JP2798118B2 - Database query statement creation device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a database query creation device.

[0002]

2. Description of the Related Art Conventionally, to make an inquiry to a relational database, an SQL (Structured Query Language) is used.
And QBE (Query By Example: Zloof M. “Qu
ery by example ”Proceedings of AFIPS '75, 197
5, pp. 431-432.), And acquire a language called a database query language, describe a query sentence using these languages, and make an inquiry using this query sentence.

[0003]

However, learning the database language and its grammar takes time, and it is very difficult for a beginner to create a query sentence.

[0004] It is an object of the present invention to provide a database sentence creating apparatus which enables even a beginner to easily create a database query sentence.

[0005]

According to the present invention, data
Display of data registered in the base, display data
Direct selection operation, based on a logical search expression for the display data
Selection operations and the selection data obtained by these selection operations
Create new data by aggregation operation
From the possible input tables, select the data selected in the input table.
Data and newly created data to the output table,
Part or all of the data required as search results on the force table
By creating a database, the database search data
An inquiry statement creation device for a database, wherein the input table and
A display means for displaying the output table, input means for performing a copy operation to the output table from the operation and the input table for the input table, and control means for controlling the said display means said input means, an operation storing means for storing information representing an operation for the input table, copying the copy operation
Only the information related to the input data is stored in the operation storage unit.
A copy operation storage unit that obtains and stores the same from the user, and interprets the meaning of the operation based on the information stored in the copy operation storage unit, thereby obtaining a query type intended by the user.
Operation interpreting means for estimating the type, and query sentence generating means for creating a query sentence corresponding to the information stored in the copy operation storage means based on the meaning interpreted by the operation interpreting means. A database query sentence creating device is obtained .

[0006]

According to the present invention, the user operates the input table directly on the display, further copies the data on the input table to the output table, and creates table data to be obtained as a result of the query, so that the system can perform these operations. The operation history of the user is recorded in the operation storage unit, meaning of the operation is performed by the operation interpretation unit from the operation history recorded in the operation storage unit, and the query sentence generation unit automatically generates a query sentence based on the result. Therefore, the user does not need to memorize the grammar of a complicated query language and the like, and can create a query sentence only by learning a simple operation method in the input table.

[0007]

Embodiments of the present invention will be described below with reference to the drawings. One embodiment of the present invention is shown in FIG. As shown in FIG. 1A, the database query sentence creating apparatus according to the present embodiment performs operations on table data on a display device 1 for displaying table data (input table and output method) on the input table. A storage device 3 having an input device 2 such as a mouse or a keyboard for operation, an operation storage unit for storing operation history and the like, and a database for storing table data, a display device 1, an input device 2 and a storage device 3. And a computer 4 for executing the above processing.

[0010] In this inquiry sentence creating apparatus, as shown in FIG.
The operation performed by the user using the input device 2 on the input table 6 displayed on the input device 6 is stored in the input table operation storage unit 8 provided in the operation storage unit 7 of the storage device 3 under the control of the computer 4. . The contents stored in the input table operation storage unit 8 are updated each time an operation is performed on the input table. Further, the copy operation from the input table 6 to the output table 9 is stored in the copy operation storage unit 10 of the operation storage unit 7.

Next, the operation interpreting section 11 of the computer 4 gives meaning to the user's operation based on the information stored in the copy operation storing section 10. Then, the query sentence generation unit 12 generates a query sentence corresponding to the information stored in the copy operation storage unit 10 based on the result interpreted by the operation interpretation unit 11. Further, if necessary, the query sentence generalization unit 13 generalizes the query sentence generated by the query sentence generation unit 12.

In this embodiment, the database 5 stores
Although described as being provided in the same storage device 3 as the operation storage unit 7, the database may be provided in another storage device connected to the computer 4 (for example, via a communication line).

