JPH056260A - Japanese data compressing system - Google Patents

Abstract

PURPOSE:To efficiently compress Japanese data composed of two byte codes according to a run length encoding system by executing the preprocessing of compression to compress continuous character strings at every two strings. CONSTITUTION:In the preprocessing of compression, a data train A to be compressed is successively read for each character (high-order byte) and compared, the codes of the continuous characters and the number of continuous characters are recorded together with a special character showing the compression, and preprocessing data B of compression are prepared. Next, compressed data C are prepared by compressing low-order bytes. Thus, compressibility can be improved in comparison with data C' compressed by normal run length encoding.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data compression system, and more particularly to a compression system for Japanese data composed of 2 bytes.

[0002]

2. Description of the Related Art In the conventional data compression system of this type, when a character is a sequence of one byte character and several characters are consecutive, a special character indicating compression, repeated character data, and the number of repetitions are used. Data was reduced by converting to.

FIG. 2 shows a conventional data compression method. The compressed data 1 is a normal run length (un-length).
th) Compressed by the compression processing unit 2 to obtain compressed data 2.

FIG. 3 is a flowchart showing an algorithm of the above conversion in the compression processing section 2. This will be described below with reference to FIG. First, the character counter Cc and the repeat counter Cr are set to 0 (steps S1 and S2). After reading one character from the original data (step S3), the character counter Cc is incremented by 1 (step S4). The value of the character counter is compared with the read character (step S5). In the first cycle, this comparison is always true, and the character read to check whether the original data is composed of four or more repeated characters is stored in the buffer (step S6). In the second and subsequent cycles, the character read from the original data is compared with the character stored in the buffer (step S7). If the original character is the same as the stored character, it is compressed because four or more same characters are repeated. When the character is equal to the stored character, the repeat counter Cr is incremented by 1 (step S8), and another character is read from the original data. If the original character is not equal to the stored character, the repeat counter Cr is compared with 4 (step S9). If less, no compression is performed because only three characters are repeated. In this way, when the repeat counter Cr is 4 or more, a compression format is created (step S10).

[0005]

In this conventional data compression method, it is possible to reduce the data size when the same character appears consecutively in a data string consisting of 1-byte code characters such as ASCII code. However, in a data string consisting of 2-byte code characters such as shift JIS code, even if the same character appears consecutively, the adjacent 1-byte data is not equal, so the data size cannot be reduced. was there.

An object of the present invention is to provide a Japanese data compression method that solves such problems.

[0007]

According to the Japanese data compression method of the present invention, the data size is reduced by converting into a special character indicating the compression of data in which the same character is continuous, repeated character data, and the number of repetitions. In the data compression method, especially in Japanese data consisting of upper and lower 2 bytes, paying attention to the upper byte, every other byte (upper byte) is a special character that indicates the repetition of the upper byte in a data string in which the same character continues. It is characterized by adding a pre-processing for converting a character, a character to be repeated, the number of repetitions, and a data string having only lower bytes.

Further, the Japanese data compression method of the present invention reduces the data size by converting special characters indicating compression, repeated character data, and the number of repetitions when consecutive data are the same character. Regarding the data compression method, particularly in Japanese data consisting of upper and lower 2 bytes, paying attention to only the upper byte, and when it matches and is continuous with the code defined in advance, it is referred to as the above data compression method. Similarly, it is characterized in that it is converted into a special character indicating compression, repeated character data, and the number of times of repetition, and then a front end for connecting the lower bytes of the compressed character is added.

[0009]

Embodiments of the present invention will now be described with reference to the drawings.

FIG. 1 is a block diagram showing an embodiment of the present invention. The compressed data 1 is 2-byte 1-character Japanese data using shift JIS, for example. After compressing this by the compression pre-processing unit 12 of the present invention, the normal run-length compression described in the related art is performed by the compression processing unit 13.
The resulting compressed data is compressed data 4
Is.

FIG. 4 shows an algorithm as a detail of the pre-compression processing unit 2. An embodiment of the compression preprocessing unit will be described with reference to FIG. First, while clearing the stack for storing the lower byte during the compression operation (step S11),
The character counter Cc and the repeat counter Cr are reset to 0 (steps S12 and S13). After reading one character (upper byte) from the compressed data 1 (step S14), another character (lower byte) is read (step S15), and the character counter is incremented by 1 (step S16). Next, when the value of the character counter is 1 (step S17), the upper byte read in the register for comparison is set as a continuous character (step S18), and the process returns to the process of reading the next character.

In the second and subsequent cycles, the character (upper byte) read from the compressed data 11 is compared with the character stored in the register (step S19). In the case of the same character in this comparison, four or more same characters are consecutive and are therefore compressed. When the character is equal to the character in the register, the repeat counter is incremented by 1 (step S20) and the loop is repeated again. If not equal to the current read character (upper) (step S2
1), the repetition counter is compared with 4, and if the number is smaller, the compression effect cannot be obtained, so the compression processing is not performed. Four
In the above cases, the compression effect can be obtained, so the data is converted to the compression format (step S22). Here, the compression format is composed of a special character 1 byte indicating compression, a repeat character 1 byte, a value of a repeat counter (1 byte), and a lower byte string stored in the stack. The lower byte string length is equal to the repeat counter value.

FIG. 5 shows an example of a data string before and after compression.
The compressed data sequence A is compressed into compressed data B by the pre-compression processing unit 2.

The compressed data B pre-processed by the compression pre-processing unit 12 is compressed into the compressed data C of FIG. 5 by the normal run-length compression processing unit 13 to obtain the compressed data 14 of FIG.
For comparison, FIG. 5 shows compressed data C ′ obtained by compressing the compressed data A by the normal run-length compression processing unit 2 of FIG. It can be seen that a character string that could not be compressed conventionally can be compressed by this embodiment.

By the way, in the restoration process, when a special character is detected, the continuous character and the continuous count following the special character are read, and the upper byte is added every time the lower byte is read by setting the continuous character as the upper byte by the value of the continuous counter. It is realized by outputting a 2-byte character.

In this way, by compressing every other consecutive character string in the pre-compression processing, consecutive 2-byte character Japanese data can be compressed.

[0017]

As described above, the present invention detects every other consecutive characters as the pre-compression processing and compresses the number of high-order bytes, so that the combination of hiragana and alphanumeric characters in the shift JIS code can be changed. There is an effect that the sentence can be compressed, and further, since the lower byte is continuous in the pre-compression process, a continuous 2-byte character of the shift JIS code can be compressed by the conventional run length encoding.

[Brief description of drawings]

FIG. 1 is a block diagram of an embodiment of the present invention.

FIG. 2 is a block diagram of a conventional example.

FIG. 3 is a flowchart of an algorithm for run length encoding.

FIG. 4 is a flowchart of pre-compression processing.

FIG. 5 is a diagram for explaining the effect of the present invention.

[Explanation of symbols]

1,11 Compressed data 2, 13 Compression processing unit 3,14 Compressed data 12 Compression pre-processing unit

Claims (2)

[Claims] 1. A data compression method for reducing the data size by converting a special character indicating the compression of data in which the same character is continuous, repeated character data, and the number of repetitions, particularly from the upper and lower 2 bytes. In Japanese data, paying attention to the upper byte, every other byte (upper byte) a data string in which the same character continues, a special character that indicates the repetition of the upper byte, a repeating character, the number of repetitions, and a data string containing only the lower byte Japanese data compression method with pre-processing added for conversion to and. 2. A data compression method of reducing the data size by converting special characters indicating compression, repeated character data, and the number of times of repetition when consecutive data are the same character, particularly in the upper order. In Japanese data consisting of the lower 2 bytes, pay attention to only the upper byte, and if it matches and continues with the code set in advance, repeat it with a special character indicating compression as in the above data compression method. A Japanese data compression method in which a character data and the number of repetitions are converted, and then a prefix is added to connect the lower bytes of the compressed character.