JP2994138B2 - Catalog Variable Management Method for Interactive Processing 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 method for managing catalog variables in an interactive processing system that takes various values in a catalog function for registering and repeatedly executing routine tasks. The present invention relates to a variable management method for easily performing and realizing variable management by reference-based management, and a program structure for realizing the variable management method.

[0002]

2. Description of the Related Art In a dialogue processing system, a catalog function is for registering a routine task to be repeatedly executed.

In general, this catalog function becomes flexible by being able to cope with various tasks by providing a variable function.

[0004] Usually, since the name of a variable (variable name) and its value are variable in length, a variable management area has a list structure in which the variables are associated with each other. At this time, two problems occur. One is the problem of "area use efficiency" of the areas that manage them, and the second is the problem of "variable search time", which is the time required to search for variables.

When managing variables that are variable length data,
The process of resetting (updating) certain data frequently occurs. At this time, there is no problem when the new data length is shorter than the data length set up to that time. However, when the data length exceeds the previously set data length,
The method used to increase the "area utilization efficiency" is to release the previous data area and allocate another contiguous area with a new data length, and then reallocate the released area later and use it. This is a method of allocating the shortage of the area up to the other area to another area and logically connecting one physically divided data.

As a method of reducing the “variable search time”,
There is a way to align variables according to certain rules. This is usually because it is difficult for the catalog function to predict which variable will be searched next at any given time,
This is to reduce the number of times of referring to irrelevant variable information at the time of retrieval by sorting.

[0007]

In the case of the "area re-allocation method", when re-allocating and using, the area length does not always match the new data length. Therefore, when such an operation is repeatedly performed, small unused areas may frequently occur.

[0008] In the case of the "data division method", the area appears to be used efficiently, but the data requires information for linking the divided data areas, and the management area is used. Will be used in vain.

In any case, it cannot be said that as a result, the "area use efficiency" is improved.

As a method of shortening the “variable search time”,
There is a way to pre-order variables according to certain rules. With this method, the search time can be reduced, but it is necessary to determine the insertion position when registering a variable in the management area. In this case, it takes time to register, and it cannot be said that this is a good method for reducing the search time.

[0011]

An interactive processing system according to the present invention.
A method of managing a catalog variable in a system includes a variable management area for managing a variable of a catalog related to a routine task to be repeatedly executed , and a means for searching a variable in the variable management area for a desired variable in response to a search request. in catalogs of variable management method in dialog processing apparatus, Name Plate length variable
Length point table consisting of unit addresses
And the name data indicated by the address and
A name index, which is information on the name data.
Name mark and setting indicated by the mark index
Value data and the set value
Index value, the empty area and the empty
And an empty area index, which is information on the
Using the used area, and using the name index
Manage the names in a name list, and use the free space index
Manages the unused area in an unused area list, and
Additional registration of a new mark with the mark length existing in
The new name mark at the top of the name list.
Registered and used as the name tag and setting value
When releasing an area, if the front is unused area, release it
The area is managed by the free space index in front, and
If the unused area is the front or the release area
Let the free space index manage the unused space behind
You.

[0012]

According to the catalog variable management method in the interactive processing system of the present invention, the unused area is managed in the same manner as the used area, and the length is managed based on the length of the name of the variable (variable name). .

The management of the unused area in the same manner as the used area means that the unused area is managed by an index having the same size as the used area, and each index is associated with the list structure in the same manner as the used area. is there.
Managing the unused area in this way prevents as much as possible the occurrence of small unused areas, which is a problem of the "area reassignment method" to improve "area utilization efficiency", and The problem of "area utilization efficiency" is solved by easily enabling the area reorganization processing in such a case.

The management based on the length of the name of a variable (variable name) solves the problem of "variable search time".

A length point table is prepared in order to perform management based on the standard length. This table is composed of address information in units of the standard length, and indicates one index of an arbitrary standard length group. This also indicates the insertion position at which a certain variable is registered in the management area. This means that when registering a variable, it is not particularly conscious of alignment by length, but as a result, the mark is aligned by its length. This shortens the time required for variable search and solves the problem of "variable search time".

[0016]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, the present invention will be described with reference to the drawings.

FIG. 1 is a conceptual diagram of variable management according to an embodiment of the present invention, FIG. 2 is a diagram showing an initial state of variable management, FIG. 3 is a diagram showing a state where a variable having a name standard 1 is registered, and FIG. FIG. 5 is a diagram showing a state in which a variable having an altitude of 3 has been registered. FIG.
FIG. 6 is a diagram showing a state in which a variable is registered, FIG. 6 is a diagram showing an unused area allocating unit, and FIG.
X) is a diagram showing an example of FIG. 8, FIG.
FIG. 9 is a diagram showing the release of the used area (a less preferable example), and FIG. 10 is a diagram showing the release of the used area (a good example).