FIG. 3 shows an example of a processing order when the present invention is carried out. As shown in FIG.
When an operation is performed from the input device 2 after the operation recording is started in step S2, it is determined in step S2 whether the operation is for the input table 6 or a copy operation. In the case of an operation on the input table 6, the operation is stored in the input table operation storage unit 8 in step S3. That is, the content stored in the input table operation storage unit 8 is updated. In the case of a copy operation, in step S4,
The contents stored in the input table operation storage unit 8 when the copy operation is performed are stored in the copy operation storage unit 10. Thereafter, if there is no query generation command, the process proceeds from step S5 to step S5.
Return to 2. The storage of these operations is repeated until there is a query generation instruction.

If it is determined in step S5 that there is a query generation command, the operation recording is terminated in step S6. Then, in step S7, the operation interpretation unit 11 interprets the operation stored in the copy operation storage unit 10. If it can be interpreted, the type of the query is determined in step S8. Then, in step S9, the query sentence generation unit 12 generates a query sentence. If it is determined in step S7 that the interpretation is not possible, the operation ends.

Next, if there is an instruction to generalize the created query sentence (step S10), at step S11,
The query sentence generalization unit 13 generalizes the query sentence. If there is no generalized query statement, the operation is terminated.

Hereinafter, the operation of the database query generation device of the present embodiment will be described in detail. Here, a personnel table and an age table as shown in FIG. 3 are considered as input tables. The first line of the table is a field column, the second line is a tabulation function column, the third line is a search condition input column, and the fourth and subsequent lines are record columns. In the following description, the field F _i of the table T _k is _defined as T _k .
Notated as _Fi . _Let the j-th record of the table T _{k be} T _k . R
_j , or “R _{j of} T _k ”, or if it is clear that the record is a record of table T _k , it is simply denoted as R _j . The records T _k . Field T _k of R _j. F
_i value of _{v (T k .F i, T} k .R j) is referred to as. Further, all the fields of the table T _k are set to T _k . F ₁ , ...,
T _k . If the _{F n, COND (T k .R} j) is, T
_k . _{F 1 = v (T k .F} 1, T k .R j) AND ... AN
DT _k . _{F n = v (T k .F} n, T k .R j) shall mean that conditions.

The query creation device of the present embodiment has a function of highlighting an arbitrary field (column) of table data on the screen of the display device 1 with respect to such an input table (table data). Function for highlighting records, function for highlighting records that satisfy search conditions, calculation by applying an aggregation function to the highlighted records
It has a function of displaying and a function of canceling all highlights. These functions are executed, for example, by moving the cursor on the screen with a mouse or the like and clicking.

For example, in the personnel table, when the fields F1 and F3 are clicked with the mouse, the entire columns of the fields F1 and F3 are highlighted as shown in FIG. Further, when the record columns R1 and R3 are clicked with the mouse, the record columns R1 and R3 can be highlighted as shown in FIG. 4B. Further, when a search condition (here, affiliation = lab, salary> 25) is input from the input device 2 to the search condition input field as shown in FIG. Is written. When a plurality of search conditions are input as in this example, the logical product of them is used as the search condition. That is, if “affiliation = lab” and “salary> 25” are entered,
"Personnel table. Affiliation = Laboratory AND Personnel table. Salary>25" is treated, and records R2 and R4 are highlighted. The execution of this search is specified from a menu or the like. In addition, a function for canceling all highlights is designated and executed from a menu or the like.

In this embodiment, the fields and records can be highlighted as described above, but the fields and records can also be highlighted at the same time. In this case, as shown in FIG.
There is a portion where the highlighted areas overlap. This part is called highlight level 2 and is distinguished from normal highlight level 1.

When the user clicks the tabulation function column in the second row of the table data with the mouse, a tabulation function menu is displayed on the screen as shown in FIG. Furthermore, when the menu is clicked with the mouse, the aggregation function is applied to the highlighted record in the table data. FIG.
In (e), the records R2 and R4 are highlighted in the personnel table, and a case where “total” is selected is shown. As a result, the highlighted record R2
And the value of the field "HR table. Salary" of R4 is added up, and the result is displayed in the cell of the second row and the third column of the personnel table. If there is no highlighted record, the aggregation function is applied to all records.

