CN117520312A

CN117520312A - Database structure comparison method and database comparison system

Info

Publication number: CN117520312A
Application number: CN202311694334.9A
Authority: CN
Inventors: 王志超; 孙林峰; 邢利; 胡伟; 潘小青; 张正叶
Original assignee: Shenzhen Meiyunji Network Technology Co ltd
Current assignee: Shenzhen Meiyunji Network Technology Co ltd
Priority date: 2023-12-11
Filing date: 2023-12-11
Publication date: 2024-02-06

Abstract

The application provides a database structure comparison method and a database comparison system, wherein the method comprises the following steps: at least acquiring a database D1 and a database D2, and respectively acquiring a structural feature set D1J of the database D1 and a structural feature set D2J of the database D2; combining the structural feature set D1J with the structural feature set D2J, performing de-duplication treatment on the combined feature set, and forming a feature union UA; respectively calculating a difference set DIF-D1 of the feature union UA and the structural feature set D1J and a difference set DIF-D2 of the feature union UA and the structural feature set D2J; and outputting the results of the difference set DIF-D1 and the difference set DIF-D2 respectively. The technical scheme can compare a plurality of databases, can compare under the condition that which database is not known as the reference, or does not need to designate the reference database, thereby effectively improving the comparison precision of the databases and ensuring the operation stability and reliability of the ERP.

Description

Database structure comparison method and database comparison system

Technical Field

The application belongs to the technical field of computers, and particularly relates to an advertisement operation adjustment method, an advertisement management module, a commodity replenishment method, a database structure comparison method and a system corresponding to the methods.

Background

With the rise of global electronic commerce, international retail trade is rapidly developed, a large number of domestic small and medium-sized electronic commerce sellers expand retail business to foreign markets, and domestic commodities are sold to foreign markets through cross-border electronic commerce platforms (such as Amazon, yibeieBay, shopping interest Wish, dried small shrimps Shupe, lazada and the like). With the development of cross-border business, an e-commerce ERP system (which may be simply referred to as "e-commerce ERP", "ERP system" or "system") developed based on the conventional ERP software is gradually developed. The e-commerce ERP system can be deeply matched with the e-commerce platform, namely, the e-commerce ERP system accesses and controls shops of the e-commerce platform through established rules, processes dynamic data of each link of shop operation, further helps domestic e-commerce sellers to uniformly manage overseas shops, solves the obstacle caused by language difference, can realize that one operator can manage a plurality of e-commerce shops at the same time, and greatly improves the efficiency of shop operation.

The e-commerce ERP system accesses and controls a store of the e-commerce platform through a set rule, processes dynamic data of each link of store operation, manages numerous and complicated data, and meets the operation convenience requirements of multiple types of users (sellers); therefore, all functional modules of the existing commercial ERP system are in the stage of gradually updating and perfecting functions, the function algorithms and rules formulated by all software enterprises when the software enterprises develop from the ERP system of the household appliances are basically different, and all functional modules continuously develop new versions along with the change of the user demands so as to be compatible with more use scenes.

In order to increase sales volume of commodities in shops, sellers usually advertise and popularize the commodities in shops to increase exposure of the commodities or improve commodity drainage effects, so that click rate and yield of buyers are increased. In addition, when the ERP system performs fine control or management on advertisements released from shops, corresponding advertisement operation data are required to be acquired from relevant interfaces of the e-commerce platform, so that the ERP system can correspondingly adjust the releasing actions of the advertisements according to established advertisement operation rules and the advertisement operation data, and the problems that errors and low efficiency are easy to occur due to manual adjustment of the releasing actions of the advertisements in a traditional mode are solved.

However, the current ERP system has a problem of low precision or inaccuracy in adjusting the advertisement operation. Specifically, the current way of invoking advertisement operation data by the ERP system is mainly to acquire and calculate and judge based on a predetermined advertisement rule through a REPORT (REPORT) interface of an e-commerce platform, so as to execute a corresponding advertisement adjustment action. The advertisement data acquired through the report interface is generated regularly, so that the mode for acquiring the advertisement data has hysteresis, namely, the advertisement data does not have real-time performance/timeliness, thereby causing the problem of low accuracy of the adjustment action of the ERP system on the advertisement operation and further affecting the reliability of the ERP system.

In the operation process of shops, sellers usually give commodity replenishment suggestions by means of an ERP system so as to reduce unnecessary capital occupation and improve commodity turnover rate. However, in the current ERP system, in the use process, in the scene of carrying out commodity follow-up sales among multiple shops and using a shared warehouse, when a replenishment suggestion is given to a user, the precision of the replenishment quantity/purchasing quantity is out of alignment. Specifically, sellers often choose to set up multiple shops in different countries (in the european union and north america) at the same time in the actual operation process, and when calculating a commodity to be sold in the same area, different countries and different shops, the current ERP system only can calculate separately based on each shop to calculate the required replenishment amount and the required purchasing amount of each shop separately, and then combine the two to replenish and purchase. Second, one commodity may be sold (followed) in multiple stores, but there may be only one or two stores with good sales, while the other stores have low sales. Therefore, in the calculation process, different replenishment strategies are required to be formulated for shops with different sales, the efficiency is very low, each shop is required to be calculated, and the time cost is very high. In addition, when several stores sell the same commodity, the commodities belong to the same SKU, and the MSKU and the SKU are in one-to-one correspondence, which results in a deduction mode of manually estimating and distributing the inventory when calculating the replenishment/purchase suggestions of different stores, otherwise, the situation that the same SKU inventory is repeatedly deducted is caused, and thus, the problem of inaccurate replenishment/purchase amount is caused.

In addition, some software or system programs (such as ERP systems) that generate large amounts of data during operation also need to be used in conjunction with external databases, i.e., by means of the databases to store and manage the generated data. When using a database to store data, in particular a relational database (e.g., mysql, ORACLE, SQL Server), it is necessary to create tables, define columns, index, and then the system program writes the data to the tables specified by the database. And when the data volume is relatively large (such as a plurality of electric shops managed in an ERP system), the excessive reading and writing pressure of a single library or a single table is avoided. In this case, a plurality of databases or a plurality of tables are often used to cooperatively store massive data, so that it is required to ensure that all tables in the plurality of databases, all columns and indexes in the tables are consistent, the system program can operate normally, and otherwise, the system can report errors. The table structure is easy to change in the development process, and in the case of many library tables (often up to tens of libraries), there are multiple tables, columns and indexes under each library, so that the situation that the tables, columns and indexes in different libraries are inconsistent due to personnel negligence or wrong operation or sql statement execution error is most likely to happen, namely, the structure of the created library is problematic.

The solution to this problem is mainly to compare the structures of the databases and find out the place where the error exists in the structures of the databases; however, the present database contrast tool has a certain limitation in the use process, for example, application number is 201610812612.X, and the name is a remote database contrast tool and method, for example, the present database verification visualization tool navicat, dbeaver; however, in the above technical solutions, one of the databases involved in comparison needs to be used as a reference database, which cannot fundamentally solve the problem of database comparison accuracy.

Other technical problems related to the present application are further described below. The foregoing is merely provided to facilitate an understanding of the principles of the present application and is not intended to represent all of the prior art.

Disclosure of Invention

The main purpose of the application is to provide an advertisement operation adjustment method, an advertisement management module and an ERP system, wherein the accuracy of operation (delivery) of advertisements is adjusted by improving the real-time and integrity of acquired advertisement operation data so as to improve the reliability of the use of the ERP system. In addition, the application also provides a commodity replenishment method and an ERP system, so that commodity follow-up sales among different shops and repeated deduction of commodity generated replenishment advice in a scene of using a shared bin are avoided, the precision of the replenishment advice is improved, and the operation cost of sellers is reduced; and providing a database structure comparison method and a database comparison system, wherein differences or errors among a plurality of different databases are compared and found in a mode of calculating a union and a difference among the multiple databases, so that the accuracy of the comparison of the multiple databases and the reliability of the use of the comparison system are improved.

In order to achieve the above objective, the present application proposes an advertisement operation adjustment method for an advertisement management module of an ERP system, the method comprising the following steps:

s1, loading advertisement rules;

s2, judging whether the advertisement rule contains a target condition, wherein the target condition is that data in the current time period is acquired, if yes, the first state information is received and the step S3 is converted, and if not, the second state information is received and the step S4 is converted;

s3, judging whether the real-time data is changed or not based on the first state information, if so, acquiring real-time data and report data generated by advertisement operation, and if not, acquiring report data generated by advertisement operation;

s4, judging the change state of the report data based on the second state information, and acquiring report data generated by advertisement operation when the report data is changed;

and S5, judging whether the report data and/or the real-time data acquired in the step S3 or the step S4 meet the condition of the advertisement rule, and executing the adjustment action on the advertisement when meeting the condition of the advertisement rule.

