A multi-database-based intelligent report pushing system and method

Abstract

The present application relates to the field of medical informatization, in particular to a kind of intelligent report pushing system and method based on multi-database.The present application is linked and cooperated through five modules of multi-database connection, parameterized SQL script engine, intelligent task scheduling, message pushing and encrypted storage, realizes the complete closed loop of "query plan generation-parameterized execution-result pushing-state feedback".Core innovation includes: task dependency arrangement adopts DAG topological sorting+semaphore concurrent control, parameterized SQL variable replacement adopts pre-compiled anti-injection mechanism, failure retry adopts exponential backoff strategy+ fuse mechanism.After testing, the system successfully replaces more than 90% of repetitive manual operations, data flow time is shortened from an average of 2 hours to 5 minutes, task execution success rate reaches 99.8%, meets the requirements of three-level security protection.

Description

Technical Field

[0001] This invention relates to the field of medical informatics, specifically to an automated medical data push system and method based on multi-database connections, parameterized SQL scripts, and intelligent task scheduling. It is particularly suitable for data acquisition, processing, and precise push from core systems such as hospital HIS (Hospital Information System), EMR (Electronic Medical Record System), LIS (Laboratory Information System), and PACS (Picture Archiving and Communication System). Background Technology

[0002] 1.1 Existing Technology Search Instructions: The search was conducted in the State Intellectual Property Office's Patent Publication Announcement System and Google Patents using keywords such as "multi-database connection," "intelligent report push," "parameterized SQL," "task scheduling," "WeChat push for enterprises," "medical data integration," and "hospital report system." The published text and bibliographic information of some entries were verified using Google Patents pages.

[0003] Technical Direction 1: Generation of Multi-Source Data Reports 1. Multi-source data intelligent analysis report generation technology (existing technology). Technical solution: Existing technologies include multi-source data intelligent analysis report generation solutions, encompassing functional modules such as semantic analysis, data modeling, visualization generation, differential configuration management, permission management, and custom components. Application scenario: Unified modeling and fully automated updating of multi-source data. Limitations: Emphasizes semantic analysis and visualization generation, but does not address multi-database connection management, task dependency orchestration, or failure retry mechanisms. 2. Multi-Source Data Integration Intelligent Reporting Technology (Existing Technology). Technical Solution: Existing technologies include intelligent report generation solutions based on multi-source data integration. These solutions achieve accurate semantic alignment and fusion of heterogeneous multi-source data through a semantic anchor system, constructing a knowledge graph for in-depth analysis. Application Scenarios: Commercial data processing and artificial intelligence technology fields. Limitations: Focuses on AI analysis and knowledge graph construction; does not involve parameterized SQL scripts, task scheduling, or WeChat push notifications. 3. Multi-database report generation technology (existing technology). Technical solution: Existing technologies offer various report generation solutions based on multiple databases, supporting multiple database types and using a unified interface for data querying and report generation. Application scenario: Enterprise-level reporting systems. Limitations: Only basic data querying and report generation are implemented; task scheduling, message push, and encrypted storage functions are lacking, thus failing to form a complete closed loop.

[0004] Technical Direction 2: Message Push System 1. CN119676305A (Kangjian Information Technology (Shenzhen) Co., Ltd.). Technical Solution: A method for information push based on WeChat for Enterprise, which generates user tags and user profiles by obtaining basic contact information and user behavior information, and sends product information in bulk based on a preset information push strategy. Application Scenario: Product information push field. Limitations: Focuses on user profiling and product push, and does not involve multi-database connections and task scheduling; 2. Message Push and Status Tracking Technology (Existing Technology). Technical Solution: Existing technologies include various intelligent message push systems that support multi-channel message push and status tracking. Application Scenarios: General message push. Limitations: Lack of a linkage mechanism with database operations, making it impossible to achieve intelligent push based on data query results.

[0005] Technical Direction 3: Task Scheduling System 1. CN111666326B (Prior Art). Technical Solution: An ETL scheduling method that automates the scheduling of data extraction, transformation, and loading. Application Scenarios: Data warehousing and big data processing. Limitations: Only involves data scheduling; it does not cover multi-database connection management, parameterized SQL scripts, or message push. 2. Distributed Task Scheduling Technology (Existing Technology). Technical Solution: Several distributed task scheduling systems exist, supporting task dependency configuration and failure retries. Application Scenarios: Distributed computing. Limitations: Lacks a coordination mechanism with SQL script engines, and cannot trigger message pushes based on task execution results.

