JPH0229861A - Data access subroutine - Google Patents

Abstract

PURPOSE:To reduce the capacity of memory and the manhour of tests by executing data access including the updating of the index values of tables in respective stages at the time of retrieving data as a subroutine. CONSTITUTION:At the time of accessing a subroutine from a main routine, the index value of tables in respective stages to know the data storing position of the final stage table for starting data retrieval and a data retrieving condition are specified by a retrieval starting means 12. Effective data are searched in a multistage table with tree structure by a data detecting means 13 on the basis of said specification. When the index value of the final stage table including the initial data retrieving position exceeds the maximum value, the operation is returned to the 3rd stage table and the 4th stage table is searched. Thus, effective data are searched while updating the indexes of tables in respective stages. After detecting the effective data, the operation is returned to the main routine and objective processing is executed by using the detected data. Consequently, the capacity of the memory and the manhour of tests can be reduced.

Description

[Detailed Description of the Invention] [Summary] Regarding a data access subroutine for performing sequential access to tree-structured multi-level table data, the data access subroutine hides the data structure by including index update during data retrieval in the subroutine. The purpose is to provide data in the final stage table, and the start address of the next stage table is stored in the tables from the first stage to the stage before the final stage, and the address or data in each table from the first stage to the final stage is stored. In a sequential access subroutine for tree-structured multi-stage table data in which the storage location is specified by an index value indicating a relative address from the start address of each table, the last stage is A search start means for specifying an index value of each stage table indicating a data search start position on the table, a search condition indicating validity as search target data, and detecting valid data matching the search condition. data detecting means for searching the tree structure multi-stage table from the data search start position while updating the index value of each stage table; a caller return means for returning to the caller of the main routine while retaining the top address and index value of the table; and upon re-calling from the main routine;
The data is detected from the data stored in the area of the index value next to the index value of the last stage table at the time of return, using the start address and index value of the each stage table held by the caller return means. The apparatus is configured to include search restart means for restarting the detection of valid data matching the search conditions.

[Industrial application field]

The present invention relates to a data access method in a computer system, and more particularly to a data access subroutine for sequentially accessing tree-structured multi-stage table data.

Among the computer systems used in various application fields, there are systems that have data in a tree-structured multi-table format. An example of this is a route definition table for routing processing in common line signaling telephone exchange processing.

FIG. 7 shows an example of the general format of a tree-structured multi-stage table.

As shown in the figure, the number of tables generally increases as the number of stages increases, such as from the first stage to the second stage, from the second stage to the third stage, and so on. All data for the system is stored in the final table, and the tables from the first stage to the pre-final stage store the start address of the next stage table. However, when "0" is stored, it indicates that the next table does not exist, that is, there is no data. Also, the storage position of the data or the next stage start address in each table starts from the first stage address of each table. specified by an index indicating the relative address of

When accessing the data stored in the final stage of the tree-structured multi-stage table data as described above, give the index for each table and write the start address from the first stage to the second stage, from the second stage to the third stage, etc. The data at the corresponding position in the final stage table is accessed using the index and the index. When accessing this, even if the data is in the final stage, it is checked whether the data is valid as the data to be accessed, or if the first stage table is accessed, it is always checked whether the next stage table exists. There is a need.

[Conventional technology]

A conventional example of the tree structure multi-stage table data access method as described above will be explained with reference to FIGS. 8 and 9.

In the access method of FIG. 8, the index of each stage table is given in exactly the same way as the explanation of FIG. 7, and for example, the last stage,
Here, the data at the position of index "io" of the four-stage table is obtained.Next, when obtaining the data at the position of index "i+1°" of the same last-stage table, the data given from the first position of the first-stage table is again used. According to the instructions above, I was accessing via almost the same path as last time.

FIG. 9 shows an algorithm for an access method in which an index indicating the starting position of data access is given and valid data is sequentially obtained one after another from that position. First, in step 1, an index indicating the starting position is set,
The first stage table is searched in step 2, the second stage table is searched in step 3, and the final stage table is searched in step 5.
In step 6, one piece of valid data in the final stage table is acquired, and processing using that data is performed within the system. Next, in step 7, the index of the final stage table is incremented, and until the index of the final stage table reaches the maximum value in step 5, valid data acquisition and processing using it in step 6 are performed sequentially.

When the index reaches the maximum value in step 5, step 8
Then, the index of the three-stage table is updated, and the process returns to step 4, where data access is again performed to the other last-stage table, here the four-stage table.

Similarly, if the index of the 3-level or 2-level table exceeds the maximum value in step 4 or step 3, step 9
, or step IO processing is performed, and the table search is performed by returning to the previous table.