Next, a copy operation from the input table to the output table will be described. Copying from the input table to the output table is permitted only for the data highlighted in the input table and the aggregation result of the aggregation function. Therefore, the copy function of the query creation device according to the present embodiment includes a function of copying all of the portions having the highest highlight level in the input table to the output table (record copy function) and a result of applying the aggregation function to the output table. This is the function of copying (counting result copy function). Such a copy function is realized by dragging and dropping from the input table to the output table with the mouse.

As shown in FIG. 5, when the input table has a highlight level 2 as shown in FIG.
All of these highlight levels 2 are copied to the record column of the output table. At this time, all the field names highlighted in the input table are also copied to the field columns of the output table. When only records (rows) are highlighted, the record copy function copies all portions of highlight level 1 of the input table to the record column of the output table as shown in FIG. Copy all the field names of the input table to the field columns of the output table. Furthermore, when only the field (column) is highlighted, as shown in FIG. 6B, all the highlighted fields are copied to the output table field, and the input table is highlighted. Copy all parts of level 1 to the record column of the output table.

As shown in FIG. 7, the counting result copy function copies the calculation result displayed in the second line of the field of the input table to the record column of the output table. At this time, a field name corresponding to the applied aggregation function is automatically given to the field column of the output table.

As can be seen from the above description, the output table is
The first line is a field column, and the second and subsequent lines are record columns. The records copied from the input table are displayed left-justified and top-justified in the output table. The position of the copy destination cell in the output table is the second row (first record row) and the first blank column. For example, FIG.
When the total of the salary is copied from the input table to the output table shown in FIG. 8 by the tally result copy function, the result becomes as shown in FIG.

The above-mentioned user operations (operations on the input table and copy operations from the input table to the output table) are stored in the operation storage unit 7 of the inquiry sentence creating apparatus. Next, what processing is performed by the operation storage unit 7 in response to a user operation will be described.

The operation in the input table is recorded in an input table operation storage section 8 provided in the operation storage section 7. The input table operation storage unit 8 is prepared for each input table.
As shown in (a), the list includes a field list, a record list, each slot of a search condition, and an arbitrary number of aggregation function structures.

A field list in the input table operation storage unit 8,
Each slot of the record list and the search condition is for storing the state of highlight in the input table. When a field is newly highlighted in the input table, the field name is added to the field list slot of the input table operation storage unit 8, and when a record is highlighted by mouse click, its record number is added to the record list slot. . When a record is highlighted by the search, the search condition at that time is stored in the search condition slot. When all highlights are released, the contents of the field list slot, record list slot, and search condition slot in the input table operation storage unit are all deleted.

The aggregation function structure is for storing the type of the applied function and the state of the highlight at the time when the function is applied. When an aggregate function is calculated, an aggregate function structure consisting of the aggregate function, field, record list, and search condition slots is generated,
It is added and stored in the input table operation storage unit 8. The type of the applied aggregation function is stored in the aggregation function slot of the aggregation function structure, and the field to which the aggregation function is applied is stored in the field slot. Further, the contents of the record list slot and the search condition slot of the input table operation storage unit when the tally function is applied are stored as they are in the record list slot and the search condition slot of the tally function structure, respectively.

The copy operation from the input table to the output table is stored in a copy operation storage unit 10 provided in the operation storage unit 7. Two types of structures are stored in the copy operation storage unit. The record structure is used when copied by the record copy function, and the aggregation function structure is used when copied by the aggregation result copy function. Each time a copy operation is performed, one of these two types of structures is stored in the copy operation storage unit 10.

As shown in FIG. 9B, the record structure includes an input table, a field list, a record list, and slots for search conditions. The input table name of the copy source is stored in the input table slot. In the record list slot and the search condition slot, the contents of the record list slot and the contents of the search condition slot in the input table operation storage unit at the time when the copy operation occurs are stored as they are. If the field list slot in the input table operation storage unit 8 is not empty, its contents are stored as it is in the field list slot, and if it is empty, all the field names in the input table are stored.

