JPH05101040A - Document processor - Google Patents

Abstract

PURPOSE:To easily prepare a layout template and to partially utilize an existent layout template again by generating partial layout structure by applying a partial layout template, which is corresponding to certain partial logical structure in a document, to this logical structure. CONSTITUTION:A partial layout template holding means 105 holds the partial layout template showing rules for laying-out the partial logical structure of the document in a logical structure holding means 101. In respect to the partial logical structure designated by a partial logical structure designating means 108, a partial layout means 109 executes a partial layout processing by applying the template corresponding to that logical structure. Assuming that the partial layout means 109 completes the partial layout processing to certain partial logical structure and further executes the partial layout processing to the other partial logical structure, at such a time, this partial layout means 109 executes the partial layout processing according to the retroactive call of the own means, and executes the layout processing of the entire logical structure of the document.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document processing apparatus capable of obtaining a document layout structure by applying a template, which is a rule for layout, to a logical structure of a document.

[0002]

2. Description of the Related Art In general, both logical expansion and actual output layout are important elements of a document.
However, the layout is not always important in the process of creating a document. For example, the layout may be indispensable at the stage when the document is completed, but the layout may not be so important at the initial stage of document creation in which the logical development is not clear. In addition, there are cases where documents with different layouts are required in the same logical development, such as when there are multiple submission destinations.

Such characteristics of a document have been pointed out in the past, and a method has been proposed in which a structure representing a layout called a layout structure is generated from a structure representing a chapter structure of a document called a logical structure. The process of generating the layout structure from the logical structure is called layout process, and is performed by a program (allocation process program) built in the document processing system.

In order to generate various types of allocation structures from the same logical structure, the operation of the allocation program must be changed. Generally, modifying the program itself is a task that requires very specialized knowledge, and is not something that a normal user can do. For this reason, there has been adopted a method of changing the operation of the program for performing the layout process by parameters such as "page size is A4" and "double column".

However, in this method, the types of parameters that can be specified are limited to those assumed by the program in advance. Further, in order to control the operation of the allocation program having a high function, it is necessary to instruct a large number of parameters, which is also disadvantageous in that it cannot be used by ordinary users.

In order to solve this problem, therefore, a method has been proposed in which the allocation program is controlled not by a simple parameter but by a data structure representing an allocation template. One of them is the international standard ODA (IS
O8613 = Information Processing-Text and Office
Systems-Office Document Architecture (ODA) and Int
erchange Format (1989)) is a common layout structure (Generic Layout Structure).

The ODA shows a data structure representing a document structure and a guideline for using the data structure, and does not describe an actual allocation process. However, OD
In order to realize the allocation process based on A, it is clear that the following functions are required. Hereinafter, the allocation process having such a function will be referred to as an ODA allocation process.

(1) Allocation processing function based on an allocation template (common allocation structure). (2) Allocation processing (top-down and bottom-up) selection function. (3) Reuse of allocation results. (4) Allocation processing function according to category.

The functions (mechanisms) of the above (1) to (4)
Is not used alone, but is used by combining a plurality of functions as necessary, and thereby the efficiency of allocation processing is achieved.

Next, the above-mentioned functions (mechanisms) will be sequentially described with reference to specific examples.

<< Allocation Template (Common Allocation Structure)
38 shows the logical structure and the allocation template (common allocation structure) based on the ODA.

In FIG. 38, the document logical structure 3810.
Each of the elements that make up is called a logical object, and includes "logical root", "1 chapter", "related article", and "1".
Chapter ", 8" rectangles filled with left diagonal lines "
Are logical objects. The eight “rectangles filled with left diagonal lines” are “A”, “B”, “C”, “a”, “b”, in order from the leftmost rectangle.
The contents of “c”, “D”, and “E” are linked. Here, each element (content) constituting the document content will be referred to as a content section. The logical object holding the content part in this way is called a basic logical object.

Each element constituting the allocation template (corresponding to a common allocation structure in ODA) 3820 is called an allocation object class, and "document template", "page" and "text frame" are allocation object classes respectively. Is. Each allocation object class has an attribute "Generator for subordinates" (attribute G
FS) has a structural formula (subordinate generator). This structural formula is a formula that expresses a structural constraint that can be connected immediately below an allocation object that belongs to the allocation object class that has the structural formula. The term SEQ, the term CHO, the term REP,
It is composed of a combination of structure generators such as the term OPT, the term OPTREP, and the term AGG, and “object class identifier” (class ID). Note that in FIG. 38, "re
p ″ is a structure generator. In the figure, 3820
Reference character a denotes a layout image represented by the layout template 3820. In this image, a solid line frame indicates a page and a dotted line frame indicates a text frame.

The logical object of the "logical root" of the logical structure 3810 specifies the allocation object class of the "document template" of the allocation template 3820 (this is the allocation root). As a result, the allocation template 3820 can be applied to the logical structure 3810 to perform the allocation process.

Here, when the allocation process based on the allocation template 3820 is performed on the logical structure 3810,
For example, the allocation result shown in FIG. 39 (a) is obtained.