[Problem to be solved by the invention]

Of the conventional access method examples shown in FIGS. 8 and 9 above, in FIG.
Even when accessing +1゛ data, access is performed from the first stage on the same bus while determining the presence or absence of the next stage table and whether the data is valid or invalid, as in the previous stage, which is inefficient and takes time. There was a problem. Particularly when high speed is required or when accessing a large amount of data, the negative effect on the operation of the entire system has been a problem.

Furthermore, in the sequential access method shown in Figure 9,
Since the index update process is performed by the program outside the data acquisition process, there is a problem in that when the data specifications are changed, the entire program needs to be modified in accordance with the change in the specifications. Also, if such programs exist in multiple locations in the system,
There were also problems such as an increase in the amount of memory, an increase in testing man-hours, and a decrease in reliability.

An object of the present invention is to provide a data access subroutine in which the data structure is hidden by including index updating during data retrieval in the subroutine.

[Means to solve the problem]

FIG. 1 shows a block diagram of the principle of the present invention. In the same figure, search start means 12 is used to search for search start position data in the last stage table when this subroutine is called for the first time from the main routine for the purpose of searching for data necessary for a specific process in the system. The index value of each stage table and the search condition for valid data are specified, for example, by the identification symbol of the data.

The data detection means 13 searches for data at a search start position using the index value of each stage table specified by the search start means 12, and updates the index value of each stage table from the data at that position to meet the search condition. The tree-structured multi-stage table is searched until matching valid data is detected.

When the data detection means 13 detects valid data on the final stage table, the caller return means 14 retains the start address and index value of each stage table at that time in preparation for the next call. Performs processing to return to the calling source of the main routine.

When this subroutine is called again from the main routine for the above-mentioned specific processing, the search restart means 15 uses the start address and index value of each stage table held by the caller return means 14 to restart the search. The data detecting means 13 is caused to detect valid data that matches the search condition, targeting data from the position next to the valid data detection position at the time.

(Function) When it becomes necessary to search for data used in a specific process within the combinator system, the subroutine of the present invention is called from the main routine. The index value of each stage table and the data search condition are search start means 1.
2.

According to this designation, the data detection means 13 searches for valid data in the tree-structured multi-stage table. For example, the final stage table has multiple stages as shown in Figure 7, so when the index value of the final stage table including the first data search position exceeds the maximum value, the table returns to the 3-stage table in Figure 7, and the next 4-stage table Valid data is searched for while updating the index of each stage table in a similar manner to searching for a full table.

After valid data is detected, the process returns to the main routine and the desired processing is performed using the detected data. This subroutine is called again when sequentially the next data is needed for the desired processing. Then, a valid data search is performed in the final stage tabulation from the position next to the previous valid data detection position.

As described above, in the present invention, data access is performed in a manner in which updating of index values of each stage table at the time of data retrieval is included in a subroutine.

〔Example〕

FIG. 2 shows an example of data access in a tree-structured multi-stage table using the data access subroutine of the present invention. In the present invention, there are two data access functions: locate entry and netast entry. Locate entry is a function used at the time of first data access, and the subroutine of the present invention is called manually using the index of each stage table indicating the starting position.

In the locate entry, data at the access start position of the final stage table 21 is retrieved via tables 18, 19, and 20 using the index of each stage table, for example, data
is accessed. Then, starting from this data and updating the index, data that matches the desired processing,
That is, valid data is searched. If no valid data is found even after searching the final stage table 21 that includes data at the start position, the process returns to the previous stage table 20 of the final stage and searches other final stage tables 22 in the same way as in FIG. However, in FIG. 2, for the sake of simplicity, it is assumed that the data (2) at the starting position itself is valid. In this case, the subroutine is stored in the common data area shown in Figure 3.
As information for accessing the data, the current number of stages (stage) 16, in this case 4 stages, is stored, as well as the start address 17 of each stage table and an index 171 indicating the relative address within each stage table. to end the process.

When the access path information is stored in the common data area by the locate entry, the index of the final stage table 21 is updated at the next access, and the final stage table 21 is searched from the position next to the data ■. Accessed as valid data. This access is called a next entry as an access function after access information has already been stored in the common data area, and after accessing data ■, the final stage table index of the common data area indicates the position ■.

Furthermore, at the next access, a table search is performed from the next position of data ■, but if there is no valid data in the same table 21, it returns to the 3-tier table 20, updates the index of the 3-tier table 20, and searches for other data. final stage table 2
Search for 2 and access data ■.

In the same way, after the data ■ is accessed, it returns to the three-stage table 20, and when the index value of the three-stage table 20 exceeds the maximum value, it returns to the two-stage table 19, passes through another three-stage table 23, and then returns to the three-stage table 20. Data (2) in the stage table 24 is accessed. Similarly, sequential access to data ■ and ■ is executed.

In this way, in the access subroutine of the present invention, at the time of the first data access, the access path is set in order from the first stage table as a locate entry, but at the time of the second and subsequent accesses, the previous valid data storage position is set as a netast entry. Data search is performed from next.

