A Data Offline Synchronization System and Method for a Distributed Experience Feedback System

Abstract

The present invention belongs to the field of distributed data storage and synchronization in information engineering, and specifically relates to a distributed experience feedback system data offline synchronization system and method. It mainly includes the following steps: application of system eigenvalue algorithm; data set export process based on JSON format; data encryption and decryption process based on eigenvalue algorithm. Its beneficial effect is: adopting a JSON-based data set mode without using SQL mode, supplemented by a data recovery algorithm, each data in the JSON data set is restored and compared with the target system data, so as to achieve the effect of securely merging data and avoid accidental deletion and overwriting of data. The data validity period is also set when the file is exported, and the system also generates a check code based on the validity period and the target system feature code. When importing, the check code can be used to judge the validity period. If the file exceeds the validity period, it cannot be imported, which improves the fault tolerance during a certain import.

Description

Technical Field

[0001] The present invention belongs to the field of distributed data storage and synchronization in information engineering, and in particular relates to a distributed experience feedback system data offline synchronization system and method. Background Art

[0002] With the continuous advancement of information technology, the demand for enterprise informatization is increasing, and the scope of application is also expanding. Some larger enterprises have many subsidiaries or departments. These structures are restricted by network infrastructure conditions or the requirements of relevant confidentiality management regulations, resulting in these subsidiaries and headquarters not being in the same network environment, and even being completely isolated at the physical level. In the years of implementation experience of status reporting systems, we have encountered many such situations. Generally, enterprises choose to deploy multiple systems that do not interfere with each other and use them separately, but this is not a perfect solution. Information silos will form between systems. Based on this background, a distributed experience feedback system data offline synchronization implementation technology is introduced. It can effectively solve the data synchronization between experience feedback systems and break the information silo effect. At the same time, the eigenvalue algorithm and multiple encryption measures are used to ensure the security of offline data transmission. Summary of the Invention

[0003] The purpose of the present invention is to provide a distributed experience feedback system data offline synchronization system and method, which can realize the data synchronization function of the system in different network environments and achieve the distributed effect of multi-network deployment of one system.

[0004] The technical solution of the present invention is as follows: a distributed experience feedback system data offline synchronization system, including an application system eigenvalue algorithm module, a data set export module based on JSON format, and a data encryption and decryption process module based on the eigenvalue algorithm.

[0005] The application system characteristic value algorithm module, in which the characteristic value serves as an identification code between different server systems, pre-calculates the characteristic value in distributed systems deployed in different network environments, stores the characteristic value with each other, and realizes subsequent mutual identification. The characteristic value will also be used as a secret key for subsequent data export and import in the encryption and decryption process of the data file. The characteristic value will change after the system hardware is replaced, and the new characteristic value needs to be re-imported into the associated system. Specifically, it includes the following: collecting the CPUID, network card MAC address, memory ID, motherboard ID, hard disk ID and IP address of the current system deployment machine, merging all the information, and combining it with the serial number of the current status reporting system for hash calculation to obtain a unique value as the basic characteristic value; storing its own characteristic value and the characteristic value of the associated system in each distributed system.

[0006] The JSON format-based dataset export module constructs a data entity class equivalent to a database business table; establishes a data entity association relationship corresponding to the association relationship between the database business tables; writes business data to convert into data entity-related logic, queries business data and its associated data from the database, and then assigns the data to the data entity one by one; including: exporting business data; encrypting business data, first performing initial encryption with the characteristic value of the target system; randomly splitting the encrypted file; then randomly merging it into a new file; recording random information during random splitting and random order during random merging, then merging the random information during file splitting and random order information during merging, and merging it with the merged new file again to form re-merged data; then calculating the secret key based on the data validity interval selected during export and the characteristic value of the source system, and encrypting the generated secret key with the re-merged data to generate the final encrypted data; file import is the inverse process of file encryption, and the exported data file is imported into the target system offline to achieve data import.

[0007] The data encryption and decryption process module based on the eigenvalue algorithm includes calling a random number function, performing a random number calculation, and using the random result value R as the file segmentation score; converting the file into a byte stream array and obtaining a length L; calling the random function R times in a loop, initially setting the random interval to [0, L]. After the loop starts, the random value RN is recorded for each loop, and the random interval is adjusted according to the value of RN in the next loop; if RN is less than the maximum length L of the byte stream, RN is used as the minimum value of the next random interval; if RN is equal to L, RN-1 is used as the maximum value of the next random interval. This process repeats until the R loops are completed, generating an array with R records; bubble sorting the array to form an array that increases in size from small to large; using the array data to segment the byte stream data of the file to form R+1 arrays; adding an array with numbers from 1 to R+1, and then calling a random sorting algorithm to shuffle the array; and recombining the R+1 segmented arrays based on the shuffled data to form a new file.

