JPH03282961A - Mutual conversion dictionary system - Google Patents

Abstract

PURPOSE:To save a memory resource necessary for conversion by providing the mutual conversion dictionary system with double structure capable of retrieving the 'KANA' (Japanese syllabary) reading of a 'KANJI' (Chinese character) character string simultaneously with the conversion of the 'KANJI' character string from a 'KANA' character string. CONSTITUTION:A 'KANA' character string is inputted or a 'KANJI' character string to be reconverted is specified through a Japanese editor. Whether a character string type set up in a conversion buffer is a 'KANA' character type or a 'KANJI' character type is discriminated, and in the case of 'KANA' character type, 'KANJI' retrieval is executed by 'KANJI' processing and converted candidates are obtained. In the case of 'KANJI' character type, the 'KANA' reading is retrieved to obtain converted candidates. Respective converted candidates are outputted to a 'KANJI' character string or a 'KANA' character string by the Japanese editor. Since 'KANJI' conversion or 'KANA' reading conversion can be selected at the time of conversion, the using efficiency of a memory can be improved.

Description

[Detailed Description of the Invention] Industrial Field of Application The present invention relates to a mutual conversion dictionary system that is capable of searching for Kanji character strings from Kana character strings and conversely searching for readings from Kanji character strings. .

In addition, since mutual searches are possible, a mutual conversion dictionary system that functions as a language dictionary such as English translation and Japanese translation will be developed.

BACKGROUND OF THE INVENTION Conventional kana-kanji conversion dictionary systems have a dictionary structure that is used only to search for kanji character strings from kana character strings.

In addition, the reconversion function for kanji strings has a structure in which the original kana string information remains in memory even after it is first converted to a kanji string, and the kana string is converted back to a kanji string. It has become.

Problems to be Solved by the Invention In the conversion dictionary system described above, it is used only as a one-way search dictionary, and even in the re-conversion function, kana character string information remains, resulting in very poor memory efficiency.

Therefore, it is desirable to be able to search between Kanji strings and Gana strings, but this requires the use of separate dictionaries for each, and the dictionary memory size is approximately 2.
The problem is that it doubles.

The present invention has been made in view of these points, and its purpose is to provide a mutual conversion dictionary system that realizes expansion of specifications in a kana-kanji conversion dictionary and improvement of memory efficiency.

Means for Solving the Problems The present invention has a dual structure in which it is possible to search for Kanji character strings from Kana character strings and conversely to search readings from Kanji character strings. The configuration has a conversion dictionary having the above two types of structures with a memory size of 10 minutes.

Example Hereinafter, the present invention will be explained according to one embodiment.

First, the present invention and the conventional technology will be compared using flowcharts.

FIG. 5 is a flowchart of the conventional kana-kanji conversion method. First, in step (11), various key operations are performed using a Japanese editor.

Next, input kana characters (step (12)) and set the kana character code in the conversion buffer (step (12)).
13)). Furthermore, set the kana character code in the reconversion buffer (step (14)) L, then step (1
After the Japanese conversion process in 5), a kanji dictionary search is performed,
After obtaining a conversion candidate (step (16)), the Japanese editor outputs a kanji character string from the kanji code of the candidate (step (17)).

FIG. 6 is a flowchart of a conventional kanji character string reconversion method, in which a character string converted into kanji by the process shown in FIG. 5 is displayed on the Japanese editor in step (21). Here, in step (22), specify the kanji string to be reconverted, and in step (24), check whether the kana character code for the specified kanji string exists in the buffer. If not, in step (23) ) to display an error message and return to the beginning.

On the other hand, if there is a corresponding kana character code, step (2)
Step 5) Set the kana character code in the conversion buffer.

Regarding the next processing steps (26), (27), and (28), exactly the same processing as the normal conversion is performed.

In this way, when trying to reconvert with conventional technology,
A kana character buffer is required for reconversion, and if there is no kana character code corresponding to this character buffer, reconversion will not be possible.

Therefore, in order to make it possible to reconvert all kanji character strings, an extremely large amount of memory is required, and from the perspective of effective use of limited memory resources, this method is extremely inefficient. There is a problem.

Therefore, in the present invention, Japanese language conversion is performed using a mutual conversion dictionary system as shown in the flowchart of FIG.