As an example of tree-structured multi-stage table data as shown in FIG. 7, FIG. 4 shows an example of a route definition table for routing processing in exchange processing of the common signal line system. In the same figure, the index of the first stage table 26 is only “0”, that is, 1
There is only one corresponding two-stage table 27.

The two-stage table 22 has 32 indexes from 0' to 31', and each area stores the start address of the corresponding three-stage table. However, if the corresponding three-stage table does not exist, or in other words, there is no data in the final stage table, the address value is "0°.The index of the second day of the three-stage table is from 0" to "1".
The index of the 4-tier table 29 with 16 items up to 5' is “
There are 256 items from Oo to "255°. The 4-stage table 29 is the final stage table, and each area of this table contains route definition data and an identification symbol indicating whether the data matches a certain search condition. (ID) is stored.

A data access method using the data access subroutine of the present invention will be explained with reference to the flowchart of FIG. 5, taking the tree-structured multi-stage table of FIG. 4 as an example. In the figure, first, in step 30, a data access request code, that is, a determination is made as to whether it is a locate entry or a netast entry. At the first data access, it is determined to be a locate entry, and in step 31, the stage 16 of the common data area in FIG. It is said that

In the locate entry, an index within each stage table is given for searching for data at the start position of data retrieval. Therefore, when the number of table stages is first determined in step 32, it is determined to be one stage, and then, in step 33, it is determined whether or not there is a next stage table.

If it is determined that there is a next table, 2
The start address of the stage table is set, the stage 16 of the common data area is updated to 2', and the process returns to step 32.

In step 32, the number of table stages is determined to be 2. Next, in step 35, it is checked whether the given index exceeds the maximum value of the two-stage table, which is 31 degrees in FIG. Therefore, in step 36, the two-stage table is searched using the given index, and in step 31, it is determined whether the first address of the next-stage table is stored at that position. If there is, step 38
The start address is set, and stage 16 is set to 3.
Then, the process returns to step 32.

If the number of table stages is determined to be "3" in step 32, the given index is less than the maximum value, and there is a 4-stage table, then steps 39 to 42 are repeated to steps 35 to 38 for the 2-stage table. The processing is performed in the same manner as in step 32, and the process returns to step 32.

For the final four-stage table, it is checked in step 43 whether the index exceeds the maximum value "255'," and then in step 44 data at the starting position is searched using the given index.Step In step 45, it is determined by the identification symbol whether the data matches the given search condition, and if it does, in step 46, the return code to the main routine is set to, for example, "0" (data present). The address of the requested data is output, and the subroutine processing ends.

In step 45, if the data at the search start position on the 4-level table does not meet the search conditions, the process moves to step 47, where the index of the 4-level table is incremented in order to access the data for the next index of the start position. , return to step 32. After step 32, steps 43 to 47 are performed on the data in the 4-level table again to find the data that matches the data search condition within the range where the index does not exceed the maximum value "255°" (determined in step 43). The process continues until one is found (step 45).

If it is determined in step 43 that the index exceeds the maximum value before data matching the search condition is found in the four-level table, the process moves to step 48. In step 48, returning to the three-stage table, stage 16 of the common data area is set to °3° in order to check the address stored in the area of the next index value on the three-stage table, and the index value of the three-stage table is set to 3°. is incremented. Further, the index of the four-stage table is set to "0" in order to perform data retrieval from the head position on the next four-stage table, and the process returns to step 32.

At step 32, it is determined that the number of tables is "3", and when it is determined that the index incremented at step 48 on the three-stage table does not exceed the maximum value (step 39), the data at that position is determined at step 40. A search is performed, and in step 41 it is determined whether the first address of the next stage table is stored. If it is determined that there is a four-stage table, the start address of the four-stage table is set and the stage 16 is advanced in step 42, and the process returns to step 32.

Here, the 4-level table for which the start address is set is:
This differs from the aforementioned four-stage table including search start position data, and corresponds to, for example, the four-stage table 22 in FIG. Then, via step 32, a data search is performed on the new four-stage table from the beginning position by the processing from step 43 onwards.

In the determination process of step 41 for the 3-stage table, if the start address of the next-stage table is not stored at the index position incremented in step 48, the 3-stage table is
In order to check the stored contents at the next index position on the tiered table, the index of the 3-tiered table is updated in step 49, and the process returns to step 32. Thereafter, the processes of steps 39, 40, 41, and 49 are continued for the three-stage table until the next stage table is detected (step 41) as long as the index does not exceed the maximum value "15". .

If it is determined in step 39 that the index exceeds the maximum value before the four-stage table is detected, the process moves to step 50. In step 50, similarly to step 48, the stage 16 of the common data area is set to "2", the index of the two-stage table is incremented, and the index of the three-stage table is further initialized, and the process returns to step 32.

