JPH05298369A - Method and device for retrieving data - Google Patents

Abstract

PURPOSE:To shorten time required for each data retrieval and to improve the efficiency of data retrieving work in the case of retrieving data from a storage medium storing much sequential data whose rewriting is not almost required. CONSTITUTION:Data are previously recorded in a storage medium 103 in the order based upon retreiving condition codes corresponding to retrieving conditions at the rate of 1 to 1, and when a certain retrieving condition is specified, data are retrieved in the medium 103 by using the retrieving condition code corresponding to the specified retrieving condition. Since data 105 can be directly read out without accessing an index 104, time required for retrieval can be shortened. Like a convensional method, retrieval based upon the index 104 is also used for retrieving conditions which are not managed by retrieving condition codes.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION The present invention is basically ordered like the contents of a dictionary and does not vary much in size.
In addition, the present invention relates to a data search method and apparatus for performing a data search on a large-capacity data storage medium storing data that does not need to be rewritten in the future.

[0002]

2. Description of the Related Art In a conventional data search for a storage medium, as shown in FIG. 1, when a host program 101 indicates a search condition to a storage medium control program 102, the storage medium control program 102 exists inside the storage medium 103. By referring to the index 104, an address or the like in the storage medium having storage information that matches the given search condition is allocated.

When referring to the corresponding data,
The following two methods are known. First, according to the specifications of the storage medium control program 102, the information such as the obtained address in the storage medium is once returned to the upper-level program 101, and the upper-level program that has received the information again designates the address in the storage medium control program 102. This is a method for instructing to refer to the data of The second is a storage medium control program 1
02 does not return control to the higher-level program 101 and uses the obtained information such as the address in the storage medium to obtain the corresponding data 1
This is a method of returning the reference result to the higher-level program 101 after performing the reference up to 05.

[0004]

However, in the above-mentioned conventional technique, the capacity of the storage medium to be searched and the number of search data are increased with the recent development of information recording technology and the increase of processing information. Is not considered,
When the number of search items increases, the number of times that the index inside the storage medium is referred to during search increases, and as a result, the average search time increases. That is, even if one piece of data is referred to, it is necessary to refer to the index on the storage medium when performing the search, and then to refer to the requested data using the result. Increasing the number of references is inevitable.

In addition, even for data that need not be rewritten in principle, such as the contents of a dictionary, the index is referenced every time one data is referenced, which improves the efficiency of data retrieval work. It is an obstacle.

The object of the present invention is that, like the contents of a dictionary, it is not necessary to rewrite it in principle, and the number of data units for the search is enormous, and the size of each data is relatively uniform. It is an object of the present invention to provide a data search method and device that can reduce the time required for each search and improve the efficiency of a data search operation when searching a storage medium storing data. ..

[0007]

In order to achieve the above object, the present invention provides a data search method for accessing a storage medium according to an instruction from a host program to search data on the storage medium, Data having almost the same size are registered in advance in the storage medium in the order according to the search condition code corresponding to the search condition, and when a certain search condition is designated, the search condition is corresponded to. Retrieving data from the storage medium using a retrieval condition code.

Further, according to the present invention, in a data retrieval method of accessing a storage medium according to an instruction from a higher-order program to retrieve data on the storage medium, data of substantially the same size is stored on the storage medium. Preliminarily registered in the order according to a series of character codes defined by JIS or a standard equivalent thereto, and when a certain character code is specified, the address inside the storage medium is calculated from the character code, and the calculation result Is to retrieve data from the storage medium.

Further, according to the present invention, in a data retrieval apparatus for accessing a storage medium according to an instruction from a higher-level program and retrieving data on the storage medium, data having substantially the same size are paired as retrieval conditions. When a certain search condition is designated with respect to the registration means for registering on the storage medium in the order according to the search condition code corresponding to 1, and the storage medium whose data is registered by the registration means, Search means for searching data from the storage medium using the corresponding search condition code.

Further, according to the present invention, in a data retrieval apparatus for accessing a storage medium in accordance with an instruction from a higher-level program to retrieve data on the storage medium, data having almost the same size is specified by JIS or a standard equivalent thereto. When a certain character code is designated with respect to the registration means for registering on the storage medium in the order according to the series of character codes defined in 1. and the storage medium having the data registered by the registration means, the character code is specified. And a search means for calculating an address and the like inside the storage medium from the storage medium and searching the storage medium for data based on the calculation result.

[0011]

The principle of data retrieval according to the present invention will now be described with reference to FIG.
Will be explained. In FIG. 1, in order to reduce the search time, the time required to check whether each data satisfies the search condition or the reading time of the data satisfying the search condition given as the search result is shortened. There are two methods of reducing the number of accesses to the storage medium 103 required for one search. However, even if the time required for one access is shortened as in the former case, the same problem may occur again due to the increase in the capacity of the storage medium brought about by the technical progress of the storage medium in the future. Therefore, in the present invention, the number of times of accessing the storage medium, which is the latter one search, is reduced as much as possible to shorten the time required for one search.

That is, according to the present invention, in order to reduce the number of accesses to the storage medium 103 required in one search, the search condition is met in the index 104 for each one search conventionally performed. The method of checking the data is stopped, and the method of associating the search condition code with the address inside the storage medium 103 is adopted. However, if the search condition is other than the managed code, the conventional search method is used. Furthermore,
In order to correctly associate the search condition code with the address inside the storage medium, the present invention employs a method of writing data in the storage medium in the order according to the search condition code. The details will be described below.

Now, as a general rule, a large amount of data that is relatively uniform in size, is not rewritten in principle, and is frequently referred to is registered in a direct access storage medium, and a search process for the data 105 is performed. I shall. Now, as shown in FIG. 1, there is a series of ordered data D ₁ , D ₂ , D ₃ , ..., D _n , which shall be determined once and never changed. That is, the order of the data 105 that has been once ordered and arranged is constant, and the i-th data D _i from the beginning can be treated as specific data that is always uniquely determined using i as a key. The search condition for this collection of data is that only a specific one data is selected from these many data, and multiple data will not be retrieved by one search. To do.

Further, when there are two different search conditions, the data satisfying the respective search conditions are never the same. That is, each search condition and data is 1
Assuming that the number of data is n, which corresponds to the pair 1, n search conditions are required to read all the data. The present invention is effective in a system that performs a data search process that satisfies the above conditions.

Hereinafter, a search method capable of reading the corresponding data in the storage medium used with the least number of accesses is called a primary search method, and the information used at that time is called primary search information. The primary search information uniquely determines the specific area to be read on the storage medium. For example, the storage medium 1 in which the data to be read is stored
If it is possible to read the data 105 without referring to the index 104 by directly specifying the internal address of 03, this is called the primary search method, and the internal storage medium used at that time is called. Information such as the address is called primary search information. That is, the search method according to the present invention is the storage medium 103 based on the given search conditions.
The primary search information is obtained directly by searching and reading the information on the internal index 104, and then the primary search information is directly derived from the given search condition as compared with the conventional method of reading the data 105 by the primary search method. The data 105 is read by the primary search method using this.

Now, search condition codes corresponding to the respective search conditions are set. That is, the data D ₁ ,
D _2, D _3, ........., a search condition corresponding to _{D n, S 1, S 2} , S 3, ........., when expressed as S _n, the number of arrangement of the set of search criteria set of It replaces and expresses. This is called a search condition code. When there are n search conditions, n search condition codes are created. Hereinafter, individual search condition codes will be represented as C ₁ , C ₂ , C ₃ , ..., C _n . Here, the subscripts 1, 2, 3, ..., N indicate that the search condition codes correspond to the search conditions S ₁ , S ₂ , S ₃ , ..., S _n , respectively. That is, the order of a series of numbers which are the search condition codes follows the order of the search conditions determined by the order of the data. Therefore, the order of the data and the order of the serial number which is the search condition code are the same.

Next, the data 105 is registered in this order at predetermined intervals on the storage medium. A series of data is arranged on the storage medium in the order of the search condition code. Further, when the data can be directly read from the information such as the address on the storage medium corresponding to the storage position of each of these data, the information such as the address is the primary search information. When data is arranged in order on the storage medium 103 at regular intervals, the temporary search information for each data can also be regarded as a series of numerical values arranged at regular intervals. Hereinafter, each primary search information is represented as d ₁ , d ₂ , d ₃ , ..., D _n . _{Here, d 1, d 2, d} 3, ........., d n are each the data _{D 1, D 2, D 3} , ........., a primary search information corresponding to the storage position of D _n. Since the primary search information and the search condition code corresponding to each other are arranged in the same order, there is always a certain relation between them: d _i = f (C _i ) (i = 1,2,3, ..., n ) ... (1) is satisfied.

