a pop-up content node within a short period of time, triggering four child node addition events within a 16-millisecond time window. Instead of triggering four separate decision processes, the system merges these child node addition events into a single child node addition event for processing.

[0060] As another concrete example in the healthcare field, on the online consultation page of the Ping An Health App, after a user clicks the "Select Department" button, the H5 page sequentially adds a mask layer node, a department list container node, a department option node, and a close button node within a short period of time, triggering four child node addition events within a 16-millisecond time window. Instead of triggering four separate decision processes, the system merges these child node addition events into a single child node addition event for processing.

[0061] Step S132: Take the nodes of the document object model involved in the single child node addition event as the new nodes and obtain the feature information of the nodes in multiple dimensions.

[0062] Specifically, after step S131 merges multiple child node addition events within a preset time window into a single child node addition event, the DOM node involved in the single event is determined, and the node is used as the new node to be judged, and the feature information of the node in multiple dimensions is obtained.

[0063] "Nodes in the Document Object Model involved in a single child node addition event" refers to the collection of all nodes added to the DOM tree within a preset time window. During merging, these nodes are uniformly considered "new nodes." In one possible implementation, if only one node is added to the window, that node is considered a new node; if multiple nodes are added, all of them are considered new nodes, and their feature information is obtained for each node's determination. It should be noted that in practical applications, multiple constituent nodes of the same pop-up interface (such as overlay nodes and pop-up container nodes) often have similar feature information (such as similar z-index, similar positions, and the same parent node), so processing these nodes uniformly will not affect the accuracy of the determination result.

[0064] Optionally, the specific method for obtaining multi-dimensional feature information of a node can be referred to the description in step S130. Feature information includes, but is not limited to: whether the class name contains pop-up semantic keywords, whether it has a role attribute with pop-up semantics, whether the stacking level value is higher than a preset level threshold, whether the viewport size ratio is greater than a preset ratio threshold, whether the deviation between the center point and the viewport center point is less than a preset deviation threshold, whether there are interactive elements within the node, whether the background color has a non-transparent visual effect, and whether there are masked style nodes with the same parent node as the newly added node, etc.

[0065] Optionally, before acquiring multi-dimensional feature information of newly added nodes, a preliminary screening can be performed. For example, check the tag name of the newly added node. If the tag name is a non-visual element such as "script" or "style", it can be skipped directly without further feature information acquisition and judgment, thereby further improving processing efficiency.

[0066] As a concrete example in the property insurance sector, on the insurance product consultation page of the Ping An Good Life App, a user sequentially adds a mask layer node and a pop-up container node within a 16-millisecond time window. The system merges these two nodes into a single child node addition event for processing. The system treats both the mask layer node and the pop-up container node as newly added nodes and obtains multi-dimensional feature information for each node. For example, for the pop-up container node, it obtains its class name (containing the keyword "modal"), z-index value (1001), size information (width 360 pixels, height 600 pixels), center point position (187, 406), background color (rgba(255,255,255,1)), and whether there are mask sibling nodes, among other feature information.

[0067] As another concrete example in the healthcare field, on the online consultation page of the Ping An Health App, a user sequentially adds a mask layer node and a pop-up container node within a 16-millisecond time window. The system merges these two nodes into a single child node addition event for processing. The system treats both the mask layer node and the pop-up container node as newly added nodes and obtains multi-dimensional feature information for each node. For example, for the pop-up container node, it obtains its class name (containing the keyword "dialog"), z-index value (2000), size information (width 340 pixels, height 500 pixels), center point position (180, 420), background color (rgba(255,255,255,0.95)), and whether there are mask sibling nodes, among other feature information.

[0068] Understandably, firstly, this embodiment processes multiple child node addition events detected within a preset time window into a single event, avoiding repeated judgments caused by the sequential addition of multiple DOM nodes in a short period. This reduces unnecessary computational resource consumption and prevents page lag caused by frequent triggering of judgment logic, thus improving system performance. Especially in some complex pages, the appearance of a single pop-up interface may trigger multiple DOM changes (such as sequentially adding multiple nodes such as overlays, pop-up containers, titles, content, and buttons). After merging the processing, only one judgment process is triggered, resulting in a significant performance improvement.

[0069] Second, this embodiment treats the nodes involved in the merged single event as newly added nodes, realizing an incremental update mechanism that "only processes the changed parts and does not scan all nodes," further reducing the amount of computation and improving processing efficiency.

[0070] Third, this embodiment, through dual optimization of merging processing and incremental updates, effectively reduces the system's computational load and response latency while ensuring the accuracy of pop-up interface recognition, thus improving the smoothness of user interaction. It is particularly suitable for interactive scenarios requiring frequent pop-up interfaces, such as insurance product consultations and online medical consultations.

[0071] Step S140: When the values ​​of feature information in multiple dimensions meet the preset conditions, the newly added node is determined to be a pop-up interface; Specifically, based on the feature information of multiple dimensions obtained in step S130, the value of the feature information of each dimension is evaluated to determine whether it meets the preset conditions. When the preset conditions are met, the newly added node is identified as a pop-up interface.

[0072] Optionally, preset conditions can be that the values ​​of one or more dimensions of feature information from multiple dimensions meet corresponding thresholds or matching rules. For example, when the z-index of a newly added node is higher than a preset z-index threshold, it can be determined as a pop-up interface; when the viewport size ratio of a newly added node is greater than a preset ratio threshold, it can be determined as a pop-up interface; when the class name of a newly added node contains pop-up semantic keywords, it can be determined as a pop-up interface. Feature information from multiple dimensions can be combined to improve the accuracy and reliability of the determination.

[0073] Optionally, a scoring mechanism can be used to determine the pop-up interface: assign corresponding scores or weights to the feature information of each dimension, calculate the comprehensive score, and determine the newly added node as a pop-up interface when the comprehensive score exceeds the preset score threshold.

[0074] As a concrete example in the property insurance sector, on the insurance product consultation page of the Ping An Good Life App, after the system obtains multi-dimensional feature information of a newly added node, it determines whether the values ​​of these feature information meet preset conditions. For example, if the system detects that the class name of the newly added node contains the keyword "modal," and its z-index value is 1001 (higher than the preset z-index threshold of 1000), and its viewport size ratio is 60% (higher than the preset ratio threshold of 50%), and the values ​​of the above feature information meet the preset conditions, then the system determines that the newly added node is a pop-up interface.

[0075] As another concrete example, in the healthcare field, on the online consultation page of the Ping An Health App, after the system obtains multi-dimensional feature information of a newly added node, it determines whether the values ​​of these feature information meet preset conditions. For example, if the system detects that the class name of the newly added node contains the keyword "dialog", and its z-index is 1500 (higher than the preset z-index threshold of 1000), and its viewport size ratio is 55% (higher than the preset ratio threshold of 50%), and the values ​​of the above feature information meet the preset conditions, then the system determines that the newly added node is a pop-up interface.

[0076] In some embodiments of the present invention, in step S140, when the values ​​of feature information in multiple dimensions meet preset conditions, the newly added node is determined to be a pop-up interface. The preset conditions meet at least two of the following conditions.

[0077] In practical applications, single-dimensional feature information may lead to misjudgment. For example, a navigation bar element may have a high z-index, but it is not a pop-up interface; an advertising pop-up may have the keyword "popup" in its class name, but its size is small, and it may just be a floating element on the page rather than a pop-up interface requiring user interaction. Therefore, this invention adopts a multi-dimensional comprehensive judgment method. Only when at least two of the feature information in multiple dimensions meet the corresponding conditions will a newly added node be judged as a pop-up interface, thereby reducing the false positive and false negative rates.

[0078] Specifically, after obtaining feature information from multiple dimensions of the newly added node, the value of each dimension's feature information is evaluated to determine whether it meets the corresponding conditions. When at least two dimensions meet the conditions, the newly added node is identified as a pop-up interface.

[0079] Optionally, the preset conditions can adopt a "at least two" judgment logic, meaning that if any two or more of the eight conditions are met, the pop-up interface is determined. In specific implementation, "at least two" can be adjusted to "at least three" or "at least four" according to the actual business scenario and accuracy requirements to further improve the accuracy of the judgment. The following provides a detailed explanation of each preset condition.

[0080] Preset condition 1: The class name of the newly added node contains the semantic keywords for pop-up windows; Specifically, in H5 page development practice, developers typically add identifying keywords with pop-up semantics to the class name (class attribute) of the root node of the pop-up interface to facilitate style control and semantic recognition. By obtaining the class attribute value of the newly added node and performing string matching or regular expression matching on that value, it can be detected whether it contains pop-up semantic keywords.

[0081] Optionally, the semantic keywords for pop-up windows may include, but are not limited to, at least one of the following: "modal", "dialog", "popup", "overlay", "toast", "drawer", "panel", "layer", "alert", "confirm", and "action-sheet".

[0082] Optionally, when performing class name matching, you can use exact matching, prefix matching, suffix matching, or inclusion matching. For example, when the class attribute value of a newly added node is "modal-content", inclusion matching can detect the keyword "modal".

[0083] Optionally, a negative filtering mechanism can be added during matching. If the class name contains keywords that are not pop-up semantics (such as "input", "nav", "header", "footer", "menu"), the weight of that dimension can be reduced or it can be excluded directly to avoid misjudging non-pop-up elements such as navigation bars and footers as pop-up interfaces.

[0084] As a concrete example in the property insurance sector, on the insurance product consultation page of the Ping An Good Life App, after a user clicks the "Product Details" button, an H5 page dynamically creates a pop-up node with the class attribute value of "modalproduct-detail". The system obtains the class attribute value of this newly added node and detects that it contains the pop-up semantic keyword "modal", thus the feature information of this dimension meets the preset conditions.

