JPH0778795B2 - Character processor - Google Patents

Description

The present invention relates to a kana-kanji conversion device for converting a kana into kanji.

[Prior art]

Conventionally, in an apparatus for converting a kana into kanji, there has been a method of converting an inputted kana into kanji by separating them with clauses.
However, in that case, the longest kana part was converted to kanji as the reading of the word from the beginning of the phrase, so when you want to convert it to a kanji with a shorter reading, it is not always necessary to separate with kanji. I got it. For example, if you want to create a phrase "feeling" and input "Kanjiji" to perform a conversion operation, there is a risk that "Kanji" and "secretary" will be converted. In other words, the longest reading from the beginning as the reading of the word is "Kanji" as compared to the reading of "Kanji". Once this reading was cut out, only the words corresponding to the reading "Kanji" were listed as homonyms, and the reading "Kan" was not cut out.

In particular, in the input method in which the delimitation of the clause is not specified, the end of the conversion unit starting from the beginning of the conversion target character string is unknown. Therefore, when trying to cut out the longest reading for the entire conversion target character string, the length of such longest reading is often much shorter than the entire conversion target character string.
It takes a long time to obtain the longest reading. If the longest reading obtained is incorrect, the conversion of the subsequent character string may be missed.

〔Purpose〕

The present invention quickly obtains a conversion candidate for a conversion target kana character string that is input without explicitly indicating a punctuation of a phrase, and at that time, not only a conversion candidate corresponding to the longest matching reading,
It is an object of the present invention to provide a character processing device that can enumerate candidates for readings shorter than that, and can flexibly cope with an error in conversion delimitation due to a difference in candidate length.

〔Explanation〕

Hereinafter, the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing the configuration of an embodiment according to the present invention. Here, the CPU is a central processing unit that controls various calculations and peripheral devices. ROM is the program memory, the third
Central processing unit as shown in FIG. 4, FIG. 6, FIG. 9 and FIG.
It stores the CPU execution program. BL is a bus line that connects other memories and peripherals to transmit signals. KB is a keyboard, which performs the input function. RAM
Is random access memory, document memory ME, homophone word buffer BU, homophone word counter CO, document end undress EA
D, cursor address CAD, insertion flag IF.
The document memory ME stores characters input from the keyboard KB, converted Kanji characters, and the like. Homophone buffer BU
Stores reading and kanji corresponding to the reading. The homophone word counter CO designates the display order of the homophone words stored in the homophone word buffer BU. The document end address EAD stores the last address where the character exists in the document memory ME. Cursor Address CAD stores the position of the cursor. The insertion flag IF plays a role of a switch as to whether or not the insertion process is performed.

The LCD is a liquid crystal display, and displays a character string input from the keyboard KB, the contents of the document memory ME, and other cursors.

The DCT is a display controller that controls the liquid crystal display LCD.

CG is a character generator that stores fonts corresponding to character codes.

The DIC stores the kanji corresponding to the phonetic kana in a kanji dictionary.

FIG. 2 is a schematic view showing details of the keyboard KB shown in FIG. Among them, the kana and symbol keys are equipped with a plurality of character and symbol keys such as hiragana, katakana, and "," and ".". The cursor left move and cursor right move keys are keys for moving the cursor left and right. UL has a function to add an upper line to the character on the cursor display with the upper line key. The SC has functions such as comparison of kana character strings with dictionary readings using the search key, creation of homophone buffers, and confirmation of homophone selection. TR is a conversion key and has a function of sequentially displaying the Chinese characters in the homophone word buffer.

With the above configuration, the operation of the embodiment of the present invention will be described with reference to the flowchart.