Additional features and technical effects of the present application are set forth in the description that follows. The technical problem solving thought and related product design scheme of the application are as follows:

Taking amazon platform as an example, in order to realize automatic operation of the ERP system on advertisement delivery adjustment, the data (REPORT data) referred by the current ERP system when adjusting advertisements is mainly obtained from the amazon REPORT (REPORT) interface, and because the REPORT is generated regularly, the REPORT has hysteresis, so the accuracy of the ERP system on advertisement delivery adjustment based on the REPORT data cannot meet the use requirement.

Applicant found that obtaining advertisement data (real-time data) via amazon's data STREAM (STREAM) interface can improve the accuracy of advertisement placement adjustment; the applicant also finds that the problem of insufficient precision still exists when real-time data is acquired through a STREAM interface and advertisement delivery is adjusted based on the real-time data.

After extensive investigation, the applicant found that in some specific scenarios, the following problems exist when real-time data is acquired through the STREAM interface: 1. the STREAM interface only pushes the subscribed data; for example, store 1 is authorized at 10 months 1 day of site time, and will subscribe to data immediately after authorization, but only have complete 1 day of data starting at 10 months 2 days of site time; therefore, when the user just authorizes the store, only a data time range smaller than the authorized days can be selected, which leads to that the new user cannot normally obtain complete advertisement data in a cold start stage. 2. The STREAM interface only pushes updated data, and because the updated data is pushed when the data is changed, the ERP system cannot automatically adjust advertisements according to the acquired data under certain conditions; for example, when an advertisement is paused, budget is consumed, no exposure is generated, etc.; in addition, the controlled object that has not changed in some scenes also needs to perform advertisement adjustment actions, for example, advertisement delivery is suspended in which the user expects to be less than 10 in the exposure of approximately 7 days. The existence of the problems still causes that the adjustment accuracy of the ERP system on advertisement running cannot reach the expectations of users.

Based on the above-mentioned problems, after further investigation, the applicant found that, in the first problem, for the situation that a new authorized (to obtain the authority of the e-commerce platform STREAM interface) store may have a loss when acquiring real-time data in the cold start stage, the loss of data has a greater influence on the hysteresis of the data, so that for the part of the data lost in the cold start stage of the store, the data acquired from the e-commerce platform REPORT interface is used for supplementing, so as to improve the accuracy of the data, that is, the accuracy influence generated by the small non-real-time (REPORT) data can be completely covered by the overall accuracy of the real-time data, and further, the accuracy of the ERP system for advertisement adjustment action can be effectively improved. Aiming at the second problem, in order to avoid that the ERP system cannot trigger corresponding advertisement adjustment actions according to the real-time data, non-real-time data can be periodically obtained from the REPORT interface to the ERP system at the moment, so that the defect caused by using the real-time data is further overcome.

According to the advertisement operation adjustment method, the corresponding advertisement management module and the ERP system, the report data are cooperatively called on the basis of calling the real-time data, so that the problem of adjustment precision caused by using only the real-time data in some specific scenes is solved on the basis of accurately adjusting advertisements on the whole by the advertisement management module or the ERP system, and the use reliability of the ERP system is further improved. Other embodiments and technical effects are set forth below.

Further, the present application also provides a method for replenishing goods and an ERP system, which improve the accuracy of replenishment suggestion and reduce the operation cost of sellers, and the method is used for the ERP system and comprises: k1, acquiring a plurality of intra-area shops based on a target large area or acquiring a plurality of domestic shops based on a target country; k2, acquiring in-area commodity inventory data and in-area commodity historical sales data of the in-area shops based on the standard identification codes, or acquiring domestic commodity inventory data and domestic commodity historical sales data of the domestic shops based on the standard identification codes; k3, performing duplicate removal processing on the commodity inventory data in the district or the domestic commodity inventory data based on the inventory unit code, and forming duplicate removal district commodity inventory data or duplicate removal domestic commodity inventory data; k4, calculating the total number of the commodity inventory data in the duplication removal area and the total number of the commodity historical sales volume data in the area, or calculating the total number of the duplication removal domestic commodity inventory data and the total number of the domestic commodity historical sales volume data; k5 generating a first replenishment proposal based on the total number of the product inventory data in the duplication removal region and the product historical sales volume data in the region, or generating a second replenishment proposal based on the total number of the duplication removal domestic product inventory data and the domestic product historical sales volume data

Additional features and technical effects of the present application are set forth in the description that follows. The technical problem solving thought of the commodity replenishing method and the ERP system and the related product design scheme are as follows:

based on the technical problems existing in the prior art of making a replenishment strategy for commodities of multiple stores, the applicant proposes a replenishment strategy based on large-area cross-store or country cross-store; specifically, the commodity replenishment strategy based on the cross-store of a large area is to screen out commodities in stores in European Union areas or all countries in North America according to a standard identification code (ASIN code) of a certain commodity, acquire corresponding inventory data and historical sales, and finally generate commodity replenishment suggestions based on the inventory data and the historical sales; the country-based cross-store replenishment strategy is to screen out the commodities in all stores in a country according to the standard identification code (ASIN code) of the commodity, obtain corresponding inventory data and historical sales, and finally generate replenishment suggestions of the commodity based on the base inventory data and the historical sales, however, the replenishment suggestions generated at this time are still inaccurate based on some specific scenes.

After a great deal of investigation, the applicant finds that when the seller selects to use the FBA logistics service provided by the E-commerce platform, the inventory of commodities in each store in the time zone is shared; in addition, the seller can select one or more stores to sell a certain product in other stores according to the requirement in actual operation, and merchant inventory codes (MSKU) of different stores are all related to the same inventory unit code (SKU). Further, in the process of generating the replenishment proposal, the problem that the same stock is repeatedly deducted is caused, so that the calculation of the replenishment quantity/purchasing quantity is inaccurate, that is, the precision of the replenishment proposal cannot meet the use requirement of a user. At this time, after the goods screened based on the standard identification code (ASIN code) are subjected to the duplicate removal processing according to the stock keeping unit code (SKU), the problem that the same goods stock is repeatedly deducted in the calculation process of the replenishment proposal can be effectively solved.

According to the commodity replenishment method and the corresponding ERP system, corresponding commodities are selected through large-area cross-store or national cross-store and based on standard identification codes, further inventory of the corresponding commodities is subjected to duplication elimination processing according to stock-keeping unit codes (SKUs), finally replenishment suggestions are generated based on inventory data after duplication elimination processing, the calculation efficiency of the commodity replenishment method and the corresponding ERP system for generating the replenishment suggestions is improved, the operation cost is reduced, and the problem that the same inventory is repeatedly deducted in the process of generating the replenishment suggestions under certain specific scenes is solved, so that the replenishment suggestion accuracy is improved. Other embodiments and technical effects are set forth below.

Further, the present application also provides a database structure comparison method and a database comparison system, which improve the accuracy of multi-database comparison and the reliability of the comparison system, and the method comprises: p1, at least acquiring a database D1 and a database D2, and respectively acquiring a structural feature set D1J of the database D1 and a structural feature set D2J of the database D2; p2, combining the structural feature set D1J with the structural feature set D2J, performing de-duplication treatment on the combined feature set, and forming a feature union UA; p3, respectively calculating a difference set DIF-D1 between the feature union UA and the structural feature set D1J, and a difference set DIF-D2 between the feature union UA and the structural feature set D2J; and P4, respectively outputting the results of the difference set DIF-D1 and the difference set DIF-D2.

Additional features and technical effects of the present application are set forth in the description that follows. The technical problem solving thought and related product design scheme of the database structure comparison method and the database comparison system are as follows:

in order to ensure reliable operation of the ERP system, the ERP system stores and manages the generated data by means of databases, and when the data volume is relatively large, a plurality of databases or a plurality of tables are often used for cooperatively storing mass data; then we need to ensure that all tables in the multiple libraries, all columns in the tables, and indexes are consistent before the system program can run properly, otherwise it will cause a program or system to report errors. The problems in the databases can be found by comparing different databases, and at present, when the databases are compared, only 2 databases are supported for comparison, and multi-database comparison cannot be supported; when two libraries are compared, one library is required to be designated as a reference library, and a place different from the reference library in the other library phase is output. In the case of multiple libraries and multiple tables, we cannot determine that the assigned reference library is necessarily correct; it is assumed that once there is an error in the structure of the reference library, the comparison is not a problem anyway. In addition, in the actual use process, there are often cases that there are multiple sub-tables in one library, and the existing database comparison tool cannot support verification of columns and indexes among multiple sub-tables in a single library.

Further, the applicant has found through a great deal of experiments and tests that when a plurality of databases are used, the composition structures of the tables, columns and indexes are necessarily the same. Based on this, the structural features of the tables, columns and indexes in the multiple database data to be subjected to the comparison verification are respectively collected, and then the feature sets of the same type are combined to form a union, and of course, only one table, column or index with the same features in the union is reserved, namely, the duplicate is removed. Then, respectively calculating difference sets of the union set and the feature sets of different libraries, and obtaining errors in tables, columns and indexes in the corresponding libraries according to the result of the difference sets; furthermore, when the result of the difference set is not null, the result of the non-null is a specific error in the corresponding library, and the research personnel can correct the corresponding error in the database according to the difference set of the non-null.

The tables in all the libraries, the columns in all the tables and the index union set in all the tables are used as the reference, and then the tables in a single library, the columns in a single table and the index union set in a single table are compared with the reference union set, so that the comparison of a plurality of databases without specifying the reference tables can be realized, and according to the comparison result, the inconsistent information that a certain table does not exist in a certain library and a certain column and index does not exist in a certain table can be obtained. The final developer can judge whether a certain table, a certain column or an index is redundant or missing according to the output information, and then perform specific processing actions. The technical scheme can compare a plurality of databases, and can compare under the condition that which database is not known as the reference or without designating the reference database, thereby effectively improving the accuracy of database comparison and ensuring the running stability and reliability of the ERP system.

The application also provides an advertisement management module for adjusting the operation of the advertisement according to the real-time data and/or the report data, wherein the advertisement management module executes the operation steps in the advertisement operation adjustment method.

The application also provides a database comparison system for checking whether an error exists in a database, and the database comparison system executes the operation steps in the database structure comparison method.

The application also provides an ERP system, which comprises the advertisement management module.

Further, the application also provides a computer device, which comprises a memory and a processor, wherein the ERP system in the application is stored in the memory, and the processor can run the operation instructions of the ERP system and execute the function implementation method of each function module.

The following is stated: for simplifying the description, the application mainly describes related technical schemes based on an Amazon platform, and it is to be noted that the e-commerce platform related to the application can be other e-commerce platforms besides Amazon. In addition, the ERP system comprises one or more of commodity modules, sales modules, purchasing modules, logistics modules, warehouse modules, financial modules, advertisement management modules, customer service modules, tool modules, permission management modules, data modules and other functional modules, wherein the functional modules can be mutually fused or independently exist, and one functional module can be used as a sub-module of another functional module. Operators of the ERP system of the present application may also be referred to as store managers, sellers, operators, or users, whose identity is not strictly defined unless specifically stated.

Meaning and description of e-commerce domain nouns in this application (letters other than english words and field symbols in this application are case-insensitive):

1. ERP (Enterprise Resource Planning) is an enterprise resource planning, which is a management platform based on information technology and provides decision operation means for enterprise decision-making layers and staff with systematic management ideas, and can also be used as an abbreviation of an ERP system in the application.

2. API (Application Programming Interface), an application programming interface, is a set of definitions, programs, and protocols that enable the communication or exchange of data between computer software and software via an API interface.

3. ASIN (Amazon standard identification number) is an amazon standard identification code, which is used primarily for product identification in a product catalog; SKU (Stock Keeping Unit) stock units, which are units for stock in-out metering; MSKU (Merchant Stock KeepingUnit), merchant/manufacturer inventory units; FBA (Fullfillment By Amazon) Amazon logistics distribution service

Drawings

The accompanying drawings are included to provide a further understanding of the application and are not to be construed as limiting the application; the content shown in the drawings can be real data of the embodiment, and belongs to the protection scope of the application.

FIG. 1 is a flow chart of an advertisement operation adjustment method according to an embodiment of the present application;

FIG. 2 is a logic diagram of an advertisement operation adjustment method according to an embodiment of the present application;

FIG. 3 is a schematic flow chart of a method for replenishing goods according to an embodiment of the present disclosure;

FIG. 4 is a schematic diagram illustrating a division of a store by an e-commerce platform according to an embodiment of the present application;

FIG. 5 is a flow chart of a database structure comparison method according to an embodiment of the present application;

FIG. 6 is a schematic diagram of a database structure according to an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the following description of the embodiments in the present application is further provided by the following detailed description with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.

As shown in fig. 1-2, the present embodiment provides an advertisement operation adjustment method, which includes steps S1-S5, specifically as follows:

s1, loading advertisement rules; taking the inferior electronic commerce platform as an example, in order to improve sales volume of commodities in a store, sellers usually perform advertisement promotion on the commodities in the store so as to increase exposure or drainage effect of the commodities, further improve click rate and success rate of buyers, and after advertisement rules are set for the delivered advertisements, the ERP system can correspondingly adjust the advertisements based on the advertisement rules, and further the advertisements can operate according to the expected directions of users. For example, the advertisement rules may include basic information and rule conditions; the basic information can specify a rule name, select a store, select the type of advertisement and select the time of effectiveness; the rule condition may specify a judgment condition, such as selecting a condition that the sum of the advertisement orders is 200 or more, and executing the price adjustment.

S2, judging whether the advertisement rule contains a target condition, wherein the target condition is to acquire data in the current time period, and the target condition in the embodiment is the time range of the current day, namely 24 hours; if yes, receiving the first state information and converting to the step S3, and if not, receiving the second state information and converting to the step S4;

and S3, judging whether the real-time data is changed or not based on the first state information, and acquiring the real-time data and report data generated by advertisement operation when the real-time data is changed. And when the real-time data is not changed, acquiring report data generated by advertisement operation. For example, assuming that the current date is 23 years 11 months 17, one rule condition in the advertisement rule is to count keywords (commodities) with advertisement click-through number greater than or equal to 500 in approximately 7 days, the advertisement is paused for 1 day, the controlled object is advertisement campaign a, which means that the advertisement campaign a is obtained for 7 days, and the advertisement campaign a is paused for 24 hours for keywords (commodities) with click-through number greater than or equal to 500; at this time, this near 7 days includes the current day, that is, the number of advertisements for the current day is to be acquired. Assuming that the date of the advertisement authorization (the access right of the STREAM interface of the e-commerce platform is obtained by the ERP system) is 11 months 15, only the real-time data from 16 to 17 can be obtained at this time based on the characteristics of the STREAM interface, but the advertisement data from 15 to 11 cannot be obtained. Further, acquiring a shop authorization date T1, an ending date T2 and a starting date T3 of time conditions in the advertisement rule, judging the relation between T1 and T2, and T3, and if T1 is in the interval between T2 and T3, adding 24 hours on the basis of the value of T1 to form T4; on the basis, real-time data in the interval from T4 to T3 is acquired through a STREAM interface, and REPORT data in the intervals from T4 to T2 is acquired through a REPORT interface. If the T1 is before the T2 and the T3, the pushed real-time data only needs to be obtained through the STREAM interface. Otherwise, when the real-time data is not monitored to be unchanged after a certain time point, the report data generated by the advertisement operation is acquired.

S4, judging the change state of the report data based on the second state information, and acquiring report data generated by advertisement operation when the report data is changed. When aiming at the following scene: assuming that the current date is 23 years, 11 months and 17, the advertisement authorization time is 11 months and 17, the advertisement rule loading time is 11 months and 17, the controlled object is advertisement activity B, and the specific condition is that the advertisement click quantity is more than or equal to 500 keywords (commodities) in the last 7 days are counted, the advertisement is put in for 1 day, and the advertisement put data in the last 2 days are excluded, namely 7 days before 11 months and 15 are acquired; at this time, the corresponding number cannot be obtained through the STREAM interface, so that the report data can only be obtained through the report interface as a basis for judgment at this stage. Furthermore, during the cold start stage of the store, the data cannot be acquired by using the STREAM interface, the ERP system cannot monitor the running condition of the advertisement and adjust the advertisement based on the running condition, and the acquired report data has certain hysteresis but can ensure the reliable running of the system.

And S5, judging whether the real-time data and/or the report data acquired in the step S3 or the step S4 meet the condition of the advertisement rule, and executing the adjustment action on the advertisement when the condition of the advertisement rule is met. Further, the ERP system calculates the acquired real-time data and report data according to the controlled object and the requirement condition of the advertisement rule (specified by the user); for example, when the keyword (commodity) with the advertisement click rate of 500 or more is counted for nearly 7 days, the advertisement delivery is paused for 1 day, when the condition is satisfied, the advertisement delivery for the corresponding commodity is paused for 1 day, and when the condition is not satisfied, the advertisement delivery is continued.