[0085] As another concrete example in the healthcare field, on the online consultation page of the Ping An Health App, after a user clicks the "Select Department" button, the H5 page dynamically creates a pop-up node with the class attribute value of "dialog department-selector". The system obtains the class attribute value of the newly added node and detects that it contains the semantic keyword "dialog", indicating that the feature information of this dimension meets the preset conditions.

[0086] Preset condition 2: The newly added node has a role attribute with pop-up semantics; In one possible implementation, the preset conditions include: the newly added node has a role attribute with pop-up semantics.

[0087] Specifically, in H5 page development practice, to improve webpage accessibility (Ally) and facilitate the identification of page element purposes by auxiliary tools such as screen readers, developers typically set the `role` attribute on the root node of a pop-up interface to declare the semantic role of that element. By obtaining the `role` attribute value of the newly added node, it can be detected whether the value corresponds to the semantic role type of the pop-up.

[0088] Optionally, the role attributes of pop-up semantics can include, but are not limited to: role="dialog" (dialog box) and role="alertdialog" (alert dialog box). The role attribute of "dialog" indicates that the element is a dialog box, which usually requires user interaction; the role attribute of "alertdialog" indicates that the element is an alert dialog box, used to convey important information to the user and require the user to respond.

[0089] Optionally, in addition to the role attribute, other relevant attributes of Accessible Rich Internet Applications (ARIA) can be detected, such as the aria-modal attribute. When aria-modal="true", it indicates that the element is a modal dialog box, further helping to determine that the node is a pop-up interface.

[0090] As a concrete example in the property insurance sector, on the insurance product consultation page of the Ping An Good Life App, after a user clicks the "Product Details" button, an H5 page dynamically creates a pop-up node with the role attribute value of "dialog". If the system detects that the role attribute value of this newly added node is "dialog", then the feature information for this dimension meets the preset conditions.

[0091] As another concrete example in the healthcare field, on the online consultation page of the Ping An Health App, after a user clicks the "Select Department" button, an H5 page dynamically creates a pop-up node with the role attribute value of "alertdialog". If the system detects that the role attribute value of this newly added node is "alertdialog", then the feature information for that dimension meets the preset conditions.

[0092] Preset condition 3: The stacking level value of the newly added node is higher than the preset level threshold; In one possible implementation, the preset conditions include: the z-index of the newly added node is higher than a preset level threshold.

[0093] Specifically, in Cascading Style Sheets (CSS), the z-index property controls the stacking order of elements along the Z-axis. Elements with higher z-index values ​​appear closer to the user visually and are more likely to float above other elements. Pop-up interfaces typically need to float above the page content, so their z-index values ​​are usually higher than those of regular page content. By retrieving the z-index value from the computed style of a newly added node, it can be determined whether it exceeds a preset layer threshold.

[0094] Optionally, the preset level threshold can be set according to the actual application scenario. In one possible implementation, the preset level threshold can be 1000. When the z-index value of a newly added node is greater than 1000, the feature information of that dimension is determined to meet the preset condition.

[0095] Optionally, if a newly added node does not explicitly set a z-index value, its z-index value may be the default value "auto" or an empty value. In this case, it can be considered as not meeting the condition, or a default low-level value (such as 0) can be assigned for comparison. In practice, the stacking order can also be determined by combining the position property (such as fixed, absolute, relative, sticky) of the newly added node.

[0096] As a concrete example in the property insurance sector, on the insurance product consultation page of the Ping An Good Life App, after a user clicks the "Product Details" button, an H5 page dynamically creates a pop-up node with a z-index value of 1001. The system retrieves the z-index value from the calculated style of this newly added node and determines that 1001 is greater than the preset level threshold of 1000, thus the feature information of this dimension meets the preset conditions.

[0097] As another concrete example, in the healthcare field, on the online consultation page of the Ping An Health App, after a user clicks the "Select Department" button, an H5 page dynamically creates a pop-up node with a z-index value of 2000. The system obtains the z-index value in the calculated style of this newly added node and determines that 2000 is greater than the preset level threshold of 1000, thus the feature information of this dimension meets the preset condition.

[0098] Preset condition 4: The viewport size ratio of the newly added node is greater than the preset ratio threshold; Among them, the viewport size percentage is the ratio of the size of the newly added node to the viewport size of the webpage view; Specifically, pop-up interfaces typically cover a large portion of the page to attract user attention or display complete content. Therefore, the size of the pop-up interface usually occupies a significant proportion of the viewport size. The system calculates the proportion of the newly added node's size to the viewport size by obtaining the node's dimensions (width and / or height) and the webpage's viewport dimensions (viewport width and / or viewport height), and determines whether this proportion exceeds a preset threshold.

[0099] Optionally, the viewport size proportions may include width proportions and / or height proportions. The width proportion is the ratio of the width of the newly added node to the viewport width; the height proportion is the ratio of the height of the newly added node to the viewport height. In one possible implementation, when the width proportion and / or height proportion are greater than a preset proportion threshold (e.g., 50%), it is determined that the feature information of that dimension meets the preset conditions.

[0100] Optionally, the preset percentage threshold can be set according to the actual application scenario. In one possible implementation, the preset percentage threshold can be 50%. When the width of a newly added node accounts for more than 50% of the viewport width, or the height accounts for more than 50% of the viewport height, the feature information of that dimension is determined to meet the preset conditions.

[0101] In practical implementation, the size information of newly added nodes can be obtained by calling the DOM API (such as getBoundingClientRect); the viewport size information can be obtained by accessing window.innerWidth and window.innerHeight. It should be noted that, since it may not be possible to completely guarantee that the size information of newly added nodes and viewports can be successfully obtained at all times during implementation, this embodiment of the invention does not limit the success rate and applicable scenarios of the above acquisition methods.

[0102] As a concrete example in the property insurance sector, on the insurance product consultation page of the Ping An Good Life App, after a user clicks the "Product Details" button, an H5 page dynamically creates a pop-up node. This pop-up node has a width of 360 pixels, a viewport width of 375 pixels (width ratio of 96%), and a height of 600 pixels, a viewport height of 812 pixels (height ratio of 74%). The system obtains the dimensions and viewport dimensions of this newly added node and calculates that the width ratio of 96% and the height ratio of 74% are both greater than the preset ratio threshold of 50%. Therefore, the feature information of this dimension meets the preset conditions.

[0103] As another concrete example in the healthcare field, on the online consultation page of the Ping An Health App, after a user clicks the "Select Department" button, the H5 page dynamically creates a pop-up node. This pop-up node has a width of 340 pixels, a viewport width of 375 pixels (width ratio of 91%), and a height of 500 pixels, a viewport height of 812 pixels (height ratio of 62%). The system obtains the dimensions and viewport dimensions of this newly added node and calculates that the width ratio of 91% and the height ratio of 62% are both greater than the preset ratio threshold of 50%, thus the feature information of this dimension meets the preset conditions.

[0104] Preset condition 5: The deviation between the center point of the newly added node and the viewport center point of the webpage view is less than the preset deviation threshold. Specifically, pop-up interfaces are typically placed in the center of the viewport (centered display) for easy user focus and operation. Therefore, the degree of deviation between the center point of a newly added node and the viewport center point can be used as one of the criteria for determining whether the node is a pop-up interface. The smaller the deviation, the closer the node is to the viewport center, and the more likely it is to be a pop-up interface. The system calculates the deviation between the coordinates of the center point of the newly added node and the coordinates of the viewport center point, and determines whether the deviation is less than a preset deviation threshold.

[0105] Optionally, the center point coordinates of a newly added node can be calculated using the coordinates of its top-left corner and dimensions. Specifically, the x-coordinate of the center point = top-left corner x-coordinate + width / 2, and the y-coordinate of the center point = top-left corner y-coordinate + height / 2. The viewport center point coordinates can be calculated using the viewport dimensions. Specifically, the x-coordinate of the viewport center point = viewport width / 2, and the y-coordinate of the viewport center point = viewport height / 2.

[0106] Optionally, the deviation can be represented using Euclidean distance, which is the straight-line distance between the center point and the viewport center point. When this distance is less than a preset deviation threshold, the feature information of this dimension is determined to meet the preset condition. In one possible implementation, the preset deviation threshold can be 10% of the viewport width or height. For example, when the deviation is less than 10% of the viewport width and less than 10% of the viewport height, the condition is determined to be met.

[0107] As a concrete example in the property insurance sector, on the insurance product consultation page of the Ping An Good Life App, after a user clicks the "Product Details" button, the H5 page dynamically creates a pop-up node. The center point coordinates of this pop-up node are (187, 406), and the viewport center point coordinates are (187.5, 406). The calculated Euclidean distance between the center point of this pop-up node and the viewport center point is approximately 0.5 pixels. The deviation is much smaller than the preset deviation threshold (e.g., 10% of the viewport width of 375 pixels, i.e., 37.5 pixels). Therefore, the feature information in this dimension meets the preset conditions.

[0108] As another concrete example in the healthcare field, on the online consultation page of the Ping An Health App, after a user clicks the "Select Department" button, the H5 page dynamically creates a pop-up node. The center point coordinates of this pop-up node are (180, 420), and the viewport center point coordinates are (187.5, 406). The calculated Euclidean distance between the center point of this pop-up node and the viewport center point is approximately 15.5 pixels. The deviation is less than a preset deviation threshold (e.g., 37.5 pixels), so the feature information of this dimension meets the preset conditions.

[0109] Preset condition 6: The newly added node contains interactive elements; Specifically, pop-up interfaces typically contain interactive controls for users, such as close buttons, OK buttons, cancel buttons, input boxes, selectors, and links. If a newly added node contains one or more interactive elements, it is more likely to be a pop-up interface. The presence of interactive elements is checked by traversing the descendant nodes of the newly added node to determine if this condition is met.

[0110] Optionally, interactive elements may include, but are not limited to, at least one of the following: buttons (button, input type="button" / "submit" / "reset"), links (a tags), input boxes (input, textarea), selection boxes (select), and close icons (elements containing the class name "close" or elements that trigger the close function when clicked).

[0111] Optionally, interactive elements can be detected through tag name matching, class name matching, role attribute matching, or event listener detection. For example, it can detect whether there are button, a, or input tags in the descendant nodes of a newly added node, or whether there are elements with role="button".

[0112] As a concrete example in the property insurance sector, on the insurance product consultation page of the Ping An Good Life App, after a user clicks the "Product Details" button, the H5 page dynamically creates a pop-up node. This pop-up node contains a close button (with the class name "close"), a "Buy Now" button (button label), and a "Later" button (button label). The system iterates through the descendant nodes of this newly added node, and if it detects the existence of interactive elements, then the feature information of that dimension meets the preset conditions.

[0113] As another concrete example in the healthcare field, on the online consultation page of the Ping An Health App, after a user clicks the "Select Department" button, the H5 page dynamically creates a pop-up node. This pop-up node contains a close icon (with the class name "close"), a "Confirm" button (labeled with "button"), and a "Cancel" button (labeled with "button"). The system iterates through the descendant nodes of this newly added node, and if it detects the existence of interactive elements, then the feature information of that dimension meets the preset conditions.

[0114] Preset condition 7: The background color of the newly added node has a non-transparent visual effect; Specifically, pop-up interfaces typically have semi-transparent or opaque background colors to visually distinguish them from the page background content and highlight the pop-up itself. By obtaining the background color attribute value from the calculated style of the newly added node and analyzing its transparency (alpha channel value), it can be determined whether it has a non-transparent visual effect.

[0115] Optionally, in CSS, background color can be represented using RGB or RGBA format. The RGBA format includes an alpha channel value to represent the color's transparency. The alpha channel value ranges from 0 (completely transparent) to 1 (completely opaque). When the alpha channel value is greater than 0, it indicates that the background color has a non-transparent visual effect (both completely opaque and semi-transparent are considered non-transparent). In one possible implementation, when the alpha channel value of the background color is greater than 0, it is determined that the feature information of that dimension meets a preset condition.

[0116] Optionally, if the newly added node does not have an explicitly set background color, its background color may be the default transparent background (alpha channel value of 0). In this case, you can further check whether the parent or sibling nodes of the newly added node have a masking background to assist in the judgment.

[0117] As a concrete example in the property insurance sector, on the insurance product consultation page of the Ping An Good Life App, after a user clicks the "Product Details" button, an H5 page dynamically creates a pop-up node with a background color of rgba(0,0,0,0.7) (a semi-transparent black background). The system retrieves the background color attribute from the calculated style of this newly added node and detects that the alpha channel value is 0.7 (greater than 0), thus the feature information of this dimension meets the preset conditions.

[0118] As another concrete example in the healthcare field, on the online consultation page of the Ping An Health App, after a user clicks the "Select Department" button, the H5 page dynamically creates a pop-up node with a background color of rgba(255,255,255,0.95) (a nearly opaque white background). The system retrieves the background color attribute from the calculated style of this newly added node and detects that the alpha channel value is 0.95 (greater than 0), thus the feature information of this dimension meets the preset conditions.

[0119] Preset condition 8: The newly added node has a masked style node with the same parent node as the newly added node.

[0120] Specifically, in the implementation of H5 pages, pop-up interfaces are usually accompanied by a mask layer (backdrop / overlay / mask) to obscure the page background content and allow the user to focus on the pop-up interface. This mask layer and the pop-up interface node typically share the same parent node in the DOM tree (i.e., they are sibling nodes). This condition is determined by checking if any of the sibling nodes of the newly added node possesses a mask style characteristic.

[0121] Optionally, a mask style node refers to a node that has one or more of the following style characteristics: its position style is fixed or absolute (position: fixed or position: absolute), its coverage area is the entire viewport (top:0, left:0, right:0, bottom:0, or width:100vw, height:100vh), its background color is semi-transparent (the alpha channel value of background-color is between 0.3 and 0.8), its stacking level is lower than or equal to the stacking level of the newly added node (the z-index value is less than or equal to the z-index value of the newly added node), and its node class name contains mask semantic keywords (such as one or more of "overlay", "mask", and "backdrop").

[0122] Optionally, when a newly added node has sibling nodes that satisfy the above masking style characteristics, the feature information of this dimension is determined to meet the preset conditions.

[0123] As a concrete example in the property insurance sector, on the insurance product consultation page of the Ping An Good Life App, after a user clicks the "Product Details" button, an H5 page dynamically creates a pop-up node. Among the sibling nodes of this pop-up node is a masking layer node. This masking layer node has a class attribute value of "modal-overlay", a position attribute of "fixed", covers the entire viewport, and has a background color of rgba(0,0,0,0.5). The system detects the existence of this masking layer node, and since it shares the same parent node as the newly created node, the feature information for this dimension meets the preset conditions.

[0124] As another concrete example in the healthcare field, on the online consultation page of the Ping An Health App, after a user clicks the "Select Department" button, the H5 page dynamically creates a pop-up node. Among the sibling nodes of this pop-up node is a masking layer node. This masking layer node has a class attribute value of "dialog-backdrop", a position attribute of "fixed", covers the entire viewport, and has a background color of rgba(0,0,0,0.6). If the system detects the existence of this masking layer node and that it shares the same parent node as the newly added node, then the feature information for this dimension meets the preset conditions.

[0125] It is understandable that, firstly, this embodiment uses multi-dimensional feature information to comprehensively determine whether a newly added node is a pop-up interface. Compared with a single-dimensional determination method (such as relying solely on class name matching), it significantly reduces the false positive rate (misjudging non-pop-up elements as pop-up interfaces) and the false negative rate (omitting pop-up elements), thereby improving the accuracy and reliability of pop-up recognition.

[0126] Second, the multi-dimensional feature information covers multiple aspects such as class name semantics, ARIA role attributes, stacking hierarchy, viewport size ratio, centering deviation, existence of interactive elements, background color transparency, and association of masked sibling nodes. It comprehensively evaluates the pop-up attributes of newly added nodes from different perspectives, making the judgment results more comprehensive and objective.

[0127] Step S150: When the newly added node is a pop-up interface, control the native input box to be hidden.

[0128] Specifically, when step S140 determines that the newly added node is a pop-up interface, it indicates that a pop-up interface requiring user attention has appeared on the H5 page. At this time, in order to avoid the pop-up interface obscuring or visually conflicting with the native input box, the system controls the native input box to perform a hiding operation, making the native input box invisible.

[0129] Optionally, hiding the native input field can be achieved by calling the native input field's hide method or setting its visibility property. For example, in Android, the native input field can be hidden by calling the `View.setVisibility(View.GONE)` method; in iOS, it can be hidden by setting the `hidden` property of the `UIView` to `YES`.

[0130] Optionally, the specific implementation of controlling the native input box to be hidden may vary depending on the operating system and framework, and this embodiment of the invention does not limit this.

[0131] Optionally, when the pop-up is closed or disappears, the system can also control the native input box to be displayed, restoring the native input box to its visible state.

[0132] As a concrete example in the property insurance sector, on the insurance product consultation page of the Ping An Good Life App, after a user clicks the "Product Details" button, a product details pop-up window appears on the H5 page. After the system determines this newly added node as a pop-up window through step S140, it controls the original input box at the bottom of the screen to be hidden, ensuring the product details pop-up window is fully displayed and not obscured by the original input box. After the user views the information in the product details pop-up window and closes it, the system controls the original input box to reappear.

[0133] As another concrete example in the healthcare field, on the online consultation page of the Ping An Health App, after a user clicks the "Select Department" button, an H5 page pops up a department selection pop-up window (pop-up interface). After the system determines this newly added node as a pop-up interface through step S140, it controls the original input box at the bottom of the screen to be hidden, ensuring the department selection pop-up window is fully displayed and not obscured by the original input box. After the user selects a department and closes the pop-up window, the system controls the original input box to reappear.

[0134] It is understandable that the embodiments of the present invention, through the architectural design of "overlaying a native input box layer on top of the webpage view," replace the approach of "fully native solutions" that require developing native components separately for different underlying operating systems. Since the native input box layer is overlaid on the webpage view, the H5 business code does not need to care about the differences in the underlying operating system. A single set of H5 code can run normally on various underlying operating systems and external deployment scenarios, fundamentally solving the problems of drastically increased development costs and loss of business flexibility that exist in fully native solutions.

[0135] Meanwhile, because the native input box layer is separated from the H5 business code, the H5 business logic can obtain the interactive experience of the native input box without modification, thus solving the problem of significant intrusion into H5 business logic. Furthermore, the native input box layer is displayed by default, ensuring that users can immediately see and use the bottom input box after entering the page.

[0136] This invention enables the system to automatically identify all types of pop-up interfaces by uniformly monitoring changes in the DOM tree. This achieves centralized management of pop-up recognition logic. All pop-up recognition is completed through the same set of monitoring and judgment mechanisms, eliminating the need to write different notification code for different types of pop-ups and fundamentally eliminating the problem of "dispersed logic".

[0137] This invention, through listening to the addition of child nodes in the document object model, automatically detects the appearance of pop-ups and intelligently identifies new nodes using a multi-dimensional feature scoring algorithm. This eliminates the need for developers to manually add any notification logic to the pop-up code. When a new node is determined to be a pop-up, the system automatically hides the native input box. The entire process is automated and does not rely on active calls from H5 business code, fundamentally eliminating the problems of "manual dependence and easy omissions."