In step 32, the number of tables is determined to be "2", so it is checked in step 35 whether the index of the two-stage table incremented in step 50 does not exceed the maximum value "31°".
If it is not exceeded, the two-stage table is searched in step 36, and it is determined in step 37 whether or not the address of the three-stage table exists in the area of the index incremented in step 50. If it is determined that there is a table, the processing in steps 38 and 32 is performed, and step 39 is executed for the other three-stage table, table 23 in FIG.
Subsequent processing is performed.

If it is determined in step 37 that there is no next-stage table, the index of the two-stage table is incremented in step 51, and the process returns to step 32. Steps 35 and 3 continue through step 32 until the next table is detected (step 37) within the range where the index of the two-stage table does not exceed the maximum value.
Processes 6, 31, and 51 are repeated.

If it is determined in step 35 that the index has exceeded the maximum value before the detection of the three-level table, there is only one two-level table, so there is no search target data, and the return code is returned in step 52. “Fo is set, the process as a subroutine ends, and the process returns to the main routine.Also, if it is determined in step 33 that there is no two-stage table for the first-stage table, the same process is performed in step 52. After that, the routine returns to the caller of the main routine.

Here, as shown in FIG. 4, the case where there is only one index in the first stage table 26, that is, there is only one two-stage table, is taken as an example, but if there are multiple two-stage tables, the fifth After it is determined in step 35 of the figure that the index exceeds the maximum value, processing similar to step 48 or 50 is performed. That is, stage 16
Set to '1°, increment the index of the first table,
The index of the two-stage table is initialized and the two-stage table start address in the first stage table is detected.

When valid data in the four-level table is accessed by the method described in detail above and the subroutine is called again after returning to the main routine, the request code is the next entry, and steps 30 to 53 in FIG. , and the final table, here 4
The index of the stage table is incremented. Then, data access is performed on the four-stage table in the same manner as described above for data subsequent to the index next to the previous effective data detection position. .

FIG. 6 shows a processing flowchart on the system side which calls the data access subroutine detailed in FIG. 5 and performs a specific process using the obtained data. The specific process performed here includes, for example, the process of developing a route assignment table from the route definition table for the routing process described above.

In FIG. 6, when it becomes necessary to call this access subroutine for a purpose processing on the main routine side, first, in step 54, the above-mentioned data search condition is inputted as input information.
The index of each stage table is set, and an access subroutine is called at step 55. When the processing in the subroutine is finished and the main routine returns, the presence or absence of data is determined in step 56 based on the return code described above.
If there is no data, the target processing is aborted.

If there is data, the desired processing is performed using the data in step 57. After that, the subroutine is called again in step 55, and step 5 continues until there is no more data.
7, the desired processing is performed.

〔Effect of the invention〕

As explained above, according to the present invention, by converting data access into a subroutine, including updating the index value of each stage table when searching for data, the amount of memory and testing man-hours are reduced, and the reliability of the system is improved. can be done. Furthermore, it becomes possible to easily deal with changes in data structure.

[Brief explanation of the drawing]

Figure 1 is a block diagram of the principle of the present invention, Figure 2 is a diagram showing an example of data access in a tree-structured multi-stage table, Figure 3 is a diagram showing a common data area, and Figure 4 is an example of a route definition table. Figure 5 is a flowchart of data access, Figure 6 is a flowchart of processing on the system side, and Figure 7 is a diagram showing an example of the general format of a tree-structured multi-stage table. Figures 8 and 9 are tree-structured multi-stage table data access. FIG. 2 is a diagram showing a conventional example of the method. 12... Search start means, 13... Data detection means, 14... Caller return means, 15... Search restart means, 18.26... First stage table, 19.27... Second stage Table, 20.23.28...3-tier table, 21.22,2
4,25.29... 4th stage (last stage) table,

Claims (1)

[Claims] Data is stored in the final stage table, and the start address of the next stage table is stored in the tables from the first stage to the stage before the final stage, and the address or data storage position in each table from the first stage to the final stage. is a sequential access subroutine for tree-structured multi-level table data specified by an index value indicating a relative address from the first address of each table, and when the first call is made from the main routine for the purpose of data retrieval, a search start means (12) for specifying an index value of each stage table indicating a data search start position and a search condition indicating validity as search target data; and detecting valid data matching the search condition. data detection means (13) for searching the tree-structured multi-stage table from the data search start position while updating the index values of each stage table; and when the data detection means (13) detects the valid data; a caller return means (14) for returning to the caller of the main routine while retaining the top address and index value of each stage table at the time; Using the start address and index value of each stage table held by the means (14),
Search restart means (13) for causing the data detection means (13) to restart detection of valid data matching the search condition from the data stored in the area of the index value next to the index value of the last stage table at the time of return; 15) A data access subroutine comprising: