JP5827260B2 - Method, apparatus and program for estimating foreign key in database - Google Patents

Description

  The present invention relates to a method, apparatus, and program for estimating a foreign key in a database.

  According to a recent IT trend survey, companies and businesses that develop new systems are on the decline, and the ratio of existing system migration cases is increasing. In order to understand the functions of the system, the current system generally starts with the current analysis. However, “There is no design document that is maintained,” “There is no designer at that time,” “System users are systems. The current situation is that there is a restriction that the function cannot be explained or sufficient time for hearing is not available. In this case, “reverse engineering technology” that obtains necessary design information by analyzing program source code and database objects becomes important.

  In database design and analysis, table definition information that defines column names and data types, and information that indicates whether a column is a primary key (Primary Key, PK) or a foreign key (Foreign Key, FK) are important. It becomes. Patent Document 1 describes a method of setting a foreign key in accordance with the designation of a relation between columns arbitrarily specified in a database design stage.

Special table 2008-524671

  Also, a technique for acquiring table definition information and the like by referring to an existing database object in database analysis is also known.

  However, the conventional technique has a problem that the foreign key cannot be specified when the database object does not include information representing the foreign key. If the foreign key is not known, it cannot be expressed as an ER diagram, so the analyst cannot grasp the relationship between the tables and it is difficult to understand the system.

  Note that it may not be practically possible to artificially estimate a foreign key based on a column name or the like, and a method using a system or an apparatus is essential. For example, there are cases where the number of tables is 700 and the number of columns is 20,000 to 30,000. In such a case, if the foreign key is estimated manually, the amount of work becomes enormous, and the estimation is caused by variations in estimation criteria. There is a risk that the accuracy may also decrease.

  The present invention has been made to solve such a problem, and an object thereof is to provide a method, an apparatus, and a program for a computer to estimate a foreign key in a database.

In order to solve the above-described problem, a method for estimating a foreign key in a database according to the present invention is as follows.
A method for estimating a foreign key in a database including a plurality of tables,
A computer obtaining one or more column names for each table;
A computer acquiring actual data recorded in each column of each table;
A column coupling pattern identifying step in which a computer identifies a pair of two columns as a column coupling pattern based on the column name;
A foreign key estimation step, wherein the computer estimates whether each column is a foreign key,
In the foreign key estimation step, for any one column included in each column combination pattern,
-The other column is the primary key,
-The one column is not the only primary key in the table; and-the actual data recorded in the one column are all contained in the actual data recorded in the other column. As a prerequisite, one column is presumed to be a foreign key referring to the other column.

The column combination pattern specifying step includes:
In the source code of a program that refers to any of the tables, the first actual data recorded in the first column of the first table and the second data recorded in the second column of the second table A step in which a computer extracts a match determination command for performing a match determination with actual data based on the column name;
For each extracted match determination instruction, the computer specifies a pair consisting of the first column and the second column as a column combination pattern;
May be included.
The column combination pattern specifying step may include a step of specifying a pair consisting of two columns having the same column name in different tables as a column combination pattern.
The method further comprises a primary key estimation step,
In the primary key estimation step, for each column,
-The computer assumes that the column is a primary key, provided that the actual data for the column does not contain undefined values, and-the actual data for the column is not duplicated within the column. Also good.
The computer stores an exclusion list including at least one column;
In the foreign key estimation step, it may be further necessary that no column is included in the exclusion list.
Further comprising the step of the computer creating an ER diagram including each table;
The ER diagram is
An entity representing each table,
For each column that is a foreign key, an association line that associates two related tables may be included.

  A database foreign key estimation apparatus according to the present invention includes a computer that executes the above-described method.

  The program according to the present invention causes a computer to execute the above-described method.

  According to the method, apparatus, and program for estimating a foreign key in a database according to the present invention, the foreign key of the database can be estimated based on the actual data of the database and the column name of the database.

It is a figure which shows the example of a structure of the foreign key estimation apparatus which concerns on Embodiment 1 of this invention. It is a figure which shows the example of a structure of table definition information. It is a figure which shows the example of a structure of real data. It is a figure which shows the example of a structure of source code. It is a flowchart showing the flow of a process of an external key estimation apparatus. It is a figure which shows the example of primary key information. 6 is a diagram illustrating an example of a configuration of a column combination list according to Embodiment 1. FIG. It is a figure which shows the example of a structure of an exclusion list. 6 is a diagram illustrating an example of foreign key information in the first embodiment. FIG. In Embodiment 1, it is an example of the ER diagram represented by ER diagram information. 10 is a flowchart showing a process flow of the foreign key estimation apparatus according to the second embodiment. 10 is a diagram illustrating an example of a configuration of a column combination list according to Embodiment 2. FIG. FIG. 11 is a diagram illustrating an example of foreign key information in the second embodiment.