As shown in FIG. 10, the contents stored in the input table operation storage unit 8 and the contents stored in the copy operation storage unit 10 shown in FIG.
In the search field of the field "Personnel table. Affiliation", "= lab"
Enter ">25" in the search field of the field "HR table. Salary" and perform a search. (2) Apply the aggregation function "count" in the field "HR table. Name". (3)
Highlight the fields "HR table. Affiliation" and "HR table. Boss", (4) highlight record R1 by mouse click, (5) Field "HR table. Salary"
Apply the aggregation function "Total" in, and copy the highlight level 2 part in the record copy, and copy the result of the aggregation function "Count" for "HR table. It describes about the case of doing.

Next, the processing in the operation interpreting section 11 will be described. The operation interpreting unit 11 gives meaning to the user's operation based on the contents of the copy operation storage unit 10 of the operation storage unit 7 and the state of the output table, and determines what kind of query sentence is to be generated. Is passed to the query sentence generation unit 12. In addition, the operation interpreter 11 passes information necessary for generating a query to the query generator 12.

The meaning of the operation in the operation interpreting section 11 is determined in accordance with the interpretation rules 1 to 5 shown in FIGS. The type of query sentence that can be generated by the query sentence generating apparatus of the present embodiment is determined by these interpretation rules. Here, four types of query sentences (search query, aggregation format, group format, and join format query sentence) are used. Can be created. Of course, if another interpretation rule is adopted, a different query sentence can be created.

[Rule 1] shown in FIG. 11 shows conditions for generating a query sentence in a search format. The condition is that the copy to the output table is made from a single input table,
This is the case where the copy is a record copy only. However, when copying is performed in a plurality of times, all highlighting conditions in the input table must be the same. For example, FIG. 16 shows an example in which the input table “Personnel Table” is copied to the output table in two separate steps. The record list and the search of the two record structures in the copy operation storage unit 10 in FIG. The content of the condition slot is the same, and the situation of the highlight of the record in the personnel table is the same. Therefore,
The example shown in FIG. 16 satisfies rule 1.

Rule 2 shown in FIG. 12 shows conditions for generating a query statement in a tabulation format. The condition is that copying to the output table is performed only for the result of the aggregation function, the record of the output table is one row, and the highlight status of the record at the time the aggregation function is applied matches all It is. For example, FIG. 17 shows an example in which two aggregation results in the personnel table are copied to the output table. The record list of the two aggregation function structures and the contents of the search condition slot are equal, and the record of the personnel table is The highlight situation is the same. Therefore, the example shown in FIG.

Rule 3 shown in FIG. 13 indicates conditions for generating a group form query sentence. The condition of Rule 3 indicates a case where the record created in the output table by grouping is only one row (one group). That is, the fields T _k . When performing grouping in F _j ,
T _k . A search must be performed using an arbitrary value appearing in F _j (search condition is “T _k .F _j = Some Value”), and an operation such as applying an aggregation function to the result must be performed. For example, in the example shown in FIG. 18, the operation for calculating the total of the count and the salary of records where “salary> 25” for each affiliation in the personnel table is performed (the field “HR table. Affiliation”). Grouping). That is, the personnel table. The tally result of one group, “Laboratory” to which the user belongs is copied to the output table.
In order to create such an output table, the records of the personnel table are searched for in "Personnel table. Affiliation = research institute" (refer to the search condition slot of the aggregate function structure), and the "Research" used in the search is further searched. The value "place" has been copied to the output table (see record structure). Therefore, the example shown in FIG.

Rule 4 shown in FIG. 14 shows conditions for generating a query sentence in a combined form. Rule 4
Is when the obtained output table has only one row of records when a plurality of input tables are copied to the output table. That is, two different only one record in each table in the input table T ₁ and T ₂ is highlighted the highlighted record _{(T 1 T 1 .R j,} T 2
Records of T ₂ . R _i ) when these records are copied as one record in the output table. In addition to this, in order to be able to join the tables, v (T ₁ .F _m , T ₁ .R _j ) = v (T ₂ .F _n ,
T ₂ . R _i ) (T ₁ .F _m ,
T ₂ . F _n ) must be present at least once. For example, the example shown in FIG. 19 is a case in which the fields “personnel table.name” and “age table.name” are combined, and v of the personnel table (personnel table.name, personnel table.name).
R ₁ ) and v of the age table (age table. Name, age table. R ₃ )
Are the same for "Suzuki", and the salary and age of "Suzuki" are combined into one table.