[0138] Upon automatically recognizing the pop-up interface, this embodiment of the invention automatically controls the native input box to be hidden, ensuring the pop-up interface is fully displayed without being obscured by the native input box. This directly solves the problem of "input boxes failing to hide in time when pop-ups appear, causing interface overlap and abnormal interaction." Furthermore, in this embodiment, the hiding action is automatically triggered by the system after pop-up detection, avoiding the manual triggering of input box hiding by H5 business code in existing technologies. This automates the process, replacing manual notification. In addition, the overlay architecture of the native input box layer ensures that the "hide" operation only applies to the native layer, without affecting the rendering of the webpage view content, thus guaranteeing a consistent user experience.

[0139] To more intuitively illustrate the collaborative control effect of the native input box in the embodiments of the present invention, Figure 3a and Figure 3b This diagram illustrates the displayed and hidden states of a native input field. Figure 3a As shown, a webpage view 200 is displayed on the screen 100 of the terminal device. The webpage view 200 is used to load and render H5 pages. A native input box layer 300 is superimposed on the bottom area of ​​the webpage view 200. The native input box layer 300 contains a native input box 310. The native input box layer 300 is displayed by default. After the user enters the page, he / she can see and use the native input box 310 at the bottom of the screen.

[0140] like Figure 3bAs shown, when a new child node event is detected in the document object model of webpage view 200, and the new node is determined to be a pop-up interface 400 based on multiple dimensions of its feature information, the native input box layer 300 is hidden. At this time, the pop-up interface 400 is displayed on top of the webpage view 200. The pop-up interface 400 contains interactive elements 410 (such as a close button, a confirm button, a cancel button, etc.). The display area of ​​the webpage view 200 is covered by a mask layer 500, which has the same parent node as the pop-up interface 400; the native input box layer 300 is hidden. Figure 3b (Illustrated by a dotted line in the middle) The pop-up interface 400 is fully displayed and is not obscured by the native input box 310, thus avoiding visual conflict and interactive interference.

[0141] When the pop-up interface 400 disappears, the system controls the native input box layer 300 to be displayed again, and the user can continue to use the native input box 310 for input.

[0142] As a specific example, in the property and casualty insurance sector, such as Figure 3a As shown, when a user opens the insurance product consultation page of an insurance app, this is an H5 page displayed in the webpage view 200 on the terminal device's screen 100. A native input box layer 300 and its contained native input boxes 310 are overlaid at the bottom of the page. Figure 3b As shown, after the user clicks the "Product Details" button, the H5 page dynamically creates a product details pop-up node in the document object model. It captures the addition event of the child node through a listening mechanism, obtaining multi-dimensional feature information of the newly added node (such as the class name containing the keyword "modal," the z-index value being higher than a preset layer threshold, and the viewport size ratio being greater than a preset ratio threshold). It then determines this as a pop-up interface 400 and controls the native input box layer 300 to be hidden, while the product details pop-up is fully displayed. The user can click interactive elements 410 such as "Insure Now" or "Talk Later" within the pop-up. After the user closes the product details pop-up, the disappearance of the pop-up interface 400 is detected, and the native input box layer 300 is restored to display, allowing the user to continue entering new information in the native input box 310.

[0143] Understandable, Figure 3a and Figure 3b The interface layout, pop-up style, and control names shown are for illustrative purposes only. This embodiment of the invention does not limit the specific type of pop-up interface or the specific form of interface elements.

[0144] In some embodiments of the present invention, such as Figure 4 As shown, after step S150, the following steps are also included.

[0145] Step S161: Maintain the parent-child relationships between each newly added node that has been identified as a pop-up interface; Specifically, during the operation of an H5 page, there may be scenarios where pop-up interfaces are nested, that is, one pop-up interface pops up another pop-up interface (for example, clicking the "Coverage Scope" button in an insurance product details pop-up window will bring up a coverage scope pop-up window). In this case, there may be a hierarchical nesting relationship between the multiple nodes that are identified as pop-up interfaces in the DOM tree, with the child pop-up interface node located inside the DOM subtree of the parent pop-up interface node.

[0146] In order to identify and process all associated child pop-ups when the parent pop-up disappears, it is necessary to record the parent-child relationships between each newly added node that has been identified as a pop-up.

[0147] Optionally, the parent-child relationship can be determined as follows: When a newly added node is identified as a pop-up, the system checks the position of the new node in the DOM tree and detects whether there is a node among its ancestor nodes that has already been identified as a pop-up. If so, the ancestor node is identified as the parent pop-up, the new node is identified as the child pop-up, and the parent-child relationship is recorded.

[0148] Optionally, parent-child relationships can be stored using a map data structure. The key of the map can be a unique identifier for the parent popup (such as a node ID, node reference, or custom identifier), and the value can be a list of child popups. When a new popup is identified as a child popup, it is added to the child list of the corresponding parent popup. When a new popup is identified as a parent popup, a new entry is created for it in the map.

[0149] Optionally, a tree structure can be used to maintain the parent-child relationship between pop-up interfaces. Each pop-up interface node can store its parent node reference and child node list, thus forming a complete pop-up interface hierarchy tree.

[0150] Optionally, when maintaining the parent-child relationship, the hierarchy depth of each pop-up can be recorded so that when the parent pop-up disappears, the child pop-ups are closed in descending order of depth.

[0151] As a concrete example in the property insurance sector, when a user clicks the "Product Details" button on the insurance product consultation page of the Ping An Good Life App, a product details pop-up window (parent pop-up) appears on the H5 page. The system identifies this new node as a pop-up window and records it. When the user clicks the "Coverage Scope" button within this product details pop-up window, a coverage scope pop-up window (child pop-up) appears from within the product details pop-up window. After identifying this child pop-up window, the system checks its position in the DOM tree and finds that it is located within the DOM subtree of the product details pop-up window node. Therefore, the system identifies the product details pop-up window as the parent pop-up window and the coverage scope pop-up window as the child pop-up window, storing this parent-child relationship.

[0152] As another concrete example, in the healthcare field, when a user clicks the "Select Department" button on the online consultation page of the Ping An Health App, an H5 page pops up a department selection pop-up window (the parent pop-up). The system identifies this newly added node as a pop-up window and records it. When the user clicks the "More Departments" button in this department selection pop-up window, the H5 page then pops up a more departments pop-up window (a child pop-up window) from within the department selection pop-up window. After identifying this child pop-up window, the system checks its position in the DOM tree and finds that it is located within the DOM subtree of the department selection pop-up window node. Therefore, the system identifies the department selection pop-up window as the parent pop-up window and the more departments pop-up window as the child pop-up window, storing this parent-child relationship.

[0153] Step S162: When the parent pop-up is detected to have disappeared, determine all child pop-ups associated with the parent pop-up based on the maintained parent-child relationship. Specifically, when a user closes a pop-up (e.g., by clicking the close button, clicking the overlay, or performing a back operation), the DOM node of that pop-up is removed from the DOM tree, or its style properties change (e.g., display changes from block to none, visibility changes from visible to hidden), causing the pop-up to visually disappear. When the system detects that the parent pop-up has disappeared, it searches for all child pop-ups associated with that parent pop-up based on the parent-child relationships maintained in step S161.

[0154] Optionally, the method for detecting the disappearance of the parent pop-up may include, but is not limited to, at least one of the following: listening to the child node removal event of the DOM tree (triggered when the parent pop-up node is removed from the DOM tree), listening to the attribute change event of the DOM tree (triggered when the display property of the parent pop-up node changes from non-none to none, or the visibility property changes from visible to hidden), or periodically checking the visibility status of the pop-up node through polling.

[0155] Optionally, when the parent pop-up is detected to have disappeared, the system searches for the entry with the parent pop-up as the key in the parent-child relationship mapping table maintained in step S161, and obtains a list of all its corresponding child pop-ups.

[0156] Optionally, if there are multiple levels of nesting between pop-up interfaces (e.g., a parent pop-up contains child pop-ups, and a child pop-up contains grandchild pop-ups), a recursive traversal can be used to search for all descendant pop-ups, starting from the parent pop-up and proceeding downwards, not just limiting them to direct children. Specifically, first, all direct child pop-ups of the parent pop-up are searched, and then for each child pop-up, its child pop-ups are recursively searched until the lowest-level descendant pop-ups are found.

[0157] As a concrete example in the property insurance sector, the Ping An Good Life App's insurance product consultation page contains a product details pop-up (parent pop-up) with a coverage scope pop-up (child pop-up) nested within it. When a user clicks the close button in the upper right corner of the product details pop-up, the H5 page removes the DOM node of that pop-up from the DOM tree. The system detects the disappearance of the product details pop-up (parent pop-up) by listening for child node removal events in the DOM tree. Subsequently, based on the parent-child relationship mapping table maintained in step S161, it finds the child pop-up associated with that parent pop-up, which is the coverage scope pop-up.

[0158] As another concrete example, in the healthcare field, the online consultation page of the Ping An Health App contains a pop-up window for selecting departments (the parent pop-up) that nests a pop-up window for more departments (the child pop-up). When a user clicks the close button in the upper right corner of the department selection pop-up, the H5 page removes the DOM node of that pop-up from the DOM tree. The system detects the disappearance of the department selection pop-up (parent pop-up) by listening for child node removal events in the DOM tree. Subsequently, based on the parent-child relationship mapping table maintained in step S161, it finds the child pop-up associated with that parent pop-up as the "more departments" pop-up.

[0159] Step S163: Control the native input box to execute the display state corresponding to the disappearance of all child pop-up interfaces.