[0006] Technology Direction 4: Medical Data Integration System 1. Hospital Multi-System Data Integration Technology (Existing Technology). Technical Solution: Existing technologies include various hospital multi-system data integration platforms to achieve data interface and unified management of systems such as HIS, EMR, and LIS. Application Scenario: Hospital data integration. Limitations: Focuses on data integration and storage, and does not involve intelligent report generation, task scheduling, and message push.

[0007] Search Summary: Existing technologies are mostly single-point solutions, and a complete system has yet to organically integrate multi-database connection management, parameterized SQL script engines, intelligent task scheduling (including dependency orchestration and failure retries), precise message push, and encrypted storage, and achieve inter-module linkage and collaboration. Existing technologies suffer from the following technical bottlenecks: 1. Existing multi-database connection solutions only implement connection management and do not integrate with the task scheduling system, making it impossible to dynamically call database connections according to task requirements; 2. Existing parameterized SQL solutions only implement variable substitution, do not coordinate with the task scheduling system to generate query plans, and lack syntax validation and performance optimization; 3. The existing task scheduling scheme only implements task execution and does not link with the SQL script engine and message push module, thus failing to achieve a complete closed loop of "query-execution-push"; 4. Existing message push solutions only send messages and do not link with task execution status, thus failing to provide real-time feedback on push status and unread reminders; 5. Existing encrypted storage solutions only encrypt data and are not integrated with connection management and task scheduling systems, thus failing to enable dynamic retrieval of encrypted connection information.

[0008] 1.2 Shortcomings of existing technologies In current hospital management, the extraction, aggregation, and distribution of data heavily rely on manual operations, leading to the following problems: 1. Inefficiency: Manual data collection is cumbersome and time-consuming, consuming a significant amount of manpower. For example, a secondary-level hospital needs to assign 3-5 dedicated personnel daily to data aggregation and report preparation, consuming approximately 2 hours per person per day. 2. Information lag: Reports are usually issued on a T+1 basis, which prevents managers from obtaining daily operational data in a timely manner, affecting the timeliness of decision-making; 3. Limited coverage: It is difficult to achieve precise point-to-point push notifications, important information is easily missed, and the information reach rate is only 60-70%; 4. Lack of security: Database connection information is stored in plaintext, posing a security risk and failing to meet the national Level 3 information security protection requirements; 5. Lack of coordination mechanism: The existing system's modules operate independently, making it impossible to achieve coordinated data collection, processing, and push, resulting in process breaks and data inconsistencies.

[0009] While existing technologies include multi-database connection schemes, parameterized SQL script schemes, task scheduling schemes, and message push schemes, these schemes operate independently, and a complete system that organically integrates these functions and achieves inter-module linkage has yet to emerge. In particular, existing schemes lack the following key functions: task dependency orchestration (DAG topology sorting + semaphore concurrency control) and automatic retry mechanisms for failures (exponential backoff + circuit breaking) in intelligent task scheduling; automatic syntax validation, performance analysis, and pre-compiled anti-injection mechanisms for SQL scripts; secure encrypted storage and key management of database connection information (AES-256-GCM + HSM); dynamic grouping, time window control, and unread notification functions for message push; and a linkage and collaboration mechanism between modules to achieve a complete closed loop of "query plan generation - parameterized execution - result push - status feedback."

[0010] Therefore, the present invention aims to solve the above problems and provide a fully functional, safe and reliable intelligent report push system and method with modular linkage. Summary of the Invention Technical issues

[0011] This invention aims to solve the problems of traditional hospital data flow, such as reliance on manual labor, low efficiency, information lag, limited coverage, and data security, and to achieve a shift from "people searching for data" to "data finding people." Specifically, it includes: 1. Solves the problems of complex management and low connection efficiency for multiple database connections; 2. Address the issues of lack of syntax validation, inflexible variable substitution, and SQL injection risks during SQL script writing and execution; 3. Resolves issues related to the lack of dependency orchestration (DAG topology sorting) in task scheduling and the inability to automatically retry after failure (exponential backoff + circuit breaker); 4. Resolves issues such as the inability to accurately locate recipients, lack of push status feedback, and unread notifications in message push notifications; 5. Resolve the issue of plaintext storage of database connection information, which poses a security risk; 6. Solve the problem of the lack of linkage and coordination mechanisms between modules, which prevents the formation of a complete closed loop. Technical solution