If the above rules 1 to 4 do not apply, it is determined that interpretation is impossible according to rule 5.

As described above, the result of interpretation by the operation interpretation unit 11 (the format of the query sentence) is
Delivered to 2. If the interpretation result is in a join format, all pairs of tables and fields (T ₁ ,
T ₂ , T ₁ . F _m , T ₂ . Information representing F _n ) is passed from the operation interpretation unit 11 to the query sentence generation unit 12.

In the query sentence generator 12, the operation interpreter 11
An inquiry sentence is generated from the result of the interpretation in step S1 and the contents stored in the copy operation storage unit 10. However, if the result of interpretation by the operation interpretation unit 11 is “uninterpretable”, no query is generated. In the following, a description will be given of a case in which a query sentence in an SQL format, which is the most standard in a database query language, is generated.

As shown in FIG. 20, a query sentence described in SQL includes a SELECT clause, a FROM clause, a JOIN clause, and a WHERE clause.
Clause and GROUP BY clause. Therefore, the query statement generation unit 12, as described later,
A clause, a JOIN clause, a WHERE clause, and a GROUP BY clause are generated, and the generated clauses are concatenated in this order to form a query statement. As a result of applying the generation rule, JOIN clause, WHER
If the E clause and the GROUP BY clause are empty, the query statement does not include these clauses.

The generation of the SELECT clause is performed by the query statement generation unit 12 scanning all the structures stored in the copy operation storage unit 10 and describing all the fields copied to the output table as the SELECT clause. Done. That is, as shown in FIG. 21, in the case of a record structure, the contents of the field list slot
In addition to the ECT clause, in the case of an aggregation function structure, the aggregation function slot “Func ₁ ” and the field slot “T ₁ .
F ₃ ”, output table field name slot“ NewNam
e ”and“ Func ₁ (T ₁ .F ₃ ) AS New
Into a format that F ₃ ", to add to the SELECT clause.

The generation of the FROM phrase is performed by the query statement generation unit 12.
Scans all structures stored in the copy operation storage unit 10 and replaces all input tables appearing in the input table slots with FRO.
This is done by describing it as an M clause. That is, FIG.
As shown in (1), the contents of the input table slot included in each structure are added to the FROM phrase.

The JOIN phrase is generated only when the result of the interpretation by the operation interpretation unit 11 is a combined form. In this case, as shown in FIG. 23, the query sentence generation unit 12 sets a table and a field (T ₁ , T ₂ , T ₁ .F ₁ , T ₁₎ to be combined with the table passed from the operation interpretation unit 11. ₂ .
F ₁ ),..., (T ₁ , T ₂ , T ₁ .F _n , T ₂ .F _n )
_{All, "T 1 JOIN T 2 ON} (T 1 .F i = T of
₂ . F _i ) "to form a JOIN clause.

As shown in FIG. 24, the generation of the WHERE phrase is performed separately for the search format, the tabulation format, the group format, and the combination format. If the result of the interpretation by the operation interpreting unit 11 is a search format or a tabulation format, the query statement generation unit 12 extracts an arbitrary structure from the copy operation storage unit 10 and the contents of the search condition slot of the extracted structure. "C
ond0 ”and the contents of the record list slot“ R ₁ ,
, R _n ", the condition statement of the WHERE clause is generated. Similarly, in the case of the group format, an arbitrary aggregate function structure is extracted, and the content" Cond0 "of the search condition slot of the extracted structure is extracted. "and the contents of the record list slot" R _1, ..., with R _n ", to generate a conditional statement wHERE clause. here, an operation of highlighting the record R _i by mouse click, COND (R
_i ) is interpreted as a record to be added to the existing highlight, and “Cond0 COND (R ₁ )... C
OND (R _n ) ”is described as a WHERE clause.

