JPS61190653A - Document processor - Google Patents

Abstract

PURPOSE:To facilitate an editing operation in clause units and to improve the function by determining a hierarchical structure in accordance with headers and holding this structure. CONSTITUTION:Document data designated by an input device is inputted to a document data storage part 7. Candidates of a header and the start position of the header of document data inputted to the document data storage part 7 are obtained by the processing of a header candidate finding part 8 and are inputted to storage parts 9 and 10. Data inputted to storage parts 9 and 10 are sent to a header determining part 11, and it is discriminated whether data is the header or not and the classification of the header is discriminated on the basis of data in a discrimination rule storage part 12 where a discrimination rule is written. When all header data cells are generated, the processing of a document structure determining part 14 is started, and a hierarchical structure determining rule storage part 15 is used to link header data cells into the hierarchical structure.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to a document processing device that generates and maintains a logical structure in the form of document data.

[Technical background of the invention and its problems]

In order to make a document easier to read, the entire document is divided into multiple ranges, each group is given a short sentence called a heading at the beginning of the group, and the divided group is further divided, It is common to have a structure in which each item has a heading. Headings generally have a heading symbol at the top of the heading to indicate the scope of the heading and its relationship with other headings.

For example, ``Chapter 1'' and ``Section 3'' are examples of heading symbols used here. When a computer attempts to process a document with such a document structure, it does not take such document structure into account at all, and due to device constraints, it is likely to be processed by another unit unrelated to the document structure. For example, when editing a document that has a document structure, such as a unit of size that can be displayed on the display screen at one time, or a unit of size that fits on one page when printed on paper, the above restrictions apply. This makes it “difficult to use” for users.

For example, when replacing chapters, with conventional equipment,
If it is necessary to indicate the movement range using a cursor or the like to indicate the beginning and end of a chapter, and if the chapter is large, the display screen must be updated many times until the beginning and end of the chapter are specified. Therefore, the specified position is often mistaken.

However, with a device that has a document structure, it is possible to easily specify the desired chapter, so the complicated specification operations described above are no longer necessary, and operability is thought to be considerably improved.

Furthermore, even if the document was created by someone else, by yourself, or if you have created it before and have forgotten its contents, conventional devices can only tell you the contents of the document file, such as the file name. First, it is often difficult to infer the content of a document from there, but if you can see just the heading of the document at once, it is easier to infer the content of the document.

In other words, the conventional apparatus has the disadvantage that it is difficult to realize more advanced document processing using the structure of a document.

[Purpose of the invention]

The present invention was developed in view of the above-mentioned circumstances, and an object of the present invention is to extract a logical structure in the form of a document and construct a hierarchical structure.

[Summary of the invention]

The present invention makes it possible to extract heading candidates from document data written in code information, determine the relationship between the headings from the symbol part of the heading, and determine the hierarchical structure of the headings. It is.

To give a specific example t-6, a character string whose length is equal to or less than - digits is extracted from the beginning of the sentence in a character string separated by line feed codes, and "Chapter -" is written at the beginning of the character string.

A character string containing a heading symbol such as r(3N rOJ) is used as a heading.

When the heading symbol contains numbers and alphabetic characters, the hierarchical structure of the heading currently being processed is determined depending on the font, order, and format, and when only symbols are included, the hierarchical structure of the heading currently being processed is determined based on the relationship between the front and back. This is done by creating a data structure with a logical structure between the top rankings and headings.

〔Effect of the invention〕

According to the present invention, the document to be processed is
If the book has a hierarchical structure, the hierarchical structure is determined and maintained from the headings, so by using that data structure, it becomes easier to edit chapters and sections, improving operability. In addition, there are more advanced functions that utilize hierarchical structures, such as creating a table of contents list, printing regularly indented chapters and nine sections to make the document easier to read, and identifying errors in heading symbols. It is easy to change the format of text, heading symbols, etc. In general, the document data input to this device does not need to have a hierarchical structure, so document data created with conventional document processing devices can be processed with this device, and new documents can be created with this device. Even in operation, there is no need to be aware of the hierarchical structure (the same operations as in the past are required, and there is no additional burden on the operator).

[Embodiments of the invention]

Please refer to all the drawings below.
The figure shows an overall block diagram of an embodiment of the invention. Here, input device 1. Output device 29 Display device 3. External storage device 4. The internal storage device 5 is one that has been widely used in document processing and will not be specified.

The block diagram of FIG. 2 shows the control device 6 in detail.

From the external storage device 4 containing the C1 document file shown in FIG. 1, a file with the name specified by the input device 1 is sent to the document data storage section in the internal storage device 5.

The document entered in the document data storage section 7 is sent to the heading candidate detection section 8
Through the process, heading candidates and starting positions of the headings are obtained and stored in the respective storage units 9.10. The entered data is sent to the next heading determining section 11, where it is determined whether this data is a heading or not, and if so, a judgment rule storage section in which the type of heading and its judgment rule are written. Processed based on 12 data. The processed data is called a heading data cell, which is created by writing various pieces of information regarding the heading, and is stored in the heading data cell storage section 13.

The processing of the heading candidate discovery section 8 and the heading judgment section 11 is as follows:
Continue until all documents currently being processed have been processed.

When all the heading data cells have been created, the processing moves to the document structure determination section 14, where the hierarchical structure determination rule storage section 15 is used to connect the heading data cells into a hierarchical structure, and all processing is completed.

Next, the operation of each part will be explained in detail.

A block diagram of the heading candidate finding section 8 is shown in FIG.

According to instructions from the read position system # section 16, the document data storage section 7
The characters are sequentially sent one by one to the comparison section 17, and at the same time, the position of each character in the document is stored in the address storage section 18.

On the other hand, the data sent to the comparing section 17 is compared with the line feed code stored in the register 1. The data is also simultaneously sent to the line buffer 19 and stored there, and the number of characters is counted by a counter 21. The comparison unit 22 compares the value of the counter 21 when a line feed code appears in the data with the value stored in the register 2 as a predetermined length of one line (for example, 40 characters), and the counter When the value matches, the judgment unit 20 enters. If the line feed code matches at the same time or precedes, this data is determined to be a heading candidate, and the data in the line buffer is stored in the heading candidate data storage unit 10, and the address The value of the counter is subtracted by the calculation unit 23 from the value of the storage unit, and the start address of the line buffer unit is determined from this value. This is stored in the header candidate data start position storage unit 9, and the line buffer,
The counter is returned to its initial state and the process of this heading discovery section is repeated again.

On the other hand, if the counter value matches the one-line sentence length value first, it is determined that the data in the line buffer section cannot be a heading, and the line buffer and counter are returned to their initial states. Return it and repeat the process of this heading discovery section again.

When the document data is completed during processing, the processing is handed over to the header determination section 11.

Next, a block diagram of the heading determination section is shown in FIG.

The judgment rule storage unit 24 contains rules for determining headings expressed using regular expressions as shown in FIG. This is applied by the determination rule application control unit 25, and if it is successful, it is determined that it is a heading, and if it is not successful, it is determined that it is not a heading.

In Figure 5, the symbols enclosed in circles are called non-terminal symbols, and the symbols enclosed in squares are terminal symbols and represent the character code data itself. Each rule states that a symbol that is not enclosed in a circle or square on the left is replaced by a nonterminal or terminal symbol on the right. Each symbol is applied in order in the direction of the arrow, and arrows above and below indicate that the upper rule will be applied first, and if the upper rule is not successful, the lower rule will be applied. Then, when these rules are applied and the non-terminal symbols of C and C are replaced with terminal symbols, this applied rule will be successful, and if the replacement does not go well in the middle, the application will not be successful. It turns out.

For example, input the document shown in Figure 6 and enter "s4" in the 5th line.
w w w l, Introduction -J (where 口 means a blank code) is input into the heading determination section,
Applying the rules as shown in FIG. 7 is successful and is determined to be a heading. Any other application will not be successful.

In this way, there are many ways of application, so if any one way of application is successful, it is considered a heading, and if all the ways of application are unsuccessful, it is not considered a heading.

When it is determined that the child is a child, the heading data cell creation unit 26
Processing will proceed to. Here, the heading data cell as shown in Figure 8? Create one for each heading. Then, according to the heading determination rule, write the result of the successful running method in the cell. In addition, for headings to which numeric parts and alphabetic parts are applied, in the case of numeric parts, the a value is called the order, and in the case of the alphabetic parts, the numeric value is called the order in the alphabetical section, and is entered there. For example, the heading symbol is "Chapter 2"
In the case of ゛, the order is 2. In the case of rcoJ, it is 3.

This creation section determines the portion λ in FIG. 8 of the header data cells. Such data cells are collected and stored in the data cell storage section 27.

For example, in the case of the heading shown in the example in Figure 7, there is no prefix, the number part is the number A (Figure 5 (e)), and the order is 1.
．． It turns out that the suffix is "," and the start position of this heading is entered in the heading candidate data start position storage unit 9 °C
Since there is, we will also move this to the data cell. With the above operations, a data cell is completed as shown in FIG. 9. If it is determined that it is not a heading, no data cell is created and this heading candidate data and data at the start position are discarded.

The above process is repeated until there is no more input data, and when there is no more input data, the process is advanced to the document structure determining section 14.

Finally, the document structure determining section 14 will be explained in detail.

Here, the heading data cells created in the previous step are connected to the hierarchical structure.

The processing procedure is shown in FIG. 10, FIG. 11, and FIG. 12.

In the case of the document shown in FIG. 6, which will be explained using an example, five cells are created as shown in FIG. 13.

First, enter the first cell "1. Introduction".

Then, according to the flow shown in FIG. 10, the first cell is connected as a child of the root cell, which is provided in advance. In other words, enter the root address where you write the start address of the parent cell of the first cell, and the start address of the next older cell,
The first child cell's starting address and error flag are currently not applicable, so enter a special value that means it is not pointing anywhere. And the floor number is 1. That is 1
Let's assume it's the top child.

Next, input the second cell "20 Features of the Present Invention".

The current cell (the 1'tr cell) is an alphanumeric character with no prefix, and if it is a numeric part, the last part is "," Order 1 The next cell (2nd cell) is the same except for the order 2. Therefore, according to the rules in Figure 12, the case becomes ■, so
Let the second cell be the older brother of the first cell. In other words, 1
Write the start address of the second cell in place of the start address of the next older cell in the second cell, and write the start address of the parent cell of the first cell into the start address of the parent cell of the second cell. Write the same address (that is, the address of the root cell) that is in the cell, write 1 for the rank, which is the same as the rank of the existing cell, and enter unapplicable values for the rest, and set C.

Next, input the third cell "(1) Prior art".

Compared to the second cell, which is the current cell, this is different in prefix, number 4, and postfix, and the current order is 1. Therefore, in the case of ■ in FIG. 12, this third cell is made a child of the second cell, and the rank is increased by one to make it 2. Then use the pointer to connect them as before. If you input the fourth cell "(2) Explanation of % characteristics" and perform the same process, you will get the third older cell.

If you input the 5th cell "3. At the end", the case of ■ in Figure 12 will occur, and the rank will be reduced by 1 to 1, which is the parent of the 4th cell, that is, the 2nd cell "20 Features of the present invention" As a result of the processing, it becomes the older brother of this cell, so we connect this fifth cell as the older brother of the second cell.

When all the cells are processed, a hierarchical structure as shown in FIG. 13 is completed, and the processing is completed.

In this process, if the header symbol is incorrect, it makes some guesses and attempts to connect the hierarchical structure.