[0160] Specifically, when the parent pop-up disappears, if its associated child pop-ups still exist in the DOM tree (i.e., the pop-ups are not automatically closed when the user closes the parent pop-up), these child pop-ups, although still active in the DOM tree, may be invisible or floating on the page and unresponsive because the parent pop-up has disappeared. This prevents the native input box from returning to its correct display state (because the system may still believe that a pop-up exists and keep the native input box hidden). In this case, all associated child pop-ups need to be closed, and the native input box should be restored to its displayed state.

[0161] Optionally, controlling the native input box to perform the display state corresponding to the disappearance of all child pop-ups can include: closing all child pop-ups associated with the parent pop-up and switching the native input box from a hidden state to a displayed state.

[0162] Optionally, closing the child pop-up can be done by removing the child pop-up node from the DOM tree, or by setting the child pop-up's display property to none or its visibility property to hidden.

[0163] Optionally, after closing all child pop-ups, you can further check if any other pop-ups exist on the current page. If no other pop-ups exist, control the display of the native input box; if other pop-ups exist (e.g., pop-ups other than the parent pop-up), keep the native input box hidden and wait for all other pop-ups to disappear before restoring its display.

[0164] As a concrete example in the property insurance sector, on the insurance product consultation page of the Ping An Good Life App, when a user clicks the close button of the product details pop-up (the parent pop-up), the system detects the disappearance of the parent pop-up and identifies the associated child pop-up as the coverage scope pop-up. The system automatically closes the coverage scope pop-up (removing the node from the DOM tree). At this point, all pop-ups have disappeared, and the system restores the original input fields from a hidden state to a visible state, allowing the user to continue entering new questions in the bottom input fields.

[0165] As another concrete example in the healthcare field, on the online consultation page of the Ping An Health App, when a user clicks the close button of the department selection pop-up (the parent pop-up), the system detects the disappearance of the parent pop-up and determines that the associated child pop-up is the "More Departments" pop-up. The system automatically closes the "More Departments" pop-up (removing the node from the DOM tree). At this point, all pop-ups have disappeared, and the system restores the original input fields from a hidden state to a visible state, allowing the user to continue entering a description of their symptoms in the bottom input fields.

[0166] It is understandable that, firstly, by maintaining the parent-child relationship between each newly added node that has been identified as a pop-up interface, this embodiment can accurately record the hierarchical nesting relationship between pop-up interfaces, providing data structure support for subsequent cascading closure, and avoiding management chaos caused by the inability to know the relationship between pop-up interfaces.

[0167] Second, when the parent pop-up disappears, this embodiment determines all child pop-ups associated with the parent pop-up based on the maintained parent-child relationship. It can automatically identify all child pop-ups that become unreachable or inoperable due to the parent pop-up being closed, thus achieving automatic discovery and location of child pop-ups.

[0168] Third, this embodiment avoids the "ghost pop-up" problem caused by the child pop-up remaining after the parent pop-up is closed by controlling the native input box to execute the display state corresponding to the disappearance of all child pop-up interfaces. It ensures that the native input box can be displayed in time after all pop-up interfaces disappear, preventing the native input box from being unusable due to abnormal pop-up interface states, and improving the reliability of the system and the continuity of user operation.

[0169] In some embodiments of the present invention, such as Figure 5 As shown, the collaborative control method for the bottom input box also includes the following steps.

[0170] Step S172: Detect the focus state change of the webpage input box in the webpage view; Specifically, the webpage view contains webpage input boxes (i.e., input elements or textarea elements in H5 pages), which users can click to enter text, numbers, or other information. When a user clicks on a webpage input box, the input box gains focus, indicating that the user is currently interacting with it; when the user clicks on an area outside the webpage input box or switches to another input box, the input box loses focus.

[0171] The system detects changes in the focus state of web page input boxes by listening to their focus and blur events.

[0172] Optionally, detecting focus state changes of webpage input boxes can include binding focus event listeners and focus loss event listeners to all webpage input box elements in the webpage view. A focus event is triggered when any webpage input box gains focus; a focus loss event is triggered when any webpage input box loses focus.

[0173] Optionally, if the input boxes in the web page view are dynamically created or destroyed (e.g., input boxes dynamically generated in a pop-up interface), event delegation can be used to uniformly listen for focus and defocus events on the root node of the web page view, avoiding binding event listeners to each input box separately, thus improving the efficiency and reliability of event management.

[0174] As a concrete example, in the property insurance sector, when a user clicks the input box at the bottom of the insurance product consultation page on the Ping An Good Life App, preparing to enter "What is the coverage of Ping An Million Medical Insurance?", the system detects that the input box on the webpage has gained focus.

[0175] In another scenario, after a user completes their input, they click on a blank area of ​​the page or click the send button, and the system detects that the input box on the webpage has lost focus.

[0176] As another concrete example, in the healthcare field, when a user clicks the input box at the bottom of the online consultation page on the Ping An Health App, preparing to type "I've been feeling dizzy and weak lately, which department should I consult?", the system detects that the input box has gained focus. In another scenario, after the user completes the input, they click on a blank area of ​​the page or click the send button, and the system detects that the input box has lost focus.

[0177] In some embodiments of the present invention, such as Figure 6 As shown, step S172 includes the following steps.

[0178] Step S1721: Maintain the global focus state value; Specifically, a global focus state value is maintained to represent the overall focus state of all webpage input boxes in the current webpage view. This global focus state value is a numeric variable whose value represents the number of webpage input boxes currently in focus.

[0179] The global focus state value can be implemented using a global variable, the lifecycle of which is related to the webpage view. Figure 1 This global variable can be stored in the context of the webpage view or in the global state management module in system memory.

[0180] Optionally, the initial value of the global focus state can be set to a preset baseline value (such as 0), indicating that no web page input box is currently in focus.

[0181] Optionally, the global focus state value can be of integer type or other data type capable of numerical operations. In one possible implementation, an integer variable is used as the global focus state value.

[0182] As a specific example, in the property insurance sector, when the insurance product consultation page of Ping An Good Life App finishes loading, the system maintains a global focus state value. The initial value of this value is 0, indicating that none of the web page input boxes on the current page have received focus.

[0183] As another concrete example, in the healthcare field, when the online consultation page of Ping An Health App finishes loading, the system maintains a global focus state value. The initial value of this value is 0, indicating that none of the web page input boxes on the current page have received focus.

[0184] Step S1722: In response to any webpage input box gaining focus, increase the global focus state value by a preset increment value; Specifically, when any webpage input box gains focus, the global focus state value is increased by a preset increment.

[0185] The preset increment value is a pre-defined numerical value used to incrementally update the global focus state value when a webpage input box gains focus. In one possible implementation, the preset increment value can be 1. When any webpage input box gains focus, the global focus state value increases by 1, indicating that the number of webpage input boxes currently in focus has increased by one.

[0186] Optionally, the preset increment value can be any positive integer (such as 1, 2, 3, etc.), and in practical applications it is usually set to 1 so that the global focus state value can accurately reflect the number of web page input boxes that are currently in focus.

[0187] As a concrete example in the property insurance sector, on the insurance product consultation page of the Ping An Good Life App, when a user clicks on the bottom input box, the input box gains focus. In response to this focus event, the system increments the global focus value from 0 to 1, indicating that one input box is currently in focus.

[0188] As another concrete example in the healthcare field, on the online consultation page of the Ping An Health App, when a user clicks on the bottom input box, that input box gains focus. In response to this focus event, the system increments the global focus value from 0 to 1, indicating that one input box is currently in focus.

[0189] Step S1723: In response to any webpage input box losing focus, the global focus state value is reduced by a preset increment value; Specifically, when any webpage input box loses focus, the global focus state value is reduced by a preset increment.

[0190] The preset increment value is the same as the preset increment value in step S1722. In one possible implementation, the preset increment value is 1. When any webpage input box loses focus, the global focus state value decreases by 1, indicating that the number of webpage input boxes currently in focus has decreased by one.

[0191] Optionally, if the value of the global focus state after reduction is less than the preset baseline value (such as 0), it can be reset to the preset baseline value to avoid negative abnormal values.

[0192] As a concrete example in the property insurance sector, on the insurance product consultation page of the Ping An Good Life App, after a user completes their input in the bottom input box and then clicks on a blank area of ​​the page, the input box loses focus. In response to this focus loss event, the system decreases the global focus value from 1 to 0, indicating that no input box is currently in focus.

[0193] As another concrete example in the healthcare field, on the online consultation page of the Ping An Health App, after a user completes their input in the bottom input box and then clicks on a blank area of ​​the page, the input box loses focus. In response to this focus loss event, the system decrements the global focus value from 1 to 0, indicating that no input box is currently in focus.

[0194] Step S1724: When the global focus state value is a preset baseline value, determine that all web page input boxes in the web page view are in a completely out-of-focus state. In the completely out-of-focus state, the native input box will be controlled to display.

[0195] Specifically, the current value of the global focus status is continuously monitored. When the global focus status value equals the preset baseline value, it indicates that no web page input box is currently in focus, meaning that all web page input boxes have lost focus.

[0196] The preset baseline value is a reference value used to determine a completely out-of-focus state. In one possible implementation, the preset baseline value can be 0. When the global focus state value is 0, it means that no web page input box is currently in focus, and the system determines it to be in a completely out-of-focus state.

[0197] When the system determines that the input box is completely out of focus, it triggers the operation to display the native input box. It should be noted that triggering the display of the native input box in a completely out-of-focus state is only performed under certain conditions (e.g., no pop-up interface currently exists). In practical applications, it is also necessary to combine this with the pop-up detection rules in steps S110 to S140 for a comprehensive judgment. Specifically, if a pop-up interface exists on the current page, the native input box should remain hidden even if the webpage input box is completely out of focus; only when no pop-up interface exists and the webpage input box is completely out of focus will the native input box be controlled to display. The existence of a pop-up interface can be determined by checking whether there are active pop-up interface nodes in the pop-up interface parent-child relationship graph maintained in step S161.