In a preferred embodiment, step S2 further includes determining whether the advertisement rule includes a target condition, where the target condition is that data in the current time period is obtained, if the advertisement rule includes the target condition, establishing a task of monitoring real-time data, and when the real-time data is monitored to be changed, generating first state information; if the advertisement rule does not contain the target condition, a monitoring task for the report data is established, and when the report data is monitored to change, second state information is generated. That is, the ERP system triggers corresponding actions according to the monitoring condition, namely, automatic adjustment of advertisements is achieved.

In a preferred embodiment, the method further comprises loading at least one controlled object based on advertisement rules, and establishing association between the controlled object and a monitoring task of real-time data and a monitoring task of report data; when the real-time data is monitored to be changed and the controlled object is contained, generating first state information; and when the report data is monitored to be changed and the report data contains the controlled object, generating second state information. Specifically, in the amazon platform, the seller often sets up a plurality of shops, and simultaneously, according to different operation demands, the seller also can choose to put in a plurality of advertisements for one or a plurality of commodities in different shops, so that in order to further improve the convenience of using the ERP system, the controlled object and the advertisement rule are mutually matched, and the fine control of advertisements can be realized. For example, assuming that the specific condition of the advertisement rule is a keyword (commodity) with advertisement click number of 500 or more in the last 7 days of store a, the delivery is suspended for 1 day; it is assumed that advertisements are respectively put on the B commodity and the C commodity in the store, and at this time, if the controlled object C is added, the advertisement put on the C commodity can be correspondingly monitored and adjusted.

In a preferred embodiment, step S3 further includes acquiring report data when the first status information is not generated or received within a preset period of time. Specifically, the STREAM interface pushes only updated data, since the data is pushed only when there is a change; furthermore, when the advertisement is paused, the budget is consumed or the exposure condition is not generated, the data is not changed and updated data is pushed at the STREAM interface, so that the specific advertisement delivery condition of the condition advertisement cannot be monitored. In order to monitor the situation of advertisement in the above scene and adjust the advertisement operation according to the monitored situation, it is necessary to obtain report data at regular time, so as to make up for the shortages caused by using the STREAM interface. For example, the user expects to pause the advertisement with an exposure less than 10 for approximately 7 days, assuming that no relevant exposure is available for 7 days, and thus no relevant data is available from the STREAM interface, and no exposure is generated for seven days, the pause is satisfied; the problem can be effectively solved by periodically acquiring corresponding data through the REPORT interface. Further, in this embodiment, when 23 points of each day are reached, the first status information is not generated or received within a preset period of time, and the ERP system requests to obtain REPORT data from the REPORT interface of the e-commerce platform, which of course can be flexibly set according to different usage requirements, for example, 22: 00. 22: 30. 23:30.

In a preferred embodiment, in step S5, the adjustment operation for performing the advertisement is pre-checked based on the controlled object and the advertisement rule, and if the condition of the pre-check is satisfied, the adjustment operation for the advertisement is performed, and if the condition of the pre-check is not satisfied, the adjustment operation for the advertisement is not performed. The pre-check condition at least comprises whether the store is authorized, whether the controlled object is added into the advertisement rule, and whether the controlled object is in the effective time, and when the store is in the authorized state, the controlled object is added into the advertisement rule, and the controlled object is in the effective time, the adjustment action of the advertisement is executed. Specifically, after the ERP system performs the monitoring task on the advertisement according to the advertisement rule and the controlled object, the store which is in the authorized state before is de-authorized and the rule is modified due to some other reasons, and then the specific controlled object is forgotten to be designated, so that the ERP system is caused to have errors when triggering the action of adjusting the advertisement, and therefore, when the ERP system needs to confirm again before triggering the action of adjusting the advertisement. In this embodiment, before the adjustment action of the ERP system on the advertisement is triggered, whether the store is authorized, whether the controlled object is added into the advertisement rule, and whether the controlled object is within the effective time are further confirmed, when all the conditions are satisfied, the advertisement adjustment action is executed, and when any one of the conditions is not satisfied, the adjustment action on the advertisement is stopped and prompt information is sent to the user, thereby improving the operation reliability of the ERP system.

In a preferred embodiment, in step S5, the reporting data and/or the real-time data are processed, comprising the steps of:

s51, screening report data and/or real-time data based on the controlled object, namely screening real-time data and/or report data related to the controlled object;

s52, acquiring a set of report data and/or real-time data in a maximum interval based on advertisement rules; assuming that the rule condition has a plurality of time spans, for example, condition 1 is a keyword (commodity) with click rate exceeding 500 in nearly 30 days, condition 2 is a keyword (commodity) with exposure exceeding 100 in nearly 15 days, and condition 3 is a keyword (commodity) with advertisement traffic exceeding 10 in nearly 7 days, in order to reduce network delay generated when each condition is independently executed, at this time, the ERP system selects and acquires all data in the maximum interval, namely 30 days, that is, the data are firstly collected locally and then are respectively processed according to each condition, and as network request, response and transmission delay which are not respectively executed are not executed, the processing speed of advertisement data can be effectively improved.

S53, assembling the collection of report data and/or real-time data into a calculation chain according to a preset rule; the calculation chain classifies parameters in the real-time data and the report data according to a plurality of preset labels and calculates the parameters respectively. Specifically, the rule data relates to calculation of a plurality of indexes in a plurality of advertisement data, such as order quantity, click quantity, exposure quantity and the like, and the calculation chain forms a plurality of independent calculation programs for the indexes respectively, when real-time data and/or report data are acquired, parameters such as the order quantity, the click quantity, the exposure quantity and the like are extracted respectively and are directly filled into the corresponding calculation programs, and then results of the order quantity, the click quantity and the exposure quantity are directly calculated, so that the processing speed of the data is improved.

S54, executing a calculation chain and generating a calculation result.

In another embodiment, when the rule condition needs to acquire data for more than 30 days, real-time data acquired through the STREAM interface can be read from the local database for data within 30 days, and REPORT data acquired through the REPORT interface can be pulled from the local data or the e-commerce platform for data over 30 days. Specifically, since the STREAM interface provides data of a delivery level, the granularity of the data is relatively thin, which means that the quantity is very large, and in order to control the input of resources, only data of about 30 days acquired in the STREAM interface is generally stored in a database under the ERP system, so that if the ERP system needs to acquire advertisement data more than 30 days ago, the data can only be acquired from the reporting interface to participate in calculation.

Aiming at the problem that a newly authorized store is in a cold start stage, because of the existence of a STREAM interface, the real-time data acquired through the STREAM interface cannot solve the problem that an ERP system or an advertisement management module can not accurately adjust the operation of advertisements, and REPORT data are acquired through a REPORT interface and are mutually cooperated with the real-time data, so that the overall accuracy of the advertisement data is effectively improved, the ERP system can accurately adjust or automatically control the operation of advertisements, and meanwhile, the reliability of the ERP system is improved.

As shown in fig. 3-4, the present embodiment provides a commodity replenishment method for an ERP system of an e-commerce platform, the commodity replenishment method comprising steps K1 to K5, specifically including:

k1, acquiring a plurality of intra-area shops based on a target large area or acquiring a plurality of domestic shops based on a target country. In the Amazon E-commerce platform, a target large area is mainly divided into European and North America areas of European Union; the European Union mainly includes French, germany, italy, spanish and Netherlands, and the United kingdom is in a unoOutlet state. The north american regions mainly include the united states, canada, and mexico. Further, in this case, the acquisition of a plurality of stores in the european union region based on the large area means that stores in all countries in the european union region are selected, assuming that the seller has opened stores in all countries in the european union region. The acquisition of a plurality of domestic shops based on the target country refers to acquisition of shops set in a specific country, such as selection of all shops set in the deceased country.

K2, acquiring in-area commodity inventory data and in-area commodity historical sales data of the in-area shops based on a standard identification code (ASIN), or acquiring domestic commodity inventory data and domestic commodity historical sales data of the domestic shops based on the standard identification code. Specifically, when a certain commodity needs to be restocked, the corresponding commodity can be selected through the standard identification code of the commodity, and further, according to the standard identification code, all commodities with the same standard identification code in a region can be selected, or all commodities with the same standard identification code in a certain country can be selected.

And K3, performing duplicate removal processing on the commodity inventory data in the district or the domestic commodity inventory data based on Stock Keeping Unit (SKU), and forming duplicate removal commodity inventory data in the district or duplicate removal domestic commodity inventory data. Specifically, when the seller uses the amazon logistics distribution service (FBA) in whole or in part, the commodities of the stores in all or part of the countries in one large area (e.g., the european union) are in shared inventory, i.e., the stores in all the countries in the area share one inventory; the same is true when a plurality of stores are opened in one country. Further, at this time, the goods with the same stock-keeping unit code are further encoded by the stock-keeping unit code (SKU) on the basis of the goods screened by the standard identification code, and finally the data of the goods with the same stock-keeping unit code are combined into one piece, namely, the duplicate removal processing is performed. Therefore, the problem of repeated deduction of inventory goods in the process of calculating the replenishment proposal is avoided, and the accuracy of the replenishment proposal is effectively improved.

And K4, respectively calculating the total number of the commodity inventory data in the duplicate removal area and the commodity historical sales data in the area, or respectively calculating the total number of the duplicate removal domestic commodity inventory data and the domestic commodity historical sales data. After calculating the total number of the inventory data of the products in the duplication eliminating area and the historical sales data of the products in the area, the corresponding total number of the inventory data of the products in the duplication eliminating area and the historical sales data of the products in the area can be generated based on the total number of the inventory data of the products in the duplication eliminating area and the historical sales data of the products in the area to generate the replenishment suggestion. The same is true for the deduplication in-district product inventory data and in-district product historical sales data.

And K5, generating a first replenishment suggestion based on the total number of the product inventory data in the duplication removal region and the product historical sales volume data in the region, or generating a second replenishment suggestion based on the total number of the duplicate removal domestic product inventory data and the domestic product historical sales volume data. The seller may choose to generate the first replenishment proposal based on a large block across stores or the second replenishment proposal based on a national across stores depending on the actual needs of the seller. For example, if the seller has opened stores in all countries in the EU, then the seller may choose to generate a first replenishment proposal based on a large area across stores or a second replenishment proposal based on a country across stores; if the seller opens a store in only one or two countries of a certain European Union, the seller at that time can directly choose to generate a second replenishment proposal across stores based on the country to reduce unnecessary system computation. Further, the improvement point in the technical solution of the present application is mainly that the pre-stage of generating the final replenishment proposal, that is, how to generate the replenishment proposal is not the technical problem to be solved by the present application; regarding the generation of the replenishment proposal, reference may be made specifically to application No. 202310272748.6, and the patent name is a corresponding technical scheme in a commodity replenishment method and system based on target inventory, so that a detailed description of the technical scheme of how to generate the replenishment proposal is not provided in this application.

In a preferred embodiment, in step K2, acquiring the in-district commodity inventory data or the domestic commodity inventory data of the in-district store includes:

k21, obtaining in-zone local in-warehouse quantity, in-zone FBA in-warehouse quantity, in-zone overseas in-warehouse quantity and in-zone overseas in-warehouse quantity, or obtaining domestic local in-warehouse quantity, domestic FBA in-warehouse quantity, domestic overseas in-warehouse quantity and domestic overseas in-warehouse quantity. The local warehouse in the middle area refers to a warehouse built by a seller; the amount of the warehouse (local warehouse in the area, FBA warehouse in the area, overseas warehouse in the area) refers to the amount of the commodity which has arrived and is stored in the warehouse; the FBA warehouse means a warehouse provided by amazon after selecting amazon logistics distribution service, and the in-transit quantity (local warehouse in a zone, FBA warehouse in a zone, oversea warehouse in a zone, overseas warehouse in a zone) means the quantity of goods in transit. Overseas bins refer to warehouses built by domestic sellers in target countries for reducing operating costs. Similarly, reference is made to the above description for domestic local in-store quantity, domestic FBA in-store quantity, domestic overseas in-store quantity, and domestic overseas in-store.

K22, calculating the total number of the local in-zone warehouse quantity, the FBA warehouse in-zone warehouse quantity, the outside-sea warehouse quantity and the outside-sea warehouse in-zone warehouse quantity, or calculating the total number of the domestic local warehouse quantity, the domestic local warehouse in-process quantity, the domestic FBA warehouse in-process quantity, the domestic overseas warehouse quantity and the domestic overseas warehouse in-process quantity. Once the corresponding inventory totals and historical sales totals are calculated, corresponding replenishment recommendations may be calculated and generated based on the data. For example, when the historical sales are calculated, the subsequent daily sales can be calculated based on the historical sales.

In the preferred implementation, step K3 further includes screening out target data in the area, which is the same as the second commodity code, in the inventory data of the commodity in the area, and merging the target data in the area into one piece when the target data in the area, which is the same as the second commodity code, has more than two pieces; or screening domestic target data with the same code as the second commodity in the domestic commodity inventory data, and combining the domestic target data into one piece when the second commodity code has more than two domestic target data. And further, the problem that the stock quantity is repeatedly deducted in the calculation process of generating the replenishment proposal is avoided.

For example, assume that stores in two countries under the eu have selected FBA services, i.e., stores under the two countries share one inventory at this time; for example, the relevant features of two commodities are ASIN1, MSKU1, store 1 (Germany), SKU1, respectively; ASIN1, MSKU1, store 3 (France), SKU1; further, assume that the stock amounts of the two commodities are 100, and in this case, in the calculation process of generating the replenishment proposal, the two screened commodity data with the same stock unit code are aggregated into one stock amount, namely, the total stock amount is calculated according to 100 pieces; for another example, according to the sequencing, the related information of the commodity of the latter item is ignored, namely, the related information does not participate in calculation; or, setting the stock quantity of the latter commodity to zero; alternatively, when the number of goods can be divided, the number of goods stock may be averaged (50 pieces) according to the number of pieces of data of which stock unit codes are the same; the problem of repeated deduction of inventory commodities in the process of calculating the replenishment proposal is avoided, and the precision of the replenishment proposal given by the ERP system is further effectively improved.

Further, when the duplicate removal processing of the stock data is performed, whether the number parameters of the acquired target data in the area are the same is judged, if so, the target data in the area are combined into one piece, and if not, the data synchronization processing is performed, so that the number parameters of the target data in the area are the same, and then the target data in the area are combined into one piece. Or judging whether the number parameters of the domestic target data are the same, if so, merging the domestic target data into one piece, and if not, performing data synchronization processing to ensure that the number parameters of the domestic target data are the same, and merging the domestic target data into one piece.

In the running process of the ERP system, an API interface of the e-commerce platform is required to be accessed and data is acquired, and the acquired quantity parameters of the target data in the region or the quantity parameters of the target data in the region are possibly inconsistent due to delay of network communication or in a refreshing blank file of system data; at this time, if the inconsistent data participates in calculation, the generated replenishment proposal is inaccurate or the system is wrongly reported; therefore, when the deduplication processing is performed, the comparison processing is performed on the acquired quantity parameters of the target data in the area, so that the problems can be effectively solved.

In a preferred embodiment, in step K2, the target large area includes a european union area and a north american area, and when acquiring the commodity inventory data of the european union area based on the store of the european union area, the method includes:

k211, acquiring a first parameter set in commodity inventory data of the European Union region through an API (application program interface) of an e-commerce platform;

k212, acquiring British commodity inventory data under a British store through an API (application program interface) of the electronic commerce platform, and acquiring a second parameter set in the British commodity inventory data;

and K213, judging whether the second parameter set is matched with the first parameter set, and if the second parameter set is successfully matched with the first parameter set, aggregating the acquired commodity inventory data of the European Union area and the acquired commodity inventory data of the British station.

Specifically, when the united kingdom is in the unobtrusive state, and when the replenishment proposal is generated across stores based on the large area, and the united kingdom actually chooses to use the FBA service or to follow-up corresponding commodities, the seller needs to try to solve the aggregation problem of inventory sharing between the eu area and the united kingdom by itself, that is, when the replenishment proposal is generated across stores based on the large area, the business data in the united kingdom can be automatically acquired and calculated, so that the inconvenience (difficult to realize the system and required to be manually processed) caused by separately calculating and adding the united kingdom data when the replenishment proposal is generated is avoided. Further, when the first parameter set or the second parameter set includes the network stock keeping unit code (fnsu), the vendor stock keeping unit code (MSKU), the standard identification code (ASIN) and the commodity total amount parameter (total quality), after the first parameter set and the second parameter set are successfully matched, the network stock keeping unit code (fnsu), the vendor stock keeping unit code (MSKU) and the standard identification code (ASIN) are the same, and the assignment of the commodity total amount parameter is non-zero, which means that the shops in the english area and the shops in the eu area have been subjected to stock sharing, then in the following calculation for generating the replenishment proposal, the commodity data of the united kingdom can be directly acquired and participated in the calculation when the commodity data is acquired across the shops based on the eu area.