[0008] The data encryption and decryption process module based on the eigenvalue algorithm includes a decryption stage, which is the inverse operation process of encryption, including a data recovery process module, which converts JSON data into a data entity data interface, performs data comparison through logic, and generates the final data process, including: converting JSON data into a data entity; matching data according to the data entity's ID, CREATE_DATE, UPDATE_DATE and other fields, generating actual add, delete, modify and query behaviors, and merging the data into the database.

[0009] A distributed experience feedback system data offline synchronization method includes the following steps:

[0010] S1: Apply the system eigenvalue algorithm;

[0011] S2: Dataset export process based on JSON format;

[0012] S3: Data encryption and decryption process based on eigenvalue algorithm.

[0013] The S1 application system eigenvalue algorithm, in which the eigenvalue serves only as an identification code between different server systems; distributed systems deployed in different network environments pre-calculate the eigenvalue and store the eigenvalues to achieve subsequent mutual identification; the eigenvalue will also be used as the secret key for the encryption and decryption process of data files when subsequently exporting and importing data. The eigenvalue will change after the system hardware is replaced, and the new eigenvalue needs to be re-imported into the associated system, specifically including the following:

[0014] S11: Collect the CPUID, network card MAC address, memory ID, motherboard ID, hard disk ID and IP address of the current system deployment machine, merge all the information, and perform hash calculation on the serial number of the current status reporting system to obtain a unique value as the basic feature value;

[0015] S12: Each distributed system stores its own characteristic value and the characteristic value of the associated system respectively.

[0016] The S2 establishes the mutual relationship between all business data in the form of data entities, then exports all data entities and converts them into JSON format for storage, including the following steps:

[0017] S21: Construct a data entity class equivalent to the database business table;

[0018] S22: Establishing a data entity association relationship corresponding to the association relationship between database business tables;

[0019] S23: Write the logic related to converting business data into data entities, query the business data and its related data from the database, and then assign the data to the data entities one by one.

[0020] The S3 data encryption and decryption process based on the eigenvalue algorithm includes:

[0021] S31: the encryption process of the data;

[0022] S32: Implementation steps of the encryption phase;

[0023] S33: the decryption process.

[0024] The encryption process of the S31 data includes the following:

[0025] S311: Exported business data;

[0026] S312: Encrypt the business data, first performing initial encryption using the target system characteristic value;

[0027] S313: Randomly split the encrypted file;

[0028] S314: randomly merge into a new file;

[0029] S315: Record the random information during the splitting process in step S313 and the random order during the merging process in step S314; then merge the random information during the file splitting process and the random order during the merging process, and merge them with the merged new file to form the merged data.

[0030] S316: The secret key is calculated based on the data validity interval selected during export and the characteristic value of the source system, and the generated secret key is encrypted with the re-merged data to generate the final encrypted data.

[0031] The implementation steps of the S32 encryption stage are as follows:

[0032] S321: Calling a random number function to perform a random number calculation, and using the value R of the random result as the file segmentation score;

[0033] S322: Convert the file into a byte stream array and obtain the length L;

[0034] S323: Call the random function for R cycles, initially setting the random interval to [0, L]. After the cycle starts, record the random value RN for each cycle, and adjust the random interval for the next cycle based on the value of RN.

[0035] If RN is less than the maximum length L of the byte stream, RN is used as the minimum value of the next random interval. If RN is equal to L, RN-1 is used as the maximum value of the next random interval. This process repeats until R cycles are completed, generating an array with R records.

[0036] S324: Bubble sort the array in step S323 to form an array that increases from small to large;

[0037] S325: Use the array data in step S324 to split the byte stream data of the file into R+1 arrays;

[0038] S326: Add an array of numbers from 1 to R+1, and then call the random sorting algorithm to shuffle the array;

[0039] S327: Based on the disordered data in step S326, the R+1 arrays divided in step S325 are reassembled to form a new file.

[0040] The S33 decryption process includes the following:

[0041] The process of converting JSON data into a data entity data interface, performing data comparison through logic, and generating the final data;

[0042] S331: Convert JSON data into data entity;

[0043] S332: Match data based on the ID, CREATE_DATE, UPDATE_DATE and other fields of the data entity, generate actual add, delete, modify and query actions, and merge the data into the database.

[0044] The beneficial effects of the present invention are: by adopting a JSON-based data set model instead of an SQL model, and supplemented by a data recovery algorithm, each piece of data in the JSON data set is restored and compared with the target system data, thereby achieving the effect of securely merging data and avoiding accidental deletion and overwriting of data. The data validity period is also set when the file is exported, and the system also generates a check code based on the validity period and the target system feature code. When importing, the check code can be used to determine the validity period. If the file exceeds the validity period, it cannot be imported, thereby improving the fault tolerance during a certain import. In addition, the check code and the imported data file must also be matched. If the check code is incorrect, the correct decryption key cannot be generated, and the data cannot be decrypted normally. DETAILED DESCRIPTION

[0045] In order to make the objectives, technical solutions, and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to examples. It should be understood that the specific examples described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0046] The present invention is aimed at the experience feedback system under the nuclear power management support platform. In an environment where the network environment is physically isolated, the method for storing and synchronizing data of the distributed deployment system in an offline manner is implemented; the data synchronization function of the system in different network environments is realized, and the distributed effect of multi-network deployment of one system is achieved.

[0047] The third-party dependencies required to implement the present invention include but are not limited to MySQL, JDK 1.8, and JSON.

[0048] The present invention adopts characteristic values as the unique identifier for mutual identification between systems and multiple encryption methods to achieve high reliability and security of offline data transmission; adopts JSON format for data synchronization to avoid the situation where the data of the two databases cannot be compared due to data isolation, resulting in the inability to generate valid SQL; finally, by reading JSON data and using data recovery algorithms to construct a complete data structure, complete data comparison is achieved, achieving the purpose of secure data merging, and ultimately realizing effective offline data synchronization of distributed systems.

[0049] A distributed experience feedback system data offline synchronization system includes an application system eigenvalue algorithm module, a data set export module based on JSON format, and a data encryption and decryption process module based on the eigenvalue algorithm.

[0050] The characteristic value algorithm module of the application system is used only as an identification code between different server systems. Distributed systems deployed in different network environments pre-calculate characteristic values and store them for mutual identification. The characteristic values will also be used as the secret key for subsequent data export and import in the encryption and decryption process of data files. The characteristic values will change after the system hardware is replaced, and the new characteristic values need to be re-imported into the associated system, which includes the following:

[0051] Collect the CPUID, network card MAC address, memory ID, motherboard ID, hard disk ID and IP address of the current system deployment machine, merge all the information, and perform hash calculation on the serial number of the current status reporting system to obtain a unique value as the basic feature value; store its own feature value and the feature value of the associated system in each distributed system.

[0052] The dataset export module based on JSON format takes into account the physical isolation between application systems. When the data of the other system cannot be known, if the export format uses SQL statements, then when importing data, it is bound to cause a certain degree of data overwriting or SQL execution failure due to data duplication, resulting in the failure of data import. Therefore, in terms of the format selection of data export, the embodiment of the present invention adopts a JSON-based dataset model, that is, all business data is established in the form of data entities with mutual associations, and then all data entities are exported and converted into JSON format for storage, which specifically includes:

[0053] Construct a data entity class equivalent to the database business table;

[0054] Establishing data entity association relationships corresponding to the association relationships between database business tables;

[0055] Write logic to convert business data into data entities, query business data and its related data from the database, and then assign the data to data entities one by one.

[0056] The data encryption and decryption process module based on the eigenvalue algorithm includes the following steps:

[0057] Export business data (DATA1);

[0058] To encrypt the business data (DATA1), first perform the initial encryption (DATA2) using the target system characteristic value;

[0059] Randomly split the encrypted file (DATA3);

[0060] Then randomly merge into a new file (DATA4);

[0061] Record the random information during random splitting and the random order during random merging, then merge the random information during file splitting and the random order information during merging, and merge them with the merged new file again to form the re-merged data (DATA5);

[0062] The key is then calculated based on the data validity interval selected during export and the characteristic value of the source system. The generated key is then encrypted with the re-merged data (DATA5) to generate the final encrypted data (DATA6).

[0063] File import is the reverse process of file encryption. The exported data file can be imported into the target system offline to achieve data import.

[0064] The algorithm in the check code algorithm module based on characteristic values and validity intervals is a reversible encryption algorithm based on the DES encryption algorithm. It is appropriately adjusted based on the DES algorithm to achieve reversible operations. If the input value is the characteristic value and validity interval, the output is the check code. If the input value is the check code and characteristic value, the output is the validity interval.

[0065] The algorithm in the random cutting and merging file recovery algorithm module mainly realizes byte streaming of files. During encryption, the files are first cut into multiple copies and then recombined. During decryption, the files are restored based on the random number information of the cutting and combining.

[0066] The encryption phase includes:

[0067] Call the random number function to perform a random number calculation, and use the value R of the random result as the file cutting score;

[0068] Convert the file into a Byte stream array and get the length L;

[0069] Call the random function R times in a loop, initially setting the random interval to [0, L]. After the loop starts, record the random value RN for each time, and adjust the random interval for the next loop based on the value of RN. If RN is less than the maximum length L of the byte stream, RN is used as the minimum value of the next random interval. If RN is equal to L, RN-1 is used as the maximum value of the next random interval. Repeat this process until R loops are completed, generating an array with R records.

[0070] Bubble sort the array to form an array that increases from small to large.

[0071] Use array data to split the byte stream data of the file to form R+1 arrays;

[0072] Add an array of numbers from 1 to R+1, and then call the random sorting algorithm to shuffle the array;

[0073] According to the above-mentioned disordered data, the R+1 arrays that were split out are reassembled to form a new file.

[0074] The decryption stage is the inverse operation process of encryption.

[0075] The data recovery process module converts JSON data into a data entity data interface, performs data comparison through logic, and generates the final data, including:

[0076] Convert JSON data into data entities;

[0077] Match data based on the data entity's ID, CREATE_DATE, UPDATE_DATE and other fields, generate actual add, delete, modify and query actions, and merge the data into the database.

[0078] A distributed experience feedback system data offline synchronization method includes the following steps:

[0079] S1: Applying the system eigenvalue algorithm

[0080] The eigenvalues serve only as identification codes between different server systems. Distributed systems deployed in different network environments pre-calculate and store eigenvalues for mutual identification. The eigenvalues also serve as the encryption key for data file encryption and decryption during subsequent data export and import. Eigenvalues can change after system hardware replacements, requiring the new eigenvalues to be re-imported into the associated systems.

[0081] The details include:

[0082] S11: Collect the CPUID, network card MAC address, memory ID, motherboard ID, hard disk ID and IP address of the current system deployment machine, merge all the information, and perform hash calculation on the serial number of the current status reporting system to obtain a unique value as the basic feature value;

[0083] S12: Each distributed system stores its own characteristic value and the characteristic value of the associated system respectively.

[0084] S2: Dataset export process based on JSON format

[0085] Considering the physical isolation between application systems, if SQL statements are used for export without knowing the data in the other system, data import will inevitably result in a certain degree of data overwriting or SQL execution failure due to data duplication, resulting in data import failure. Therefore, in terms of data export format selection, this technology adopts a JSON-based dataset model, which establishes the relationship between all business data in the form of data entities, then exports all data entities and converts them into JSON format for storage. The specific steps include:

[0086] S21: Construct a data entity class equivalent to the database business table;

[0087] S22: Establishing a data entity association relationship corresponding to the association relationship between database business tables;

[0088] S23: Write the logic related to converting business data into data entities, query the business data and its related data from the database, and then assign the data to the data entities one by one.

[0089] S3: Data encryption and decryption process based on eigenvalue algorithm, including encryption process and decryption process.

[0090] S31: The data encryption process includes the following:

[0091] S311: Exported business data (DATA1);

[0092] S312: Encrypt the service data (DATA1), first performing initial encryption (DATA2) using the target system characteristic value;

[0093] S313: Randomly split the encrypted file (DATA3);

[0094] S314: randomly merge into a new file (DATA4);

[0095] S315: Record the random information during the splitting process in step S313 and the random order during the merging process in step S314; then merge the random information during the file splitting process and the random order during the merging process, and merge them with the merged new file to form the merged data (DATA5);

[0096] S316: The key is calculated based on the data validity interval selected during export and the characteristic value of the source system. The generated key is then encrypted with the merged data (DATA5) to generate the final encrypted data (DATA6).

[0097] File import is the reverse process of file encryption. The exported data file can be imported into the target system offline to achieve data import.

[0098] A reversible encryption algorithm based on the DES encryption algorithm is a checksum algorithm that is adapted to achieve reversibility. If the input value is the eigenvalue and validity interval, the output is the checksum. If the input value is the checksum and eigenvalue, the output is the validity interval.