[0198] Optionally, the determination of a completely out-of-focus state can also be combined with other auxiliary conditions. For example, before determining a completely out-of-focus state, it can be checked whether there is a pop-up interface on the current page. If a pop-up interface exists, even if the global focus state value is a preset baseline value, the native input box will not be triggered to display, so as to avoid visual conflict between the pop-up interface and the native input box.

[0199] In one possible implementation, a page contains two input fields, one for username and one for password. When the user clicks the username field and the other for password, the system responds by changing the global focus value from 0 to 1 when the username field gains focus, and from 1 to 2 when the password field gains focus. At this point, both input fields are in focus. If the user clicks on a blank area of ​​the page, the password field loses focus, and the global focus value changes from 2 to 1. If the user clicks on a blank area again, the username field loses focus, and the global focus value changes from 1 to 0. At this point, the global focus value equals a preset baseline value (0), and the system determines that the input field is completely out of focus, triggering the display of the native input field.

[0200] As a concrete example in the property insurance sector, the Ping An Good Life App's insurance product consultation page currently lacks a pop-up interface. The user previously clicked the webpage input box at the bottom, with a global focus value of 1. After completing the input, the user clicked a blank area of ​​the page, causing the webpage input box to lose focus, and the global focus value changed from 1 to 0. The system detected that the global focus value was the preset baseline value of 0, determined that all webpage input boxes were completely out of focus, and triggered the native input box to be displayed, allowing the user to continue using the recording function of the native input box.

[0201] As another concrete example in the healthcare field, on the online consultation page of the Ping An Health App, there is currently no pop-up interface. The user previously clicked the webpage input box at the bottom, and the global focus value was 1. After the user completed their input, they clicked on a blank area of ​​the page, causing the webpage input box to lose focus, and the global focus value changed from 1 to 0. The system detected that the global focus value was the preset baseline value of 0, determined that all webpage input boxes were completely out of focus, and triggered the native input box to be displayed, allowing the user to continue using the recording function of the native input box.

[0202] It is understandable that, firstly, this embodiment maintains a global focus state value, abstracting the focus state of multiple complex and ever-changing input boxes into a single quantifiable value, which enables the representation of the entire focus state of the webpage view with extremely low computational cost, avoiding complex DOM traversal and state query.

[0203] Secondly, this embodiment increases or decreases the global focus state value in response to any input box gaining or losing focus, thereby achieving real-time synchronous updates of the focus state value and ensuring that the global focus state value can always accurately reflect the overall focus state of all web page input boxes.

[0204] Third, this embodiment determines the state as completely out of focus when the global focus state value is a preset benchmark value. It replaces the complex logic of checking the focus state of each input box one by one with a single numerical comparison, which simplifies the detection process of the completely out of focus state and improves the detection efficiency. At the same time, it can accurately identify the state of multiple input boxes being completely out of focus, avoiding misjudgment caused by some input boxes being in focus.

[0205] Fourth, when a user switches focus between different input boxes, the global focus state value always remains greater than the preset benchmark value (such as 1 or above), and the judgment of a completely out-of-focus state will not be triggered during the switching process. This avoids the flickering and state jump problems of the native input box during focus switching, and improves the smoothness of user interaction and the stability of the interface state.

[0206] In some embodiments of the present invention, such as Figure 7 As shown, before step S172, the following steps are also included.

[0207] Step S1711: In response to the triggering of a focus event, the detection of a focus state change is postponed to the microtask queue for execution. Specifically, during the operation of a webpage view, certain operations may generate momentary focus events. However, these focus events are not triggered by the user's actual input, but rather by internal browser mechanisms or system-level operations. For example, when a user clicks the "copy" button, the browser may create a temporary input box at the underlying level to perform the copy operation. This temporary input box immediately gains focus after its creation and is immediately removed after the copy is completed. This process triggers focus and defocus events, but these focus events do not reflect the user's actual input intent. If the system responds to such focus events, it will cause the native input box to incorrectly appear or disappear, a problem known as "false focus."

[0208] To address this issue, embodiments of the present invention postpone the detection of focus state changes to be performed in a microtask queue.

[0209] In JavaScript's event loop mechanism, tasks are divided into macrotasks and microtasks. Macrotasks include the entire script, `setTimeout`, `setInterval`, etc., while microtasks include `Promise.then()`, `MutationObserver` callbacks, etc. The microtask queue is executed after the current macrotask has completed and before the next macrotask begins. Leveraging this characteristic, when a focus event is triggered, the focus state change detection is not executed immediately. Instead, the detection logic is placed in the microtask queue, ensuring that the detection logic is invoked after all synchronous code in the current macrotask has finished executing and before the next macrotask begins.

[0210] Optionally, the detection of focus state changes can be deferred to the microtask queue, which can be achieved using the `Promise.resolve().then()` method. Specifically, in the callback function of a focus event (such as a focus or blur event), create a `Promise.resolve()`, and execute the focus state change detection logic in its `then` method. This way, the detection logic will be placed in the microtask queue and executed after the synchronous code of the current macrotask has finished executing.

[0211] Optionally, by deferring the detection of focus state changes to the microtask queue, the browser has sufficient time to process all synchronous operations in the current macro task before the detection is executed, including the creation, gaining focus, losing focus, and removal of temporary input boxes. When the detection logic is executed in the microtask queue, the DOM node of the temporary input box has already been removed, thus accurately identifying whether the input box corresponding to the focus event still actually exists.

[0212] As a concrete example in the property insurance sector, a user clicks a "copy" button on the insurance product consultation page of the Ping An Good Life App to copy the current insurance product code. The browser's underlying layer creates a temporary input box, which immediately gains focus upon creation and is immediately removed after copying. In response to this focus event, the system postpones the detection of the focus state change to the microtask queue, rather than immediately detecting the focus event.

[0213] As another concrete example in the healthcare field, a user clicks a "copy" button on the online consultation page of the Ping An Health App to copy the current consultation number. The browser's underlying layer creates a temporary input box, which immediately gains focus upon creation and is immediately removed after copying. In response to this focus event, the system postpones the detection of the focus state change to the microtask queue, rather than immediately detecting the focus event.

[0214] Step S1712: In the microtask queue, check whether the web page input box that generated the focus event is connected to the document object model of the web page view. Specifically, when the detection logic is executed in the microtask queue, the system obtains a reference to the web page input box that generated the focus event and checks whether the input box is still connected to the document object model of the web page view.

[0215] "Connection status" refers to whether the input box node on the webpage still exists as part of the DOM tree. In the Document Object Model (DOM), each node has an `isConnected` property, which indicates whether the node is connected to the DOM tree. If the node is currently connected to the DOM tree, the `isConnected` property is true; if the node has been removed from the DOM tree, the `isConnected` property is false.

[0216] Optionally, you can check whether the input field that generated the focus event is still connected to the document object model of the webpage view by accessing the isConnected property of the input field node. If the isConnected property is true, it means that the input field still exists in the DOM tree; if the isConnected property is false, it means that the input field has been removed from the DOM tree.

[0217] Optionally, in addition to checking the `isConnected` property, you can further check whether the input box's parent node exists, whether the input box is set to a hidden style (such as `display:none` or `visibility:hidden`), etc., to comprehensively determine whether the input box is truly valid. In one possible implementation, you can check both the `isConnected` property and the element's visibility state (e.g., by getting the `display` and `visibility` property values ​​using `getComputedStyle`). If `isConnected` is false, or `display` is none, or `visibility` is hidden, then it can be determined that the input box is not in a valid connection state.

[0218] As a concrete example in the property insurance sector, on the insurance product consultation page of the Ping An Good Life App, a user clicking the "Copy" button triggers a focus event on a temporary input box. When the system executes the focus state change detection logic in the microtask queue, it checks whether the temporary input box that generated the focus event is still connected to the Document Object Model. Since the temporary input box has been removed from the DOM tree after copying, its isConnected property is false, therefore the system detects that the input box is not connected.

[0219] As another concrete example in the healthcare field, on the online consultation page of the Ping An Health App, a user clicking the "Copy" button triggers a focus event on a temporary input box. When the system executes the focus state change detection logic in the microtask queue, it checks whether the temporary input box that generated the focus event is still connected to the Document Object Model. Since the temporary input box has been removed from the DOM tree after copying, its isConnected property is false, so the system detects that the input box is not connected.

[0220] Step S1713: If the web page input box is not in a connected state, discard the focus event.

[0221] Specifically, when step S1712 detects that the input box that generated the focus event is not in a connected state (i.e., the input box has been removed from the DOM tree), it indicates that the focus event is a "fake focus" generated by a temporary input box, and not the user's real input intention. At this time, the focus event is discarded, and subsequent focus state change detection and native input box visibility control are not triggered.

[0222] "Discarding a focus event" means that the system does not process the focus event as a valid change in the focus state of the input box, that is, it will not trigger "controlling the native input box to hide" in step S173 or "controlling the native input box to show" in step S174. For discarded focus events, the system neither changes the display state of the native input box nor updates the focus state value.

[0223] Optionally, if step S1712 detects that the input box that generated the focus event is still in a connected state (i.e., the input box actually exists in the DOM tree), the system will process the focus event normally, treat it as a valid focus state change, and trigger the subsequent native input box show / hide control logic.

[0224] Optionally, discarding a focus event can be done silently, meaning no user-visible feedback is generated and no logs are recorded; or it can be done non-silently, meaning a debug log is generated when the focus event is discarded, so as to track the occurrence of fake focus events during the development or testing phase.

[0225] As a concrete example in the property insurance sector, on the insurance product consultation page of the Ping An Good Life App, after a user clicks the "copy" button, the system detects in the microtask queue that the temporary input box that generated the focus event is not in a connected state, and therefore discards the focus event. The display state of the original input box does not change, and the user does not perceive any flickering or abnormal interface changes.

[0226] As another concrete example in the healthcare field, on the online consultation page of the Ping An Health App, after a user clicks the "copy" button, the system detects in the microtask queue that the temporary input box that generated the focus event is not in a connected state, and therefore discards the focus event. The display state of the original input box does not change, and the user does not perceive any flickering or abnormal interface changes.

[0227] It is understandable that, firstly, this embodiment delays the detection of focus state changes to the microtask queue, so that the detection logic is executed only after all synchronous operations of the current macro task (including the creation, gaining focus, and removal of temporary input boxes) are completed. This allows for an accurate distinction between focus events generated by the user's actual input behavior and focus events generated by temporary system operations, thus solving the problem of false focus caused by browser internal mechanisms or system-level operations.

[0228] Secondly, this embodiment detects whether the input box that generates a focus event remains connected to the document object model in the micro-task queue, and uses the isConnected property of the DOM node to objectively verify the authenticity of the focus event, providing a reliable and efficient method for identifying false focus, thus avoiding complex event source tracing and intent inference.

[0229] Third, the focus event is discarded when the input box is not in a connected state, which avoids unnecessary switching of the native input box due to false focus events, prevents poor user experience such as flickering and accidental switching caused by false focus, improves the smoothness of user interaction and the stability of the interface state, and also avoids the inconsistency between the native input box state and the actual state of the page caused by false focus.

[0230] Step S173: When the webpage input box is detected to have focus, control the native input box to hide. Specifically, when a user clicks on the webpage input box to begin typing, the webpage input box gains focus. At this point, the native input box is hidden. The purpose of hiding the native input box is to avoid visual overlap and interaction conflicts caused by both input boxes (the webpage input box and the native input box) existing simultaneously at the bottom of the screen. When the webpage input box gains focus, the user will use it to input (via the system keyboard), and the native input box does not need to be displayed, therefore it is hidden.

[0231] Optionally, the specific implementation for hiding the native input field can vary depending on the operating system and framework. For example, in Android, the native input field can be hidden by calling the `View.setVisibility(View.GONE)` method; in iOS, it can be hidden by setting the `hidden` property of the `UIView` to `YES`.

[0232] Optionally, when controlling the native input box to be hidden, other UI elements associated with the native input box (such as the record button, quick reply button, etc.) can also be hidden at the same time to maintain the consistency and cleanliness of the interface.

[0233] Optionally, before controlling the native input field to be hidden, you can first check if the native input field is currently displayed. If the native input field is already hidden, the hiding operation can be skipped to avoid repeated execution.

[0234] As a concrete example in the property insurance sector, when a user clicks on the bottom input box of the insurance product consultation page on the Ping An Good Life App, the system detects that the webpage input box has gained focus and hides the native input box at the bottom of the screen. At this point, the user can see that the webpage input box has gained focus and the system keyboard pops up, while the native input box is no longer visible, resulting in a clean interface without any overlapping input boxes.

[0235] As another concrete example in the healthcare field, when a user clicks on the input box at the bottom of the online consultation page on the Ping An Health App, the system detects that the webpage input box has gained focus and hides the native input box at the bottom of the screen. At this point, the user can see that the webpage input box has gained focus and the system keyboard pops up, while the native input box is no longer visible, resulting in a clean interface without any overlapping input boxes.

[0236] Step S174: When it is detected that the web page input box has lost focus, control the native input box to be displayed.

[0237] Specifically, when a user completes input and clicks on a blank area of ​​the page, switches to another input box, or performs other operations that cause the web page input box to lose focus, the system controls the native input box to be displayed, restoring the native input box to its visible state.

[0238] Optionally, the specific implementation for controlling the display of the native input field can vary depending on the operating system and framework. For example, in Android, the native input field can be displayed by calling the View.setVisibility(View.VISIBLE) method; in iOS, it can be displayed by setting the hidden property of the UIView to NO.

[0239] Optionally, before controlling the display of the native input box, it is also possible to check whether there are any pop-up interfaces on the current page. If there are still pop-up interfaces on the current page, the native input box should remain hidden and should not be restored. Only when all pop-up interfaces have disappeared and the web page input box loses focus should the native input box be controlled to display. The determination of the existence of pop-up interfaces can be achieved by checking whether there are still active pop-up interface nodes in the pop-up interface parent-child relationship graph maintained in step S161, or by traversing the DOM tree to determine whether there are currently nodes that are determined to be pop-up interfaces. The checking method here can be based on the pop-up window determination logic used in step S140 (such as checking whether there are nodes that are determined to be pop-up interfaces in the DOM tree), or other methods can be used. This embodiment of the invention does not limit the specific implementation method.

[0240] Optionally, when controlling the display of the native input box, other UI elements associated with the native input box (such as the record button, quick reply button, etc.) can also be restored at the same time to maintain the consistency and cleanliness of the interface.

[0241] As a concrete example in the property insurance sector, when a user enters "What is the coverage of Ping An Million Medical Insurance?" in the bottom webpage input box on the Ping An Good Life App's insurance product consultation page, and then clicks on a blank area of ​​the page, the system detects that the webpage input box has lost focus and restores the display of the native input box at the bottom of the screen. The user can then see the native input box reappear at the bottom of the screen and continue to use its recording or quick input functions.

[0242] As another concrete example, in the healthcare field, on the online consultation page of the Ping An Health App, after a user enters "I've been feeling dizzy and weak lately" in the bottom webpage input box and clicks on a blank area of ​​the page, the system detects that the webpage input box has lost focus and restores the native input box at the bottom of the screen. The user can then see the native input box reappear at the bottom of the screen and continue to use its recording or quick input functions.

[0243] It is understandable that, firstly, this embodiment can detect changes in the focus state of the webpage input box in real time to sense whether the user is interacting with the webpage input box, providing an accurate trigger signal for subsequent native input box control.

[0244] Secondly, when the web page input box gains focus, this embodiment automatically controls the native input box to be hidden, avoiding visual overlap and interaction conflicts caused by the simultaneous display of the web page input box and the native input box at the bottom of the screen. This allows users to focus on text input through the web page input box and the system keyboard, improving the smoothness of the input experience and the cleanliness of the interface.

[0245] Third, when the webpage input box loses focus, this embodiment automatically controls the native input box to be displayed, restoring the usability of the native input box and allowing users to continue to use the recording, quick reply and other functions of the native input box, realizing seamless switching between different input methods.

[0246] Fourth, the control path based on focus state changes and the control path based on pop-up detection work together to form a complete collaborative control system. When a pop-up appears, the native input box is hidden; after the pop-up disappears, the native input box is displayed based on the focus state; when the web page input box gains focus, the native input box is hidden; and when the web page input box loses focus, the native input box is displayed again without a pop-up. This allows the display state of the native input box to adaptively respond to different types of state changes on the page, further improving the consistency of the user experience and the intelligence level of the system.

[0247] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0248] It should be noted that if any AI models, software tools, or components not belonging to this company appear in the embodiments of this invention, they are merely illustrative examples and do not represent actual use. All user personal information involved in the embodiments of this invention is authorized (with the knowledge and consent) by the relevant parties or fully authorized by all parties, and the executing entity may obtain it through various legal and compliant means. The collection, storage, use, processing, transmission, provision, and disclosure of the information, data, and signals involved all comply with relevant laws and regulations and do not violate public order and good morals.

[0249] In one embodiment, a collaborative control device for a bottom input box is provided, which corresponds one-to-one with the collaborative control method for the bottom input box in the above embodiments. For example... Figure 8As shown, the collaborative control device for the bottom input box includes an overlay module 810, an event listening module 820, an information acquisition module 830, a node determination module 840, and a control module 850. Detailed descriptions of each functional module are as follows: The overlay module 810 is used to overlay a native input box layer on the bottom area of ​​the webpage view when the webpage view is in the display state, wherein the native input box layer is in the display state by default and the native input box layer contains native input boxes; The event listening module 820 is used to listen for new child node events in the document object model of the webpage view; The information acquisition module 830 is used to acquire multi-dimensional feature information of the newly added node when the newly added child node event is detected. The node determination module 840 is used to determine the newly added node as a pop-up interface when the values ​​of the feature information of the multiple dimensions meet the preset conditions. The control module 850 is used to control the native input box to be hidden when the newly added node is the pop-up interface.

[0250] In one embodiment, the preset condition satisfies at least two of the following: The class name of the newly added node contains pop-up semantic keywords; The newly added node has a role attribute with pop-up semantics; The stacking level value of the newly added node is higher than the preset level threshold; The viewport size ratio of the newly added node is greater than a preset ratio threshold, wherein the viewport size ratio is the proportion of the size of the newly added node to the viewport size of the webpage view; The deviation between the center point of the newly added node and the viewport center point of the webpage view is less than a preset deviation threshold. The newly added node contains interactive elements; The background color of the newly added node has a non-transparent visual effect; The newly added node has a masked style node that has the same parent node as the newly added node.

[0251] In one embodiment, the information acquisition module 830 is specifically used for: Multiple child node addition events detected within a preset time window are merged into a single child node addition event for processing. The nodes of the document object model involved in the single child node addition event are taken as the newly added nodes, and the feature information of the nodes in multiple dimensions is obtained.