In a preferred embodiment, obtaining an in-area to-be-processed order of a store in an area based on a standard identification code, calculating the total number of commodities in the in-area to-be-processed order based on the standard identification code, and forming an in-area floating replenishment quantity; or acquiring domestic to-be-processed orders of domestic shops based on the standard identification codes, calculating the total number of commodities in the domestic to-be-processed orders based on the standard identification codes, and forming the floating replenishment quantity in the area.

Specifically, on the e-commerce platform, when a new product of a certain burst is always kept at a higher sales level in a certain period, that is, a large number of orders which cannot be quickly digested in a preset time are generated, the ERP system also participates in calculation of the part of the to-be-processed orders when the replenishment proposal is generated, that is, forms a floating replenishment amount, so that the precision of the generated replenishment proposal is further improved.

In a preferred embodiment, in step K2, the method further includes performing aggregate display on the acquired plurality of pieces of product inventory data in the district based on the standard identification code, or performing aggregate display on the acquired plurality of pieces of domestic product inventory data based on the standard identification code. And carrying out aggregation display on the acquired inventory data of the commodities in the plurality of areas based on the inventory unit code, or carrying out aggregation display on the acquired inventory data of the domestic commodities based on the inventory unit code. The commodities screened based on the standard identification codes and the stock unit codes are displayed in an aggregation mode, so that sellers or operators can intuitively know the states and stores, namely the commodities in the stores are calculated in a mode of crossing the stores based on the region, and the commodities in the stores are calculated in a mode of crossing the stores based on the state.

The method selects corresponding commodities by crossing shops in a large area or crossing shops in a country and based on standard identification codes, further carries out duplication elimination processing on the stock of the corresponding commodities according to stock keeping unit codes (SKUs), and finally generates a replenishment suggestion based on stock data after duplication elimination processing, so that the problem that the same stock is repeatedly deducted in the process of generating the replenishment suggestion under certain specific scenes is solved on the basis of improving the calculation efficiency of a commodity replenishment method and a corresponding ERP system for generating the replenishment suggestion and reducing the operation cost, and the replenishment suggestion precision is improved; meanwhile, compared with the traditional mode, the method avoids the mode that the goods supplementing quantity required by each store is calculated firstly and finally the goods supplementing quantity required by each store is calculated in a summarizing mode, and therefore time efficiency of calculating and generating goods supplementing suggestions is improved effectively.

As shown in fig. 5-6, the present embodiment provides a database structure comparison method for use in an ERP system or other system using multiple databases, the method comprising the steps of:

p1, at least acquiring a database D1 and a database D2, and respectively acquiring a structural feature set D1J of the database D1 and a structural feature set D2J of the database D2. The database D1 and the database D2 herein refer to databases created under MYSQL database management system in database software, for example, when the usage scale of the ERP system is large, a large amount of data is often generated, and at this time, a plurality of databases are used to store the data so as to ensure the operation performance of the ERP system. Further, in the present embodiment, for the sake of more clear description of the technical solution, only two databases D1 and D2 are used for description, and in the actual use process, more than two databases may be compared according to different use requirements.

P2, combining the structural feature set D1J with the structural feature set D2J, performing de-duplication treatment on the combined feature set, and forming a feature union UA. In particular, it is assumed that in the case where databases D1 and D2 are free of errors, the structural feature sets D1J and D2J in the data D1 and D2 are identical, i.e., there are two for each feature after merging; further, the feature set is subjected to a deduplication process, i.e., only one feature is taken from the same two features, so that a feature union UA is obtained. Specifically, in the absence of errors, the structural feature sets D1J and D2J in data D1 and D2 are identical, i.e., there are two for each feature after merging; further, the feature set is subjected to a deduplication process, i.e., only one feature is taken from the same two features, so that a feature union UA is obtained.

P3, respectively calculating a difference set DIF-D1 of the feature union UA and the structural feature set D1J, and a difference set DIF-D2 of the feature union UA and the structural feature set D2J; specifically, it is assumed that in the case where databases D1 and D2 have no errors, the difference set of the union UA subtracted from the feature set D1J must be a null value at this time. When the structural feature set D1J or D2J is missing or different, there is a difference set that is not a null value, and the difference set that is not a null value is a problem that needs to be removed by a developer.

And P4, outputting the results of the difference set DIF-D1 and the difference set DIF-D2 respectively.

In the preferred embodiment, since there are multiple features in the database, i.e., one database mainly includes tables and columns and indexes in the tables, performing comparison feature by feature can ensure that the accuracy of the comparison is maximized when performing database comparison. Further, the structural feature set D1J includes the table name set UD1 in the acquisition database D1, and the structural feature set D2J includes the table name set UD2 in the data D2. The table name set UD1 and the table name set UD2 are combined and subjected to de-duplication processing to form a table name union UD, and a table name difference set DIF-UD1 of the table name union UD and the table name set UD1 and a table name difference set DIF-UD2 of the table name union UD and the table name set UD2 are calculated respectively. Judging whether the table name difference set DIF-UD1 and the table name difference set DIF-UD2 are null values, and when the table name difference set DIF-UD1 and the table name difference set DIF-UD2 are not null values, respectively outputting table name comparison results of non-null values.

For example, table name set UD1 (card 1, card 2, table 1) in database D1 is acquired, and table name set UD2 (card 1, card 2, table1, table 2) in database D2 is acquired; combining the table name sets UD1 and UD2 to obtain (board 1, board 2, table1, board 2, table1 and table 2), and performing de-duplication treatment to obtain a table name union UD (board 1, board 2, table1 and table 2); calculating and obtaining a table name difference set DIF-UD1 (table 2) of a table name union UD (table 1, table2, table 1) and a table name set UD1 (table 1, table2, table 1), namely, subtracting the same characteristics in the table name union UD and the table name set UD 1; similarly, a table name difference set DFI-UD2 () of the table name union UD (card 1, card 2, table1, table 2) and the table name set UD2 (card 1, card 2, table1, table 2) is calculated and obtained. Further, according to the result of the table name difference set DIF-UD1 (table 2) is not empty, at this time, it can be known that the table2 is absent in the database D1; and the result of the table name difference set DFI-UD2 () is a null value indicating that the table in database D2 is not missing.

In a preferred embodiment, whether the table name difference set DIF-UD1 and the table name difference set DIF-UD2 are null values or not is judged respectively, if not, table names identical to those in the table name difference set DIF-UD1 and the table name difference set DIF-UD2 are removed from the table name set UD1, a column name set UDC1 under each table in the table name set UD1 after the duplication removal is obtained respectively, table names identical to those in the table name difference set DIF-UD1 and the table name difference set DIF-UD2 are removed from the table name set UD2, and a column name set UDC2 under each table is obtained respectively; before a table (a table corresponding to an output value in a table name difference set) which may be problematic is not corrected, calculation is not participated in so as to ensure accuracy of comparison of the next columns, and therefore the table (the table in the difference set) which may be problematic needs to be temporarily excluded before the column comparison.

Further, merging the column name set UDC1 and the column name set UDC2, and performing de-duplication treatment to form a column name union UDC; respectively calculating a column name difference set DIF-UDC1 of a column name union UDC and a column name set UDC1 and a column name difference set DIF-UDC2 of the column name union UDC and a column name set UDC2; judging whether the column name difference set DIF-UDC1 and the column name difference set DIF-UDC2 are null values, and outputting a column name comparison result with non-null values when the column name difference set DIF-UD1 and the column name difference set DIF-UD2 are not null values.

For example, the table name difference set DIF-UD1 (table 2) has a result that table2 is not empty, that is, table name set UD2 (table 1, table2, table1, table 2) contains table2, so table2 needs to be excluded from table name set UD2, and table name set UD1 and table name set UD2 that participate in comparison are matched. Next, column name sets UDC1 and UDC2 in the table name set UD1 and UD2 are obtained, respectively, further, it is assumed that the specific column name set UDC1 is { card 1 (id, log, create _time), card 2 (id, log, create _time), table1 (id, name, age) }, UDC2 is { card 1 (id, log), card 2 (id, log, create _time), table1 (id, name) }.

Further, the column name set UDC1 and the column name set UDC2 are combined and de-duplicated to form a column name union UDC { board 1 (id, log, create _time), board 2 (id, log, create _time), table1 (id, name, age) }, and column name difference sets DIF-UDC1{ board 1 of the column name union UDC and the column name set UDC1 and the column name set UDC2 are calculated and obtained respectively

() Board 2 (), table1 () }, and a column name difference set DIF-UDC2{ Board 1 }

(create_time), board 2 (), table1 (age) }; further, the calculation result of the column name difference set DIF-UDC1 is a null value, so that the situation that no column deletion occurs in the database D1 can be judged; the column name difference set DIF-UDC2 results in { board 1 (create_time), board 2 (), table1 (age) }, and thus we can determine that the "create_time" column is missing in the board 1 table, the board 2 () is null, the no missing column is in the board 2 table, and the "age" column is missing in the table1 table of the database D2. And according to the output result of the database comparison, a developer can conduct corresponding error checking and correction on the database.

In the preferred embodiment, whether the table name difference set DIF-UD1 and the table name difference set DIF-UD2 are null values or not is judged respectively, if not, the same table names as the table name difference set DIF-UD1 and the table name difference set DIF-UD2 are removed from the table name set UD1, index sets UDI1 under the tables in the table name set UD1 after the duplication removal process are obtained respectively, the same table names as the table name difference set DIF-UD1 and the table name difference set DIF-UD2 are removed from the table name set UD2, and the index sets UDI2 under the tables are obtained respectively. Before a table (a column corresponding to an output value in a table name difference set) which may be problematic is not corrected, calculation is not participated in temporarily, so that accuracy of comparison of the subsequent indexes is ensured, and therefore the table (the table in the difference set) which may be problematic needs to be temporarily excluded before the index comparison.

Further, merging and deduplicating the index set UDI1 and the index set UDI2 to form an index union UDI; respectively calculating an index difference set DIF-UDI1 of the index union UDI and the index set UDI1 and an index difference set DIF-UDI2 of the index union UDI and the index set UDI2; and judging whether the index difference set DIF-UDI1 and the index difference set DIF-UDI2 are null values, and outputting an index comparison result of non-null values when the index difference set DIF-UDI1 and the index difference set DIF-UDI2 are not null values.

For example, the table name difference set DIF-UD1 (table 2) has a result that table2 is not empty, that is, table2 is a result, table name set UD2 (table 1, table2, table1, table 2) contains table2, and therefore table2 needs to be removed from table name set UD2, so that table name set UD1 and table name set UD2 that participate in comparison agree. Next, the index set UDI1 and the index set UDI2 in the table name set UD1 and the table name set UD2 are acquired respectively, further, it is assumed that the specific index set UDI1 is { card 1 (), card 2 (log), table1 (id, name) }, UDI2 is

{shard1(),shard2(log),table1(id、name-age)}。

Further, merging and de-duplicating the index set UDI1 and the index set UDI2 to form an index union UDI { card 1 (), card 2 (log), table1 (id, name, name-age) }, and then respectively calculating and obtaining an index difference set DIF-UDI1{ card 1 (), card 2 (), table1 (name-age) }, and an index difference set DIF-UDI2{ card 1 (), card 2 (), table1 (name) }, respectively; further, the result of the index difference set DIF-UDI1 is { Board 1 (), board 2 (), table1 (name-age) }, so that the index with the name-age as the missing field in the Table1 can be judged; the result of the index difference set DIF-UDI2 is { Standard 1 (), standard 2 (), table1 (name) }, thereby judging that the missing field in the database D2 is the index of the name. Further, the developer performs corresponding error checking and correction on the database according to the index comparison result.

Further, in the preferred embodiment, the field types in the database may be further compared, specifically, the data will be selected to be different field types according to different usage requirements during the usage process, for example, a field of numeric type is generally selected for the age in the table, a field of text type is generally used for the name, and the limiting field length is 10. The type of the field is often modified from text type to digital type during development due to misoperation and other factors, and the length of the text type field is modified by mistake. It is assumed that the length of the field type in a certain table is modified to 20 due to 10 errors, and at this time, although the problem of error reporting does not occur in the running process of the database, after the length of the field is increased, the occupation of the storage space is increased. The comparison implementation of field types in data may refer to the comparison implementation of indexes, and thus will not be described in detail herein.

In the preferred embodiment, when the comparison of multiple data additions is completed and the sub-tables in a single database are required to be compared, it is assumed that 1 hundred million pieces of data need to be stored in the database D1, and if only one table is used to store 1 hundred million pieces of data, the data amount appears to be relatively large for the single table; in order to alleviate the storage pressure of a single table on data, multiple sub tables can be used to cooperatively store data, for example, 10 tables are used, so that each table only needs to store tens of millions of data. Similarly, when there are multiple sub-tables, it is also necessary to maintain consistency between sub-tables, that is, further comparing sub-tables in a single library may further improve possible errors in the database structure. Specifically, the comparison among the databases is transverse comparison, and the sub-table comparison among the single databases belongs to longitudinal comparison; for example, the database D1 has the tables of the card 1, the card 2, and the card 3, and the database D2 has the tables of the card 1, the card 2, and the card 3; the comparison between the multiple libraries can solve the structural consistency of the data D1 and D2, namely the data D1 and D1, the data D2 and D2, and the data D3 and D3, and the comparison between the single libraries can solve the structural consistency of the data D1, the data D2 and D3. Further, in some extreme cases, it is assumed that the structure of the card 1 table in the database D1 is the same as the structure of the card 1 table in the database D2, and in this case, the error cannot be found by cross-library comparison, that is, the inter-library comparison, and the error can be found by sub-table comparison in a single library.

Further, when the single library comparison operation is performed, at least the sub-table FUD11 and the sub-table FUD12 are acquired from the database D1, at least the sub-table name set FUD1C1 under the sub-table FUD11 and the sub-table name set FUD1C2 under the sub-table FUD12 are acquired, or at least the sub-table FUD21 and the sub-table FUD22 are acquired from the database D2, and the sub-table name set FUD2C1 under the sub-table FUD21 and the sub-table name set FUD2C2 under the sub-table FUD22 are acquired, respectively.

The split list name set FUD1C1 and the split list name set FUD1C2 are combined and subjected to de-duplication treatment to form a split list name union FUD1C, or the split list name set FUD2C1 and the split list name set FUD2C2 are combined and subjected to de-duplication treatment to form a split list name union FUD2C.

The partial list name difference sets DIF-FUD1C1, and the partial list name difference sets DFI-FUDC2, or the partial list name difference sets DIF-FUD2C1, and the partial list name difference sets FUD2C2, are calculated respectively.

Judging whether the differential list name set DIF-FUD1C1 and the differential list name set DIF-FUD1C2 are null values, when the differential list name set DIF-FUD1C1 and the differential list name set DFI-FUD1C2 are not null values, or judging whether the differential list name set DIF-FUD2C1 and the differential list name set DIF-FUD2C2 are null values, and when the differential list name set DIF-FUD2C1 and the differential list name set DIF-FUD2C2 are not null values, respectively outputting differential list name comparison results of non-null values. Further, the developer performs corresponding error checking and correction on the database according to the comparison result of the columns

In the preferred embodiment, the sub-table index set FUD1I1 under each column of the sub-table name set FUD1C1 and the sub-table index set FUD1I2 under each column of the sub-table name set FUD1C2 are obtained, respectively, or the sub-table index set FUD2I1 under each column of the sub-table name set FUD2C1 and the sub-table index set FUD2I2 under each column of the sub-table name set FUC2C2 are obtained, respectively.

The method comprises the steps of merging and de-duplicating a sub-table index set FUD1I1 and a sub-table index set FUD1I2 to form a sub-table index union FUD1I, or merging and de-duplicating a sub-table index set FUD2I1 and a sub-table index set FUD2I2 to form a sub-table index union FUD2I.

The difference sets DIF-FUD1I1 of the sub-table index union FUD1I and FUD1I1 of the sub-table index set FUD1I1 and DIF-FUD1I2 of the sub-table index union FUD1I and FUD1I2 of the sub-table index set FUD1I2 are calculated respectively, or, the difference set DIF-FUD2I1 of the sub-table index union FUD2I and the sub-table index set FUD2I1 and the difference set DIF-FUD2I2 of the sub-table index union FUD2I and the sub-table index set FUD2I2 are calculated respectively.

Judging whether the differential set DIF-FUD1I1 and the differential set DIF-FUD1I2 are null values, when the differential set DIF-FUD1I1 and the differential set DIF-FUD1I2 are not null values, or judging whether the differential set DIF-FUD2I1 and the differential set DIF-FUD2I2 are null values, and when the differential set DIF-FUD2I1 and the differential set DIF-FUD2I2 are not null values, respectively outputting non-null value differential index comparison results. Further, the developer performs corresponding error checking and correction on the database according to the index comparison result

Further, the process of performing single-library comparison is the same as the process of performing multi-library comparison, and the main difference is that there is no comparison between tables, i.e., the single-library comparison is only required to compare columns and indexes in each sub-table; therefore, a detailed description of the comparison process will not be made here.

The present embodiment also provides a database comparing system for checking whether there is an error in a database, and the database comparing system performs the operation steps in the database structure comparing method as described above. In the preferred embodiment, when a table is wrong in the multi-library comparison process, the next column and index comparison are not involved in calculation, that is, if the table with problems is not excluded, the next column and index comparison are caused to have problems; thus, when a comparison is performed, errors in the table, column, and index are corrected; the comparison operation is re-performed one or more times again until all errors in the database have been corrected.

The tables in all the libraries, the columns in all the tables and the index union set in all the tables are used as the reference, and then the tables in a single library, the columns in a single table and the index union set in a single table are compared with the reference union set, so that the comparison of a plurality of databases without specifying the reference tables can be realized, and according to the comparison result, the inconsistent information that a certain table does not exist in a certain library and a certain column and index does not exist in a certain table can be obtained. The final developer can judge whether a certain table, a certain column or an index is redundant or missing according to the output information so as to process. The technical scheme can compare N databases, can compare under the condition that which database is not known as the reference or does not need to designate the reference database, thereby effectively improving the comparison precision of the databases and ensuring the operation stability and reliability of ERP.

The foregoing description is only of the preferred embodiments of the present application and is not intended to limit the scope of the claims, and all equivalent modifications made by the specification and drawings of the present application, or direct/indirect application in other related technical fields are included in the scope of the claims of the present application.

Claims

1. A database structure comparison method, comprising the steps of:

p1, at least acquiring a database D1 and a database D2, and respectively acquiring a structural feature set D1J of the database D1 and a structural feature set D2J of the database D2;

p2, combining the structural feature set D1J with the structural feature set D2J, performing de-duplication treatment on the combined feature set, and forming a feature union UA;

p3, respectively calculating a difference set DIF-D1 between the feature union UA and the structural feature set D1J, and a difference set DIF-D2 between the feature union UA and the structural feature set D2J;

and P4, respectively outputting the results of the difference set DIF-D1 and the difference set DIF-D2.

2. The database structure comparison method according to claim 1, wherein the structural feature set D1J includes a table name set UD1 in the database D1, and the structural feature set D2J includes a table name set UD2 in the data D2;

Merging the table name set UD1 and the table name set UD2 and performing de-duplication processing to form a table name union UD, and respectively calculating a table name difference set DIF-UD1 of the table name union UD and the table name set UD1 and a table name difference set DIF-UD2 of the table name union UD and the table name set UD2;

judging whether the table name difference set DIF-UD1 and the table name difference set DIF-UD2 are null values, and when the table name difference set DIF-UD1 and the table name difference set DIF-UD2 are not null values, respectively outputting table name comparison results of non-null values.

3. The database structure comparing method according to claim 2, wherein whether the table name difference set DIF-UD1 and the table name difference set DIF-UD2 are null values is determined, if not, the same table names as the table name difference set DIF-UD1 and the table name difference set DIF-UD2 are excluded from the table name set UD1, column name sets UDC1 under the tables in the table name set UD1 after the duplication removal process are acquired, table names as the table name difference set DIF-UD1 and the table name difference set DIF-UD2 are excluded from the table name set UD2, and column name sets UDC2 under the tables are acquired, respectively.

4. A database structure comparison method according to claim 3, wherein the column name set UDC1 and the column name set UDC2 are combined and de-duplicated to form a column name union UDC;

Respectively calculating a column name difference set DIF-UDC1 of the column name union UDC and the column name set UDC1 and a column name difference set DIF-UDC2 of the column name union UDC and the column name set UDC2;

judging whether the column name difference set DIF-UDC1 and the column name difference set DIF-UDC2 are null values, and outputting a column name comparison result of non-null values when the column name difference set DIF-UD1 and the column name difference set DIF-UD2 are not null values.

5. The database structure comparing method according to claim 2, wherein whether the table name difference set DIF-UD1 and the table name difference set DIF-UD2 are null values is determined, if not, the same table names as the table name difference set DIF-UD1 and the table name difference set DIF-UD2 are excluded from the table name set UD1, index sets UDI1 under respective tables in the table name set UD1 after the duplication removal process are acquired, the same table names as the table name difference set DIF-UD1 and the table name difference set DIF-UD2 are excluded from the table name set UD2, and index sets UDI2 under respective tables are acquired, respectively.

6. The database structure comparison method according to claim 5, wherein the index set UDI1 and the index set UDI2 are combined and de-duplicated to form an index union UDI;

Respectively calculating an index difference set DIF-UDI1 of the index union UDI and the index set UDI1 and an index difference set DIF-UDI2 of the index union UDI and the index set UDI2;

and judging whether the index difference set DIF-UDI1 and the index difference set DIF-UDI2 are null values, and outputting an index comparison result of non-null values when the index difference set DIF-UDI1 and the index difference set DIF-UDI2 are not null values.

7. The database structure comparison method according to claim 1, wherein at least a sub-table FUD11 and a sub-table FUD12 are obtained from the database D1, at least a sub-table name set FUD1C1 under the sub-table FUD11 and a sub-table name set FUD1C2 under the sub-table FUD12 are obtained, or at least a sub-table FUD21 and a sub-table FUD22 are obtained from the database D2, and a sub-table name set FUD2C1 under the sub-table FUD21 and a sub-table name set FUD2C2 under the sub-table FUD22 are obtained, respectively;

combining and de-duplicating the sub-table name set FUD1C1 and the sub-table name set FUD1C2 to form a sub-table name union FUD1C, or combining and de-duplicating the sub-table name set FUD2C1 and the sub-table name set FUD2C2 to form a sub-table name union FUD2C;

Calculating a partial list name difference set DIF-FUD1C1 of the partial list name union FUD1C and the partial list name set FUD1C1 and a partial list name difference set DFI-FUDC2 of the partial list name union FUD1C2 and the partial list name set FUD1C2 respectively, or calculating a partial list name difference set DIF-FUD2C1 of the partial list name union FUD2C and the partial list name set FUD2C1 and a partial list name difference set DIF-FUD2C2 of the partial list name union FUD2C2 and the partial list name set FUD2C2 respectively;

judging whether the differential list name set DIF-FUD1C1 and the differential list name set DIF-FUD1C2 are null values, when the differential list name set DIF-FUD1C1 and the differential list name set DFI-FUD1C2 are not null values, or judging whether the differential list name set DIF-FUD2C1 and the differential list name set DIF-FUD2C2 are null values, and when the differential list name set DIF-FUD2C1 and the differential list name set DIF-FUD2C2 are not null values, respectively outputting differential list name comparison results of non-null values.

8. The database structure comparing method according to claim 7, wherein the sub-table index set FUD1I1 under each column of the sub-table name set FUD1C1 and the sub-table index set FUD1I2 under each column of the sub-table name set FUD1C2 are obtained respectively, or the sub-table index set FUD2I1 under each column of the sub-table name set FUD2C1 and the sub-table index set FUD2I2 under each column of the sub-table name set FUC2C2 are obtained respectively;

Combining and de-duplicating the sub-table index set FUD1I1 and the sub-table index set FUD1I2 to form a sub-table index union FUD1I, or combining and de-duplicating the sub-table index set FUD2I1 and the sub-table index set FUD2I2 to form a sub-table index union FUD2I;

respectively calculating a sub-table index difference set DIF-FUD1I1 of the sub-table index union FUD1I and the sub-table index set FUD1I1, and a sub-table index difference set DIF-FUD1I2 of the sub-table index union FUD1I2, or respectively calculating a sub-table index difference set DIF-FUD2I1 of the sub-table index union FUD2I and the sub-table index set FUD2I1, and a sub-table index difference set DIF-FUD2I2 of the sub-table index union FUD2I and the sub-table index set FUD2I2;

judging whether the differential table index differential set DIF-FUD1I1 and the differential table index differential set DIF-FUD1I2 are null values, when the differential table index differential set DIF-FUD1I1 and the differential table index differential set DIF-FUD1I2 are not null values, or judging whether the differential table index differential set DIF-FUD2I1 and the differential table index differential set DIF-FUD2I2 are null values, and when the differential table index differential set DIF-FUD2I1 and the differential table index differential set DIF-FUD2I2 are not null values, respectively outputting non-null value differential table index comparison results.

9. A database comparison system for checking whether an error exists in a database, characterized in that the database comparison system performs the operational steps of the database structure comparison method according to any one of claims 1-8.

10. The database comparison system of claim 9, wherein the steps of performing the database structure comparison method are repeated one or more times and the database is modified one or more times based on non-null comparison results generated in the database structure comparison method.