In this embodiment, as shown in FIG.
It comprises a point table 1, an area management information table 8, and a variable management area 9.

FIG. 1 shows a state in which four names representing four variables and their values are managed as use areas, and an empty area is physically divided into two in the variable management area. It is.

A name mark is composed of name mark data 3 and a name index (SIDX) 2 as its information. The set value is set value data 5 and a set value index (VIDX) 4 as its information.
The unused area is composed of a free area 7 and a free area index (FIDX) 6 as its information. Also,
Name index (SIDX) 2 and free space index (FID
X) 6 are associated with each other by a list structure.

The length point table 1 is constituted by address information in units of a standard length, and indicates one index of an arbitrary standard length group.

The area management information table 8 has information such as sizes of used areas and unused areas, and head addresses of a name index (SIDX) 2 and a free area index (FIDX) 6.

In any case, any area of the variable management area 9 is always included in the index area or the data area indicated by the index area. “Variable registration processing”, “unused area allocating means”, and “used area releasing means” will be sequentially described.

First, the variable registration processing will be described.

The state of the variable management will be described with reference to FIGS.

FIG. 2 shows an initial state of variable management.

In the initial state, no variable is managed, and the variable management area 9 has only one free area index (FIDX) 6 which is information on free area. Since the variables are not managed, the address information of the length point table 1 is all null, and the address information indicating the head of the name index (SIDX) of the area management information table 8 is also null.

At this point, a variable having a name length of 1 is registered. FIG.
Indicates a state in which a variable having a standard length of 1 has been registered, and a bold frame in the figure indicates information on the registered variable.

When a variable is registered, a mark index (SIDX) 2, mark data 3, a set value index (VIDX) 4, and a set value data 5, which are information on the variable, are stored in a free space index (FIDX) 6 A free area 7 (unused area) to be managed is allocated. Since the size of the free area 7 is changed by “unused area allocating means” that performs this allocation process, the free area index (F
The information of (IDX) 6 is updated.

When the variable information is registered in the area, the address information of the corresponding name length in the length point table 1 is changed to the information of the registered name index (SIDX) 2 (). Further, since the variable registered here is the first one, the address information indicating the head of the name index (SIDX) in the area management information table 8 is also updated ().

Next, the variable of the name standard 3 is registered. FIG.
Indicates a state in which a variable having a standard length of 3 has been registered, and a bold frame in the figure indicates information on the registered variable.

As in the previous process of registering the variable of the standard length 1,
Allocate free space, length point table 1
The address information of the name standard 3 is changed (). Further, the name index (SIDX) 2 is associated with the name index (SIDX) of the previously registered variable of the name 1 to form a list structure of the name index (SIDX).

Further, a variable having a name length of 1 is registered. However, the name mark is different from the variable of the name length 1 registered for the first time. FIG. 5 shows a state in which a second variable having a standard length of 1 has been registered, and a bold frame in the figure indicates information on the registered variable.

As before, the information on the variables is registered by the “unused area allocating means”, and the address information of the mark length 1 of the length point table 1 is registered.
Change to the registered name index (SIDX) 2 address (). At this time, length point table 1
Changes the address information about the name index (SIDX) that has been shown so far (). Changing this address information means inserting it into the list structure. That is, the insertion position of the registered name index (SIDX) in the list structure is determined by the value of the length point table 1. With this process,
Each name index is managed on the basis of the name length.

The variable registration processing has been described above.
In the variable registration process, the insertion position in the list structure of the name index (SIDX) is determined based on the address information of the name index (SIDX) indicated by the length point table. In the variable registration process, it is not necessary to particularly perform the insertion position determination process, and it is possible to perform management based on the standard length. Further, as a result, the management based on the standard length can reduce the time for searching for an arbitrary variable.

Second, the use area allocating means will be described with reference to FIGS.

FIG. 6 shows an unused area allocating means.

The unused area allocating means is performed by searching a free area index (FIDX) for managing the free area.

When the unused area allocating means finds a free area index (FIDX) for managing the allocable size, it performs an area allocation process 10. In the area allocation processing 10, when the requested size (request length) is allocated, a new free area index is set in the remaining area.
If (FIDX) can be created, a request length is allocated and the remaining area is managed as a free area (step 11). If there is not enough space left to create a free area index (FIDX), it is assigned according to the management length (step 12).

For this reason, in the area allocation processing 10, there is not left an extremely small area that cannot be managed without creating an index.

At this time, the newly created free space index (FIDX) is added to the list structure.

FIG. 7 shows an example of a name index (SIDX) as an example of the index.

Due to the above-described allocating means, the name index (SIDX) 2 includes, in addition to the address information before and after the list structure, the area length allocated for the name mark and the actual name. It contains information on the effective length, which is the length of the target.

The means for allocating the unused area has been described above. The unused area allocating unit prevents a state in which the unused area cannot be managed by the index. Therefore, any area of the variable management area is always included in the index area or the data area indicated by the index area in any case.

Third, the use area releasing means will be described with reference to FIGS.

FIG. 8 shows the means for releasing the used area.

Release of the used area is completed when the area is managed by the free area index (FIDX).

An easy method is to create a free area index (FIDX) at the head of the area to be released. In this case, if there is an unused area in front of or behind the area to be released as shown in the release of the used area (a less preferable example) in FIG. 9, a useless free area index (FIDX) is generated. Become. This not only causes waste in index list processing, but also reduces the efficiency of area allocation processing in the unused area allocation means.

Therefore, the use area releasing means is constituted by the front area confirmation 13 and the rear area confirmation 14, like the use area release means of FIG. In the forward area confirmation 13, when the area ahead of the area to be released is an unused area, the released area is managed together with the forward free area index (FIDX). Conversely, if the front is the used area, a new free area index (FIDX) may be created in the free area.

Since the size of the name index (SIDX) is equal to the size of the free area index (FIDX), when the area previously used as the name index (SIDX) is released, the free area is stored in the free area. Index (FIDX)
There is no such thing as not being able to create.

In the rear area check 14, when the rear area is an unused area, the free area index (FIDX) of the front or free area determined in the front area check 13 is used.
To manage the rear unused area together.

By using such a use area releasing means, the state shown in FIG. 9 does not exist, and a state similar to the use area release (good example) shown in FIG. 10 is obtained. This means that the Free Space Index (FIDX) manages unused space.
Is to always maintain the optimal free area management state, and reduce the list processing for maintaining the index list structure when the used area is released.

The means for releasing the used area has been described above. In the used area releasing means, the optimum free area management is always maintained by checking the states of the front and rear areas. This also reduces index list processing.

[0054]

As described above, according to the present invention,
The following effects can be obtained. * Since the management based on the length of the mark can be performed by a simple program structure without considering the alignment, the time for searching for any variable can be reduced. * The unused area is managed in the same way as the used area, and the arbitrary area is always managed by either of them, so that a program structure that facilitates the area reorganization processing is enabled. * Since the optimal free area management state is always maintained by checking the front and rear areas in the area release processing for area reuse, the efficiency of the area allocation processing is not reduced, and there is no useless index.

[Brief description of the drawings]

FIG. 1 is a conceptual diagram illustrating variable management according to an embodiment of the present invention.

FIG. 2 is a diagram showing an initial state of variable management.

FIG. 3 is a diagram showing a state where a variable having a name length of 1 is registered.

FIG. 4 is a diagram showing a state where a variable having a name length of 3 is registered.

FIG. 5 is a diagram showing a state where a second variable having a standard length of 1 is registered.

FIG. 6 is a diagram showing an unused area allocating unit.

FIG. 7 is a diagram illustrating an example of a name index (SIDX).

FIG. 8 is a diagram illustrating a use area releasing unit.

FIG. 9 is a diagram illustrating release of a use area (a less preferable example).

FIG. 10 is a diagram showing release of a used area (a good example).

[Explanation of symbols]

 1 length point table 2 target index (SIDX) 3 target data 4 set value index (VIDX) 5 set value data 6 free area index (FIDX) 7 free area 8 area management information table 9 variable management area 10 area allocation Processing 11 Assignment of request length 12 Assignment of management length 13 Confirmation of front area 14 Confirmation of rear area

Claims (1)

(57) [Claims] 1. An interactive processing apparatus comprising: a variable management area for managing a variable of a catalog related to a repetitively executed routine task; and means for searching a variable in the variable management area for a desired variable in response to a search request. in catalogs of variable management method in the configuration of the address of the Name Plate length unit of the variable
Length point table and the address
The information of the name data and the name data
A name mark composed of a name mark index and the name mark
Set value data indicated by the index and the setting
A setting consisting of a set value index,
The fixed value, the free area, and the free space information
Use the unused area consisting of the area index and
Manage the above-mentioned names in a name list with the name index
And the unused area is identified by the free area index.
Manages in the unused area list and already exists
When registering a new name mark with
Register the new name mark at the top of the
To release the area used as the
If the front area is unused, the free area is set to
Area index, and if the back is an unused area,
Forward or to the free space index of the free space
A method of managing a catalog variable in an interactive processing system, wherein a rear unused area is managed .