JPS5866141A - Character processing device - Google Patents

Abstract

PURPOSE:To prevent a search word table from overflowing, by generating a connection table on the basis of a table DTBL for phonetical reading elements registered in a dictionary and registering only words required for genetation in the search word table. CONSTITUTION:Data read out from a keyboard controller 6 is written in a buffer SWB of RAM 14. It is discriminated whether Kana(Japanese syllabary)- Kanji (Chinese character) conversion start is designated or not; and if the conversion start is designated, a data string in the buffer SWB is devided to phonetical reading elements and is retrieved by a dictionary. When a desired word exists as the result, read data is registered as a word code in a table DTBL. If this word does not exist, it is discriminated whether the dictionary retrieval is completed or not; and if it is not completed, the next phonetical reading element is generated. If it is completed, the table DTBL is used to generated a connection table CTBL. Only words of reading used in the connection table CTBL are registered in said search word table.

Description

[Detailed description of the invention] Separated for each onyomi from the present @Korean input card data string II
The present invention relates to a character processing device that derives a desired O word from a dictionary whose keywords are reading elements based on given elements.

Conventionally, the kana-kanji conversion method K performs a ``Sengokuyaho'' search on an input character data string, and then searches a dictionary on a storage device for each Kuon reading element separated by the ``Shugoyaho'' for the input data. All the searched words were temporarily stored in a table (hereinafter this table will be referred to as the search word table).

All subsequent processing is performed based on the search word table.

In this case, if the search word table overflows, the kana-kanji conversion process will not be performed. For example, if you enter (Kikaishinkoukai), the dictionary will be searched for 145 types of on-yomi elements as shown in Figure 1, but for the on-yomi element 1'*6 in the same figure, , ki, existing, wood, ki, and multiple words correspond. There are about 40 phonetic reading elements such as Hiro io' or 4. If stored as is in the search word table, S
+S8%O This is a huge number for the on-reading element. Therefore, in the case as shown in FIG. 1, for example, at the stage where the on-reading element 43 is searched for from the dictionary and stored in the search word table, the all search word table becomes
If it is set to 0-, even though there is a word 'kai' corresponding to 7 of the on-yomi element 44 in the dictionary, it will be registered in the search word f-full and a dangerous situation will occur.

As a result, all the necessary words for converting the input character data string into Kanji are available in the dictionary (although there is a drawback that it cannot be converted into Kanji).

The present invention solves the above-mentioned drawbacks, and when searching a dictionary memory that stores words using on-yomi as a keyword, %. ◇ Another object of the present invention is to store the first aS word IDQ word obtained by one on-yomi when searching a dictionary memory in which words are stored using on-yomi as a keyword. Provided is a character processing device that stores (registers) codes and their reading elements.

Other objects of the present invention will become apparent from the following description of embodiments with reference to the drawings.

The present invention will be described in detail below with reference to the drawings.

FIG. 2 is a block diagram showing one embodiment according to the present invention.

In the illustrated configuration, 1 is a microprocessor, 2
is an address bus, which specifies the address of each person's output memory. In the front, 3 is a data bus, which is a bidirectional path used for transferring various data, and 4 is a control path f1, which is used for reading each memory, etc. It is used to transmit information such as ing. Furthermore, 5 is a keyboard, on which various keys such as alphabet keys and hiragana keys are arranged, and 6 is a keyboard controller. Keyboard control 96 inputs 9 data by keymate 5
At the same time, it sends an inter-2 signal to the control bus 4. Next, 7 is the controller C'J s CTR device 8, character generator 9% refresh memo 910, ,,IJ
Fresh memory controller) controls the low jll. Also,
Reference numeral 8 is a CRT device, which can perform a display similar to that of a normal television receiver, and 9 is a character generator, which allows character patterns to be displayed by inputting codes such as character codes and line addresses of character patterns. 10 is a refresh memory that stores the character code to be displayed on the CRT mount 8 and repeatedly outputs the character code in response to display instructions from the CRT controller 7. It is to do
llal mirefresh memory controller
In response to instructions from the CRT controller 7, the above-described refresh cycle of rounding of the CRT display and data writing and output from the microprocessor 1 are controlled. Furthermore, 12 is a memory exclusively for bladder removal, that is, a so-called ROM, which is used to store the control procedures shown in FIGS. 3 and 4, as well as various processing procedures.
is a two-row 2 of 80 yen 12, and is a 14-bit random access memory, that is, so-called RAM.
Used for temporary storage of various data. DTBL in the random access memory 14 is a word table memory, and as shown in FIG. Store. One reading element consists of two words in total: one word of reading data and one word of word code. The total number of words is determined by your needs.

