JP5300655B2 - Information storage search method, information storage search device, and program - Google Patents

Description

  The present invention relates to an information storage search method, an information storage search device, and a program.

  RDB (relational database) is one of information storage and retrieval systems (database). In general, the basic role of such a basic system is to provide a common function that does not depend on the application as a general-purpose system separately, thereby reducing the development man-hours of each application, and is common to the applications. The aim is to make it available for use.

  Provides a basic information storage and retrieval system similar to the one described above for building and searching for applications that detect real-world events with various sensors or operate with actuators and handle the information detected by the sensors. There is merit to do. One such sensor information storage and retrieval system is uTupleSpace (see, for example, Non-Patent Document 1).

  One of the features of uTupleSpace is that it is a history type information storage and retrieval system. That is, all data from the sensor (in the case of uTupleSpace, EventActual (EA)) is accumulated as a history in the system. In addition, a search request (in the case of uTupleSpace, EventForm (EF)) specifies a range condition such as a time or a value, so that information from a single sensor or multiple sensors can be specified in the past history. You can search and obtain all that matches. As described above, since the operation for examining the past history of the sensor information can be easily realized by the application, the history type information storage and retrieval system has an advantage.

  One of the desired characteristics of the sensor information storage and retrieval system is a schemaless retrieval function. In RDB, generally, a type definition of information stored in each table is given as a schema, and information having the same type is accumulated in a single table.

  However, when handling information from various sensors, a wide variety of sensors generate a variety of combinations of information, so it is required to accumulate them and search across them. For this reason, a function for storing and searching without requiring a schema definition (schemaless) is required. As a result, the system can be provided with the versatility generally required for basic systems.

"Evaluation of dynamic scaling method in real-world data sharing mechanism uTupleSpace", Arakawa et al., IEICE Technical Report, June 2009

  When trying to construct a typical application using a history type sensor information storage / retrieval system, there is a problem to be pointed out below that an application cannot be easily constructed with the retrieval function provided conventionally.

  For the sake of explanation, as an example, consider an application that manages an article having a state sensor and a communication function. Here, for example, it is assumed that the article regularly transmits the position coordinates of the article by its communication function. Further, as an example of the basic management function, it is assumed that the management target articles are displayed as a list. At that time, in addition to the article name, article number, owner, etc., the state of the article acquired from the state sensor is included in the table and displayed.

  In order to realize the list function, it is necessary to obtain the latest value of the state for each row of the table, that is, each target article, and draw it alongside other information such as the article name. Similar functionality is required in many other applications as well.

  There has been a problem that the operation for obtaining the latest value of the state cannot be realized directly using, for example, the EF of u TupleSpace. The reason for this is that in a historical sensor information storage and retrieval system, what can be obtained directly by the basic operation provided is the entire history or a partial history specifying a time range, and the latest of the target object of interest. This is because it is not a value.

  Alternatively, the application program, for example, uses a separate RDB or the like to receive sensor information and updates the latest value of the corresponding article each time according to the contents without using a historical sensor information storage and retrieval system. And the above functions can be realized.

  However, in this realization method, in another type of application, the procedure for receiving the sensor information using RDB or the like and updating the latest value of the corresponding article each time according to the contents is realized as a separate program each time. Must. Therefore, in this realization method, there is a problem that it is not possible to enjoy the merits of the basic role generally obtained in the basic system.

  Further, the above-described method for receiving sensor information using RDB or the like and simply implementing the procedure for updating the latest value of the corresponding article each time according to the contents as an independent program and constructing a basic system Can be considered.

  However, the above method has a problem that the versatility inherent in the schemaless search function is impaired. The reason is that the concept of “corresponding article” included in the procedure in the above method is different for each application.

  What attributes are stored in the system as an attribute of an identifier of an article, or what is the concept of an article in the first place, that is, when multiple pieces of information are equal among stored individual information Is to be recognized as information relating to a single article, and it depends heavily on the design of the application, and cannot be assumed by a general-purpose system.

  Therefore, for example, as an example of implementation of the method, “the identification method of using the value indicated by the attribute“ article ID ”as the identifier of the article is used, and the latest value of the corresponding article is updated each time according to the identification method” Even if the implementation method is adopted, such a storage / retrieval system cannot be said to have sufficient versatility that the system originally has, and there is a problem that the versatility is impaired.

  In addition, an information storage / retrieval system generally provides a function of receiving an argument for limiting a search target during a search and performing a search within the target range. Further, a function is provided for receiving a conditional sentence at the time of search and searching for a sentence that matches the condition. It can be understood that these two functions provide the same kind of semantics in the sense of specifying a range of search results.

  For example, in the SQL language of RDB, the table to be searched can be specified by the FROM phrase of the SELECT statement, and thereby the range of the search process can be defined. Furthermore, in the WHERE clause of the SELECT statement, a conditional statement can be specified, thereby limiting the search result. These two phrases have the same description method and the target that can be specified, but the same in terms of narrowing down the data to be returned as a result by limiting the search target among all the data held by the RDB. It has the functionality.

  However, in the history-type sensor information storage and retrieval system, a function for searching for the latest value of an article is provided, and at that time, a function for simply designating a search target range meaning that a search target is limited is simply provided. Then, as pointed out below, there was a problem that the necessary functionality for realizing the application could not be provided.

  That is, for example, in the application for managing an article, it is assumed that the article is carried and moved, and history of different position information is accumulated one after another. Here, it is assumed that the application executes a search operation for acquiring “a list of articles currently in the specified position range”. The application calls a function for searching for the latest value of the article provided by the history type sensor information storage and search system. At that time, the search target is “the transmission information of the state sensor and within the specified position range”. Is included as a condition.

  The sensor information accumulation / retrieval system searches only the information that matches the information and the position range condition of the search target, and acquires target data. Then, it is assumed that a list of articles is created from the acquired data and is returned to the application as a search result of “list of articles currently in the specified position range”.

  In this case, the implementation method is incorrect and cannot return a correct result. Because some articles may exist in the location range in the past and then exist outside the location range, such articles should not be included in the list. However, in the realization method, such an article is also included as a search result, resulting in an error.

  In this way, the function of specifying the search target range in the sense of limiting the target of the search scan is a necessary function, but this function alone is a necessary function for realizing the application as pointed out above. Can't provide sex.

  The present invention has been made in consideration of such circumstances, and its purpose is to acquire the latest information while maintaining versatility by schemaless search in a history type information storage and search system. Is to provide an information storage search method, an information storage search device, and a program.

The present invention includes a storage unit, a history search unit, a latest value search means and information storage retrieval method information storing retrieval apparatus performs equipped with the storage means, the unit der Ri attributes of the storage retrieval target information A tuple including an array of value pairs as input, an accumulation procedure for storing the tuples, and the history search means from the tuple group stored in the accumulation means for at least one attribute-value range set A history search procedure for obtaining a tuple group of search results by selecting a target that matches a search expression including a list, and a tuple group in which the latest value search means performs a search operation from the tuple groups stored in the storage means Select a target that matches a search expression that includes a search target that is information that limits the range of a key, a key attribute that is a sequence of one or more attributes, and a conditional expression that is a sequence of one or more pairs of attribute-value ranges To obtain the tuple group of search results And a procedure, the accumulation procedure, the storing means, the tuple input, said time attribute as an attribute in the list if that does not contain a set of well time attributes and values after adding comprises the steps of storing previously in addition to information stored in the prior SL history search procedure, the history search means, for each of the attributes in list included in the input search expression comprises sure procedures and included as attributes in the tuple that attribute stores the value thereof corresponding to obtain as a result the corresponding select the one that matches the value range condition search in the search expression, before Symbol In the latest value search procedure , the latest value search means is input from a procedure for selecting a tuple that matches the search target included in the input search formula and a tuple that matches the selected search target. The key included in the search expression A procedure for extracting a value corresponding to a gender and a tuple that is associated with the key attribute-value pair is stored for each tuple having the same key attribute-value pair based on the extracted value. Included in the inputted conditional expression from the procedure for updating the tuple stored in the latest value table by overwriting with a tuple with a newer time attribute value and the tuple stored in the updated latest value table Re that always has that attribute for each attribute in the list included as an attribute in the tuples are stored and also select to the corresponding values match the corresponding value range condition in said condition characterized in that it comprises a procedure for obtaining a search result.

The present invention, in the storage procedure, the same or different and sequence of attributes other one tuple sequence of attributes, including the tuple to be accumulated is already stored comprises, in the history search procedure, the A history search is performed on the accumulated tuples, and in the latest value search procedure , the sequence of attributes included in each tuple in the tuple group included in the range as a search target including the search expression is the same or different between the tuples. A search result is obtained by inputting the search target for designating such a range.