[0252] In one embodiment, the control module 850 is further configured to: Maintain the parent-child relationships between each of the newly added nodes that have been identified as part of the pop-up interface; When the parent pop-up is detected to have disappeared, all child pop-ups associated with the parent pop-up are determined based on the maintained parent-child relationship. Control the native input box to execute the display state corresponding to the disappearance of all child pop-up interfaces.

[0253] In one embodiment, the control module 850 is further configured to: Detect changes in the focus state of the webpage input box in the webpage view; When the webpage input box is detected to have focus, the native input box is controlled to be hidden. When the webpage input box is detected to have lost focus, the native input box is controlled to be displayed.

[0254] In one embodiment, the control module 850 is further configured to: In response to the triggering of a focus event, the detection of the focus state change is postponed to the execution of the microtask queue; In the microtask queue, it is detected whether the web page input box that generated the focus event is still connected to the document object model of the web page view; If the webpage input box is not in a connected state, then the focus event is discarded.

[0255] In one embodiment, the control module 850 is further configured to: Maintain global focus state value; In response to any webpage input box gaining focus, the global focus state value is increased by a preset increment value; In response to any webpage input box losing focus, the global focus state value is reduced by the preset increment value; When the global focus state value is a preset baseline value, it is determined that all the web page input boxes in the web page view are in a completely out-of-focus state, wherein the completely out-of-focus state will trigger the control of the native input box to perform display.

[0256] It is understandable that the embodiments of the present invention, through the architectural design of "overlaying a native input box layer on top of the webpage view," replace the approach of "fully native solutions" that require developing native components separately for different underlying operating systems. Since the native input box layer is overlaid on the webpage view, the H5 business code does not need to care about the differences in the underlying operating system. A single set of H5 code can run normally on various underlying operating systems and external deployment scenarios, fundamentally solving the problems of drastically increased development costs and loss of business flexibility that exist in fully native solutions.

[0257] Meanwhile, because the native input box layer is separated from the H5 business code, the H5 business logic can obtain the interactive experience of the native input box without modification, thus solving the problem of significant intrusion into H5 business logic. Furthermore, the native input box layer is displayed by default, ensuring that users can immediately see and use the bottom input box after entering the page.

[0258] This invention enables the system to automatically identify all types of pop-up interfaces by uniformly monitoring changes in the DOM tree. This achieves centralized management of pop-up recognition logic. All pop-up recognition is completed through the same set of monitoring and judgment mechanisms, eliminating the need to write different notification code for different types of pop-ups and fundamentally eliminating the problem of "dispersed logic".

[0259] This invention, through listening to the addition of child nodes in the document object model, automatically detects the appearance of pop-ups and intelligently identifies new nodes using a multi-dimensional feature scoring algorithm. This eliminates the need for developers to manually add any notification logic to the pop-up code. When a new node is determined to be a pop-up, the system automatically hides the native input box. The entire process is automated and does not rely on active calls from H5 business code, fundamentally eliminating the problems of "manual dependence and easy omissions."

[0260] Upon automatically recognizing the pop-up interface, this embodiment of the invention automatically controls the native input box to be hidden, ensuring the pop-up interface is fully displayed without being obscured by the native input box. This directly solves the problem of "input boxes failing to hide in time when pop-ups appear, causing interface overlap and abnormal interaction." Furthermore, in this embodiment, the hiding action is automatically triggered by the system after pop-up detection, avoiding the manual triggering of input box hiding by H5 business code in existing technologies. This automates the process, replacing manual notification. In addition, the overlay architecture of the native input box layer ensures that the "hide" operation only applies to the native layer, without affecting the rendering of the webpage view content, thus guaranteeing a consistent user experience.

[0261] Based on the collaborative control method of the bottom input box mentioned above, such as Figure 9 As shown in the diagram, this embodiment of the invention also provides a structural schematic of an apparatus for a collaborative control method for a bottom input box. The apparatus includes a processor 91 and a memory 92 coupled to the processor 91. The memory 92 stores a computer program, which, when executed by the processor 91, causes the processor 91 to perform the steps of the collaborative control method for the bottom input box described in the above embodiment.

[0262] For further details regarding the implementation of the above technical solution by the processor 91 in the device for the collaborative control method steps of the bottom input box, please refer to the description in the collaborative control method of the bottom input box provided in the above embodiments of the invention, which will not be repeated here.

[0263] The processor 91 can also be called a CPU (Central Processing Unit). The processor 91 may be an integrated circuit chip with signal processing capabilities. The processor 91 can also be a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor, or the processor 91 can be any conventional processor.

[0264] like Figure 10 As shown in the diagram, this embodiment of the invention also provides a schematic diagram of a computer-readable storage medium, on which a readable computer program 101 is stored. The computer program 101 can be stored in the storage medium in the form of a software product, including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in various embodiments of the invention. The aforementioned storage medium includes various storage media capable of storing program code, such as USB flash drives, portable hard drives, magnetic disks or optical disks, ROM (Read-Only Memory), RAM (Random Access Memory), or terminal devices such as computers, servers, mobile phones, and tablets.

[0265] In the embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, apparatuses, or modules, and may be electrical, mechanical, or other forms.

[0266] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0267] Furthermore, the functional modules in the various embodiments of this invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium.

[0268] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product.

[0269] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available storage medium that a computer can store or a data storage device such as a server or data center that integrates one or more available storage media. The available storage medium may be a magnetic storage medium (e.g., floppy disk, hard disk, magnetic tape, optical storage medium) or a semiconductor storage medium (e.g., SSD (solid state disk)).

[0270] The technical solution provided by the present invention has been described in detail above. Specific examples have been used to illustrate the principle and implementation of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation and application scope based on the idea of ​​the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

[0271] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, etc.) containing computer-usable program code.

[0272] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0273] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0274] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0275] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A collaborative control method for a bottom input box, characterized in that, include: When the webpage view is displayed, a native input box layer is overlaid in the bottom area of ​​the webpage view. The native input box layer is displayed by default and contains native input boxes. Listen for new child node events in the document object model of the webpage view; When a new child node event is detected, the feature information of the new node in multiple dimensions is obtained; When the values ​​of the feature information in the multiple dimensions meet the preset conditions, the newly added node is determined to be a pop-up interface; When the newly added node is the pop-up interface, control the native input box to be hidden.

2. The collaborative control method for the bottom input box according to claim 1, characterized in that, The preset conditions must satisfy at least two of the following: The class name of the newly added node contains pop-up semantic keywords; The newly added node has a role attribute with pop-up semantics; The stacking level value of the newly added node is higher than the preset level threshold; The viewport size ratio of the newly added node is greater than a preset ratio threshold, wherein the viewport size ratio is the proportion of the size of the newly added node to the viewport size of the webpage view; The deviation between the center point of the newly added node and the viewport center point of the webpage view is less than a preset deviation threshold. The newly added node contains interactive elements; The background color of the newly added node has a non-transparent visual effect; The newly added node has a masked style node that has the same parent node as the newly added node.

3. The collaborative control method for the bottom input box according to claim 1, characterized in that, When a new child node event is detected, the system obtains multi-dimensional feature information of the newly added node, including: Multiple child node addition events detected within a preset time window are merged into a single child node addition event for processing. The nodes of the document object model involved in the single child node addition event are taken as the newly added nodes, and the feature information of the nodes in multiple dimensions is obtained.

4. The collaborative control method for the bottom input box according to claim 1, characterized in that, After controlling the native input box to hide when the newly added node is the pop-up interface, the method further includes: Maintain the parent-child relationships between each of the newly added nodes that have been identified as part of the pop-up interface; When the parent pop-up is detected to have disappeared, all child pop-ups associated with the parent pop-up are determined based on the maintained parent-child relationship. Control the native input box to execute the display state corresponding to the disappearance of all child pop-up interfaces.

5. The collaborative control method for the bottom input box according to claim 1, characterized in that, Also includes: Detect changes in the focus state of the webpage input box in the webpage view; When the webpage input box is detected to have focus, the native input box is controlled to be hidden. When the webpage input box is detected to have lost focus, the native input box is controlled to be displayed.

6. The collaborative control method for the bottom input box according to claim 5, characterized in that, Before detecting the focus state change of the webpage input box in the webpage view, the method further includes: In response to the triggering of a focus event, the detection of the focus state change is postponed to the execution of the microtask queue; In the microtask queue, it is detected whether the web page input box that generated the focus event is still connected to the document object model of the web page view; If the webpage input box is not in a connected state, then the focus event is discarded.

7. The collaborative control method for the bottom input box according to claim 5, characterized in that, The detection of focus state changes in the webpage input box in the webpage view includes: Maintain global focus state value; In response to any webpage input box gaining focus, the global focus state value is increased by a preset increment value; In response to any webpage input box losing focus, the global focus state value is reduced by the preset increment value; When the global focus state value is a preset baseline value, it is determined that all the web page input boxes in the web page view are in a completely out-of-focus state, wherein the completely out-of-focus state will trigger the control of the native input box to perform display.

8. A collaborative control device for a bottom input box, characterized in that, include: An overlay module is used to overlay a native input box layer on the bottom area of ​​a webpage view when the webpage view is in display mode. The native input box layer is in display mode by default and contains native input boxes. The event listener module is used to listen for new child node events in the document object model of the webpage view; The information acquisition module is used to acquire multi-dimensional feature information of the newly added node when the newly added child node event is detected. The node determination module is used to determine the newly added node as a pop-up interface when the values ​​of the feature information of the multiple dimensions meet the preset conditions. The control module is used to control the native input box to be hidden when the newly added node is the pop-up interface.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the collaborative control method for the bottom input box as described in any one of claims 1 to 7.

10. A computer-readable medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the collaborative control method for the bottom input box as described in any one of claims 1 to 7.