DM is a dictionary memory, and as shown in FIG. 6, one grammatical information EI is stored for each word, reading ogi, and kanji.

The word code TC indicates what position the word is registered in the dictionary.

CTM is a memory that stores combinations of readings of input data strings as shown in FIG.

7-1 is an example showing how to combine input character data strings.

7-2 shows the form in which the above example is actually registered in the memory. For (Kikai)-(Shinko)-(Kai), 08, 44.16 are registered, and these indicate the relative addresses where the corresponding reading data 0 is registered in the word table BTBL.

04 indicates how far the join has been made, 0 indicates the relative address in the X direction in the join table. For each join, there are up to 5 single @ marks, 5 words, and 1 word to represent the number of joins, for a total of 6 words. It is the composition. The number of bonds can be changed as needed.

The 8WB is used to store the character data string input in the search word buffer.The size of the 08WB is N words, and the size can be changed as necessary.

is entered, and the input end code /)/ is entered at the end. 15 more
1 is an 8mm M14 controller, and 16 is a flombi disk, which stores a dictionary having the structure shown in FIG.

All l words are composed of 8 words, and each l word is, for example, 1
Consists of 6 bits.

In one word with such a structure, the first 8 bytes contain VT information, and then the next 3 words following the first 4 words are
As shown in the figure, the word contains a kanji code, and the last word contains grammatical information.

That is, in the structure of this dictionary, the reading information can be up to 8 characters, and the kanji can be up to 3 characters.O17 is a floppy disk drive device and a link controller.

The character processing device of the present invention, which is composed of the above-mentioned components, is operated by input from the keyboard 5, and when an input is made from the keyboard 5, an interwoven signal is transmitted to the microprocessor IK. , ROM
Various control operations necessary for character processing are performed in accordance with control procedures stored in the controller 12.

Next, the above-mentioned control operation of the character processing apparatus of the present invention will be explained with reference to the flowchart shown in FIG. First, in step 3-1, a series of data is read from the keyboard controller 6 and stored in the buffer in the RAM 14.
write to.

Next, in step 3-2, it is determined whether or not the start of kana-kanji conversion is specified, and if it is not the beginning of kana-kanji conversion, step 3-3
Next, various input processes, such as inputting normal English characters, kana characters, and inputting character codes, are performed.On the other hand, if it is the beginning of kana-kanji conversion, step 3-4
Move to.

In step 3-4, the input character data string in the buffer is divided into on-yomi elements as shown in FIG. The above search is performed.In step 3-6, it is determined whether the word sought as a result of the dictionary search is ``Atsushi''.If it is not found, step 3-6
Move on to 8. If it exists, proceed to step 3-7 and register the reading data and word code in table DTBLK as shown in FIG. 2.

In step 3-8, it is determined whether or not the dictionary search has come to an end. If not, the next on-yomi element is created in step 3-4 as shown in FIG. In the example in Figure 1, 45
This will create an element of reading the seeds aloud.

In step 3-9, a join table CTBL as shown in FIG. 7 is created using the table DTBL.

In step 3-10, only the reading words used in the combination table CTRL are registered in the search word table. Next, FIG. 4 will be explained. This is a detailed explanation of step 3-10 in Figure 3.0 In step 4-1, the join table l'c'rBL
Shift the row counter CL to OK-.

In step 4-2, the column counter CCt01C of the CTBL is similarly set.

In step 4-3, the row counter CL.

The value in the CTRL table determined by the column counter CC is stored in the register DL.

In step 4-4, the value of the register DL is increased by 2 by 1.

In step 4-5, the data indicated by the value of register L in the DTBL table is stored in register fiJcO.
Store in DB. A word code will be stored in this register JCODE.

In step 4-6, the dictionary is searched for the word code O stored in the register JCODE. - In step 4-7, it is determined whether or not to write the searched word in the dictionary to the search word table 8WDK.

If the reading data is the same as that registered in the DTBL, it is registered and the process moves to step 4-8; otherwise, step 4-1lIIC@ru〇In step 4-8,
The searched word is registered in the search word table.

In step 4-9, the data in register JCODF; is increased by 1.

In step 4-10, it is determined whether the dictionary has been searched to the end, and if the dictionary has been searched to the end, the process returns to step 4-11. Otherwise, step 4-7
Wc move.

In step 4-11, the column kakunta CC of the CTRL table is amplified by 2.

In step 4-12, one connected string is checked to the end to determine whether or not they are friends.

If it has not been completed yet, go to step 4-3; if it has been completed, go to step 4-13.

In step 4-13, the value of the column counter CL is incremented.

In step 4-14, it is determined whether all lines of the CTRL have been checked.

If completed, the series of tasks of registering the word in the search word table is completed. If you haven't done so yet, go to step 4.
-Move to 2.

That’s 14! Finish the O movement.

Next, other embodiments of the present invention will be described.

That is, in the embodiment described above, the reading information of the dictionary is limited to 8 characters, and the join table has a maximum of 5 connections.
Although only reading data and word codes are registered in the word table DTBL, the number can be increased as necessary.

As is clear from the above description, according to the present invention, for input character data strings, wt-reading elements are separated by round-robin or rule, and are registered in the dictionary. The word code of the first word registered in the data and the first dictionary is O, and the word code of the first word registered in the data is O, which is registered in the DTBL table.
To create a join table, as mentioned above, only the words necessary to create a compound word are registered in the search word table based on the %DTBL table and the join table, so the search word table will not overflow and will not contain any kana-kanji. We were able to realize a character processing device with significantly improved conversion efficiency.

[Brief explanation of the drawing]

FIG. 1 is a diagram illustrating a divided string of input character data strings based on the on-yomi reading. FIG. 2 is a diagram illustrating the overall configuration of the present@mei character processing device. FIG. 3 is a flowchart sequentially illustrating the S states of its operation. FIG. 4 is a detailed flowchart of step 8-1O in FIG. 8 marked *1+. Figure 5 is a diagram showing a DTBL table in which the word code of one of the words corresponding to the reading data is registered, Figure 16 is a diagram showing the form of the dictionary in which words are registered, and Figure 7 is a combination. A diagram showing the form of a table - FIG. 8 is a diagram showing the configuration of a search word buffer 1 that stores input characters and data strings. 5... Keyboard 6... Keyboard code/draw 27... CRT controller 8.・CR
T9...Character generator 10...Refresh, upper memory 11...Refresh memory controller) o-912...
・Read-only memory (R, OM) 13... ROM controller 5 14... Random assignment" 18... Flora kettle drive controller 19... Printer 20... Printer controller applicant Canon Co., Ltd. Otsu 8 !Roasted=゛4 7-2

Claims (1)

[Claims] A means for dividing the input data string into each sound wL to create a 9-on-yomi index; a 9-dictionary memory for storing words having on-yomi elements; and a base K1 for the on-yomi elements from the means for removing the on-yomi elements.
a search means for searching for 1i-ki dictionary memo'JO 薊KI dansho;
◇ A character processing device comprising: a storage means for storing one word for each of the reading elements;