[0012] This invention proposes an intelligent report push system based on multiple databases, comprising the following core modules: 1. Multi-database connection module: Supports unified configuration and management of multiple databases such as SQL Server, MySQL, Oracle, and Kingbase, providing connection testing, encrypted storage, and connection pool management functions. The connection pool management unit sets the configurable maximum number of connections, minimum number of idle connections, and connection timeout, effectively improving connection reuse efficiency; 2. SQL Script Engine Module: Supports parameterized SQL template definition and dynamic variable substitution, automatically injecting actual parameters at runtime. Dynamic variable tags use the format [[variable name]] and support multiple variable types such as date, numeric, string, and enumeration. The SQL syntax verification unit checks the syntax correctness of the SQL statement before execution and provides error location and correction suggestions. Parameterized SQL variable substitution uses regular expression matching + type-safe conversion + pre-compilation mechanism to effectively prevent SQL injection attacks; 3. Intelligent Task Scheduling Module: Supports both timed triggering and interval triggering, and supports task dependency orchestration. Timed triggering allows setting fixed execution times by hour or minute, and supports multiple time slots; interval triggering allows setting cyclic execution intervals by hour or minute. Task dependency orchestration uses a directed acyclic graph (DAG) topology sorting algorithm for task ordering, combined with a semaphore mechanism to achieve concurrency control, ensuring the orderly execution of complex processes such as cleaning followed by aggregation. When a task fails, an exponential backoff strategy is used for automatic retrying, with a maximum number of retries and a circuit breaker threshold set. A notification is sent to the administrator when the maximum number of retries is reached or the circuit breaker is triggered. The intelligent task scheduling module also supports task priority settings, allowing high-priority tasks to preempt the execution resources of low-priority tasks. 4. Message Push Module: Integrates with WeChat Work API, supporting dynamic subscriptions by role, department, or individual, and displays key indicators using card-style message templates. It supports multiple message types including text, images, and cards. Card messages support titles, descriptions, images, and jump links. It supports real-time feedback on push status and unread reminders. In case of push failure, it automatically retryes using the same exponential backoff strategy as the intelligent task scheduling module. The message push module also supports setting message sending time windows, allowing users to configure the time periods during which messages can be sent to avoid sending messages outside of working hours. 5. Encrypted Storage Module: Employs AES-256-GCM encryption to encrypt and save database connection information, generating an encrypted result containing encrypted data, random numbers, and authentication tags. The key management unit is responsible for key generation, storage, and rotation, and supports Hardware Security Module (HSM) protection.

[0013] Core collaboration mechanism: The core innovation of this invention lies in the linkage and collaboration between the modules, and the specific implementation method is as follows: 1. Task Scheduling and SQL Engine Collaboration: The intelligent task scheduling module dynamically generates parameterized SQL query plans based on task dependencies and sends these plans to the SQL script engine module. The SQL script engine module automatically replaces variable markers and executes the SQL query based on the query plan, returning the query results to the intelligent task scheduling module. This collaborative mechanism ensures the orderly execution of query tasks and the correct transmission of parameters. 2. Task Scheduling and Message Push Integration: The intelligent task scheduling module triggers the message push module based on task execution results. The message push module works in tandem with the intelligent task scheduling module to provide real-time feedback on push status and unread notifications. This linkage mechanism ensures the timeliness and accuracy of message pushes. 3. Encrypted Storage and Connection Management Integration: Connection information from multiple database connection modules is encrypted by the encrypted storage module and then dynamically retrieved by the intelligent task scheduling module during task execution. This integrated mechanism ensures the secure storage and dynamic use of database connection information.

[0014] This invention offers the following advantages: Increased efficiency: Actual testing shows it replaces over 92% of repetitive manual operations, freeing up personnel for high-value analysis. For example, in a secondary-level hospital, data aggregation and report creation time was reduced from 2 hours to 5 minutes after implementation, saving approximately 83.3% in labor costs; Improved timeliness: The time from the generation of key operational data to its reach of decision-makers was reduced from an average of 2 hours to within 5 minutes, with a task execution success rate of 99.8%, ensuring managers receive operational data promptly; Expanded coverage: It enables the automatic generation and distribution of refined and personalized reports, increasing information reach from 60-70% to over 99%; Enhanced security: Database connection information is encrypted using AES-256-GCM, meeting national Level 3 security protection requirements and effectively protecting sensitive data; Flexible configuration: It supports multiple databases, triggering methods, and push channels, adapting to various needs. Meeting the hospital's IT needs; Task priority scheduling: By setting task priorities and a preemption mechanism, critical business tasks are ensured to be executed first, improving system responsiveness; Message push time window: By setting message sending time windows, users are avoided from being disturbed during non-working hours, improving user experience; Module synergy: The linkage and collaboration mechanism between modules realizes a complete closed loop of "query plan generation - parameterized execution - result push - status feedback". Compared with the existing single-point solutions, the overall system efficiency is improved by more than 38%, and the task execution success rate is improved by more than 5%; Enhanced security protection: The parameterized SQL variable replacement adopts a pre-compilation mechanism, effectively preventing SQL injection attacks and significantly improving system security. Attached Figure Description

[0015] Figure 1 System overall architecture diagram.

[0016] Figure 2 Flowchart of multi-database connection.

[0017] Figure 3 Intelligent task scheduling flowchart.

[0018] Figure 4 Message push flowchart.

[0019] Figure 5 : Schematic diagram of task dependency orchestration DAG topology sorting. Detailed Implementation

[0020] Implementation Environment This invention can run in a server environment, supports multiple operating system platforms, and can be implemented using multiple programming languages. It requires the following basic functional components: a database connection management component for establishing and managing connections with multiple databases; a task scheduling component for scheduling and executing timed and interval tasks; an encryption component for implementing data encryption and decryption functions; a network request component for calling external instant messaging platform APIs; and a log management component for recording system operation logs.

[0021] Implementation steps

[0022] 1. Database connection configuration: Users select the database type (SQL Server, MySQL, Oracle, Kingbase, etc.) in the data source configuration interface; enter the server address, port number, database name, username, and password; the system automatically detects connection connectivity, and the test timeout is configurable; after the test passes, the connection information is encrypted and stored using AES-256-GCM encryption mode; the connection information is added to the connection pool management, and the maximum number of connections, minimum number of idle connections, and connection timeout can be set.

[0023] 2. SQL Script Configuration: Users enter business SQL in the report configuration interface, supporting variable marking (e.g., [[b_date]] represents the start date, [[e_date]] represents the end date); the system supports three query methods: data tables, views, and stored procedures; the SQL syntax verification unit automatically verifies the syntax correctness of SQL statements, providing error location and correction suggestions; configure parameter mapping relationships, set default values ​​and data types for variables; the system uses regular expression matching + type safe conversion + pre-compilation mechanism for variable replacement to prevent SQL injection attacks.

[0024] 3. Task Scheduling Configuration: Select triggering method: timed triggering or interval triggering; Set execution time (timed triggering) or interval time (interval triggering), supporting multi-time period timing; Configure task dependencies, using the Directed Acyclic Graph (DAG) topology sorting algorithm to set the execution order dependencies between tasks, supporting parallel and serial task orchestration; Set task priorities, high-priority tasks can preempt the execution resources of low-priority tasks; Set failure retry strategy: adopt an exponential backoff strategy, set the maximum number of retries (default 3 times) and circuit breaker threshold (default 5 failures within 5 minutes trigger circuit breaker).

[0025] 4. Message Template Configuration: Configure instant messaging platform message templates in the message template configuration interface; set recipient parameters (user ID, department ID, tag ID), supporting dynamic group push; configure message title and content variables, supporting multiple message types such as text, images, and cards; set message sending time windows to avoid sending messages outside of working hours.

[0026] 5. Instant Messaging Platform API Configuration: Configure application credential information; the system automatically obtains and caches access tokens, and the caching time is configurable; configure message sending frequency limits to avoid triggering API call restrictions.

[0027] 6. Task Execution and Message Push: The system executes tasks based on configured trigger conditions. The intelligent task scheduling module dynamically generates parameterized SQL query plans based on task dependencies. The SQL script engine module uses regular expression matching, type-safe conversion, and pre-compilation mechanisms to automatically replace variables and execute queries. Query results are cached to improve the efficiency of repeated queries. The intelligent task scheduling module triggers the message push module based on the task execution results. The message push module assembles and sends message content, supporting data formatting and conditional rendering. The push status is recorded, and unread users are reminded. When a push fails, an exponential backoff strategy is used to automatically retry. Thread pool technology is used to improve concurrent execution capabilities, combined with a semaphore mechanism to implement task concurrency control.

[0028] 7. Log Viewing and Analysis: Users can query push records based on multiple dimensions such as date, task type, and status; view detailed information such as push time, content, and recipient; the system provides task execution statistical analysis, including execution success rate, average execution time, and reasons for failure.

[0029] Algorithm Flow Description The core algorithm flow of this invention includes the following key steps:

[0030] Step S1: Database connection encryption storage algorithm. The AES-256-GCM symmetric encryption algorithm is used to encrypt the database connection information, generating an encrypted result containing encrypted data, a random number (nonce), and an authentication tag (tag); the encryption process is as follows: (1) Generate a 12-byte random number (nonce); (2) Encrypt the connection information using the key and the nonce; (3) Generate a 16-byte authentication tag (tag); (4) Combine and store the encrypted data, the nonce, and the tag; during decryption, the same key and the nonce are used for decryption verification, and the authentication tag is verified to ensure data integrity and authenticity; the key is uniformly managed by the key management unit and supports periodic rotation, and the key storage is protected by a hardware security module (HSM).

[0031] Step S2: Parameterized SQL Variable Replacement Algorithm. Iterate through the variable markers in the SQL template (in the format [[variable name]]), matching the variable markers using regular expressions; perform type-safe conversion based on variable type: convert date types to database date format, numeric types to their corresponding numeric types, string types are escaped, and enumeration types are validated for validity; execute the SQL statement using a prepared statement mechanism, passing the variables as parameters to effectively prevent SQL injection attacks; perform SQL syntax validation before replacement to ensure the generated SQL statement conforms to syntax rules.

[0032] Step S3: Task scheduling algorithm. Timed trigger mode: Trigger task execution at a specified time according to the configured execution time (hour, minute); Interval trigger mode: Trigger task execution cyclically according to the configured interval time; Task dependency orchestration: Use the directed acyclic graph (DAG) topology sorting algorithm to sort tasks. The specific steps are: (1) Construct a task dependency graph; (2) Calculate the in-degree of each node; (3) Add nodes with an in-degree of 0 to the queue; (4) Take out the queue nodes one by one for execution and decrement the in-degree of their adjacent nodes by 1; (5) Repeat steps (3)-(4) until all nodes have been executed; Combine semaphore mechanism to implement concurrency control and set the maximum concurrency limit to limit the number of tasks executed at the same time; Task priority scheduling: Allocate execution resources according to task priority. High priority tasks can preempt the execution resources of low priority tasks. Use priority queues to implement task scheduling.

[0033] Step S4: Failure Retry Algorithm. When a task fails, the failure status and error information are automatically recorded. An exponential backoff strategy is used for automatic retries. Specifically, the retry interval for the nth time = base interval × (2^n), where the base interval is configurable (default 1 minute), and the maximum retry interval is configurable (default 30 minutes). A maximum number of retries is set (default 3 times). If the maximum number of retries is reached and the task still fails, a notification is sent to the administrator. A circuit breaker threshold is set (default 5 failures within 5 minutes trigger circuit breaker). After the circuit breaker is triggered, task execution is paused. The circuit breaker duration is configurable (default 30 minutes). During the circuit breaker period, the task will not be executed. Task execution will resume after the circuit breaker ends.

[0034] Step S5: Message Push Algorithm. Assemble the message content based on the configured message template and recipient parameters; determine if the current time is within the allowed sending time window; if not, delay sending until the start of the time window; call the instant messaging platform API to send the message, supporting various message types such as text, images, and cards; record the push status and remind unread users; automatically retry using the same exponential backoff strategy as in Step S4 if the push fails.

[0035] Step S6: Result caching algorithm. The task execution results are cached for a configurable time. When the same task is executed again within the cache's validity period, the cached result is returned directly. If the cache expires or is manually cleared, the task is re-executed and the cache is updated.

[0036] Experimental data This invention underwent a 6-month practical test at a Grade II Class A hospital. The test data are as follows (the specific data has been anonymized for confidentiality reasons): Data aggregation time: 2 hours / day before implementation, 5 minutes / day after implementation, a reduction of 95.8%; Labor costs: Before implementation, it was 3-5 people / day; after implementation, it was 0.5 people / day, a saving of 83.3%. Information reach rate: 60-70% before implementation, over 99% after implementation, an increase of 41.4%; Task execution success rate: 95% before implementation, 99.8% after implementation, an improvement of 4.8%; Data flow timeliness: Before implementation, it was an average of 2 hours; after implementation, it was within 5 minutes, a reduction of 95.8%.

[0037] Data Description: The testing period was 6 months; the testing environment was the production environment of a Class II Grade A hospital's information system; the data source was the concurrent business data provided by the hospital's information department, which was anonymized and used to verify the effectiveness of this invention; the testing method adopted the before-and-after comparison method, with the data before implementation being manually collected data for the 6 months prior to the system's launch, and the data after implementation being automatically collected data for the 6 months following the system's launch.

[0038] Test results show that the present invention effectively solves the problem of traditional hospital data transfer relying on manual labor, significantly improves work efficiency and information timeliness, and ensures data security.

Claims

1. A smart report push system and method based on multiple databases, characterized in that, include: The multi-database connection module is used to configure and manage multiple database connections, supports multiple database types, and adopts a connection pool management mechanism to improve connection reuse efficiency. The SQL script engine module supports parameterized SQL template definition, dynamic variable tag setting, parameter mapping configuration, and runtime dynamic parameter injection. It also supports automatic verification of SQL syntax correctness and provides error location and correction suggestions. The intelligent task scheduling module supports two task triggering methods: scheduled triggering and interval triggering. It also supports task dependency orchestration to ensure the orderly execution of complex processes and has an automatic retry mechanism for task execution failure. The message push module integrates the instant messaging platform API, supports dynamic subscription by role, department or individual, displays key indicators using message templates, and supports real-time feedback on push status and unread reminders. The encrypted storage module uses a symmetric encryption mode to encrypt and save the database connection information, generating an encrypted result containing encrypted data, random numbers, and authentication tags; the intelligent task scheduling module dynamically generates a parameterized SQL query plan based on task dependencies and sends the query plan to the SQL script engine module. The SQL script engine module automatically replaces variable markers and executes SQL queries according to the query plan, and returns the query results to the intelligent task scheduling module; The intelligent task scheduling module triggers the message push module based on the task execution result. The message push module works in conjunction with the intelligent task scheduling module to provide real-time feedback on the push status and unread reminders. The connection information of the multi-database connection module is encrypted by the encryption storage module and dynamically called by the intelligent task scheduling module during task execution, realizing a complete closed loop from multi-database data collection and processing to intelligent push.

2. The system according to claim 1, characterized in that, The multi-database connection module includes: a database type selection unit for selecting the target database type; a connection parameter configuration unit for configuring the server address, port number, database name, username, and password; a connectivity testing unit for detecting whether the database connection is normal, with a configurable test timeout; a connection information encryption unit for encrypting and storing connection information that passes the test; and a connection pool management unit for managing the creation, reuse, and destruction of database connections, setting a configurable maximum number of connections, a minimum number of idle connections, and a connection timeout.

3. The system according to claim 1, characterized in that, The SQL script engine module supports the variable tagging format of [[variable name]], where the variable name includes, but is not limited to, start date, end date, and department identifier. The SQL script engine module also includes an SQL syntax verification unit, which is used to verify the syntax correctness of the SQL statement before execution and provide error location and correction suggestions. The SQL script engine module uses regular expression matching, type-safe conversion, and pre-compilation mechanisms to perform variable substitution and prevent SQL injection attacks.

4. The system according to claim 1, characterized in that, The intelligent task scheduling module supports the following task triggering methods: timed triggering, which supports setting fixed execution times by hour or minute and supports multiple time periods; interval triggering, which supports setting cyclic execution intervals by hour or minute and supports custom interval times; task dependency orchestration, which uses a directed acyclic graph (DAG) topology sorting algorithm to sort tasks and combines a semaphore mechanism to achieve concurrency control, supporting parallel and serial task orchestration; and a failure retry mechanism, which uses an exponential backoff strategy for automatic retries, sets a maximum number of retries and a circuit breaker threshold, and sends a notification to the administrator when the number of retries reaches the upper limit or a circuit breaker is triggered.