The present invention is a unit der Ri attributes of the storage retrieval target information - as input tuple containing a sequence of pairs of values, and storing means for storing the tuple from the tuple group stored in said storage means, the attribute -A search operation is performed from a history search unit that selects a target that matches a search formula including an array of one or more sets of value ranges and obtains a tuple group of search results, and a tuple group stored in the storage unit. Target that matches a search expression that includes information that limits the range of tuple groups, a key attribute that is an array of one or more attributes, and a conditional expression that is an array of one or more pairs of attribute-value ranges and a latest value search means for obtaining a tuple group selected search results, said storage means, said tuple input, the well is a time attribute as an attribute in the list if that does not contain after you add a set of time attributes and values, already stores And stores in addition to have information, before Symbol history search means, is always the attribute for each attribute in the list included in the input search expression contained as attributes in the tuple for storing and corresponding the value is to select the one that matches the corresponding value range condition in the search expression acquired as the search result, before Symbol latest value retrieval means, the tuple the key attributes - stored in correspondence with the set of values The key value included in the input search formula from the latest value table, means for selecting a tuple matching the search target included in the input search formula, and the tuple matching the selected search target For each tuple having the same key attribute-value pair based on the extracted value, and the time attribute value of the tuple stored in the latest value table is calculated based on the extracted value. Overwrite with new tuple Means for updating Te, said the updated tuple said stored the latest value table, even the tuple always have the attribute for each attribute in the list that is part of the inputted said condition stored included as attributes and characterized in that it comprises a means for obtaining also the as search results by selecting the corresponding value matches the corresponding value range condition in the conditional expression.

  According to the present invention, the storage means has the same or different attribute list included in another tuple already stored in the attribute list included in the tuple to be stored, and the history search means stores the stored A history search is performed on the tuples, and the latest value search means selects a range in which the sequence of attributes included in each tuple in the tuple group included in the range as a search target including a search expression is the same or different between the tuples. A search result is obtained by inputting the search target to be specified.

The present invention is an information storage search program for causing a computer to function as each means included in the information storage search device .

  According to the present invention, in the history type information storage and retrieval system, it is possible to realize a function of acquiring the latest information while maintaining versatility by schemaless retrieval.

It is a block diagram which shows the structure of the information storage search apparatus by embodiment of this invention. It is a conceptual diagram which shows the content of the data table 108 by this embodiment, and the mode of a process of a storage request. It is a conceptual diagram which shows the search formula of a history search by this embodiment. It is a conceptual diagram which shows the search formula of the latest value search by this embodiment. It is a flowchart which shows the procedure of the process of the newest value search by this embodiment. It is a conceptual diagram which shows the content of the newest value table | surface by this embodiment.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
The present invention relates to an information retrieval method for selecting a tuple that matches a retrieval formula inputted from a tuple group stored in a history type storage system. A tuple includes a plurality of “attribute = value” including a time attribute. The search expression includes search target information that is information that limits a search operation range from the stored tuple group, and one or more “attribute = And a conditional expression consisting of a set of “value range”. When searching for a tuple, all the attributes included in the search target specified by the search target information and specified by the conditional expression are included from the stored tuple group, and the specified attribute is included. All tuple groups whose values corresponding to the conditional expression match the conditional expression are selected, and a plurality of tuples having the same “attribute = value” tuples except for the time attribute exist among the selected tuple groups. In this case, the latest tuple is selected from the plurality of tuple groups based on the time attribute.

  FIG. 1 is a block diagram showing a configuration of an information storage / retrieval apparatus according to an embodiment of the present invention. In the figure, an information storage / retrieval apparatus 100 includes a request accepting unit 101, a response sending unit 102, a storage processing unit 103, a history search processing unit 104, a latest value search processing unit 105, and a storage device 107. The latest value search processing unit 105 holds the latest value table 106. The storage device 107 holds a data table 108.

  The request reception unit 101 receives an external request 110 and instructs the accumulation processing unit 103, the history search processing unit 104, or the latest value search processing unit 105 to perform processing according to the request content. The response sending unit 102 sends the processing results of the accumulation processing unit 103, the history search processing unit 104, or the latest value search processing unit 105 to the outside as a response 111 as necessary. The accumulation processing unit 103, the history search processing unit 104, or the latest value search processing unit 105 reads / writes data from / to the storage device 107 as necessary.

  FIG. 2 is a conceptual diagram showing the contents of the data table 108 according to the present embodiment and how the accumulation request is processed. A tuple is information expressed in an array format of “attribute = value” pairs such as “attribute 1 = value 1, attribute 2 = value 2, attribute 3 = value 3,... It is not always necessary to provide an upper limit for the length of the array except for restrictions due to storage capacity or the like (multidimensional).

  The data table 108 holds the arrangement of tuples in order. At this time, the attributes do not necessarily match between the tuples. The value type may be a general data type such as a numerical value or a character string, and it is desirable that the magnitude relationship be determined for convenience of range specification, but is not limited thereto. Moreover, it is desirable that the same attribute type be used for the same attribute.

  The request reception unit 101 receives the tuple A 201 as an accumulation request. When the time attribute is not included in the tuple A 201, the accumulation processing unit 103 adds information of “time = (current time)” to the inside, and when the time attribute is included, The value is replaced with the current time, and the tuple A 201 is added 210 to the end of the data table 108.

  By the accumulation process described above, all tuples in the data table 108 include a time attribute, and it is guaranteed that a new time will be reached from the top to the bottom.

FIG. 3 is a conceptual diagram showing a search expression for history search according to this embodiment.
The request receiving unit 101 receives the search formula A301 as a history search request. The search formula A301 includes a set of “attribute = value range” pairs. The history search processing unit 104 checks the tuples in the data table 108 one by one from top to bottom, collates with the search formula A301, and acquires a set of matching items as a search result.

The above verification is performed according to the following rules.
First, for each “attribute = value range” pair in the search expression, the attribute must be included in the tuple being collated in the data table 108. Next, the value corresponding to the attribute included in the tuple being collated must be included in the range compared to the value range in the search expression. If the value does not match the data type of the value range, the verification is not established. Further, the value range may be expressed as a set of opening price to closing price, or may be a single value.

  As described above, it is determined that the tuples being matched match only when both of the two conditions are satisfied for all the pairs in the search formula. The response sending unit 102 returns the search result as a response 111.

  In the storage and history search shown in this embodiment, type definitions corresponding to the schema in the RDB are not required (schemaless). Arbitrary attribute sequences can be entered as tuples when storing, and only the attributes that you want to include at least can be specified as search expressions when searching for history. Even if other information that is not included is included, a search result that matches the search formula can be obtained without any problem.

  In the present invention, even a storage retrieval system including the concept of a table in RDB or the concept of a schema can be implemented without any problem. On the other hand, in this embodiment, the concept of the table in RDB and the concept of schema are not necessarily required, and all kinds of data can be held in a single data table, and all kinds of data are stored. It is possible to search the history horizontally.

  If you want to acquire all data of a certain type, you can understand that the attribute characterizing that type is in the tuple, so the range of “the attribute = −∞ to + ∞” is used as the search expression. The purpose can be achieved by specifying.

  FIG. 4 is a conceptual diagram showing a search expression for the latest value search according to this embodiment. FIG. 5 is a flowchart showing the latest value search processing procedure according to this embodiment.

  The request receiving unit 101 receives the search formula B401 as the latest value search request (step S1). The search formula B 401 includes a search target 402, a key attribute 403, and a conditional formula 404 as shown in FIG. The search target 402 includes a sequence of “attribute = value range”, the key attribute 403 includes a sequence of “attribute”, and the conditional expression 404 includes a sequence of “attribute = value range”.

  The latest value search processing unit 105 obtains a set of tuples in the data table 108 that match the search formula B401 by the following procedure and obtains it as a search result. First, the latest value search processing unit 105 may perform initialization to delete all the contents of the latest value table 106 (step S2). Next, the latest value search processing unit 105 checks the tuples in the data table 108 one by one from the top to the bottom (step S3), collates with the search target 402 (step S4), and applies only to each matching tuple. On the other hand, the following procedure is performed. The collation rule is the same as that used when collating a search expression when a history search request is made.

  Next, the latest value search processing unit 105 extracts, for each matching tuple, a set of “attribute = value” corresponding to the attribute included in the tuple for each attribute included in the key attribute (step S5), obtained as the sequence. If the tuple to be extracted does not exist in the tuple, a processing error occurs. The error can be avoided if the search target is appropriately specified at the time of the request and the tuple always exists.

  Next, using the list as a key, the entire tuple is included as a corresponding value, and the latest value table 106 is updated (step S6). If the latest value table 106 already includes the key as an element, the corresponding value is updated with the value, and if not, it is added to the latest value table 106. Here, FIG. 6 is a conceptual diagram showing the contents of the latest value table according to the present embodiment. The latest value table 106 includes a key column 601 and a value column 602. The above is repeated for all tuples that have been successfully verified (steps S3 to S6).

  The latest value table 106 finally obtained in this way is stored in order so that the tuples in the data table 108 become new times as they move from the top to the bottom. For each object, the old value is always overwritten in the latest value table 106 with the new value, resulting in the latest value for each object.

  In this way, in the finally obtained latest value table 106, all value column values, that is, tuples, are checked against the conditional expression 404, and only matching tuples are selected (step S7). The collation rule is the same as that used when collating a search expression when a history search request is made. The selection result becomes the search result. The response sending unit 102 returns the search result as a response (step S8).

  In the present embodiment, the collation with the search target 402 functions as specifying a search target range that means that the search target is limited. On the other hand, the collation with the conditional expression 404 is performed on the tuple in the data table 108 beyond the range of the condition, and collation with the condition is performed at the time when the final result is returned. Function.

  In this way, by providing a range designation function having two different meanings together and realizing it by the procedure described above, it is a function necessary for realizing a wide range of applications that could not be achieved in the past. Made it possible to provide sex.

  In the present embodiment, the latest value search processing unit 105 updates the latest value table 106 after receiving the latest value search request. The scope of the present invention is not limited to this, and the latest value search processing unit 105 may perform the update process in advance before receiving the request.

  After receiving the latest value search request, it is natural that the latest value search processing unit 105 initializes the contents of the latest value table 106 (step S2 in FIG. 5) before starting to update the latest value table 106. . However, if the updated tuple in the data table 108 has not been updated or deleted, and the latest value search request has the same contents as the previous request, the initialization of the contents of the latest value table 106 can be omitted. In that case, the results obtained are the same. In particular, when the update processing is performed in advance before the latest value search processing unit 105 receives the request, it is advantageous in processing efficiency to minimize the initialization of the contents of the latest value table 106. .

  In the present embodiment, the key attribute is included in the latest value search request. Thereby, the versatility of the apparatus can be enhanced. The scope of the present invention is not limited to this. The information storage / retrieval apparatus 100 determines information corresponding to the key attribute in advance, receives a request not including the key attribute as the latest value search request, By using the determined information, the subsequent processing can be executed as if it was received as a request including the key attribute. For example, “identifier”, “sensor ID”, “article number”, “name”, and the like are listed as candidates for the fixed key attribute. It is also possible to search for the attribute example from the attributes included in the search target 402 included in the request, and to make the fixed key attribute candidate.

  According to the present invention, it is often used in a typical application that handles sensor information while achieving an application that can easily implement an operation for examining past history of sensor information, which is an advantage of a history type sensor information storage and retrieval system. While providing an operation to obtain the latest value of the state, it is possible to arbitrarily specify a “concept for identifying an article” that is different for each application, and maintain the versatility inherent in the schemaless search function At the same time, in addition to the search target meaning to limit the search scan target, it is possible to specify a conditional expression that narrows down the search results according to the latest value, providing the functionality necessary for the application, Fundamental role of the basic system by providing common functions as a general-purpose system separately. It is possible to satisfy the.

  In the above-described embodiment, a program for realizing the functions of the storage processing unit 103, the history search processing unit 104, and the latest value search processing unit 105 of the information storage / search apparatus is recorded on a computer-readable recording medium. The program recorded on this recording medium may be read into a computer system and executed. Here, the “computer system” may include an OS and hardware such as peripheral devices. Further, the “computer system” includes a homepage providing environment (or display environment) if a WWW system is used. The “computer-readable recording medium” means a flexible disk, a magneto-optical disk, a ROM, a writable nonvolatile memory such as a flash memory, a portable medium such as a CD-ROM, a hard disk built in a computer system, etc. This is a storage device.

  Further, the “computer-readable recording medium” means a volatile memory (for example, DRAM (Dynamic DRAM) in a computer system that becomes a server or a client when a program is transmitted through a network such as the Internet or a communication line such as a telephone line. Random Access Memory)), etc., which hold programs for a certain period of time. The program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium. Here, the “transmission medium” for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line. The program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, what is called a difference file (difference program) may be sufficient.

DESCRIPTION OF SYMBOLS 100 Information accumulation search apparatus 101 Request reception part 102 Response sending part 103 Accumulation process part 104 History search process part 105 Latest value search process part 106 Latest value table 107 Storage apparatus 108 Data table 201 Tuple 302 Search formula A
401 Search formula B
402 Search target 403 Key attribute 404 Conditional expression 601 Key column 602 Value column

Claims (5)

An information storage search method performed by an information storage search device comprising storage means, history search means, and latest value search means,
Said storing means, the unit der Ri attributes of the storage retrieval target information - and storage procedures as input tuple, and stores the tuple containing a sequence of value pairs,
A history search in which the history search means selects a target that matches a search expression including an array of one or more attribute-value ranges from the tuple group stored in the storage means to obtain a tuple group of search results Procedure and
The latest value search means is a search target that is information that limits the range of tuple groups to be searched from the tuple group stored in the storage means, a key attribute that is an array of one or more attributes, and an attribute A latest value search procedure that selects a target that matches a search expression including a conditional expression that is a list of one or more sets of value ranges and obtains a tuple group of search results ;
Wherein the storage step, the storage means, the tuple input, after the case that does not contain the time value as one of the attributes in the list to add a pair of said well time attributes and values, already Including the procedure to store in addition to the stored information ,
The front Symbol history search procedure, the history search means, is always the attribute for each attribute in the list included in the input search expression contained as attributes in the tuple that stores and its desired value Includes a procedure of selecting a search result that matches a corresponding value range condition in the search formula ,
The front Symbol latest value search procedure, is the latest value search means,
Selecting a tuple that matches the search target included in the input search expression;
A procedure for extracting a value corresponding to the key attribute included in the input search expression from a tuple that matches the selected search target;
Based on the extracted value, for each tuple having the same key attribute-value pair, the tuple stored in the latest value table storing the tuple in association with the key attribute-value pair is time Update the attribute value by overwriting it with a newer tuple,
Tuples stored in said latest value table and the update is always the attribute for each attribute in the list that is part of the inputted said condition is included as an attribute in the tuples stored and that information storage retrieval method characterized by corresponding values and a procedure for obtaining even of a search result by selecting to meet the corresponding value range condition in the conditional expression. In the accumulation procedure , the sequence of attributes included in the tuple to be stored is the same as or different from the sequence of attributes included in another already stored tuple.
In the history search procedure , a history search is performed on the accumulated tuples,
In the latest value search procedure , the search target that designates a range in which the array of attributes included in each tuple in the tuple group included in the range as the search target including the search expression is the same or different between the tuples is input. The information storage and retrieval method according to claim 1 , wherein a retrieval result is obtained. A storage means as input tuple, and stores the tuple containing a sequence of pairs of values, - a unit der Ri attributes of the storage retrieval target information
A history search unit that selects a target that matches a search formula including a list of one or more attribute-value ranges from the tuple group stored in the storage unit, and obtains a tuple group of search results;
From the tuple group stored in the storage means, a search target that is information that limits the range of the tuple group to be searched, a key attribute that is a list of one or more attributes, and one or more attribute-value ranges A latest value search means for selecting a target that matches a search expression including a conditional expression that is an array of pairs and obtaining a tuple group of search results ;
It said storage means,
The input the tuples, the time value as one of the attributes in the list after adding said set of well time attribute and value if that does not contain, already stored in addition to information stored ,
Before Symbol history search means,
For each attribute in the list included in the input search expression, the attribute is also included as an attribute in the tuple stored in the attribute, and the corresponding value matches the corresponding value range condition in the search expression. Select a thing to get as a search result ,
Before Symbol latest value search means,
A latest value table storing the tuple in association with the key attribute-value pairs;
Means for selecting a tuple that matches the search target included in the input search expression;
Means for extracting a value corresponding to the key attribute included in the input search expression from a tuple that matches the selected search target;
Based on the extracted value, for each tuple having the same key attribute-value pair, the tuple stored in the latest value table is updated by overwriting with a tuple having a newer time attribute value. ,
Tuples stored in said latest value table and the update is always the attribute for each attribute in the list that is part of the inputted said condition is included as an attribute in the tuples stored and that means for corresponding values obtained as a result also supports matching the value range condition to the selected and the search in the conditional expression
Information storage retrieval apparatus comprising: a. The storage means is the same as or different from the attribute sequence included in the other one tuple already stored in the attribute sequence included in the tuple to be stored,
The history search means performs a history search on the accumulated tuples,
The latest value search means uses the search target that designates a range in which the array of attributes included in each tuple in the tuple group included in the range as the search target including the search formula is the same or different between the tuples as a search result. The information storage / retrieval apparatus according to claim 3 , wherein: An information storage / retrieval program for causing a computer to function as each means included in the information storage / retrieval apparatus according to claim 3 or 4 .

Images