First, on the Japanese editor in step (31), a kana character string is input or a kanji character to be reconverted is specified (step (32)). Here, determine whether the character string type set in the conversion buffer (step (33)) is a kana character or a kanji character (step (34)).
, if it is a kana character, a kanji search is performed by the kanji conversion process in step (35) to obtain conversion candidates (step (36)).

On the other hand, if it is a kanji character, the reading is searched through the reading conversion process in step (41) to obtain conversion candidates (step (42)).

Each conversion candidate is output by the Japanese editor as a kanji character string in step (37) or a kana character string in step (43).

In this way, in the present invention, it is possible to select whether to convert kanji or readings at the time of conversion, and by using a mutually compatible dictionary, a Japanese conversion method with excellent memory usage efficiency is realized. It becomes possible.

FIG. 2 shows the structure of an embodiment of the mutually compatible dictionary of the present invention.

This dictionary is classified into two parts: a dictionary index part (101) and a dictionary data part (102). The dictionary index section (101) first has the start address (103) of the kanji conversion directory, which is aligned with the start address (104) of the reading directory.

Further, (105) indicates the start address of the dictionary data section (102), and (109) indicates the start address of dictionary block 1.

Here, (106) contains information such as kanji notation. Also, (107) contains the start address of each blank for kanji conversion and a kana heading. And (108
) is the start address of each block of reading conversion, and
Contains kanji headings.

In reality, each data in the reading block is irregularly located in each kanji block, but each data has an attribute and a serial number for reading and writing conversion, making it possible to divide the data into blocks. . (109) blocks the grouping for each Kanji conversion. This blocking is configured only for Kanji conversion, and the blocking for reading Kana conversion is irregular on the surface.

FIG. 3 shows the data structure of the mutual dictionary.

That is, part of speech information (201) is at the beginning, and (202
) and the number of kana characters (203) are lined up. Next, there is a kana header 204), which is expressed in 1 by 1 ~ kana code. Kanji character string data is set in (205) and is represented by a 2-byte Kanji code. (206) is set to the kanji block number in which the next candidate reading is located, and (207) represents the number of bytes from the beginning of the block. (
208) contains information as to whether there is a next candidate. (209) and (210) contain attributes as a reading dictionary, and represent reading block numbers and kanji classification numbers.

FIG. 4 is a flowchart of reading dictionary search.

First, in step (61), the first address of the directory is obtained from (104) shown in Figure 2, and in step (62)
In step (108) shown in FIG. 2, it is determined to which reading block the kanji to be subjected to reading conversion processing belongs.

Next, the start address of the block is found, and in step (64) it is checked whether there is a candidate.

If there is a candidate here, find the length of the converted character string specified (step (32)) on the Japanese editor in step (31) shown in Figure 1 (step (65)), and add the candidate heading and kanji character. Compare column lengths (step (66)
).

Next, in step (67) it is determined whether the lengths are equal, and if the lengths are equal, the kanji codes are compared (step (68)), and if they match (step (69)).
, check whether the number of kanji characters in the dictionary matches the number of kanji characters specified in the Japanese editor (step (70))
.

If there is a match at this time, the pronunciation search is successful. Therefore, step (74)) sets the searched kana character code in the output buffer, returns to the Japanese editor, and outputs the reading kana character string.

As described in detail, according to the mutual dictionary system of the present invention,
Not only Kana-Kanji conversion but also Yomi-Kana conversion can be used in one dictionary, making it possible to selectively perform either conversion at the time of conversion.

Furthermore, memory resources required for conversion can be saved significantly, and their usage efficiency is greatly improved.

[Brief explanation of drawings]

FIG. 1 is a flowchart of a Japanese language conversion process using a mutual conversion dictionary method according to an embodiment of the present invention, FIG. 2 is a structural diagram of a mutual conversion dictionary according to an embodiment of the present invention, and FIG. FIG. 4 is a flowchart of the reading/kana dictionary search process, FIG. 5 is a flowchart of the conventional kana-kanji conversion process, and FIG. 6 is a flowchart of the conventional kanji character string reconversion process.

Claims (1)

[Claims] It has a dual structure that allows you to search for kanji strings from kana strings and conversely search for readings from kanji strings. A mutual conversion dictionary method with a conversion dictionary.