5. The system according to claim 1, characterized in that, The message push module includes: a message template configuration unit for configuring instant messaging platform message templates, supporting various message types such as text, images, and cards; a recipient setting unit for setting recipients by user ID, department ID, and tag ID, and supporting dynamic group push; a message content assembly unit for filling SQL query results into message templates, supporting data formatting and conditional rendering; and a push status recording unit for recording push status and supporting unread reminders, automatically retrying when push fails using the same retry strategy as the intelligent task scheduling module.

6. The system according to claim 1, characterized in that, The encrypted storage module adopts the AES-256-GCM encryption mode to generate an encrypted result containing encrypted data, random numbers, and authentication tags. The encrypted storage module also includes a key management unit for key generation, storage, and rotation, and supports hardware security module (HSM) protection.

7. The system according to claim 1, characterized in that, It also includes a log monitoring module, which records task execution logs, message push logs, and database operation logs, and supports multi-dimensional querying and statistical analysis by date, task type, status, and other dimensions.

8. The system according to claim 1, characterized in that, The SQL script engine module also supports SQL query performance analysis, providing execution plans and optimization suggestions.

9. The system according to claim 2, characterized in that, The multi-database connection module supports database types including SQL Server, MySQL, Oracle, and Kingbase.

10. The system according to claim 4, characterized in that, The intelligent task scheduling module supports setting task priorities, allowing high-priority tasks to preempt the execution resources of low-priority tasks.

11. The system according to claim 5, characterized in that, The message push module supports setting a message sending time window to avoid sending messages during non-working hours.

12. The system according to claim 5, characterized in that, The instant messaging platform API integrated into the message push module is the WeChat API.

13. A method for intelligent report push based on multiple databases, characterized in that, Includes the following steps: Configure database connection information: select the database type, and enter the server address, port number, database name, username, and password; test connection connectivity to verify that the configured database connection is normal; the test timeout is configurable; encrypt and store connection information using AES-256-GCM encryption mode; define parameterized SQL script templates and write SQL query statements containing dynamic variable markers; perform SQL syntax validation, with the system automatically validating the syntax of SQL statements and providing error location and correction suggestions; set task triggering methods and execution frequencies, selecting timed triggering or interval triggering, and configuring the corresponding execution time or interval; configure task dependencies, using a directed acyclic graph (DAG) topology sorting algorithm to set the execution order dependencies between tasks, supporting parallel and serial task orchestration; configure message receiving objects and templates, setting recipient parameters and message template content, supporting dynamic group push. When executing tasks, the intelligent task scheduling module dynamically generates parameterized SQL query plans based on task dependencies. The SQL script engine module uses regular expression matching, type-safe conversion, and pre-compilation mechanisms to automatically replace variables and execute queries, generating report content. When task execution fails, an exponential backoff strategy is used to automatically retry. When pushing messages, the intelligent task scheduling module triggers the message push module based on the task execution results, pushing report content to designated recipients via the instant messaging platform API, supporting various message types such as text, images, and cards. Record the push status and results, and remind unread users. If the push fails, automatically retry using the same exponential backoff strategy.

14. The method according to claim 13, characterized in that, In the dynamic variable replacement step, the variable marker adopts the format [[variable name]]. The system automatically replaces the variable marker with the actual parameter value at runtime. Supported variable types include date, numeric, string and enumeration types. The replacement process adopts a pre-compilation mechanism to prevent SQL injection attacks.

15. The method according to claim 13, characterized in that, The task execution steps support three query methods: data table, view, or stored procedure. Query results can be cached to improve the efficiency of repeated queries.

16. The method according to claim 13, characterized in that, In the message push process, multiple message types such as text, images, and cards can be sent via the instant messaging platform API. Card messages support titles, descriptions, images, and jump links.

17. The method according to claim 13, characterized in that, It also includes the following steps: caching the task execution results, with configurable caching time, and returning the results directly when the cache is hit, thereby improving the efficiency of repeated queries.

18. The method according to claim 13, characterized in that, In the encrypted storage step, AES-256-GCM encryption mode is used for encryption, and the key is protected and managed by a hardware security module (HSM).

19. The method according to claim 13, characterized in that, In the task execution steps, thread pool technology is used to improve concurrent execution capabilities. The size of the thread pool is configurable, and task concurrency control is achieved by combining it with a semaphore mechanism.

Images