[0099] The main implementation principle of the random cutting and merging file recovery algorithm is to stream the file byte by byte. During encryption, the file is first cut into multiple parts and then recombined. During decryption, the file is restored based on the random number information of the cutting and combining.

[0100] S32: Implementation steps of the encryption phase:

[0101] S321: Calling a random number function to perform a random number calculation, and using the value R of the random result as the file segmentation score;

[0102] S322: Convert the file into a byte stream array and obtain the length L;

[0103] S323: Call the random function for R cycles, initially setting the random interval to [0, L]. After the cycle starts, record the random value RN for each cycle, and adjust the random interval for the next cycle based on the value of RN.

[0104] If RN is less than the maximum length L of the byte stream, RN is used as the minimum value of the next random interval. If RN is equal to L, RN-1 is used as the maximum value of the next random interval. This process repeats until R cycles are completed, generating an array with R records.

[0105] S324: Bubble sort is performed on the array in step S323 to form an array that increases from small to large.

[0106] S325: Use the array data in step S324 to split the byte stream data of the file into R+1 arrays;

[0107] S326: Add an array of numbers from 1 to R+1, and then call the random sorting algorithm to shuffle the array;

[0108] S327: Based on the disordered data in step S326, the R+1 arrays divided in step S325 are reassembled to form a new file.

[0109] S33: The decryption process includes the following: converting JSON data into a data entity data interface, performing data comparison through logic, and generating final data;

[0110] S331: Convert JSON data into data entity;

[0111] S332: Match data based on the ID, CREATE_DATE, UPDATE_DATE and other fields of the data entity, generate actual add, delete, modify and query actions, and merge the data into the database.

[0112] This invention uses signature technology, encrypting signatures during both import and export. Without knowing the machine information and algorithm principles, the signatures are inaccessible to outsiders. Furthermore, the invention employs random encryption and recombinant file merging methods. Without the signatures, even brute force cracking cannot reveal the true contents of files. Therefore, even if data is leaked during offline transmission, there is no risk of data leakage.

[0113] The present invention adopts a JSON-based data set model instead of an SQL model, and is supplemented by a data recovery algorithm to restore and compare each piece of data in the JSON data set with the target system data, thereby achieving the effect of securely merging data and avoiding accidental deletion and overwriting of data. The data validity period is also set when the file is exported, and the system also generates a check code based on the validity period and the target system feature code. When importing, the check code can be used to determine the validity period. If the file exceeds the validity period, it cannot be imported, thereby improving the fault tolerance during a certain import. In addition, the check code and the imported data file must also be matched. If the check code is incorrect, the correct decryption key cannot be generated, and the data cannot be decrypted normally.

[0114] In general, the comprehensive utilization of various means makes the present invention have very high reliability and security, and can efficiently and securely handle offline data synchronization of various distributed systems of an enterprise.

Claims

1. A distributed experience feedback system data offline synchronization system, characterized by: It includes the application system eigenvalue algorithm module, the data set export module based on JSON format, and the data encryption and decryption process module based on the eigenvalue algorithm; The application system characteristic value algorithm module, in which the characteristic value serves as an identification code between different server systems, pre-calculates the characteristic value in distributed systems deployed in different network environments, stores the characteristic value in each other, and realizes subsequent mutual identification. The characteristic value will also be used as a secret key for subsequent data export and import to participate in the encryption and decryption process of the data file. The characteristic value will change after the system hardware is replaced, and the new characteristic value needs to be re-imported into the associated system. Specifically, it includes the following: collecting the CPUID, network card MAC address, memory ID, motherboard ID, hard disk ID and IP address of the current system deployment machine, merging all the information, and combining it with the serial number of the current status reporting system for hash calculation to obtain a unique value as the basic characteristic value; Each distributed system stores its own eigenvalues and the eigenvalues of the associated system respectively; The JSON format-based dataset export module constructs a data entity class equivalent to a database business table; establishes a data entity association relationship corresponding to the association relationship between the database business tables; writes business data to convert into data entity-related logic, queries business data and its associated data from the database, and then assigns data to data entities one by one; including: exporting business data; encrypting business data, first encrypting it with the characteristic value of the target system; randomly splitting the encrypted file; then randomly merging it into a new file; recording random information during random splitting and random order during random merging, then merging the random information during file splitting and random order information during merging, and merging it with the merged new file again to form re-merged data; then calculating the secret key based on the data validity interval selected during export and the characteristic value of the source system, and encrypting the generated secret key with the re-merged data to generate the final encrypted data; file import is the inverse process of file encryption, and the exported data file is imported into the target system offline to realize data import; The data encryption and decryption process module based on the eigenvalue algorithm includes calling a random number function, performing a random number calculation, and using the value R of the random result as the file segmentation score; converting the file into a byte stream array and obtaining a length L; performing R cycles of calling the random function, initially setting the random interval to [0, L], recording each random value RN after the cycle starts, and adjusting the random interval of the next cycle according to the value of RN; if RN is less than the maximum length L of the byte stream, RN is used as the minimum value of the next random interval; if RN is equal to L, RN-1 is used as the maximum value of the next random interval, and so on and so forth until the R cycles are completed to generate an array with R records; performing bubble sorting on the array to form an array that increases from small to large; using the array data to segment the byte stream data of the file to form R+1 arrays; adding an array with numbers from 1 to R+1, and then calling the random sorting algorithm to shuffle the array; based on the shuffled data, recombining the segmented R+1 arrays to form a new file; The data encryption and decryption process module based on the eigenvalue algorithm includes a decryption stage, which is the inverse operation process of encryption, including a data recovery process module, which converts JSON data into a data entity data interface, performs data comparison through logic, and generates the final data process, including: converting JSON data into a data entity; matching data according to the ID, CREATE_DATE, and UPDATE_DATE fields of the data entity, generating actual addition, deletion, modification, and query behaviors, and merging the data into the database.