For example, after "Chapter -", instead of "Chapter 2", say "Chapter 3".
In other words, if the order is flying,
If the order is the same, such as "Chapter - Chapter" followed by "Chapter - Chapter", this is assumed to be a continuation of the order, and an error flag is set on the cell and linked as the older cell in the same hierarchy. do. Then, the original correct order is written in a predetermined location of the cell, and subsequent processing is continued using this value. Also, if r(1)J is followed by rC2:lJ, that is, the shape is different but the order continues, and the parent does not have that shape, the shape is wrong. Assuming that the two are used as older brothers, set an error flag and add the first form to the cell.

Even with the above error processing, if there is a cell whose hierarchy is unknown, an error flag is set to indicate that the cell is the current sibling, and the shape and order of the older cell are inherited.

In the above-mentioned embodiment, all heading candidates are found and judged from C whether they are headings or not, and the hierarchical structure is determined after all the judgments are completed. It is also possible to carry out the process repeatedly, such as performing all the processes from starting the process to determining the hierarchical structure, and once the process is completed, finding heading candidates again.

In addition, in the heading candidate determination section, sentences whose length is one digit are not selected as heading candidates, but all sentences separated by sentence delimiting codes such as line feed codes are checked. Assuming it as a dashi candidate, the next processing section is,
The heading determining section determines the heading and "C" is also good.

In addition, in the above example, the hierarchical structure of the obtained headings is stored as a data structure by having the start addresses of the heading cells of its own parent, th child, and Xth child. However, the data island structure may be realized in other ways.

Furthermore, the contents of each heading cell are not limited to this example,
For example, you could tie the text that follows the heading to this cell, but that is fine.

This is not limited to heading hierarchy, but can also be applied to data that has a hierarchical structure, such as an organization chart, by changing the data of rules such as heading determination rules and document structure determination rules. It is.

This example deals with Japanese documents, but it is not limited to Japanese, but documents in other languages can also be used if they have headings.
Judgment Rule 1 If the hierarchical structure determination rule is rewritten for the Japanese language, it can be implemented in the same way.

[Brief explanation of drawings]

FIG. 1 is an overall configuration diagram of this device. FIG. 2 is a detailed configuration diagram of the control section. FIG. 3 is a configuration diagram of a heading candidate discovery section in the control section. FIG. 4 is a configuration diagram of the heading determining section in the control section. FIG. 5 is a diagram showing a regular expression of the judgment rules stored in the judgment rule storage section of the heading determining section. FIG. 6 is a diagram showing an example of a document having a document structure to be input to the present apparatus. FIG. 7 is a diagram showing an example of how the determination rule of FIG. 5 is applied to one of the heading candidates in the document example of FIG. 6. Figure 8 is a data acquisition diagram of the heading data cell. FIG. 9 is a diagram showing an example of how one heading in the document example of FIG. 6 is expressed in a data cell. FIG. 10, FIG. 11, and FIG. 12 are diagrams showing processing operations in a document structure determination section in the control section. FIG. 13 is a diagram showing an example of how the heading data cells of the document example in FIG. 6 are connected through the document structure determining section. 1... Input device, 2... Output device, 3... Display device. 4...external storage device, 5...internal storage device, 6...
·Control device. 3. Attorney's Attorney ±1 Rule Chika M Yu (1 person) Figure 1 Figure 7 Figure 3 Figure 4 Figure 5 Figure 5 Figure 5 Department: No.
6 Figure 10 Figure S Figure 11 Figure 9 Figure 12 Figure 13

Claims (1)

[Claims] an input means for inputting document data written in code information; an extraction means for extracting heading candidates from the document data input from the input means based on the length of a sentence measured by a line feed code, etc.; means for extracting a symbol at the beginning of the heading candidate for each of the heading candidates extracted by the extracting means, determining and storing a hierarchical structure such as an inclusion relationship, an ordering relationship, etc. of the heading candidate based on the format of the symbol; A document processing device characterized by: