JPH08263264A - Data processor - Google Patents

Abstract

PURPOSE: To efficiently store numerical data as variable length data and to convert a decimal operation into a BCD code so as to operate it. CONSTITUTION: When numerical data is inputted, a decimal character discrimination part 13 discriminates whether a JIS code is a decimal character or not. When it is the decimal character, numerical data which is inputted next is inputted from a JIS code conversion part 12 and it is discriminated to be the decimal character or not. When it is the decimal character, the decimal character of two bytes is converted into a BCD of one byte in a BCD generation part 14. A +1BCD generation part 15 adds '01' (hexadecimal value) and generates a +1BCD. A +1BCD output part 16 adds a delimeter to a +1BCD and writes it into a memory. At the time of executing the decimal operation, a +1BCD conversion part subtracts '01' (hexadecimal value) from the data and converts it into a BCD. A BCD output part 20 outputs the BCD to an arithmetic unit.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data processing device in an electronic computer.

[0002]

2. Description of the Related Art In office processing by a computer, numerical calculation is generally performed by a decimal operation although it is disadvantageous in that it is faster than binary operation.

Decimal operation is of course operation between decimal data, and if it is, for example, addition operation, both the addition number and augend are decimal data, and they are added.

The representation format of decimal data is, for example, B
There is a pack format in which a numerical value of one decimal digit from "0" to "9" is represented by 4 bits by a CD code and two decimal digits are stored in one byte.

FIG. 6A shows a BCD pack format which has been conventionally used. In the same figure (a), one word represents a 32-bit packed decimal number, and the decimal point is the least significant byte B.
Place it in the right P of the upper 4 bits of 1, and place the least significant byte B
The code is represented by the least significant bit of the lower 4 bits S of 1. For example, positive when the least significant bit is "0",
When it is "1", it is a negative decimal number. Also, from the upper 4 bits dl of the least significant byte B1 to the upper 4 of the most significant byte B4.
A 4-bit binary number is stored in bit d7, and a fixed length of 32 bits is "d7, d6, d5, ... d3, d2, d1".
Represents a 7-digit decimal number. A specific example of a packed decimal number is shown in FIG. FIG. 10B is a BCD pack format that represents a five-digit negative decimal number "198003", and the least significant bit of the least significant bit S of the least significant byte B1 is "1". Also, since it is a 5-digit decimal number, the most significant byte B4 is "00" (hexadecimal),
Binary codes "1001" (d5), "1000" in the upper 4 bits of the bytes B3, B2 and the least significant byte B1
(D4), "0000" (d3), "0000" (d
2) and "0010" (dl) are stored.

[0006]

Conventionally, a decimal number (BCD) is represented by a fixed length as shown in FIG. 6 (a). Therefore, when a fixed-length 32-bit 7-digit decimal number is used, if a decimal number of 6 digits or less is stored, an unnecessary digit (“0000”) must also be stored, and the memory utilization effect is poor. For example, in the case of FIG. 6B, the most significant byte B4 is "0000000" which is meaningless as data.
Since it is 0 ", it is unnecessary. Therefore, if a decimal number is expressed in a variable length, unnecessary data is not included, so that the decimal number data can be compressed and stored, and the memory utilization effect is improved. Conceivable.

However, in operating systems such as UNIX (C language system), NUL is added to the end of the character string.
There is a promise to add an L code ("00000000") to indicate the end of the string. In UNIX, various functions are provided as standard in accordance with the above promise to configure the system. Therefore, if a code other than the NULL code is used as a delimiter code, various inconveniences occur and it is wasteful. Therefore, when a decimal number is expressed in variable length,
Decimal data 00 (“00000000”) is NU
I regard it as LL. For example, if the fixed length 4-byte "-98003" shown in FIG. 6B is expressed as the variable length 3-byte "-98003" as shown in FIG. 6C, the byte B2 becomes ("00000000"). ) And the operating system considers byte B2 to be NULL. Also, codes other than the NULL code are already defined as character codes, numerical codes, and symbol codes, and therefore cannot be used as delimiter codes.

Therefore, in the conventional data processing device, BC
There is an inconvenience that the numerical data by the D code cannot be processed with a variable length.

An object of the present invention is to convert numerical data represented by BCD code into a code which avoids a delimiter code of an operating system and efficiently store it as variable length data in a memory when performing a decimal operation. Is to convert the code read from the memory into the original BCD code so that it can be operated.

[0010]

The data processing apparatus of the present invention is a pack type B which indicates a 2-digit numerical value in 1 byte.
Code conversion means for adding a specific numerical value to each byte of the numerical data represented by the CD code and converting the BCD code into a code different from the delimiter code used in the operating system, and the code conversion means. A means for storing the code in the memory, and subtracting the specific numerical value added by the code converting means from each byte of the code read out from the memory when performing the decimal operation to convert into a BCD code And means for outputting to the arithmetic means.

Therefore, the numerical data represented by the BCD code can be converted into a code avoiding the delimiter code of the operating system and stored in the memory as the variable length data, and read out from the memory when the decimal operation is performed. The code can be converted into the original BCD code and operated.

[0012]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram of an embodiment of the data processing apparatus of the present invention. In the figure, a data input unit 11 is an input device such as a keyboard, and the key input data input from the data input unit 11 is handled by a JIS code conversion unit 12 such as a keyboard driver or the like. It is converted into a code and sent to the decimal character discrimination unit 13. The decimal character discriminating unit 13 discriminates whether or not the input JIS code is a decimal character (hexadecimal, a value of "30" to "39"), and if it is a decimal character, the JIS code is converted into a BCD generating unit. It outputs to 14. The BCD generation unit 14 converts the input 2-byte JIS code into a 1-byte BCD and outputs it to the + 1BCD generation unit 15. The + 1BCD generation unit 15 adds “01” (16
) Is added and stored sequentially. Hereinafter, the + 1BCD generation unit 1
The data generated by 5 will be referred to as + 1BCD.

When the decimal character discriminating unit 13 inputs a JIS code other than the decimal character, the + 1BCD output unit 16
Reads + 1BCD from the + 1BCD generator 15,
A delimiter indicating the end of the variable length data is added to + 1BCD and written in a predetermined area of a memory (not shown).

On the other hand, + 1B from an external device (not shown)
When a CD-to-BCD conversion command is applied to the data processing device, the + 1BCD input unit 17 reads the variable-length + 1BCD stored in the memory in byte units and the delimiter determination unit 18
Output to. The delimiter discrimination unit 18 is input + 1BC
If it is determined whether D is a delimiter and + 1BCD read from the memory by the + 1BCD input unit 17 is not a delimiter by the delimiter determination unit 18, BC
The D conversion unit 19 subtracts 1 from the + 1BCD and converts it into BCD. The above operation is performed by the delimiter discrimination unit 18 by + 1B.
The process is repeated until it is determined that + 1BCD read by the CD input unit 17 is equal to the delimiter. When the delimiter that is the delimiter code of the variable length + 1BCD is detected by the delimiter determination unit 18, that is, when all the bytes of the variable length + 1BCD are converted to the BCD by the BCD conversion unit 19, the BCD output unit 20 performs the BCD conversion. The BCD converted by the unit 19 is output to an arithmetic unit (not shown).

The decimal character discriminator 13, the BCD generator 14, and the + 1BCD generator 1 surrounded by the one-dot chain line in FIG.
The 5, + 1BCD output unit 16, + 1BCD input unit 17, delimiter determination unit 18, BCD conversion unit 19, and BCD output unit 20 can be configured by a microcomputer.

Next, a method of storing decimal data in the data processor of this embodiment will be described with reference to FIGS. 2 (a) to 2 (c).