If the result of the interpretation in the operation interpreting section 11 is a combined form, the two combined records are represented by T ₁ . R _i , T ₂ . From R _j "COND (T _1.
R _i ) COND (T ₂ .R _j ) "is generated and described in the WHERE clause.

The generation of the GROUP BY phrase is performed only when the result of the interpretation by the operation interpretation unit 11 is a group format. As shown in FIG. 25, the query sentence generation unit 12 concatenates the contents of the field lists of all record structures in the operation storage unit 11 to form a GROUP BY phrase.

The SELECT clause and FR generated as described above
When the OM clause, JOIN clause, WHERE clause, and GROUP BY clause are linked, a query sentence as shown in FIG. 26 is obtained. For example,
The example shown in FIG. 16 interpreted as the search format by the operation interpreting unit 11, the example shown in FIG. 17 as the aggregation format, the example shown in FIG. 18 as the group format, and the example in FIG. 19 as the combined format When the query sentence generation unit 12 generates a query sentence using SQL, respectively, FIGS.
29 and 30 are obtained.

In response to the generalization instruction, the query sentence generalizing unit 13 sends a query sentence including a JOIN clause and a GROUP BY
Generalize query statements that contain clauses. In the query creation device of the present embodiment, when joining tables, the user performs an operation of changing two records into one record on the output table. Therefore, a WHERE clause is described in the query sentence generated by this device in order to identify the two records to be operated. If this WHERE clause is deleted, the query statement becomes a query statement that combines all the recordable pairs. Therefore, when the query sentence includes a JOIN clause, the query sentence generalizing unit 13
As shown in FIG. 1 (a), the WHERE clause is deleted, and a query sentence for performing a join is generated for all the sets of records that can be joined.

Similarly, in the case of the GROUP BY clause, the operation performed by the user is for a specific value (group) of the grouped fields, and
A term that specifies a grouped field is included in the ERE clause. Therefore, when the query sentence includes the GROUP BY clause, the
As shown in (b), the term including the grouped fields is deleted from the WHERE clause, and a query sentence for performing the tally for all the values of the grouped fields is generated.

For example, when the query sentence in FIG. 29 and FIG. 30 is generalized, the query sentence is as shown in FIG. 32 and FIG. 33, respectively.

The query sentence obtained as described above can be easily converted to a query sentence described in another query language by a query language converter (not shown). Therefore, various inquiry languages can be generated by the inquiry sentence creation device.

[0051]

As described above, according to the present invention, the user operates the input table directly on the display, further copies the data on the input table to the output table, and converts the table data to be obtained as a result of the inquiry. By creating, the operation history is recorded in the operation storage unit, the meaning of the operation is performed by the operation interpretation unit from the operation history recorded in the operation storage unit, and the query sentence is automatically generated in the query sentence generation unit based on the result. Therefore, the user does not need to memorize the grammar of a complicated query language and the like, and can create a query sentence only by learning a simple operation method in the input table.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration of an embodiment of the present invention,
(A) is a schematic diagram, (b) is a block diagram showing details.

FIG. 2 is a flowchart for explaining the operation of the database query language creating device of FIG. 1;

FIG. 3 is a conceptual diagram of an input table.

FIG. 4 is a diagram for explaining an operation on an input table.

FIG. 5 is a diagram illustrating a copy operation from an input table to an output table.

FIG. 6 is a diagram illustrating a copy operation from an input table to an output table.

FIG. 7 is a diagram illustrating a copy operation from an input table to an output table.

FIG. 8 is a diagram illustrating a copy operation from an input table to an output table.

FIG. 9 is a diagram for explaining storage contents of an input table operation storage unit and a copy operation storage unit;

FIG. 10 is a diagram for explaining an operation of an input table and a copy operation corresponding to storage contents shown in FIG. 9;

FIG. 11 is a diagram for explaining rule 1 to be interpreted by an operation;

FIG. 12 is a diagram for explaining rule 2 to be interpreted by an operation interpreter.

FIG. 13 is a diagram for explaining rule 3 to be interpreted by an operation interpreter.

FIG. 14 is a diagram for explaining rule 4 to be interpreted by an operation interpreter.

