JP3202341B2 - Database system - 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 system, and more particularly to a database system for directly collating fixed-length records by full search of text format data.

[0002]

2. Description of the Related Art Conventionally, widely used database systems are roughly classified into a database (relational database) using a relational data model and a database (network database) using a net data model. The relational database has a conceptual model in which each item of a record is expressed in a table format, and a reference operation of the table is performed by a relational algebra operation based on a set operation. Further, the mesh database has a conceptual model represented by a graph expression in which data is represented by nodes and the parent-child relationship thereof is represented by directed edges. These are intended to perform high-speed data retrieval of a huge amount of data and unify data management by associating accumulated data records and data items therein.

[0003]

In a conventional database, data management is unified by associating data with each other, but there are some problems in use.

[0004] First, it is difficult to construct and operate a database unless the relationship between each record and internal item data is sufficiently understood. In particular, a net-type database or the like requires knowledge of the physical data structure and the access path to data, and is difficult to use for those who do not have knowledge of the database. Further, since the database holds the relationship between data and the like, much information is required other than the data itself, and the size of the database itself tends to be larger than the actual data amount.

[0005] In the sense of data retrieval,
A method of searching the entirety of the accumulated data from the beginning to the end and extracting a corresponding record is also conceivable, but the search speed is not practical in a conventional device, and is not practical.

[0006]

Means for Solving the Problems The present invention, or a plurality of items
A data storage unit for storing data in a record format, a keyword input unit for inputting a keyword to be searched for a specific item of the plurality of items, and a keyword stored in the data storage unit. No record
Key with the same format as the
The input keyword is applied and other items
A keyword conversion unit that converts the key to a wildcard, and a conversion by the keyword conversion unit.
A collating unit for collating the generated key with the data of the data accumulating unit, a retrieval result accumulating unit for accumulating corresponding record information matching the retrieval condition as a result of collation by the collating unit, A database system comprising: a search result display unit for presenting a search result by a storage unit to a user , wherein the matching unit is a character string matching processor, and the data of the data storage unit is searched directly as a key. Is possible.

[0007]

Embodiments of the present invention will be described with reference to the drawings.

FIG. 1 shows an embodiment of a database system according to the present invention.

The data storage unit 11 stores data registered as a database.

The keyword input unit 12 is a device for setting a keyword or search condition to be searched by a database user.

The keyword conversion unit 13 generates data in a format corresponding to the data format of the data storage unit 11 from the keywords and search conditions input from the keyword input unit 12.

The collation unit 14 is a character string collation processor, and collates the converted keyword with the data in the data storage unit 11. The present processor has the ability to collate not only one character string but a plurality of character strings at the same time. Further, the processor can obtain the end address or the start address of the character string that matches as a result of the comparison.

The search result storage unit 15 stores data record information that matches the search conditions as a collation result.

The search result display section 16 presents information of the search result to the user.

Next, the operation will be described with reference to the flowchart of FIG. First, the user of the database inputs his / her desired keyword from the keyword input unit 12. At this time, you can specify multiple keywords, search conditions (~
AND ~, ~ OR ~) can be specified (Step 2)
1). The keyword conversion unit 13 includes the keyword input unit 12
Based on the keyword and the search condition input from, data to be a key at the time of collation in the collation unit 14 is generated (step 22). Since the operation of the keyword conversion unit 13 and the data format of the data storage unit 11 are closely related, they will be described later in detail with reference to FIGS.
The data converted and generated by the keyword conversion unit 13 is sent to the matching unit 14 as a search key, and the matching unit 14 performs matching with the data in the data storage unit 11 (step 2).
3). As a result of the collation, information on the result data searched as data to be extracted is stored in the search result storage unit 15 (step 24), and the search result can be displayed on the search result display unit 16 (step 25).

Next, a search method in the present database system will be described. In the present database system, a character string collation processor is used for the collation unit 14, so that the data retrieval is a method of extracting a match in character string collation. Therefore, it is characterized by the conversion and generation processing of the data format of the database and the keyword data serving as a search key at the time of collation.

Therefore, an example will be described in which a database such as a list having "name", "hometown", "sex" and "date of birth" is applied to this database. First, a method of expressing database data stored in the data storage unit 11 will be described with reference to FIG. The record composed of the items shown in the example can be conceptually expressed in a format as shown in FIG. 3A, but in this database, the data is expressed in a text format as shown in FIG. 3B. I do. This is a fixed-length record in which a “start code” indicating the start of a record is added at the start position of each record, and a “end code” indicating the end of the record is added at the end position.

Next, the operation of the keyword conversion unit 13 for generating a key suitable for the data format from the input keywords and search conditions will be described with reference to FIG. FIG.
It is assumed that the following search conditions and keywords are input to the text format file as in (B) (in this example, "?" Indicates a wildcard).

Last name = mountain? AND Hometown = (Chiba OR Kanagawa) Then, the keyword conversion unit generates two keys shown in FIG. 4 according to the record data format. Since the two keys converted and generated here can be regarded as one character string, these are passed to the matching unit 14 as keys. In the character string matching processor of the matching unit, FIG.
The data of the data storage unit 13 having the text format shown in FIG. The processor used in the present database system can match a plurality of keywords and use wildcards, so that these keys are matched in one data search. Since the collating unit can know the start or end address of the matched character string, both the stored data and the key are used as in the record format shown in FIG. 3B or the key format shown in FIG. If the start and end codes are added to, the address of the matching record can be always obtained. The obtained information is stored in the search result storage unit as shown in the flowchart of FIG.

[0020]

As described above, according to the present invention, a high-speed character string collation is possible because the collating unit uses a dedicated processor for character string collation. Therefore, it is possible to directly search the character string data in the database, and it is not necessary to perform index registration, and registration and updating of the database as a normal text file can be easily performed. Further, since there is no redundant information such as an index and a relationship between data, it is possible to store data with a smaller disk usage compared to a conventional database.

[Brief description of the drawings]

FIG. 1 is a block diagram showing one embodiment of a database system according to the present invention.

FIG. 2 is a flowchart illustrating an operation of the database system of FIG. 1;

FIG. 3 is a diagram for explaining a data format of a data storage unit included in the database system of FIG. 1;

FIG. 4 is a diagram for explaining an operation of a keyword conversion unit constituting the database system of FIG. 1;

[Explanation of symbols]

 11 data storage unit 12 keyword input unit 13 keyword conversion unit 14 collation unit 15 search result storage unit 16 search result display unit

────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Yoko Miyao 2-17-11 Takanawa, Minato-ku, Tokyo Within NEC Software Corporation (56) References JP-A-1-265321 (JP, A) “Introduction to awk”, Monthly ASCII, ASCII Corporation, 1989.4.1, Vol. 13 No. 4, p. 317-322 Kazuo Uozumi, "Stranger than an MS-DOS personal computer is the life of a man-a responsible person!" Ascii Monthly ASCII, ASCII Corporation, 1991.1.1 Vol. 15, No. 1 p. 313-320 Masami Yamatani, "Introduction to File Organization", 1st edition, Ohmsha Co., Ltd., July 25, 1980, p. 1-26 Sueharu Miyahara et al., "Full-text search using SMID-type parallel processor", Transactions of Information Processing Society of Japan, IPSJ Vol. 33 No. 3 1992.3, p. 397-404 Vice Information Center, "First Use dBASE III", 1st edition, Technical Review Co., Ltd., November 1987, p. 90-102, "Software Library MS-DOS 3.3 Users Reference Manual", NEC Corporation, 1988, p. 5-11 (58) Field surveyed (Int. Cl. ⁷ , DB name) G06F 12/00

Claims (3)

(57) [Claims] A data storage unit for storing data in a record format including a plurality of items; and a keyword input unit for inputting a keyword to be searched for a specific item among the plurality of items. The input keyword is converted to the data of the data storage unit.
A key with the same format as the record format,
The keyword entered above is applied to the eyes,
A keyword conversion unit that converts the key into a key in which a wildcard is applied to an item ; a matching unit that matches the key converted and generated by the keyword conversion unit with the data in the data storage unit; As a result, a database system comprising: a search result storage unit that stores applicable record information that matches the search condition; and a search result display unit that presents a search result by the search result storage unit to a user. A database system, wherein the collation unit is a character string collation processor, and the data in the data storage unit can be searched directly as a key. 2. The keyword input unit according to claim 1 , wherein
Enter multiple keywords and link these keywords.
And the keyword conversion unit determines that the search condition is “OR”.
The number of keywords connected by "OR"
Key and the keywords linked by "OR"
Generate a key applied to the corresponding item of the key
The database system according to claim 1, wherein: Wherein said data storage unit, the data base according to claim 1 or 2, characterized by using a text file consisting of fixed-length records having a record start code and end code as data representation scheme <br/>