In FIG. 39 (a), the layout structure 390.
The elements that make up 0 are called layout objects. The "layout root", the three "pages", the three "text boxes", and the eight "rectangles with diagonal grid fill" are the layout objects. It is an object. 8
The two "rectangles with diagonal grid fills"
In order from the leftmost rectangle, "A", "B", "C",
The content parts of "a", "b", "c", "E", and "D" are connected. The layout object holding the content part in this way is called a block.

FIG. 39 (b) shows a layout image represented by the layout result shown in FIG. 39 (a). In this image, pages P1 to P3 are shown in FIG. 39 (a). It corresponds to the substructure (layout structure) of the layout objects P1 to P3 of the “page” of the layout structure 3900, respectively.

Since each block of the final document allocation structure holds the content part, the logical part 3810 of the document and the document allocation structure 3900 share the content part as shown in FIG. Corresponding.

Each allocation object of the allocation structure holds the following information as attributes.

That is, (1) Two-dimensional position from the higher-level allocation object. (This information is held by the attribute "position" in the above ODA) (2) Information on the size of oneself. (This information is the above OD
In A, it is held with the attribute "dimensions". (3) An allocation object that is not a block is an identifier or pointer of an allocation object that is connected to a lower level.
(This information is held by the attribute "subordinates" in the above ODA.) (4) In the case of a block allocation object, the identifier or pointer of the content to be connected. (This information is the above OD
In A, it is held with the attribute "content portions")
Is held as an attribute.

Next, the outline of the processing by the conventional allocation processing program will be described. (1) Generate a minimum allocation structure that satisfies the restrictions of the allocation template. (2) An allocation object (block) corresponding to a logical object (basic logical object) holding contents is generated. Hereinafter, this processing will be referred to as content allocation processing. (3) The generated block is arranged at the allocation node (the lowest frame) corresponding to the lowest node of the allocation template. Hereinafter, this processing will be referred to as pouring. (4) If necessary, change the layout structure within the range allowed by the layout template. For example, when the area for arranging blocks in the area of the lowest frame is exhausted, the number of pages is increased to generate the lowest frame.

Note that the content allocation and pouring processing is performed by generating the target allocation allocation object and changing the information held by the object. In the case of content allocation, the connection of the content part and the size of itself are performed. Is changed, and in the pouring, the connection to the higher-level allocation object and the change of the position of itself are performed.

The processing by such an allocation processing program will be described more concretely with reference to the flowchart of FIG.

As shown in FIG. 41, the allocation processing program generates a minimum allocation structure that satisfies the restriction of the allocation template (step 1). Here, the generated one is referred to as "LayStruct". Next, it is judged whether or not all the basic logical objects have been allocated (step 2), and if all have been allocated, "LayStruct" is returned as a result. On the other hand, if the allocation is not completed, the allocation is not completed. The first basic logical object is fetched (step 3) and the content allocation process is called. This called content allocation process creates a block (step 4). Here, the basic logic object taken out is "LogNode" and the generated block is "LayObject". The allocation processing program which has completed the above step 4 determines whether or not there is a sufficient area for pouring "LayObject" in the lowest frame of "LayStruct" (step 5). If it remains, "LayObject" is poured into the lowest frame of "LayStruct" (step 6). On the other hand, if not
The structure permitted by the allocation template is added to "LayStruct" to generate a new lowest frame, and then the process proceeds to step 6 above. After step 6, after step 2 above
Return to and execute the steps after this step.

Next, the logical structure 381 shown in FIG.
Generation of a block and filling of the block into the lowest frame when the allocation process is performed by applying the allocation template 3820 to 0 will be briefly described.

For example, layout template 3820 results in a minimum layout structure 4200 as shown in FIG. Next, when content allocation processing is performed on the basic logical object holding the content part A, a block 4320 corresponding to the basic logical object 4310 is generated as shown in FIG. 43 (a). Next, when the filling process is performed, the block 4320 is arranged below the lowest frame (that is, the body frame) 4330. This is shown in FIG.
As shown in (b), it means that the content portion A is poured into the area of the body frame.

Next, when content allocation processing is performed on the basic logical object holding the content part B, FIG.
A block 4420 corresponding to the basic logical object 4410 is generated as shown in FIG. Next, when the filling process is performed, the block 4420 is arranged in the lower order of the lowest frame 4330. This means that as shown in FIG. 44 (b), the content portion B is poured into the empty area of the text frame.

Subsequently, in order to allocate the content part C, a content allocation process is performed on the basic logical object holding the content part C. As shown in FIG. Corresponding block 4520
Is generated, but the allocation fails if the pouring process is performed on this block 4520. This is because, as shown in FIG. 45 (b), there is no free area in the area of the text frame, and it is impossible to pour the content portion C into it. In such a case, the allocation structure may be changed within the restrictions of the allocation template. For example, when the allocation to the content section C fails, the allocation template 3820 (see FIG. 38) restricts the allocation structure as shown in FIG.
As shown in (a), the new lowest frame (text frame)
4610 can be generated. When the filling process is performed on the lowest frame 4610, the content part C that has already been generated is generated at the lower level of the lowest frame 4610.
A block 4520 corresponding to is arranged. As shown in FIG. 46 (b), this is in the text frame of a page (next page) different from the page in which the content parts A and B are poured,
It means that the content part C is poured.

