JPH0855011A - Method for converting data between different machine kinds - Google Patents

Abstract

PURPOSE:To reduce a storage capacity required for data conversion by rewriting the expression of the exponential part of data to be converted into the expression of the exponential part of corresponding conversion data and rearranging a digit number so as to turn the decimal part of the data to be converted into the same value as the decimal part of the conversion data. CONSTITUTION:A real number type data file 3 is transferred to the file 4 of another computer 2 and the exponential part of the exponential part form of the data to be converted is rewritten to the exponential form of the conversion data by the exponential part rewrite means 6 of a conversion program 5 housed on the side of the other computer 2. Then, the decimal part of hexadecimal 6 digits of the data to be converted is matched with the binary 23 digits of the decimal part to be the same value and bits are rearranged by a decimal part rearrangement means 7. Real number data converted in such a manner are housed in the file 8 and used on the other computer 2. Thus, a disk capacity required for the conversion is reduced.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data conversion method between computers having different internal representations of real number type data.

[0002]

2. Description of the Related Art Conventionally, data in a format as shown in FIG. 3 (a) written in a file by a process computer (hereinafter abbreviated as "pro computer") is converted into incompatible data as shown in FIG. If you try to use it in another computer with a data format according to the IEEE standard, the exponent part of the real number type data,
It is not possible to read and write directly because of the difference in internal representation such as the decimal part.

Therefore, by the configuration shown in the block diagram of the conventional conversion system of FIG. 4, a file 42 of real number type data which is generally desired to be diverted is once converted into text data 44 by a conversion program 43 on the processor 40. Then, a method of returning from the text data file 45 to the real number type data 47 on another computer 41 by the conversion program 46 is used. In this case, for example, if the text data input to the file 45 is “123.4”, 1 is read in order from the left and the decimal point does not exist, so the digit is increased and then 2 is read. Increase the digit of 1 and 2 based on the decimal point, and move to the next 3
Is read as it is, and the next 4 is next to the decimal point, so the digit is lowered to the power of −1, and 123.4 is converted into real number data represented by the exponent part and the decimal part.

[0004]

However, the above-mentioned conventional example is versatile because it can be widely read and written in the form of text data by general computers.
Real number type data that can be expressed in 4 bytes on a bit computer becomes 7 to 8 bytes at the stage of conversion to text data, and the capacity of the files 44 and 45 is about 2N, which is a storage capacity close to twice the normal capacity N. There is a problem that requires.

Therefore, an object of the present invention is to make it possible to directly convert the data of the process computer into a format that can be handled by another computer based on the internal representation of the real number type data in the standard format, and the storage capacity required for the data conversion. It is to provide a heterogeneous data conversion method that can reduce

[0006]

In order to achieve the above object, the present invention addresses the representation of the exponent part of the converted data in a data conversion method between models that are incompatible due to the difference in internal representation. It has exponent part rewriting means for rewriting to the representation of the exponent part of the converted data, and decimal part rearranging means for rearranging the numbers so that the decimal part of the converted data has the same value as the decimal part of the converted data. .

[0007]

According to the above construction, the exponent part rewriting means rewrites the exponent part of the converted data so as to correspond to the exponent part of the converted data, and the decimal part rearranging means makes the decimal part of the converted data the same value. Since the converted data is rearranged to the digit of the decimal part, the converted data can be directly converted into the representation format of the converted data.

[0008]

An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram of a heterogeneous data conversion method according to an embodiment of the present invention. FIG. 2 is a diagram showing a conversion processing procedure of the embodiment shown in FIG.

In FIG. 1, the file 3 of the real number type data which is the converted data created by the processor 1 has a code S =
1 digit (1 bit), exponent part = 7 bits, decimal part = 24
This is the conventional type shown in FIG. 3 (a), which takes a bit (hexadecimal 6-digit) representation format.

The real number type data file 3 is transferred to the file 4 of the other computer 2 and converted into the converted data by the exponent part rewriting means 6 and the decimal part rearranging means 7 constituting the conversion program 5 stored in the other computer 2 side. Figure 3
The representation format according to the IEEE standard of the other computer 2 shown in (b), code S = 1 bit, exponent part = 8 bits (2
Exponentiation, 0-256), fraction = 23 bits (binary 2
(3 digits) is directly converted and stored in the converted real number type data file 8 to be used by another computer 2.

Next, the operation of the conversion process will be described.
In the real number type data of the process controller 1 shown in FIG. 3A, which is the data to be converted, the decimal part is a normalized absolute value display and is a 6-digit hexadecimal display in which 24 bits are grouped by 4 bits. , The exponent part expresses 0 to 127 with 7 bits, but exponents of -64 to 63 are actually expressed by this with 64 characters. The decimal point is the leftmost digit of the decimal part.

Real number type data file 4 of such a format
2 is converted to the IEEE standard, as shown in the conversion procedure of FIG. 2, first, it is checked whether 31 bits of the code part of the converted data are [0]: positive number or [1]: negative number, and the IEEE standard The code part of the converted data is copied to S = 31 bits (S2
0).

Subsequently, the exponent part rewriting means 6 makes the exponent part format of the data to be converted into 7 bits (0 to 127, -64 to 6).
The exponent part of 3) is rewritten into the exponent part format of the IEEE standard of the converted data, which is 8 bits (0 to 256, -127 to 128) (S21).

This rewriting of the exponent part can be performed, for example, by using a computer 1
The expression of the exponent on the side is [1000000 = 64, 011
1111 = 63], and the exponent part of the IEEE standard is
[01111111 = 127, 10000000 = 12
8] is the rewriting process during that period.

Next, the decimal part rearrangement means 7 makes the hexadecimal 6-digit decimal part of the data to be converted into the IEEE equivalent.
The bits are rearranged according to the 23-digit binary number in the decimal part of the standard. Note that in this case, the decimal point is the leftmost position of the decimal part, and the normalized expression in which the bit assumed on the left side of the decimal point is always 1 is noted (S22).

This rearrangement is simply performed between hexadecimal and binary, for example, hexadecimal [3CFH] is converted into binary [1111].
[001111] is rearranged between 6 hexadecimal digits and 23 binary digits.

The real number data thus converted to the IEEE standard is stored in the file 8 and used on another computer 2.

As described above, in this embodiment, as compared with the conventional example shown in FIG. 4, the conversion program required for the conversion processing between the computer 1 and the other computer 2 is only the conversion program 5 on the other computer 2 side. , Memory capacity is also PROCON 1
Since the text data 44 of the capacity 2N is unnecessary on the side, and the capacity of the other computer 2 can be reduced from the capacity 2N to the capacity N, the storage capacity required for the data conversion processing can be significantly reduced.

Further, the conversion program 5 in this case can be stored in an external storage device such as an FD or the like because it can be converted into the representation format of the IEEE standard by a simple and reliable method without data loss. If it is configured so as to be portable, the system operability is improved in an environment where the number of computers of the conventional IBM system and the standard type computer of the IEEE standard are mixed, and the number of computers is increasing recently.

[0020]

As described above, according to the data conversion method between different models of the present invention, the exponent part of the converted data is used in the data conversion between the models which are not compatible due to the difference in the internal representation of the real number type data. Exponent part rewriting means for rewriting the expression of the exponent part of the corresponding converted data, and decimal part rearranging means for rearranging the number of digits so that the decimal part of the converted data has the same value as the decimal part of the converted data. Therefore, only one conversion program is required, and there is no need to create text data between the process computer and another computer, and the disk capacity required for conversion can be reduced.

[Brief description of drawings]

FIG. 1 is a block diagram of a heterogeneous data conversion method according to an embodiment of the present invention.

FIG. 2 is a diagram showing a conversion processing procedure of the embodiment shown in FIG.

FIG. 3 is a diagram showing a format of data in a file of a conventional process computer.

FIG. 4 is a block diagram of a conventional conversion system.

[Explanation of symbols]

 1 process computer 2 other computer 3,4 real number type data file 5 conversion program 6 exponent part rewriting means 7 fractional part rearranging means 8 conversion real number type data file

Claims (1)

[Claims] 1. An exponent part rewriting means for rewriting the exponent part of the converted data to the corresponding exponent part of the converted data in a data conversion method between models that are incompatible due to a difference in internal representation. A heterogeneous data conversion method comprising a decimal part rearranging means for rearranging digits so that the decimal part of the converted data has the same value as the decimal part of the converted data.