FIG. 15 is a diagram for explaining rule 5 to be interpreted by an operation interpreter.

FIG. 16 is a diagram illustrating an output table determined as a search format and storage contents of a copy operation unit.

FIG. 17 is a diagram illustrating an output table determined to be in an aggregation format and storage contents of a copy operation unit.

FIG. 18 is a diagram illustrating an output table determined as a group format and storage contents of a copy operation unit.

FIG. 19 is a diagram illustrating an output table determined to be a combination format and storage contents of a copy operation unit.

FIG. 20 is a diagram illustrating an example of a query sentence in a search format, a tally format, a group format, and a join format described using SQL.

FIG. 21 is a diagram illustrating a procedure for generating a SELECT phrase.

FIG. 22 is a diagram for describing a procedure for generating a FROM phrase.

FIG. 23 is a diagram illustrating a procedure for generating a JOIN phrase.

FIG. 24 is a diagram for explaining a procedure for generating a WHERE clause.

FIG. 25 is a diagram for explaining a procedure for generating a GROUP BY clause.

Fig. 26: SELECT clause, FROM clause, JOIN clause, WHERE clause, and
FIG. 9 is a diagram illustrating an example of a query sentence obtained by connecting GROUP BY clauses.

FIG. 27 is a diagram showing a query sentence obtained for the example shown in FIG. 16;

FIG. 28 is a diagram showing a query sentence obtained for the example shown in FIG. 17;

FIG. 29 is a diagram showing a query sentence obtained for the example shown in FIG. 18;

FIG. 30 is a diagram showing a query sentence obtained for the example shown in FIG. 19;

FIGS. 31A and 31B are diagrams for explaining generalization of a query sentence, in which FIG. 31A shows generalization of a query sentence in a join format (JOIN format), and FIG. 31B shows a query in a group format (GROUP BY format). Shows a generalization of a sentence.

FIG. 32 is a diagram showing a query sentence obtained when the query sentence shown in FIG. 29 is generalized.

FIG. 33 is a diagram showing a query sentence obtained when the query sentence shown in FIG. 30 is generalized.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 display device 2 input device 3 storage device 4 computer 5 database 6 input table 7 operation storage unit 8 input table operation storage unit 9 output table 10 copy operation storage unit 11 operation interpretation unit 12 query sentence generation unit 13 query sentence generalization unit

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-7-36919 (JP, A) JP-A-7-78176 (JP, A) JP-A-63-156229 (JP, A) 289920 (JP, A) JP-A-4-38563 (JP, A) JP-A-5-2611 (JP, A) JP-A-6-68157 (JP, A) JP-A-8-171508 (JP, A) Proc. Of the 36th National Convention of IPSJ, P.S. 515-516 (58) Field surveyed (Int. Cl. ⁶ , DB name) G06F 12/00 G06F 17/30

Claims (2)

(57) [Claims] 1. A table of data registered in a database
Display, direct selection operation on display data, the display data
Operations by using a logical search expression for, and their selection
By aggregation operation on the selected data obtained by the operation
From the input table where you can create new data,
In the selected data and newly created data
Copy it to the output table and use it as a search result on the output table.
By creating some or all of the data,
A query generator for a database that performs a base search
There are, display means for displaying the input table and the output table, input means for performing a copy operation to the output table from the operation and the input table for the input table, and the display means and the input means Control means for controlling the operation of the input table, and an operation storage means for storing information representing an operation on the input table ;
A copy operation storage unit for obtaining and storing only the operation from the operation storage unit; and interpreting the meaning of the operation based on the information stored in the copy operation storage unit .
Operation interpretation means for estimating the type of inquiry, and inquiry sentence generation means for creating an inquiry sentence corresponding to the information stored in the copy operation storage means based on the meaning interpreted by the operation interpretation means. An apparatus for creating a database query sentence. 2. When a part of necessary data is created as a search result on an output table by the copy operation, the query sentence created by the query sentence generating means at this time is replaced with the entire search result. 2. The database query sentence creating apparatus according to claim 1, further comprising a query sentence generalizing unit that converts the query sentence into an obtainable query sentence.