When the power is turned on, this apparatus first proceeds to step S1 in FIG. At step S1, the document memory ME in the random access memory RAM, the homophone word buffer BU, the homophone word counter CO are cleared, the document end address EAD in the same is set 0, the cursor address CAD is set 1 and the insertion flag IF is initialized. Perform processing. After that, the process proceeds to step S2, and the state of waiting for key input from the keyboard KB shown in FIG. 2 is maintained. So, if any key is entered, step
In step S3, the entered key is discriminated and the process proceeds to any one of steps S4 to S10. A step S4 carries out a process when the character / symbol key is operated.

A detailed flow chart of the character input processing step S4 is shown in FIG. First, an upper line bit is added to the character code input in step S4.1.

Here, the difference in the internal code between the normal display character and the character with the upper line will be described.

Generally, the document memory of a character processing device is composed of one word or 16 bits per character. The code for each letter is 15
Bits are assigned to each bit in a one-to-one correspondence, but the most significant bit, bit 0, is the upper line bit UB.
It plays a role as. For example, as shown in Fig. 5 (a), the code "a" is displayed in hexadecimal as "242".
2 ”, but“ a ”with an upper line attached to it has a code system of“ A422 ”as shown in FIG. 5 (b).

Now, in step S4-2, the random access memory RA
It is determined whether the insertion flag IF in M is in the ON state or the OFF state. If the flag is OFF, the process proceeds to step S4.4. If it is ON, the process proceeds to step S4.3.

When the insertion flag is ON, in the next step S4.3, in the document memory ME, the character array after the address indicated by the cursor address CAD is shifted backward by one character.

Next, in step S4.4, the cursor address CAD is shown. The input character code is stored in the address in the document memory ME. Then, in step S4.5, 1 is added to the cursor address CAD, and the character input process S4 in the flowchart of FIG. 3 is completed.

Step S5 is a process when the search key SC is operated,
FIG. 6 shows details of the search processing in step S5. FIG. 7 is a diagram showing a process in the case of making a sentence “Make a machine product.” The search key SC is operated from FIG. 7 (a) to FIG. 7 (b), FIG. (D) to FIG. 7 (e) to FIG. 7 (f), FIG. 7 (g) to FIG. 7 (h) to FIG. 7 (i), and FIG. 7 (i) to FIG. FIG. 7 (h), FIG. 7 (l) to FIG. 7 (m), FIG. 7 (n), and FIG. 7 (n) to FIG. 7 (o). The above seven steps are shown in FIG. 7 (p), FIG. 7 (p) to FIG. 7 (q), and then FIG. 7 (r).
The conversion key TR, which will be described later, is used to display the kanji candidates for the phonetic kana.

First, in step S5.1, it is determined whether or not the character on the cursor has an upper line. When there is no upper line, that is, in the case of FIG. 7 (a) and FIG. 7 (q), the process proceeds to step S5.4. When the upper line is attached, that is, FIG. 7 (d), FIG. 7 (g), and FIG.
Figure (i), Figure 7 (l), Figure 7 (n), Figure 7 (p)
If the character to be determined has an upper line, proceed to step S5.2 and check the character on the cursor and all the character codes of the character string with the upper line continuous to the left. The upper line bit is set to 0, and on the display screen, FIG. 7 (e), FIG. 7 (h), and FIG.
Drawing (j), Drawing 7 (m), Drawing 7 (o), Drawing 7 (q)
Erase the upper line as shown in.

By this processing, the same phoneme is selected and confirmed. Next, the process proceeds to step S5.3, and all the character strings that have the same reading as the kanji selected and confirmed in step S5.2 and have upper lines attached to all the characters of the reading are searched from the beginning of the document memory, If they exist, all those character strings are selected as upper line bit 0 selected in step S5.2.
Replace with the Kanji code of.

Therefore, kanji without upper lines are displayed. Of course, at this time, when the number of Kanji characters to be replaced is smaller than the number of characters to be read, the document memory is filled and a process that does not leave a gap is also performed at the same time.

