JPH0363099B2 - - Google Patents

Description

[Detailed Description of the Invention] Technical Field The present invention relates to a kana-kanji conversion device that converts a kana character string input according to the pronunciation of the characters into a kanji or a kanji-kana mixed sentence, and in particular, the invention relates to a kana-kanji conversion device that converts a kana character string input according to the pronunciation of the character into a kanji or kanji-kana mixed sentence. Concerning a radical index dictionary used when searching for kanji whose readings are unknown.

Technical Background Currently, Kana-Kanji conversion devices in general use use an onkun input method in which the reading of a Kanji is input as a method of specifying any Kanji.
However, this method has the drawback that it is not possible to specify unknown kanji or kanji whose readings are unknown. Therefore, one possible method to quickly output even kanji whose pronunciation is unknown is to search for kanji by specifying the radical, which is a characteristic of the kanji's shape.
To give one possible example, a kana-kanji conversion device basically includes an input device 1, as shown in FIG.
It is composed of a processing device 2 and a display device 3. The input device 1 includes a radical key for instructing a kanji search mode and a numeric keypad for giving a numerical value,
To specify a radical, first press and hold the radical key and enter 0 on the numeric keypad. The radical code table 4 shown in FIG. 2 is displayed on the display device 3, and the code number of the corresponding radical is determined from this radical code table and the code is input. When the radical is designated in this manner, a group of Kanji characters having the designated radical are displayed on the display device 3 together with the identification symbol given to each Kanji character.
The Kanji can be selected by inputting the identification symbol of the Kanji to be used.

In order to search for a desired kanji by specifying a radical in this way, the processing device 2 needs to include a radical index 5 and a radical index dictionary 6, which are made up of a ROM as shown in FIG. 3, for example. . The radical index 5 stores the radical code, the radical character code, and the start address of the block in the radical index dictionary. The character code of each kanji in that kanji group is stored divided into blocks. By the way, if a radical index dictionary is constructed by storing the character codes of all kanji characters, a large capacity memory is required. This inevitably results in an increase in equipment costs. On the other hand, although radical designation is not applied very often, it is a problem that this function increases the cost of the device.

Purpose of the invention This invention was made in view of the above problems,
The purpose of this paper is to propose a data compression method that can reduce the amount of data, and thereby provide a radical index dictionary that uses a small amount of memory.

Summary of the Invention The radical index dictionary of the present invention classifies a group of Kanji characters defined by 2-byte data of the upper bit and the lower byte, in which the most significant bit of the byte is 0, into a first group of Kanji characters having the same prefix. Then, this first group of kanji is divided into a second group of kanji that have a common upper byte, and the common upper byte is converted into an identification code indicating data compression, and this converted identification code and the following The second kanji group is compressed with data regarding the lower byte of each kanji of the second kanji group.

Preferably, the identification code is such that the most significant bit of the common upper byte is changed to 1.
This allows data read from the dictionary to be easily restored into Kanji codes.

More preferably, the data regarding the lower byte stored consecutively in the identification code is in at least one of two different formats.
That is, the first format is a case where the lower byte of each kanji in the second kanji group does not change regularly in ascending or descending order, and the lower byte of each kanji is the data itself. The second format is when the lower byte of each kanji in the second kanji group changes regularly in ascending or descending order. The data consists of a second identification code and the number of the second kanji group to be stored consecutively in the second identification code. If a second group of Kanji characters having data in the second format is mixed with the first group of Kanji characters, the compression ratio is further improved.

Hereinafter, the present invention will be explained with reference to Examples.

Embodiment FIG. 4 is an explanatory diagram of a data compression method using "Onnahen" as an example of a radical. The kanji code is C6226 according to the JIS standard. A single Kanji character in this standard consists of two bytes, an upper byte and a lower byte, and the most significant bit of each byte is both "0". For example, "娃" is
Although it is expressed as "00110000 00100011", it is expressed here as "3φ23" in hexadecimal for simplicity.

The first data compression method, as shown in Figure 4a, omits the upper byte for a group of kanji characters in which the upper byte of each kanji is common and represents only the lower byte, and the most significant bit of the common upper byte is An identification code with 1 is added to the beginning of the consecutive code group of the lower byte. For example, "娃",
The kanji groups of "姶", ..., "耶" have a common high-order byte "3φ", so the low-order bytes "23", "28", ...,
Represents only “79” and memorizes it consecutively. Since the common upper byte is "3φ", the most significant bit is changed to 1 and added in front of "23" as an identification code of "10110000", that is, "Bφ". “Uba”…
is a different kanji group because the upper byte is “31”.
Therefore, the common upper byte is converted into an identification code "B1" and added before "38" which is the beginning of the lower byte group. Then, the last byte of the code group following "Bφ" and the identification code of "B1" are successively stored. However, the break between code groups can be determined by the identification codes "Bφ" and "B1".
With this method, a group of n kanji characters can be compressed into (n+1) bytes.