By the way, for related articles not directly related to the flow of the text, for example, when it is desired to lay out the text in a two-column set so as to be distinguished from the text, for example, the layout template 3820 shown in FIG. 38 and the layout template 4700 shown in FIG. You can change to. 4700a is a layout image represented by the layout template 4700. As shown in FIG. 47, the logical object of the “related article” of the logical structure 3810 points to the allocation object class of the “small frame” of the layout template 4700, so that the “related article” is poured into the structure below the “small frame”. Will be done. This designation is made by the designation of the "layout object class" attribute in the ODA.

When the allocation process based on the allocation template 4700 is performed on the logical structure 3810 shown in FIG. 47, the allocation result shown in FIG. 48 (a) is obtained. In FIG. 48A, the layout structure 4800 includes “layout roots”, three “pages”, three “print face frames”, three “text frames”, one “small set frame”, and one “left frame”. Each layout object is composed of one “right frame” and eight “rectangles with diagonal grid filling”.
FIG. 48B shows a layout image expressed by the layout result shown in FIG. 48A. In this image, pages P1 to P3 are shown in FIG.
The layout structure 4800 corresponds to the images represented by the substructures of the layout objects P1 to P3 of the “page” of the layout structure 4800 shown in FIG.

Here, a brief description will be given of the processing when a two-column set of "related articles" is poured. In this case, FIG.
By applying the allocation template 4700 to the logical structure 3810 shown in (a) and performing the allocation process, the allocation process for the basic logical object holding the content parts A and B ends, and then the content part In the allocation process for the basic logical object holding C, as shown in FIG. 49A, the block 4 corresponding to the basic logical object 4910 holding the content part C is used.
920 is generated. Allocation template 4700
Due to the above restriction, a printing plate frame having a body frame and a sub-set frame is generated on a new page. Then, in the filling process, as shown in FIG. 49A, a block 4920 is arranged below the text frame 4930. This is shown in FIG.
As shown in (b), it means that the content portion C is poured into the empty area of the body frame.

Next, when content allocation processing is performed on the basic logical object holding the content part a, FIG.
A block 5020 corresponding to the basic logical object 5010 is generated as shown in FIG. When the pouring process is further performed, the block 5020 is arranged below the lowest frame (left frame in this case). This is shown in FIG.
As shown in (b), it means that the content part a is poured into the empty area of the left frame. Similarly, by performing the allocation process on the content parts b and c, the content of the page to be processed is. Page P2 shown in FIG. 48 (b)
become that way.

<< Selection Function of Allocation Process (Top Down or Bottom Up) >> In the ODA, for example, the allocation template 3820 is applied to the logical structure 3810 shown in FIG. 38 to perform the allocation process. In this case, the following events may occur.

(1) When the amount of a certain content part is large and the content does not fit in the generated lowest frame. (2) When a certain content part is a chart or the like, and the content cannot fit into the generated lowest frame, and the chart cannot be divided. However, even when such an event occurs, the allocation is made by taking the following measures.
That is, at the time of the above event (1), that is, when the contents do not fit in the lowest frame, a plurality of blocks are generated for one basic logical object. This processing means that when the block is generated during the content allocation processing, the size of the area of the block is determined to be top-down from the remaining area of the upper allocation object.

For example, it is assumed that the allocation result shown in FIG. 51 is obtained when the allocation process is performed on the logical structure 3810 (see FIG. 38). In this case, corresponding to the basic logical object holding the content part B of the logical structure 3810,
In the layout structure 5100 shown in FIG.
Two blocks 5110 and 5120 are generated so that the content portions B1 and B2 obtained by dividing the above are respectively poured. FIG. 51B shows a layout image represented by the allocation result shown in FIG.

On the other hand, at the time of the above-mentioned event (2), that is, in the case where it is not necessary to divide in the middle like the chart, the size of the block area is changed from the area required for allocating the chart to the bottom-up. It is determined.

For example, suppose that the allocation result shown in FIG. 52 is obtained when the allocation process is performed on the logical structure 3810. In this case, the content part B that composes one chapter of the logical structure 3810 is a “figure”, and as a result of performing the allocation process for the basic logical object holding the content part “figure”, the content part “figure” ”Did not fit in the area of the text frame on the previous page.
As shown in the layout structure 5200 shown in (a), it is arranged below the text frame 5210 on the next page. Figure 5
2B shows the layout image represented by the allocation result shown in FIG.

In the above-mentioned ODA, which of the bottom-up and top-down allocation processing is to be selected, since the target of the flow is the block, for example, the text is top-down, and the chart is bottom-up. I am trying to distinguish by the type. In this way, the size of the area of the block to be generated is determined to be top-down in the case of text, and bottom-up in the case of chart. The content will be poured into the area of the block thus determined. Since the block is generated by the content allocation process, the fact that the block is divided into a plurality of blocks means that the content allocation process is applied the number of times of the blocks.

<< Reuse of Allocation Result >> In the ODA, the allocation structure is generated by applying the allocation process to the logical structure. Therefore, when the logical structure is changed, the allocation structure that reflects the change in the logical structure cannot be obtained unless the allocation process is performed again. In a large document, it is inefficient to perform reallocation whenever the logical structure changes.

Although the above ODA does not directly touch on the solution of this problem, as shown in FIG. 40, each basic logical object of the logical structure 3810 and the allocation structure 390 are shown.
Each block of 0 is associated by sharing the content part. Therefore, it is clear that the efficiency of reallocation can be improved by using this correspondence. That is, a method of reallocating only the part of the change in the logical structure is conceivable. For example, when the content part C of the logical structure 3810 shown in FIG. 40 is changed, the reallocation may be performed from the content part C. That is, the content section "C",
It is sufficient to re-allocate the basic logical objects corresponding to "a", "b", "c", "D", and "E". Note that the content section may be divided, so that FIG.
The correspondence relationship between the basic logical object and the block is not always one-to-one like the correspondence relationship shown in FIG.

<< Allocation processing function according to category >> Further, in the allocation processing based on the ODA, there is a function of selecting the lowest frame into which the basic logical object is poured. By assigning a category to each of the basic logical object and the lowest frame, the basic logical object is poured into the lowest frame in which the category matches. As for the category, in the above ODA, the “layout category” attribute is
The "permitted categories" attribute is designated for the lowest frame.

For example, as shown in FIG. 53, logical structure 5
When only the “related article” portion 5311 of 310 is to be passed to the subsequent page, the logical object of “logical root” may specify the layout template 5320. 5320a is an allocation template 5320
Is the layout image expressed by.

As described above, the common allocation structure in the ODA has an excellent feature that various allocation structures can be generated without adding / changing a program.
By retaining the common layout structure and reusing it, it is possible to allocate a plurality of logical structures with a common layout. Also, by enabling top-up and bottom-up allocation processing and allocation processing according to categories, various layouts for the logical structure are possible. Furthermore, the backtrack mechanism is used to improve the efficiency of reallocation.

[0044]

However, the above-described conventional allocation processing system has the following problems in the functions (mechanisms) (1) to (4).

<< Allocation template (common allocation structure)
Allocation processing function based on the above >> In the above conventional allocation processing method, a new allocation template must be prepared in order to allocate a logical structure with a layout different from that specified by the existing allocation template. There wasn't. For example, when it is desired to set the sub-assemblies to three stages instead of two, for example, the allocation template shown in FIG. 54 (a) is prepared in place of the allocation template 4700 shown in FIG. 47, and either two stages or three stages are prepared. However, when adopting such a layout, it is necessary to prepare the layout template shown in FIG. Note that FIG.
54B shows a layout image represented by the layout template shown in FIG.

As described above, in the above-described conventional allocation processing method, the allocation processing for the entire logical structure is controlled by using the allocation template for the entire allocation structure. Therefore, the allocation template for the entire document is prepared before the allocation processing. Must. Therefore, even if the layout templates are partially common, different layout templates must be prepared as a whole by combining them, and a new layout template must be created for each slight layout change. Is very complicated,
It took a lot of work. On the other hand, it is extremely difficult to prepare an allocation template that combines all possible layout possibilities, and it also complicates the structure and reduces the efficiency of the allocation process. ..

<< Allocation Process (Top-Down and Bottom-Up) Selection Process >> In addition, in the above-described conventional allocation process method, the selection of the bottom-up / top-down allocation process was determined only by the contents. That is, if the content is text, top-down allocation is always applied, and if it is a figure, bottom-up allocation processing is always applied.

By the way, if a table is considered as the content, for example, it can be generally divided. Therefore, it may be better to allocate the whole page to the next page when it does not fit in the empty area of the page, or it is better to fill the contents of the table as much as it fits in the empty area and fill the overflowed content to the next page. Sometimes it's good. In this way, bottom-up and top-down allocations may not be judged based on the type of content.

However, in the above-mentioned conventional allocation processing method, it is not possible to select (designate) the top-down allocation or the bottom-up allocation for the same content type.

<< Reuse of Allocation Result >> Further, in the above-mentioned conventional allocation processing method, since the allocation processing is executed for the entire logical structure of the document, when a part of the logical structure is changed, That is, when a certain content part is changed, the allocation process (re-allocation) for the basic logical objects after the content part must be performed.

Here, for example, when obtaining the allocation result shown in FIG. 48 (a), focusing on the fact that the structure below the "related article" is not directly related to the structure of the text, the "related article" of the logical structure is When the following parts are changed, "Related articles"
It should be sufficient to divide only the following parts. On the other hand, when only a part of the text that is not related to the "related article" is changed, it is not necessary to reallocate the parts below the "related article", and it is possible to correspond to the "related article". All you have to do is move the allocation structure below the small frame.

However, in the above-mentioned conventional allocation processing method, for example, in the logical structure 3810 shown in FIG.
For example, when the content “C” which is a part of the text is changed, the content portion “C”, “a”, “b”, “c”, “D”,
If the basic logical object corresponding to "E" has to be reallocated, and if the content "a" which is a part of the related article is changed, the content parts "a", "b",
The basic logical objects corresponding to “c”, “D”, and “E” must be reallocated, and the efficiency of the reallocation process is poor.