When the primary search information corresponding to each search condition code on a one-to-one basis is determined by this function f (C _i ), the data to be read according to the search condition indicated by each search condition code is searched. It exists in a specific area on the storage medium 103 that is uniquely determined by the primary search information corresponding to the condition code,
It is possible to directly read without referring to the index 104.

However, since the registration method of individual data may be registered in association with the index information as in the conventional method, the storage medium created for performing the search by this method is the same as the conventional method. You can also search. Therefore, even if exceptional data or a search condition such that the function f (C _i ) is not satisfied can be dealt with by performing the search in the conventional method.

As described above, by arranging the data 105 to be searched and read in the storage medium 103 in a predetermined order and using the function f (C _i ), the search condition corresponding to the search condition can be obtained. It is possible to obtain the primary search information directly from the code and read the required data with the minimum number of accesses.

Therefore, according to the above configuration, when the data search condition is indicated by the higher-level program, if it is the search condition code (or character code) under which the search condition is managed, An address or the like in the recording medium, that is, an address or the like in the recording medium is calculated by a function associating the above-mentioned primary search information and shown to the upper program 101. If the search condition code is an exceptional search condition that does not have a search condition code, the search is performed by referring to the index held by the recording medium as in the conventional case, and the data stored in the recording medium in which the data corresponding to the condition is stored is searched. The address etc. are shown in the upper program. The upper program uses the returned information to read the corresponding data from the storage medium. In this way, when the search condition indicated by the higher order is a managed code, it is not necessary to refer to the index of the recording medium, and therefore the time required for the search can be shortened.

In the search according to the present invention, a large amount of data to be searched does not need to be basically rewritten, the data is registered in a predetermined order, and the size does not vary. This is a format for selecting a specific one from the data group, and is effective when frequent search is performed in a system in which the search condition and the individual data are associated one-to-one.

Further, in the storage medium used in the present invention, each data must be registered at a predetermined position and in a predetermined order, but once created, a dictionary that basically does not rewrite is created. Data such as contents can be easily created. In addition, the storage medium created by the present invention can be searched by a conventional method that simultaneously refers to the index 4, and can be used as it is in a conventional system.

[0024]

Embodiments of the present invention will be described in detail below with reference to the drawings. Here, the data retrieval method and apparatus of the present invention will be described by taking a document creation system using an optical disc as an example. The document creation system creates a document using character fonts registered as image information on the optical disc. The input / output configuration is as shown in FIG. Middle software 203 for optical disk image processing,
An optical disk device 204, a display device 205, and a laser beam printer (LBP) 206 are provided.

The procedure of input / output and operation of this system is as follows. The user inputs the character to be printed from the keyboard 201. The document creation system basic program 202 determines the number of input characters, the type of characters, etc., and calculates the layout of characters at the time of printing. Then, the document creating system basic program 202 searches the character font registered as image information on the optical disk on the optical disk device 204 for a required character, and instructs the optical disk image processing middle software 203 to retrieve it. .. The optical disc image processing middle software 203 accesses the optical disc 204 in accordance with an instruction from the host computer to search for and read an image character font. The document creation system basic program 202 determines the character size and position according to the calculated layout information, and instructs the optical disc image processing middle software 203 to edit the image. The optical disc image processing middle software 203 which receives the image editing instruction edits the image by performing operations such as enlarging, reducing, moving, and rotating the read image character font. After repeating the above steps for all the characters input from the keyboard 201, the completed image is displayed on the display 205 or the LBP 20.
Printed at 6.

Image data is registered in the optical disc on the optical disc device 204 in a hierarchical structure of the format shown in FIG. Cabinet 302 is an optical disc 301
It is equivalent to one side of. A plurality of files 303 are stored in the cabinet 302, and a plurality of documents 304 also exist in each file 303. The document 304 is composed of several pages 305. One of the registered image data for one page 305
Corresponds to the sheet. Cabinet 302, file 303,
Each document 304 has a name.

In the document creation system of this embodiment, the character font to be used is used as image data in the optical disk 30.
Registered on 1. Cabinet 302 and file 3
A fixed name is given to 03, and a shift JI corresponding to the image character font stored in the document 304 is given to the document 304.
The S code is given as the document name. That is, each document 304 has only one page 305, and the page 305 corresponds to the image character font. Since the names of the cabinet 302 and the file 303 are fixed, only the document name is required as a key when searching for an image character font corresponding to a specified character.

Now, the procedure for searching for image character fonts by the conventional method will be described with reference to FIG. The user inputs the character to be printed from the keyboard 201. Here, it is assumed that "brush" is input. The document creation system basic program 202 first issues an instruction to the optical disc image processing middle software 203 to search for a document "hair". The optical disc image processing middle software 203 accesses the optical disc 301, reads the document ID of the document named “hair”, and uses it to read the system basic program 2
Return to 02. The document ID is a series of codes added by the system to each document to identify the document and is unique within the file. A plurality of documents having the same name can exist in the same file. However, in this system, the document name is also made unique. Therefore, there is always only one document ID returned here. The document creation system basic program 202 gives an instruction to read the image data stored in the first page (fixed) of the document of the returned document ID. The optical disc image processing middle software 203 searches for the first page of the document having the instructed document ID, and reads the contents into the memory. When the reading of the "hair" is completed, the above-mentioned steps 1 through 3 are repeated for the "brush".

The problem with the above conventional search method is that each time a character font reading process occurs, the document ID of the document must first be accessed by searching the optical disk 301, and then again. Since the optical disk 301 is accessed to read the image data, the number of times the optical disk 301 is accessed is increased and the processing speed is reduced. Especially when the number of characters is large, it takes a long processing time.

Therefore, in this embodiment, the problem of the conventional search method is solved by the new search method. As a basic policy, speeding up of the process is achieved by reducing the number of accesses to the optical disc 301. For inputting and outputting data to and from the optical disc 301, the optical disc image processing middle software 203 is used.
We do not develop our own input / output program. The user interface is the same as in the conventional method and is not changed. 3 items were set. That is, the number of accesses to the optical disc 301 is minimized by using the optical disc image processing middle software 203.

First, the optical disk image processing middle software 203 has a function of reading image data corresponding to a number called a registration number as a key. This registration number is a sequence of registration numbers of image data, that is, a series of unique numbers added to the page each time a new page is created. The registration number has nothing to do with the information contained in the document or page, and is added only depending on the order of registration. That is, the registration number, the document name, the document ID, the page number within the document. There is no relationship with. Regardless of which page of which document the image data is, the first registered image data is given the registration number 1, and the second registered image data is given the registration number 2. If this registration number is designated, the corresponding image data can be read by calling the optical disc image processing middle software once.

However, since the registration numbers are only added in the order of registration, they cannot generally be derived directly from the characters (shift JIS code) input by the user from the keyboard 201. This is different from the conventional method in which the character code is used as the document name for retrieval. Furthermore, the above-mentioned basic policy does not allow the user to directly specify the registration number, and such a method increases the burden on the user and is not practical.

It should be noted that the registration number depends only on the order in which the image data is registered and has nothing to do with document information or the like. Where JI
The sequence of the character codes defined as the S first and second levels can be considered as a series of numbers in a fixed order. That is, if the image character fonts are registered on the optical disc 301 in the same order as defined by JIS, the registration numbers will also be added in that order. For example, according to the order specified in JIS,
Suppose that the image character fonts of "A", "But", "Shou" ... are registered in this order.
2,3 ... will be given.

If these characters are expressed by using a division code, they can be written as 1601, 1602, 1603, ... That is, if the image character codes are registered in the order of the character codes defined by JIS, it is possible to have one relationship between the registration numbers and the character codes. In this example, it is understood that the relation of (registration number) = (department point code) -1600 ... (2) holds between the division point code and the registration number. Actually, the last two digits of the group code take only values in the range of 1 to 94, so this relational expression takes a more complicated form, but here it is in the above form for simplicity.

Generally, the character codes defined as the JIS first and second levels are not limited to the division code, but even if they are represented by the JIS code or the shift JIS code, they are all in the order determined by the standard. Can be treated as a series of numbers. Therefore, if the image character fonts are registered in the same order as the determined character code order, the order of the registration numbers added according to the registration order is the same as the order of the character codes. When such a registration method is used, it becomes possible to introduce a function that associates a series of character codes with registration numbers in a one-to-one correspondence. The above formula (2) is an example, but generally, when the registration number is d and the character code is C, a functional relationship of d = f (C) ... (3) is established.

Here, when a certain functional relationship between the registration number and the character code is established in this way, the procedure for reading the image character font of the "writing brush" using this relationship will be described with reference to FIG. become that way.

Document Creation System Basic Program 20
Reference numeral 2 calculates the registration number of the image character font from the division mark code of “hair” and transfers the calculation result to the optical disc image processing middle software 203 to instruct the reading of the image character font. The optical disc image processing middle software 203 reads the image information corresponding to the registration number passed from the document creation system basic program from the optical disc 301 onto the memory. Next, repeat the above for'brush '.

As described above, by using this new search method, it becomes unnecessary to read the information of the document ID from the index on the storage medium as in the conventional method, and the number of accesses to the optical disk 301 can be reduced. In addition, the processing can be simplified and the processing efficiency can be improved.

Unless the registration number is consciously associated with each other, the document format is generally completely unrelated to the information about the document. Therefore, the format of the document is the same as that used in the conventional method. 1 image character font)
Then, the document may be left in the form that the shift JIS code of the character is given as the document name. By doing so, it becomes possible to use the optical disk produced in this embodiment with the conventional search method.

In the document creating system of this embodiment, the JIS
The order of the character code is completely determined by, the characters of JIS 1st and 2nd level, etc. are made to be able to be searched quickly by this method, and are not specified in JIS, so they are arranged in a fixed order and serial numbers are set. When it is desired to assign information with a character that is not in JIS as an external character and assign an arbitrary character code to the information, for example, the conventional search method of searching by document name is used. The processing flow in this case is shown in FIG. That is, in FIG. 4, it is determined in step 402 whether the character code received in step 401 is a JIS level 1 or level 2 character, and the JIS level 1
If the character is a second-level character, the registration number is calculated directly from the character code in step 403. If it is not a JIS first-level or second-level character, a search process is performed on the optical disk in step 404 to find the corresponding document. The document ID of is acquired.
Then, in step 405, the image font is read from the optical disc.

[0041]

As described above, according to the present invention,
When performing a data search on the storage medium, it is not necessary to refer to the index on the storage medium, which has been performed every time a search request is made, so the search time can be shortened. Especially when the amount of data to be searched is extremely large,
It is also effective when the access speed to the storage medium is slower than the execution speed of the upper program. Further, as for the search condition that is not managed by the search condition code, since the search by the index on the storage medium is also used as in the conventional case, it is possible to prevent the data from being overlooked during the search. From the above, it is possible to improve the efficiency of the search operation at the time of data search.

[Brief description of drawings]

FIG. 1 is a diagram showing a method of accessing a storage medium.

FIG. 2 is an input / output configuration diagram of a document creation system using an optical disc.

FIG. 3 is a logical structure diagram of registration data on an optical disc.

FIG. 4 is a flowchart showing a search process of a document creation system using an optical disc.

[Explanation of symbols]

 101 Upper Program 102 Storage Medium Control Program 103 Storage Medium 104 Storage Medium Index 105 Storage Medium Data 201 Keyboard 202 Document Creation System Basic Program 203 Optical Disk Image Processing Middle Software 204 Optical Disk Device 205 Display Device 206 Laser Beam Printer 301 Optical Disk 302 Cabinet 303 files 304 documents 305 pages

Claims (5)

[Claims] 1. A data search method for accessing a storage medium according to an instruction from a higher-level program to search for data on the storage medium, wherein data having substantially the same size on the storage medium is used as a search condition. It is pre-registered in the order according to the search condition code corresponding to the pair 1, and when a certain search condition is designated, data is searched from the storage medium using the search condition code corresponding to the search condition. Characteristic data retrieval method. 2. The data search method according to claim 1, wherein when a search condition that is not managed by the search condition code is designated, the data is searched using the index of the recording medium. retrieval method. 3. A data retrieval method of accessing a storage medium according to an instruction from a higher-level program to retrieve data on the storage medium, wherein data having substantially the same size is compliant with JIS or its equivalent on the storage medium. Pre-registered in the order according to a series of character codes defined by the standard, when a certain character code is designated, the address etc. inside the storage medium is calculated from the character code, and the storage medium is calculated according to the calculation result. A data search method, characterized in that data is searched from. 4. A data search device for accessing a storage medium according to an instruction from a higher-level program to search for data on the storage medium, wherein data having substantially the same size correspond to a search condition on a one-to-one basis. When a certain search condition is specified for the registration means for registering on the storage medium in the order according to the search condition code and the storage medium for which data is registered by the registration means, the search condition corresponding to the search condition And a search means for searching data from the storage medium by using a code. 5. A data retrieval device for accessing a storage medium according to an instruction from a host program to retrieve data on the storage medium, wherein data having substantially the same size is defined by JIS or a standard conforming thereto. When a certain character code is designated with respect to the registration means for registering on the storage medium in the order according to a series of character codes and the storage medium having the data registered by the registration means, the inside of the storage medium from the character code is designated. And a search unit that searches the storage medium for data based on the calculation result.