The second data compression method utilizes the characteristics of the second level kanji arrangement of JIS C6226, as shown in FIG. 4b. For example, from "奸" to "孀"
The 32 arrays leading to , have the upper byte in common and the lower bytes are arranged in ascending order. Therefore, both the upper byte and lower byte are omitted, and the upper byte and lower byte of the first consecutive kanji code are converted into a first identification code and a second identification code, respectively, and after the second identification code, This is a method of memorizing the number of consecutive kanji groups (32).
The first identification code is one in which the most significant bit of the upper byte is set to 1, as in the first compression method described above. The second identification code is similar to this, and the most significant bit of the lower byte is set to 1. This second identification code allows this second data compression technique to be distinguished from the first data compression technique described above. Therefore, compressed data to which both the first compression method and the second compression method are applied can be mixed in the radical index dictionary of the embodiment. In addition,
If the storage area of the dictionary is defined to be partitioned by the first compression method and the second compression method, this second identification code is not particularly necessary. According to the second data compression method described above, a group of m kanji can be compressed into 3
Can be compressed into bytes.

Note that when applying the above two types of compression methods, 3
The second compression method is applied to codes in which kanji codes are consecutive for more than one character, and the first compression method is applied to other codes. In the case of three consecutive characters, if there is another kanji code with the same upper byte, the first
This is because the second format provides a compression effect of 1 byte less if it does not exist, as is the case with the second format.
This is also advantageous when restoring kanji codes.

Next, a method for restoring kanji codes from data read from a compressed radical index dictionary will be briefly described.

It is assumed that reading is performed sequentially in 1-byte units. For the group of kanji characters stored using the first compression method, an identification code is first read out, and data obtained by inverting the most significant bit of this identification code to 0 is placed in the replacement means. When the next 1 byte is read, the most significant bit is checked, and if it is not 1, it is combined with the above digit data and transferred as a 2-byte Kanji code. When the next byte is read, first check the most significant bit, and if it is not 1, add 2 to the above digit data.
Transfer as a byte Kanji code. This is repeated every time a byte is read unless the most significant bit is 1.

When the most significant bit is 1, the numeral data of the numeral means is replaced with data obtained by inverting the most significant bit of the 1-byte data just read to 0. The next 1 byte is read, and if the most significant bit is not 1, it is transferred together with the new digit data as a 2-byte Kanji code. However, when the most significant bit of this byte data is 1, it is a kanji group stored using the second compression method. Therefore, at that time, data obtained by inverting the most significant bit of this byte data to 0 is set in the first counter means, and the next byte data is read out and set as number data in the second counter means. next,
The data of the number setting means and the data of the first counter means are combined to form a 2-byte Kanji code and transferred. At the same time, the second counter means counts down and the first counter means counts up. This operation is repeated as long as the content of the second counter means is not 0. When it becomes 0, the reading of the kanji group to which the second compression method is applied ends. After that, the above-mentioned operation is repeated. Then, when the first address of the next Kanji block is reached, the reading of the Kanji block with the same radical is completed.

Effects As described above, the radical index dictionary of the present invention converts a common high-order byte into an identification code indicating data compression, and uses this converted identification code and data regarding the low-order byte of each consecutive kanji. Since it is a compressed group of kanji characters that have a common upper byte, it can be configured with a small memory capacity and is advantageous in terms of cost.

[Brief explanation of drawings]

Figure 1 is a block diagram showing the basic configuration of the ephemeral-kanji conversion device, Figure 2 is a diagram showing the radical code table, and Figure 3 is a diagram showing the radical code table.
Figure a is a schematic diagram of a radical index, and Figure b is an explanatory diagram of a radical index dictionary. FIGS. 4a and 4b are diagrams for explaining the first data compression method and the second data compression method, respectively. 1... Input device, 2... Processing device, 3... Display device, 5... Radical index, 6... Radical index dictionary.

Claims (1)

[Scope of Claims] 1. In a radical index dictionary that is indexed based on the designation of the radical of a kanji in order to search for a target kanji, the upper byte and the lower byte of which the most significant bit of the byte is 0 are: A kanji group defined by byte data is divided into a first kanji group having the same prefix, and this first kanji group is divided into a second kanji group having a common high-order byte, and at the same time A second kanji group is compressed by converting the upper byte into an identification code indicating data compression, and using the converted identification code and data regarding the lower byte of each kanji of the second kanji group that follows the converted identification code. A radical index dictionary featuring . 2. The radical index dictionary according to claim 1, wherein the identification code is obtained by changing the most significant bit of the common upper byte to 1. 3. The data regarding the lower byte following the identification code is the data of the lower byte of each kanji in the second kanji group, or the lower byte of each kanji in the second kanji group is arranged regularly in ascending or descending order. If the number of the second kanji group changes, the second identification code is obtained by converting the most significant bit of the lower byte of the first kanji character to 1, and the number of the second kanji group. Radical index dictionary described in Section 2.