Next, in step S5.4, the character string to the right of the cursor is examined to determine whether any character with an upper line exists. If it exists, that is, FIG. 7 (e), FIG. 7 (h), FIG. 7 (j),
In the case of FIG. 7 (m) and FIG. 7 (o), step S5
If it does not exist, that is, if it is as shown in FIG. 7 (q), proceed to step S5.5, check the character string on the left side of the cursor, and find that even one character with an upper line exists. If it exists, that is, if it is as shown in FIG. 7 (a). When it does not exist, that is, when it is as shown in FIG. 7 (q), the process proceeds to step S5. Therefore, in step S5 and S6, since there is no character with an upper line on the cursor and on the left and right of the cursor, the upper line bits of all the character codes on the document memory are 0. At this time, in order to prepare for the next character input, the cursor address is set to the document end address + 1 so that the cursor is positioned one digit ahead of the document end address.
It is displayed as shown in FIG. On the other hand, when it is determined that the upper line character exists only on the left side of the cursor in step S5 / 5, in step S5 / 7, the cursor address is set to 1 in preparation for searching the character with the upper line from the beginning of the document memory. set. Further step S5
In the case of 8, there is at least one character with an upper line in the character string on the right side including the character on the cursor.
Characters are extracted and compared with the reading in the Kanji dictionary DIC. If they do not match, the last one character is deleted from 7 characters and the reading is compared again with the reading in the Kanji dictionary DIC. When they do not match, the last character is deleted again and the same process is repeated until they match. So, if there is a matching Yomi in the Kanji dictionary DIC,
Homophone buffer BU for the reading and all the corresponding Kanji
To store. Further, all the kanji in the kanji dictionary DIC corresponding to the reading obtained by deleting the last one character of the reading is also extracted and stored in the homophone word buffer BU together with the reading.

The procedure described above is repeated to extract all the kanji in the kanji dictionary DIC corresponding to the first character among the characters with upper lines and store them in the homophone word buffer BU together with the reading. A check code is stored at the end of the homophone word buffer BU. or,
If the beginning of a character with an upper line is "o" and there is no reading corresponding to the Kanji dictionary, only that one character is stored in the homophone word buffer BU and the check code CCD is stored in the next area. This situation will be described with reference to FIG. First, in the homophone buffer BU, the area in which one word is inserted is composed of 7 words. This corresponds to searching from the above 7 characters. Therefore, three kanji corresponding to "kikai" and its reading are stored as "opportunity", "machine", and "strange",
Next, “Kika” and the corresponding Chinese characters “naturalization” and “geometry” are stored, and further, “Ki” and the corresponding “tree” and “ki”.
"Tree" is stored.

Finally, the check code CCD is entered. When the above process is completed, the homophone word counter CO is set to 1 and the cursor is displayed at the same character position as the last character of reading stored in the first area of the homophone word buffer BU. The character position on the memory is set under "I" in "Kikai" as shown in FIG. 7 (a), and then the cursor is displayed.

Next, the procedure detailing step S6 conversion processing described above is explained in the ninth section.
It will be described with reference to the drawings. First, in step S6.1, the homophone word counter CO is incremented by 1. This is to indicate the homophone area next to the currently displayed homophone. Then, in step S6.2, the character code in the new homophone area is checked in step S6.3.
Then, the homophone word counter CO is set to 1 as a preparation for redisplaying from the beginning of the homophone word buffer BU. On the other hand, when it is other than the check code CCD in step S6.2, the process proceeds to step S6.4, and the character code in the area corresponding to the homophone word counter CO is stored in the document memory ME. At this time, when the number of characters is small, the document memory ME is packed so that there is no gap in the document memory ME. Further, at the last character position of the homonym newly transferred to the document memory ME at the cursor address CAD, for example, in the case of "tree" in FIG. 8 next, it is set at the position where the kana character "ki" was present. Then, the document memory ME and the cursor are displayed.

Process 7 is a process when the cursor left movement key or the cursor right movement key is operated. That is, in the case of the cursor left movement key, 1 is subtracted from the cursor address in the random access memory RAM, and in the case of the cursor right movement key, 1 is added on the contrary. After such processing, move the cursor to the LCD
To display.