2. A distributed experience feedback system data offline synchronization method, characterized in that: The steps include: S1: Apply the system eigenvalue algorithm; The eigenvalues are used only as identification codes between different server systems. Distributed systems deployed in different network environments pre-calculate and store eigenvalues to achieve subsequent mutual identification. The eigenvalues are also used as secret keys for subsequent data export and import, and are incorporated into the encryption and decryption processes of data files. Eigenvalues change after system hardware replacement, requiring the new eigenvalues to be re-imported into the associated systems. This is as follows: S11: Collect the CPUID, network card MAC address, memory ID, motherboard ID, hard disk ID and IP address of the current system deployment machine, merge all the information, and perform hash calculation on the serial number of the current status reporting system to obtain a unique value as the basic feature value; S12: storing the characteristic values of the distributed system and the characteristic values of the associated system in each distributed system; S2: Dataset export process based on JSON format; Establish the relationships between all business data in the form of data entities, then export all data entities and convert them into JSON format for storage. This includes the following steps: S21: Construct a data entity class equivalent to the database business table; S22: Establishing a data entity association relationship corresponding to the association relationship between database business tables; S23: Write the logic related to converting business data into data entities, query the business data and its related data from the database, and then assign the data to the data entities one by one; S3: Data encryption and decryption process based on eigenvalue algorithm; include: S31: the encryption process of the data; These include: S311: Exported business data; S312: Encrypt the business data, first performing initial encryption using the target system characteristic value; S313: Randomly split the encrypted file; S314: randomly merge into a new file; S315: Record the random information during the splitting process in step S313 and the random order during the merging process in step S314; then merge the random information during the file splitting process and the random order during the merging process, and merge them with the merged new file to form the merged data. S316: A secret key is calculated based on the data validity interval selected during export and the characteristic value of the source system. The generated secret key is then encrypted with the merged data to generate the final encrypted data. S32: Implementation steps of the encryption phase; The implementation steps are as follows: S321: Calling a random number function to perform a random number calculation, and using the value R of the random result as the file segmentation score; S322: Convert the file into a byte stream array and obtain the length L; S323: Call the random function for R cycles, initially setting the random interval to [0, L]. After the cycle starts, record the random value RN for each cycle, and adjust the random interval for the next cycle based on the value of RN. If RN is less than the maximum length L of the byte stream, RN is used as the minimum value of the next random interval. If RN is equal to L, RN-1 is used as the maximum value of the next random interval. This process repeats until R cycles are completed, generating an array with R records. S324: Bubble sort the array in step S323 to form an array that increases from small to large; S325: Use the array data in step S324 to split the byte stream data of the file into R+1 arrays; S326: Add an array of numbers from 1 to R+1, and then call the random sorting algorithm to shuffle the array; S327: Based on the out-of-order data in step S326, the R+1 arrays split in step S325 are reassembled to form a new file; S33: the decryption process; These include: The process of converting JSON data into a data entity data interface, performing data comparison through logic, and generating the final data; S331: Convert JSON data into data entity; S332: Match the data according to the ID, CREATE_DATE, and UPDATE_DATE fields of the data entity, generate actual add, delete, modify, and query actions, and merge the data into the database.