First, the data input unit 11 causes "9810".
When the 2 "decimal data is keyed in, the JIS code conversion unit 12 converts it into the 5-byte" 98102 "JIS code of B1 to B5 shown in FIG. 2 (a). , It is actually "39,3
The JIS code of "8, 31, 30, 32" is stored.
Next, the BCD generation unit 14 uses the 5-byte J shown in FIG.
The IS code is converted into the 3-byte BCD “09, 81, 02” shown by B1 to B3 in FIG. 3-byte “09, 81, 02” generated by the BCD generation unit 14
In contrast, the + 1BCD generator 15 adds "01" (hexadecimal), and the three bytes "0" of B1 to B3 shown in FIG.
A, 82, 03 ”+ 1BCD. Then, a delimiter D indicating a data delimiter code is added to the end of the generated variable length + 1BCD “0A, 82, 03” and stored in the memory.

Next, the operation of the data processing apparatus configured as described above will be described with reference to the flowchart of FIG.

For example, in the spreadsheet processing, when the operator inputs decimal value data through the data input unit 11 into the value item column displayed on the CRT, the data processing device operates as follows.

First, when decimal value data is input by the data input section 11, the JIS code conversion section 12 converts the input decimal value data into a JIS code.
The decimal character determining unit 13 inputs one character of the first JIS code of the decimal number data from the JIS code converting unit 12 (process ST1), and the JIS code changes the decimal character (from “30” to “39”). It is determined whether or not it is a hexadecimal value up to "" (process ST2). The decimal character discriminating unit 13 determines whether the numerical data input next by the data input unit 11 is a JIS code converting unit 12 or not.
Is input (process ST3) and it is determined whether or not it is a decimal character (process ST4). If it is determined that it is a decimal character, the 2-byte decimal character input in the processes ST1 and ST3 is converted into a 1-byte B character by the BCD generation unit 14.
Convert to CD (process ST5), and then + 1BCD generation unit 1
In step 5, the BCD converted by the BCD generator 14 is set to “0.
1 "(hexadecimal value) is added (process ST6).

The processes ST1 to ST6 are the same as the process ST.
2 or the process ST4 is repeated until the decimal character discriminating unit 13 discriminates a JIS code other than a decimal character.

Therefore, for example, when decimal data of even digits of "4321" is inputted by the data input unit 11, the numerical data is "0100,0100,0010,0010" (442) by processing ST1 to processing ST6.
It is converted into 2 bytes + 1BCD of 2).

On the other hand, when a character other than a decimal character (for example, a NULL code indicating the end of data) is discriminated by the decimal character discriminating unit 13 in the process ST4, the decimal number input from the data input unit 11 becomes Since it is an odd digit, B
The CD generator 14 displays the character "0" [JIS code 30
(Hexadecimal)] (process ST7), the decimal character last input in process ST1 is the lower 4 bits of the byte, and the decimal character “0” input in process ST7 is the upper 4 bits of the byte. BCD to be generated is generated (process ST8). Next, the + 1BCD generation unit 15 adds "01" (16
The + 1BCD output unit 16 generates the variable length + 1B generated by the + 1BCD generation unit 15.
A 1-byte delimiter is added to the CD and written in a predetermined area of the memory (process ST10).

Therefore, when the data input unit 11 inputs, for example, an odd-numbered decimal number "123", the above-mentioned processes ST1 to ST4 and ST7 to ST1 by the data processing device.
0 Nyori “00000010,00100100” (2
24) is converted into 2 bytes + 1BCD.

When the decimal character discriminating unit 13 discriminates a character other than a decimal character, for example, a NULL code in the process ST2, the + 1BCD output unit 16 adds a delimiter to the + 1BCD generated by the + 1BCD generating unit 15. Writing to a predetermined area of the memory (process ST10).

Therefore, for example, when an even-digit decimal number such as "2222" is input from the data input unit 11, the data processing device outputs "00100011,00".
Converted to 2 bytes + 1BCD of 1000011 "(2323).

Next, the operation of converting + 1BCD to BCD when the data processing device performs a decimal operation such as spreadsheet calculation is shown in FIG.
This will be described with reference to the flowchart of FIG.

First, the + 1BCD input unit 17 reads the first 1-byte data of + 1BCD from a predetermined area of a memory (not shown) (process SU1). Next, the delimiter discrimination unit 18 discriminates whether or not the read 1-byte data is a delimiter (process SU2), and if it is not the delimiter, the BCD conversion unit 19 extracts "01" from the data.
Subtract (hexadecimal) (processing SU3). The above processing SU1
~ SU3 is repeated until the delimiter is determined by the delimiter determination unit 18 in the process SU2.

Therefore, for example, "00110011,
The + 1BCD of “00110011” (3333) is “00110010,00110” by the processes SU1 to SU3.
It is converted into a BCD of 010 ″ (3232).

On the other hand, in the processing SU2, the delimiter discriminating unit 1
When 8 determines the delimiter, the BCD output unit 20 reads the BCD converted by the BCD conversion unit 19 and outputs it to a computing device (not shown) or the like.

As described above, according to this embodiment, the decimal number input by the operator from the keyboard is a variable length + 1B.
It can be changed to a CD and stored efficiently in memory.
In the case of decimal arithmetic, + 1BCD may be read from the memory, converted into BCD, and then arithmetic operation may be performed.

The decimal number of the BCD code is "00", "F" which is the delimiter code of the operating system.
In order to convert the code to avoid D, "FE", "FF" (above hexadecimal value), in addition to adding 1 as in the present embodiment, "02", "11" ( The above hexadecimal value) or the like may be added.

Further, in this embodiment, + 1BC without an exponent part is used.
Although it is converted to D, it is also possible to convert it to a variable-length + 1BCD having a 1-byte SE consisting of a code and an exponent as shown in FIG. In the figure, dl to dn are +1
BCD and D are delimiters. The sign SE and the byte SE of the exponent part are, for example, as shown in the enlarged view of FIG.
Bits bl to b5 represent the exponent value, the sixth bit Se represents the sign (positive or negative) of the exponent, and the seventh bit Sn is a variable length +.
1BCD code is shown. For example, “2 shown in FIG.
4,46,68,8A, 01 "is 234567789 ×
10 ⁰ = represents the decimal 23456789.

[0034]

As described above in detail, according to the present invention, the numerical data represented by the BCD code is converted into a code which avoids the delimiter code of the operating system, such as the NULL code (hexadecimal "00"). The converted data can be efficiently stored in the memory as variable length data, and the code read from the memory is used as the original BC when performing decimal operation.
It can be converted to D code and operated.

[Brief description of drawings]

FIG. 1 is a block diagram of an embodiment of a data processing device according to the present invention.

2A to 2C are diagrams illustrating a data processing method according to the present embodiment.

FIG. 3 is a diagram for explaining an operation of + 1BCD conversion according to the present embodiment.

FIG. 4 is a diagram illustrating an operation of conversion from + 1BCD to BCD according to the present embodiment.

5 (a), (b) are converted according to the present invention +
It is a figure which shows the other data format of 1BCD.

6A to 6C are diagrams for explaining a conventional BCD expression method.

[Explanation of symbols]

 11 Data Input Section 12 JIS Code Conversion Section 13 Decimal Character Discrimination Section 14 BCD Generation Section 15 + 1BCD Generation Section 16 + 1BCD Output Section 17 + 1BCD Input Section 18 Delimiter Discrimination Section 19 BCD Conversion Section 20 BCD Output Section

Claims (1)

[Claims] 1. A specific numerical value is added to each byte of numerical data represented by a BCD code in a pack format, which indicates a 2-digit numerical value in 1 byte, and the BCD code is different from a delimiter code used in an operating system. Code conversion means for converting into a code, means for storing the code converted by the code conversion means in a memory, and each byte of the code read from the memory when performing a decimal operation by the code conversion means. A data processing device comprising: a unit that subtracts the added specific value, converts it to a BCD code, and outputs the BCD code to a calculation unit.