Embodiments of the present invention will be described below with reference to the accompanying drawings.
Embodiment 1 FIG.
FIG. 1 shows the configuration of foreign key estimation apparatus 10 according to Embodiment 1 of the present invention. The foreign key estimation device 10 functions as a device that receives data relating to a database including a plurality of tables composed of records and columns and estimates whether each column is a foreign key in the database. This function is achieved by performing the method described herein. The external key estimation apparatus 10 includes a configuration as a well-known computer, and includes a calculation unit 20 that performs a calculation and a storage unit 30 that stores information. The computing means 20 includes a CPU (Central Processing Unit), and the storage means 30 includes a storage medium such as a semiconductor memory and an HDD (Hard Disk Drive).

  Although not shown in particular, the external key estimation device 10 includes an input unit and an output unit. The input means is means used by the user to input information, and is, for example, a mouse or a keyboard. The output means is means for outputting information to the user, and is a display device such as a liquid crystal display, but may be a printing device such as a printer.

  The storage means 30 stores table definition information 31, actual data 32, source code 33, column combination list 34, exclusion list 35 and ER diagram information 36. The storage unit 30 stores an external key estimation program (not shown), and the computer (particularly the calculation unit) functions as the external key estimation device 10 by executing the external key estimation program.

  FIG. 2 shows an example of the configuration of the table definition information 31. FIG. 2 shows four tables included in one database. In this specification, each table is referred to by a table name, and is referred to as a “order” table, a “order details” table, a “product” table, and a “user” table, respectively. The table definition information 31 includes, for each table, a table name that uniquely identifies the table and column definition information of the table. In the example of FIG. 2, the column definition information includes a column name, a data type, a data length, and a constraint. The table definition information 31 can be expressed using, for example, DDL (Data Definition Language). As shown in FIG. 2, the column definition information includes one or more columns for each table.

  The column name is information for uniquely identifying a column in the table, for example. For example, the “order received” table includes an “order received number” column, a “first order number” column, an “order received date” column, and an “updater” column. The data type and data length are information that defines the characteristics of the actual data of the column. The constraint represents a constraint condition imposed on the actual data of the column. In the example of FIG. 2, a “not null” parameter is specified as a constraint on some columns. The “not null” parameter represents a constraint that none of the actual data in the column should be null or undefined values, and a blank indicates that no such constraint is imposed. Other constraints may be specified, for example, a “unique” parameter indicating a constraint that the actual data of the column should not be duplicated in the column.

  FIG. 3 shows an example of the configuration of the actual data 32. FIG. 3 shows only an example of an “order” table and a “product” table. The actual data 32 includes data representing a value recorded in each column (that is, a value recorded in each cell) for each record of each table.

  FIG. 4 shows an example of a part of the configuration of the source code 33. The source code 33 is a source code of a program that refers to any of the tables included in the database. Multiple source codes may be included. Examples of such programs include programs such as views and DAO (Data Access Objects). FIG. 4 shows three SELECT statements, each specifying a column join condition (sometimes called a table join condition) in the WHERE clause.

  The column join condition is a match for determining the match between the first actual data recorded in the first column of the first table and the second actual data recorded in the second column of the second table. It can be said that it is expressed by a determination instruction. In FIG. 4, the SELECT statement that appears first is described as an example. A combination condition is specified that combines the “Product No” column of the “Order Details” table and the “Product Code” column of the “Product” table. The SELECT statement is executed based on the result of the coincidence determination of the actual data in these columns. Note that the column join condition can be expressed as long as it is an instruction that makes an equivalent match determination, not limited to the SELECT statement and the WHERE clause.

  The column combination list 34, the exclusion list 35, and the ER diagram information 36 will be described later along with the operation of the foreign key estimation apparatus 10.

The processing flow of the foreign key estimation apparatus 10 configured as described above will be described below.
FIG. 5 is a flowchart showing a process flow of foreign key estimation apparatus 10 according to the first embodiment. For convenience of explanation, part of the input / output of each step is also illustrated using broken lines.