Step S8 of performing the upper line process is a process when the upper line key UL is operated.
Bit 0 of the character code stored in the address on the document memory ME shown in CAD, that is, the upper line bit is set to 1 to bring it to the state before the selection is confirmed, and 1 is added to the cursor address CAD and the cursor is displayed.

Step S9 is a process when the insert key is pressed, and OF when the insert flag IF in the random access memory RAM is ON.
Set to ON when F or OFF.

Step S10 is a process when the delete key SK is pressed. One character at the address on the document memory ME indicated by the cursor address CAD is deleted, and the character codes after that character are packed by one address.

The cutout of the above-mentioned kana with upper line is 7 characters, but 8 characters or more or 6 characters or less does not impair the gist of the present invention. However, in that case, after the same sound buffer
Each BU area must consist of as many words as there are characters. Furthermore, the mark of the part converted to kanji is the upper line, but the display method is reverse character,
The gist of the present invention is not impaired even by the method of color characters, underlining, etc. Further, the display of the device of the present invention is a liquid crystal display, but a display such as a CRT may be used.

〔effect〕

As described above, according to the present invention, with respect to the conversion target kana character string that is input without explicitly indicating the delimitation of the clause,
Appropriate conversion candidates can be promptly presented, and in addition to conversion candidates corresponding to the longest matching reading, in response to candidate change instructions,
It is possible to enumerate candidates for readings shorter than that, and it is possible to flexibly deal with an error in conversion delimitation due to a difference in reading length of conversion candidates.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of the embodiment of the present invention. Second
The figure is an external view of the keyboard. Fig. 3 is a flow chart showing the treatment process. FIG. 4 is a flow chart showing a character input process. 5 (a) and 5 (b) are diagrams showing the upper line bit of the character code. FIG. 6 is a flow chart showing the processing steps of the search processing. FIGS. 7A to 7R are diagrams showing examples of sentences. FIG. 8 is a diagram showing a state of the homophone word buffer. FIG. 9 is a diagram showing a process of conversion processing. BU ... Bofua. CO ... Homophone counter. SC ... Search key. TR ... Conversion key. UL ... A top line key.

Claims (1)

[Claims] 1. An input means for inputting a kana character, a dictionary means for storing a reading and a notation of a word in association with each other, and a part of a kana character string input by the input means that matches a predetermined number of characters from the beginning. A search unit that searches the dictionary unit for a word having a reading and outputs the corresponding word notation; and the search unit performs the search by sequentially subtracting the predetermined number of characters from a predetermined maximum value. Search control means for controlling the above, a candidate storage means for storing the word notation output by the search means as a notation candidate for the inputted kana character string, and one of the candidates stored in the candidate storage means. The candidate display means to be displayed, and the candidate display means preferentially displays the word expression for the reading in which the input kana character string and the number of matching characters from the beginning are the largest, and sequentially displays the finger expressions. Based on the candidate change instruction, the currently displayed candidate is changed to the undisplayed candidate if there is an undisplayed candidate with the same reading as the candidate, and if there is no undisplayed candidate, the reading is performed. And a first control means for controlling to sequentially display word notations for readings in which the number of matching characters from the beginning of the inputted kana character string is smaller than the candidate currently being displayed, and the candidate display means. Selecting means for selecting one candidate from the candidates; and when the selecting means selects a word notation for a reading less than the reading having the largest number of matching characters from the beginning, the searching means determines that the number of matching characters is the largest. In the kana character string excluding the reading of the selected word notation from
From the beginning of the kana character string obtained by adding the kana character string obtained by removing the reading with the maximum number of matching characters from the input kana character string, a word having a reading that matches a portion of a predetermined number of characters from the beginning of the kana character string, the dictionary means A character processing device, comprising: a second control means for controlling so as to search more.