  First, the foreign key estimation device 10 acquires table definition information 31 (step S1). In this processing, for example, the table definition information 31 may be extracted with reference to a database object as shown in the figure, or information corresponding to the table definition information 31 may be directly acquired. As shown in FIG. 2, the acquired table definition information 31 includes one or more column names for each table.

  Next, the foreign key estimation device 10 acquires the actual data 32 recorded in each column of each table (step S2).

Next, the foreign key estimation device 10 estimates whether or not each column is a primary key based on the acquired actual data 32 (step S3, primary key estimation step). Here, the foreign key estimation apparatus 10 estimates that each column of each table is a primary key, using the following two conditions as necessary conditions.
-Primary key condition 1: The actual data of the column does not contain an undefined value.
-Primary key condition 2: The actual data of the column is not duplicated in the column (that is, they are all different from each other in the column).

In the present specification, the terms “necessary conditions” and “sufficient conditions” are understood to have ordinary meanings in the field of mathematics or logic. Or, for example, it may be interpreted as follows.
The expression “estimate X with A, B, and C as prerequisites” requires that all of A, B, and C be satisfied in order to be estimated as X, but still other conditions Does not deny the possibility of being additionally required. That is, when all of A, B, and C are satisfied, it may be estimated as X or may not be X depending on other conditions. On the other hand, when any of A, B, and C is not satisfied, it is estimated that it is not X regardless of other conditions.
The expression “estimate X with A as a sufficient condition” means that in order to be estimated as X, only A needs to be satisfied and no other conditions may be required. That is, when A is satisfied, X is always estimated regardless of other conditions. On the other hand, if A is not satisfied, it may be estimated as X or may not be X depending on other conditions.
Of course, for a certain estimation process, a certain condition (or a combination of a plurality of conditions) can be a necessary condition and a sufficient condition.

The primary key condition 1 can be determined based on, for example, whether or not the “not null” parameter is specified for the constraint of the column in the table definition information 31. Alternatively, it may be determined by directly referring to all the actual data of the column and checking whether or not an undefined value is included.
The primary key condition 2 can be determined based on, for example, whether or not the “unique” parameter is specified for the constraint of the column in the table definition information 31. Alternatively, all the actual data of the column may be directly referred to, and it may be determined by checking whether there is duplicate data.

  In the first embodiment, in this estimation, it is a sufficient condition that the above two conditions are satisfied. In the example of FIG. 3, the actual data in the “order number” column of the “order” table and the actual data in the “product code” column of the “product” table satisfy both of the above two conditions. Presumed to be the primary key.

As a modification, additional necessary conditions other than the above two conditions may be imposed. For example, one or any combination of the following conditions may be set as an additional requirement.
-An index exists for the column.
-The data type of the column does not represent a date or date / time.
-The column name of the column satisfies a predetermined condition. For example, a predetermined character string that is highly likely to represent a primary key is included. A person skilled in the art can appropriately determine what kind of character string is likely to represent the primary key. For example, it is a character string such as “code”, “ID”, “serial number”, “No”, “number”, etc. .

In step S3, the foreign key estimation device 10 outputs primary key information indicating whether each column is estimated to be a primary key.
FIG. 6 shows an example of primary key information. In this example, the primary key information is represented in the form of a flag added to the column definition of each column in the table definition information 31, but may be represented in other formats, for example, the table definition information 31 It may be stored as an independent file.

  Steps S4 to S6 executed next constitute the column combination pattern specifying step in the first embodiment. In the first embodiment, the column combination pattern specifying step is performed as a step of specifying a pair of two columns as a column combination pattern based on the source code 33.

After step S3, the foreign key estimation device 10 acquires the source code 33 (step S4).
Next, the foreign key estimation device 10 extracts a column match determination command from the source code 33 (step S5). For example, based on the column name included in the table definition information 31 (or based on the table name and the column name), in the source code 33, the first actual data recorded in the first column of the first table, the first The coincidence determination command for performing the coincidence determination with the second actual data recorded in the second column of the second table is extracted. As a result, for example, three SELECT statements shown in FIG. 4 are specified.

Next, the foreign key estimation device 10 specifies and stores a column combination pattern for each of the extracted match determination instructions (step S6). The column combination pattern is specified as a pair consisting of two columns included in the match determination instruction, and is stored in, for example, the column combination list 34. In this way, the column combination pattern specifying step in the first embodiment is completed.
FIG. 7 shows an example of the configuration of the column combination list 34 in the first embodiment. In this example, column combination patterns corresponding to the three SELECT statements in FIG. 4 are shown.

Next, the foreign key estimation device 10 estimates whether or not each column is a foreign key for each column combination pattern (step S7, foreign key estimation step). For example, for each column combination pattern, the following four conditions are required for one column (hereinafter sometimes referred to as “foreign key candidate”), and the foreign key candidate is a foreign key that refers to the other column. Presume that there is.
Foreign key condition 1: the other column is the primary key. For example, in the first of the three column combination patterns shown in FIG. 7, the “product code” column in the “product” table is the primary key (see FIG. 6), so “product No.” in the “order details” table. The column satisfies the foreign key condition 1 as a foreign key candidate.
-Foreign key condition 2: A column that is a foreign key candidate is not the only primary key in the table. That is, the table does not include a primary key or includes a primary key in addition to a foreign key candidate.
-Foreign key condition 3: The actual data recorded in the column that is a foreign key candidate is consistent with the actual data recorded in the other column. That is, the actual data recorded in the column that is a foreign key candidate is all included in the actual data recorded in the other column.
Foreign key condition 4: None of the columns are included in the exclusion list 35.

  FIG. 8 shows an example of the configuration of the exclusion list 35. The exclusion list 35 is prepared in advance by a user, for example. The exclusion list 35 includes information for specifying one or more columns (for example, a combination of a table name and a column name). In the example of FIG. 8, the exclusion list 35 includes the “updater” column of the “user” table.

  In this estimation, the above four conditions may be satisfied sufficiently, and additional necessary conditions may be imposed in addition to these conditions. As an additional necessary condition, for example, a condition that “a data type and a data length of a column that is a foreign key candidate, and a data type and a data length of the other column all match” can be used.

  As described above, since the information of the source code 33 is used in step S7, the column names of the paired columns do not match between the tables, or the character string common to the column names is not included at all. It is also possible to extract the column binding pattern. Conversely, even if the column names match, if the actual data is not consistent, it can be correctly estimated that the key is not a foreign key.

In step S7, the foreign key estimation device 10 outputs foreign key information indicating whether each column is estimated to be a foreign key.
FIG. 9 shows an example of foreign key information in the first embodiment. In this example, the foreign key information is represented in the form of a flag added to the column definition of each column in the table definition information 31.

  Regarding the “Order Number” column in the “Order Details” table, the other column in the column combination pattern (“Order Number” column in the “Order” table) is the primary key and satisfies the foreign key condition 1. The “order details” table includes two primary keys and satisfies the foreign key condition 2. Here, the actual data recorded in the “Order Number” column in the “Order Details” table is consistent with the actual data in the other column (“Order Number” column in the “Order” table), and the foreign key Assume that condition 3 is satisfied. Here, it is assumed that the “order number” column of the “order details” table is not included in the exclusion list 35 and satisfies the foreign key condition 4. In this case, this column is assumed to be a foreign key as shown in FIG.

  For the “product No” column of the “order details” table, the other column in the column combination pattern (the “product code” column of the “product” table) is the primary key and satisfies the external key condition 1. The “order details” table includes two primary keys and satisfies the foreign key condition 2. Here, the actual data recorded in the “Product No.” column of the “Order Details” table is consistent with the actual data of the other column (“Product Code” column of the “Product” table), and the foreign key Assume that condition 3 is satisfied. Here, it is assumed that the “product number” column of the “order details” table is not included in the exclusion list 35 and satisfies the foreign key condition 4. In this case, this column is assumed to be a foreign key as shown in FIG.

Strictly speaking, the information shown in FIG. 9 is only a part of the foreign key information, and in order to sufficiently specify the relationship between the columns, in addition to the information shown in FIG. It is necessary to specify the column referred to by each foreign key with reference to FIG.
The external key information may be expressed in other formats, and may be stored as a file independent of the table definition information 31, for example.

  Next, the foreign key estimation device 10 creates an ER diagram (step S8). This process is performed, for example, by outputting ER diagram information 36. The ER diagram information 36 may be output and stored in the storage unit 30 as shown in FIG. 1, or may be output to a display device or a printing device (not shown).

  FIG. 10 shows an example of the ER diagram represented by the ER diagram information 36. In the example of FIG. 10, the ER diagram includes an entity 36a representing each table, a column name of a column included in each table, a display indicating whether each column is a primary key, and each column having a foreign key. It includes a display indicating whether or not there is, and association lines 36b and 36c for associating two related tables for each column which is a foreign key. Here, the “related table” means a table including a column that is a foreign key and a table including a column that is referenced by the foreign key.

  In the example of FIG. 10, a column with the notation “(primary key)” is a primary key, and a column without this notation is not a primary key. The same applies to foreign keys. Further, a group of columns that are primary keys and a group of columns that are not primary keys are separated from each other (in the example of FIG. 10, they are separated vertically by a horizontal line).

  Further, the ER diagram includes an expression for specifying the table on the side including the foreign key for the relation lines 36b and 36c. In this example, the end point on the foreign key side is greatly expressed. In addition, the table including the foreign key includes an expression indicating whether or not it depends on the other table. Here, “dependent” means that the foreign key is also a primary key, while “independent” means that the foreign key is not the primary key. In the example of FIG. 10, since the “order number” column in the “order details” table is a foreign key and a primary key, the related line 36b corresponding to this foreign key is represented by a solid line as representing a “dependent” relationship. ing. On the other hand, since the “Product No.” column of the “Order Details” table is a foreign key but not a primary key, the relation line 36c corresponding to this foreign key is represented by a broken line as representing an “independent” relationship. Yes.

  As described above, according to the method for estimating a foreign key, the foreign key estimating apparatus 10 and the foreign key estimating program according to the first embodiment of the present invention, the actual data 32 of the database and the source of the program referring to the database Based on the code 33 (particularly the column name), the foreign key of the database can be estimated.

  It should be noted that the present invention does not simply systemize the work performed by humans. The external key estimation work may be enormous in the amount of work, and in such a case, it is practically impossible for a human to perform the work, so implementation by a computer is indispensable.

Embodiment 2. FIG.
The second embodiment changes the configuration of the column combination pattern specifying step (steps S4 to S6 in FIG. 5) in the first embodiment. Hereinafter, differences from the first embodiment will be described.
FIG. 11 is a flowchart showing a process flow of the foreign key estimation apparatus according to the second embodiment. For convenience of explanation, part of the input / output of each step is also shown.

  Step S14 executed after step S3 constitutes the column combination pattern specifying step in the second embodiment. In step S14, the foreign key estimation apparatus refers to the column definition information of each table, and extracts two columns having the same column name in different tables. And the pair which consists of the extracted column is specified as a column coupling | bonding pattern.

  Note that if there are columns having the same column name across three or more different tables, all possible pairs are identified as a column combination pattern (ie, n · ( n-1) / 2 column coupling patterns will be identified). However, as a modification, the column combination pattern to be specified may be narrowed down using some standard.

Hereinafter, in the second embodiment, an operation when foreign key estimation is executed based on the table definition information 31 shown in FIG. 2 will be described.
FIG. 12 shows an example of the configuration of the column combination list 34 created in the second embodiment. All matching column name pairs are extracted and specified from the table definition information 31 of FIG.

  FIG. 13 shows an example of foreign key information in the second embodiment. The “order number” column in the “order details” table is presumed to be a foreign key as in the first embodiment, but the “commodity number” column in the “order details” table is the column combination list 34. It is presumed that it is not a foreign key because it is not included in.

  Thus, the source code 33 is not referred to in the second embodiment. Therefore, when the creation of the column combination list 34 based on the source code 33 cannot be executed or is difficult due to a part or all of the source code 33 not being saved, as in the second embodiment. In some cases, the method based on column names can better estimate foreign keys.

  As described above, according to the method for estimating a foreign key, the foreign key estimation device 10 and the foreign key estimation program according to the second embodiment of the present invention, the actual data 32 of the database and the column names of each table are used. Based on this, the foreign key of the database can be estimated.

In the first and second embodiments described above, the following modifications can be made.
In the first and second embodiments, the column definition information includes a column name, a data type, a data length, and constraints, but may include other information. For example, information indicating whether each column is a primary key may be included. When information indicating whether or not each column is a primary key is included, the primary key determination step may be omitted (that is, information relating to the primary key is acquired in step S1, and step S3 is omitted). You may).

  Further, the column definition information may not include any or all of the data type, the data length, and the constraint. In principle, including a column name is sufficient for a minimal implementation of the present invention.

  A configuration in which the exclusion list 35 is not used may be used. For example, the storage unit 30 may not store the exclusion list 35 and may not use the foreign key condition 4 (no columns are included in the exclusion list 35) in step S7.

Processing related to the ER diagram may be omitted. That is, the process may be terminated when the foreign key is estimated, and step S8 may not be executed.
Further, the ER diagram may include information other than that shown in FIG. 10, or may not include a part of the information shown in FIG. If at least an entity representing each table and a relation line that associates two tables related to each of the columns that are foreign keys are included, it can be interpreted as corresponding to the ER diagram.

  In the second embodiment, the column combination pattern may be specified based on the column name, and may be included even if the column names do not exactly match. For example, when the column names of two different tables include matching character strings of a predetermined length, those pairs may be extracted as a column combination pattern. The judgment criteria can be appropriately designed by those skilled in the art.

  Embodiments 1 and 2 may be combined. For example, the column combination pattern specified in steps S4 to S6 of the first embodiment and the column combination pattern specified in step S14 of the second embodiment may be combined. That is, each column combination list may be ORed. In this case, in the example of FIGS. 7 and 12, a total of 12 column combination patterns are specified and included in the column combination list 34.

  In addition, in steps S4 to S6 in the first embodiment and in step S14 in the second embodiment, only the pairs specified as the column combination pattern may be included in the column combination list 34. That is, each column combination list may be ANDed. In this case, in the example of FIGS. 7 and 12, only one column combination pattern is specified and included in the column combination list 34 for only the second one of FIG. 7 (first one in FIG. 12). become.

  In creating the column combination list 34, it may be possible to specify whether to use an OR combination or an AND combination. For example, the foreign key estimation apparatus 10 accepts a designation input relating to a method of creating the column combination list 34 (designation of using an OR combination or designation of using an AND combination), and the column combination list 34 under different conditions according to this. May be created.

10 external key estimation device (computer), 20 computing means, 30 storage means, 31 table definition information, 32 actual data, 33 source code, 34 column join list (column join pattern), 35 exclusion list, 36a entity, 36b, 36c Related lines,
S3 primary key estimation step, S4 to S6, S14 column combination pattern identification step, S7 foreign key estimation step.

Claims (8)

A method for estimating a foreign key in a database including a plurality of tables,
A computer obtaining one or more column names for each table;
A computer acquiring actual data recorded in each column of each table;
A column coupling pattern identifying step in which a computer identifies a pair of two columns as a column coupling pattern based on the column name;
A foreign key estimation step, wherein the computer estimates whether each column is a foreign key,
In the foreign key estimation step, for any one column included in each column combination pattern,
-The other column is the primary key,
-The one column is not the only primary key in the table; and-the actual data recorded in the one column are all contained in the actual data recorded in the other column. As a prerequisite, a method for estimating a foreign key in a database in which one column is estimated to be a foreign key referring to the other column. The column combination pattern specifying step includes:
In the source code of a program that refers to any of the tables, the first actual data recorded in the first column of the first table and the second data recorded in the second column of the second table A step in which a computer extracts a match determination command for performing a match determination with actual data based on the column name;
For each extracted match determination instruction, the computer specifies a pair consisting of the first column and the second column as a column combination pattern;
The method of claim 1 comprising:   The method according to claim 1, wherein the column combination pattern specifying step includes a step of specifying a pair of two columns having the same column name in different tables as a column combination pattern. The method further comprises a primary key estimation step,
In the primary key estimation step, for each column,
-The computer assumes that the column is the primary key, provided that the actual data for the column does not contain undefined values, and-the actual data for the column is not duplicated within the column, The method as described in any one of Claims 1-3. The computer stores an exclusion list including at least one column;
The method according to any one of claims 1 to 4, wherein the foreign key estimation step further requires that no column is included in the exclusion list. Further comprising the step of the computer creating an ER diagram including each table;
The ER diagram is
An entity representing each table,
6. The method according to any one of claims 1 to 5, comprising a relation line that associates two related tables for each column that is a foreign key.   An apparatus for estimating a foreign key of a database, comprising a computer that executes the method according to claim 1.   The program which makes a computer perform the method as described in any one of Claims 1-6.

Images