<< Allocation Processing Function According to Category >> Further, in the above-described conventional allocation processing method, since only the block can be specified in a category and poured into a specific lowest frame, originally, the entire related article is displayed on the back page. I had to give each category to a basic logical object, even though I wanted to turn it into. For this reason, when the basic logical object is transferred from another part in the editing work, it is necessary to adjust the category given to the transferred logical object. In this way, management of categories in units of basic logical objects is complicated, and when adjustment of categories fails, only part of them is assigned unexpectedly, and the desired layout is obtained. I couldn't do that.

An object of the present invention is to provide a document processing apparatus which can very easily and easily create an allocation template and can partially reuse an existing allocation template. ..

Another object of the present invention is to provide a document processing apparatus which can easily obtain a desired document layout.

A further object of the present invention is to provide a document processing device capable of improving the reallocation processing efficiency.

[0057]

In order to achieve the above object, the document processing apparatus of the present invention corresponds to each partial logical structure when the logical structure of a document is classified according to a predetermined relationship. Designated by the partial division template holding means for holding the partial division template showing the rule for allocating the structure, the partial logical structure designation means for designating a predetermined partial logical structure in the logical structure, and the partial logical structure designation means. The partial logical structure is applied to the partial logical structure by applying the partial partitioning template corresponding to the partial logical structure.

Further, in the above invention, the allocation processing of the entire logical structure is performed by repeatedly or recursively executing the allocation processing by the partial partitioning means.

Further, in the above invention, the partial logical structure designating means extracts a logical node string composed of a basic logical object holding document contents and a logical object representing a lower logical node string from the logical structure. And a logical node sequence management unit for managing the logical node sequence extracted by the logical node sequence extraction unit.

Further, in the above invention, the partial division attaching means determines the area of the partial division structure obtained as the partial division result from the remaining area of the upper layout object to top-down, and the partial division result. A bottom-up part dividing means for determining the area of the part-dividing structure obtained as a bottom-up part from a region in which lower-level layout objects are combined, and any of the top-down part dividing means or the bottom-up part dividing means. Section selection means for selecting a part division means based on allocation means selection information regarding whether to apply.

Further, the document processing apparatus of the present invention corresponds to each partial logical structure when the logical structure of the document is classified according to a predetermined relationship, and shows a rule for allocating the partial logical structure. With respect to the partial logical structure designating means for designating a predetermined partial logical structure in the logical structure, and the partial logical structure designated by the partial logical structure designating means. A partial partitioning unit that performs the partial partitioning process by applying the partial partitioning template corresponding to the structure, and the partial logical structure and the partial partitioning structure as the partial partitioning processing result by the partial partitioning processing unit are managed in association with each other. It is equipped with logic and allocation management means.

Furthermore, the document processing apparatus of the present invention corresponds to each partial logical structure when the logical structure of the document is classified according to a predetermined relationship, and shows a rule for allocating the partial logical structure. With respect to the partial logical structure designating means for designating a predetermined partial logical structure in the logical structure, and the partial logical structure designated by the partial logical structure designating means. A plurality of partial partitioning means for performing the partial partitioning processing by applying the partial partitioning template corresponding to the structure, and performing the processing by each of the partial partitioning means in parallel to perform the allocation processing of the entire logical structure. It is characterized by

[0063]

According to the document processing apparatus of the present invention, the partial partitioning means applies the partial partitioning template corresponding to the partial logical structure designated by the partial logical structure designating means to the partial logical structure. When performing the allocation process, ending the partial partitioning process for a certain partial logical structure, and executing the partial partitioning process for another partial logical structure, the partial partitioning process is performed by a recursive call (recursive call of the partial partitioning means). By executing it, allocation processing of the entire logical structure of the document is performed. As described above, since the partial layout template is applied in units of partial logical structure, when changing the layout of an existing document, only the existing partial layout template corresponding to the layout to be changed is changed, Other existing templates with partial divisions can be reused.

Further, the subdivision means may apply the corresponding subdivision template to the logic node sequence managed by the logic node sequence management means and execute the subdivision process. It can be carried out.

Further, the partial division assigning means executes the partial division processing by either the top-down partial division attaching means or the bottom-up partial division attaching means selected by the partial division attaching selection means on the basis of the allocation means selection information. .. In this case, the logical structure is allocated top-down if the top-down part dividing means executes the part-dividing processing, while the bottom-up part dividing means executes the part-dividing processing, and the logical structure is the bottom structure. Assigned to up. Therefore, a desired document can be created by explicitly designating the sentence processing allocation method according to the content of the logical structure.

According to the document processing apparatus of the present invention,
As described above, the partial partitioning means performs the allocation processing of the entire logical structure of the document by recursively executing the partial partitioning processing, and the logical / allocation management means uses the partial logical structure and the partial partitioning processing means. Since the partial allocation structure as the allocation processing result is managed in association with each other, the partial correspondence relationship between the logical structure and the allocation structure becomes clear. Therefore, for example, when the logical structure is changed, only the allocation structure corresponding to the partial logical structure including the changed part needs to be re-allocated, so that the re-allocation process can be performed quickly.

Further, according to the document processing apparatus of the present invention, the allocation processing of the entire logical structure is performed by executing the partial allocation processing by the plurality of partial allocation means in parallel. It can be carried out.

[0068]

Embodiments of the present invention will be described below with reference to the accompanying drawings.

A first embodiment of the present invention will be described with reference to FIGS.

FIG. 1 shows a first embodiment of the document processing apparatus according to the present invention.
Is a functional block diagram, and the apparatus of the embodiment shown in this figure is specifically applied to an apparatus having a document processing function, such as a word processor, a workstation, or a computer.

In the figure, the document processing apparatus is obtained by applying a logical structure holding means 101 for holding the logical structure of a document, a content holding means 102 for holding the content of the document, and an allocation process to the logical structure. An allocation structure holding means 103 for holding an allocation structure, and a content allocating means 104 for generating an allocation object (that is, a block) having contents to be arranged in the lowest frame from a logical object having contents (that is, a basic logical object). A template holding unit 105 for holding a template with partial division (details will be described later) and a logical node sequence extracting unit 106 for extracting a logical node sequence (details will be described later) from the logical structure.
And a partial logical structure designating unit 108 including a logical node sequence management unit 107 that manages the extracted logical node sequence.
And a partial partitioning means 109 for allocating a logical node sequence for one layer based on the rule of the partial partition template.

Means 101, Means 102, Means 1 described above
03, means 105, and means 107 are respectively realized as storage devices such as a hard disk and a main memory.
The means 104, means 106, and means 109 described above are
It is realized by a control means such as a processor or a central processing unit executing software (program) for performing the function of each means. Of course, each of these means may be realized by configuring with hardware or firmware.

FIG. 2 shows an example of the logical structure of a document, document contents, and a template with copy division.
10 is the logical structure holding means 101, document content 230 is the content holding means 102, and the templates 240, 2
50 are respectively held by the template holding means 105 with copy division.

The logical structure 210 is a logical object 211.
.. 222, and in each logical object,
Logical objects 213-215, 217-219, 2
21 to 222 are called basic logic objects. Each basic logical object holds a corresponding content part among the content parts 231 to 238 forming the document content 230 (actually, the basic logical object points the content part by a pointer).

The subdivision templates 240 and 250 are rules for allocating a logical node sequence for one layer respectively. The subdivision template 240 is composed of allocation object classes 241-243, and the subdivision template 250 is It is composed of allocation object classes 251 to 253. Allocation object class 241 (document template) that is the allocation root
Is a logical object 2 that is the root of the logical structure 210
11 (logical root), and the allocation object class 251 (small frame template) that is an allocation root is specified from the logical object 216 (related article). Also, the allocation object classes 243, 252, 25
Reference numeral 3 is an allocation object class for generating the lowest frame. In addition, the template with each division is the above OD.
In A, it is expressed as a partial structure of the common allocation structure.
By the way, the template 240 with partial division shown in FIG.
The allocation template 3820 shown in FIG. 8 is reused. In FIG. 2, 240a is a layout image represented by the partial layout template 240, and 250a is a layout image represented by the partial layout template 250.

In this embodiment, among the logical objects forming the logical structure, the basic logical object and the logical object designating the template with partial division are called logical nodes. In addition, a basic logical object or a logical object is also described as needed. The logical node is the partial division assigning means 109 or the content assigning means 1
It will play the role of designating the activation of any one of 04.

Next, in this embodiment, when calling the partial division attaching means 109 for executing the partial division processing, the logical node extraction means 106 extracts a logical node sequence from the logical structure of the document as a preprocess, and this logical node sequence is extracted. Are managed by the logical node sequence management means 107 as information for indicating the range of the logical structure to be processed. The logical node sequence clarifies the range to which the partial layout template is applied. When there is a logical node in the lower structure of a certain logical node sequence, the upper logical node
The lower logical node sequence is regarded as a nested logical node sequence. Since the root of the logical structure is specified with a partial partitioning template for the entire logical structure, if logical node groups exist below the root, those logical nodes form a logical node sequence nested in the logical node of the root. To do. The logical nodes forming the logical node sequence further nest lower logical nodes, and finally nest a logical node sequence composed of basic logical objects having contents that are the leaves of the logical structure.

Here, FIG. 3 shows a logical node sequence obtained from the logical structure 210 shown in FIG. In FIG. 3, the logical node sequence has two layers, that is, a logical node sequence forming the entire logical structure and a logical node sequence forming related articles. First nested in the highest logical node 211 (logical root)
The logical node sequence of the layer is a logical node (basic logical object) 213 to 215 having content parts A, B, C, D and E,
221, 222, and a logical node (logical object) 216 indicated by "related article". The logical node sequence of the first layer is a portion to which the partial layout template 240 (actually, the "document template" that is the layout root is associated) associated with the highest logical node is applied. Allocation is done. The logical node 216 (related article) further includes logical nodes (basic logical objects) 217 to 21 having content parts a, b, and c.
The second layer logical node sequence consisting of 9 is nested. The logical node 216 specifies the partial layout template 250 (actually, the "small frame template" which is the layout root is specified), but this means that the nested logical node sequence of the second layer is Allocation template 250
Is applied and the partial division is performed.

The logical node sequence manages the logical objects to be subjected to the parting process or the content allocating process, and not all logical objects in the logical structure are logical nodes. For example, in the example of FIG.
The logical objects appearing in the logical structure 210 shown in FIG. 2 (logical objects of “Chapter 1” and “Chapter 2”) are not included in the logical node sequence. That is, even if the logical objects of Chapters 1 and 2 logically play the role of classification, they are not reflected in the allocation result.

Here, the data of the logical structure, the template with partial division, and the logical node sequence for performing the allocation process at the time when the logical node sequence as the preprocessing is extracted and managed (the state before the partial division process). The structure is shown in FIG.

In FIG. 4, reference numerals 211 to 222 are logical objects constituting the logical structure 210 shown in FIG. 2, reference numerals 231 to 238 are content parts constituting the document content 230 shown in FIG. 2, and 240 and 250. Is the template with partial divisions shown in FIG.

As shown in FIG. 4, in the data structure of the logical structure 210, the upper logical object points to the lower (immediately below) logical object, and each basic logical object points to the content part connected to itself. Have respectively. Further, the logical object (logical node) 211 has a pointer to the template with partial division 240, and the logical object (logical node) 216 has a pointer to the template with partial division 250.

Further, in FIG. 4, reference numerals 431 to 438 are the same as those in FIG.
It is a node corresponding to each logical node constituting the logical node sequence shown in FIG. This is an example of a data structure as a means for realizing a logical node sequence. The correspondence between each logical node and each node is as follows.

That is, as shown in FIG. 4, the node 431 corresponds to the logical node 211 and the node 432 corresponds to the logical node 2.
13, the node 433 corresponds to the logical node 214, the node 434 corresponds to the logical node 215, the node 435 corresponds to the logical node 216, the node 436 corresponds to the logical node 217, and the node 437 corresponds to the logical node. Two
18, the node 438 corresponds to the logical node 219, the node 439 corresponds to the logical node 221, the node 440 corresponds to the logical node 222, and each node has a pointer to the corresponding logical node. ing. In addition, the logical node 2 specifying the template 240
The node 431 corresponding to 11 is the node 432, 43.
A logical node 2 which has a pointer for nesting the node sequence of the first layer consisting of 3, 434, 435, 439, and 440, and which designates the template with partial division 250.
The node 435 corresponding to 16 has nodes 436 and 43.
It has a pointer for nesting the node sequence of the second layer consisting of 7, 438.

As means for implementing the logical node sequence, besides the structure having the structure shown in FIG. 4 as the data structure, the logical structure is interpreted each time it is needed without reflecting it in the data structure, and The column information may be extracted, that is, the logical node column may be virtually present.

For example, when the allocation process for the logical node 211 which is the root of the logical structure in FIG.
A logical object 212 (which is not a logical node) that is a subordinate of
Logical node (basic logical object) 2 that is subordinate to 2
This means that 13 to 215 may be recognized and a logical node sequence (nodes 432 to 434 shown in FIG. 4) corresponding to the recognized basic logical objects may be held as internal information.

Further, while the allocation processing is being performed on the logical nodes (basic logical objects) 213 to 215, the logical node extracting means 106 causes the logical node 211 to operate.
Of the logical nodes 216 to 219 which are lower than the logical node 216, and the logical nodes 217 to 219 which are lower than the logical node 216 are recognized. A sequence of logical nodes corresponding to a logical object (node 435 shown in FIG. 4, nodes 436 nested in this node).
438) may be included as internal information.

As another implementation means, it is conceivable to configure a document processing device that holds a logical node sequence as a logical structure instead of the normal logical structure. That is, in the above-mentioned example, the content shown in FIG. 2 (content before extracting the logical node sequence) is created, and the content shown in FIG. 3 (content after extracting the logical node sequence) is obtained from this content. After that, Fig. 3
Although the allocation process is performed by referring to the contents shown in FIG. 3, the contents shown in FIG. 3 (configuration contents in which the logical node sequence exists, that is, the contents shown in FIG. This means that the allocation processing may be performed by creating (structure contents in which the logical objects of “Chapter 1” and “Chapter 2” do not exist).

In any means, it is important that the subdivision template and the application range of this template are managed so as to be extracted as a sequence of hierarchical logical objects.

Now, FIG. 5 shows a layout structure of a document obtained by applying the template with partial division shown in FIG. 3 to the logical structure shown in FIG.

In FIG. 5A, the layout structure 510.
Are allocated objects 511 to 528, and in particular, allocated objects 514, 515, 518, and 5
Reference numerals 21, 522, 524, 527 and 528 are blocks. Each block (allocation object) is connected to the corresponding content part. FIG. 5B shows a layout image represented by the layout result shown in FIG. 5A, page P1 is a layout image corresponding to the structure below the layout object 512 (page), and page P2 is A layout image corresponding to the structure below the layout object 517 (page), and a page P3 is a layout image corresponding to the structure below the layout object 526 (page). The data structure of the allocation structure 510 shown in FIG. 5A has the contents as shown in FIG. 6, the upper allocation object is a pointer to the lower allocation object, and each block is a content part connected to itself. Each has a pointer to.

Therefore, the logical structure 210 shown in FIG. 2 and the allocation structure 510 shown in FIG. 5A are associated with each other via the contents section. That is, the content part is referred to by the logical structure and the allocation structure.

In the example shown in FIG. 6, the basic logical objects and the blocks have a one-to-one correspondence, but a plurality of allocation objects (blocks) may be generated from one basic logical object. It is possible. At this time, the content part is divided by the content allocating means 104.

Next, the generation of the layout structure by the partial layout unit 109 will be described.

The subdivision assigning means 109 is obtained by invoking the content allocating means 104 or by calling the subdivision assigning means 109 (recursive invocation) for the logical nodes constituting the logical node sequence for one layer. The partial division structure that is used is poured into the lowest frame. That is, in the past, it was done only for blocks,
This is a process in which the process of pouring allocation nodes into the lowest frame is extended to frames.

By the way, as shown in FIG. 6, a logical node that nests a logical node sequence always indicates a template with partial division. When the partial-partitioning unit 109 is activated for the partial-partitioning template designated by a certain logical node and the logical node sequence in which the logical node is nested, the partial-partitioning unit 109 performs the process described below. ..

(1) If the logical nodes forming the logical node sequence are basic logical objects, the content allocating means 104
To call the block generated by the content allocating means 104 into the lowest frame of the layout structure generated from the partial layout template.

(2) When the logical nodes forming the logical node sequence point to the template with partial division, the logical node sequence nested in the logical node sequence and the partial division with the logical node instructed With respect to the template, self (partitioning unit 109) is recursively called to perform the parting process, and the frame obtained as a result of this execution is poured into the lowest frame of the already generated layout structure.

Here, the outline of the partial division processing by the partial division attaching means 109 will be described. The partial division attaching means 109 refers to the logical node sequence management means 107, and selects the highest logical node sequence among the logical node sequences managed here. The allocation process is sequentially executed from the first logical node of the logical node sequence of the upper layer. Here, the top logical node of the logical node sequence of the highest layer is a root logical object in the logical structure, and the logical object specifies a template with partial division corresponding to the entire logical structure. Therefore, by applying the partial division unit 109 to the root logical node, the partial division unit 109 is sequentially applied to the logical node sequence nested in the root logical node, and finally the content allocation unit 104 is To be launched.

Next, the process of adding the partial division by the partial division attaching means 109 will be described with reference to the flowchart of FIG.

As shown in FIG. 7, the copy division means 109
Generates a minimum structure with partial division that satisfies the constraint of the template with partial division designated by the root logical node (step 701). Here, the generated structure with partial division is referred to as "LayStruct".

Next, it is judged whether or not all the logical node sequences have been allocated (step 702), and if all have been allocated, "LayStruct" is returned as a result. On the other hand, if the allocation has not been completed, From the logical node sequence, the first logical node whose allocation is not completed is taken out (step 703). Here, the extracted logical node is called "LogN
ode ”.

Then, it is checked whether or not this "LogNode" is a basic logical object (step 704), and if it is a basic logical object, the content allocating means 104 is called. The called content allocating means 104 generates an allocation object (block) (step 705). On the other hand, if it is not the basic logical object in step 704, the logical node sequence lower than “LogNode” and “LogNode”
With respect to the template with partial division specified by, the self (partial division processing 109) is called (recursive call) to generate an allocation object (frame) (step 706). The generated block and frame are referred to as “LayObject”. Note that step 7
In 06, the recursively called parting means 1
In the same manner as the processing of step 701, “09
The minimum structure with partial division that satisfies the constraint of the template with partial division specified by "is generated.

Now, step 705 or step 7
The section dividing unit 109 that has finished 06 is "LayStruct".
It is judged whether or not there is a sufficient area for pouring "LayObject" in the lowest frame of "LayObject".
"" Into the lowest frame of "LayStruct" (step 708).
uct "is added with a structure permitted by the template with partial division in step 701 (that is, the template with partial division specified by the root) to generate a new lowest frame (step 709), and the process proceeds to step 708. After completion of step 708, the process returns to step 702 and the steps after this step are executed.

Next, the partial division adding process by the partial division adding means 109 will be described again with reference to FIG. 4 as a specific example.

For example, in FIG. 4, the partial division attaching means 109 refers to the head node 431 of the node group corresponding to the logical node sequence, then to the logical node 211 corresponding to the node 431, and further to the logical node. Reference is made to the template 240 with partial division designated by 211 (see FIG. 3). Then, the minimum structure with partial division that satisfies the constraint of the template 240 with partial division is generated.

Next, partial partitioning processing is sequentially performed on the logical nodes corresponding to the nodes 432, 433, 434, 435, 439, 440 in which the node 431 is nested. At this time, the template 240 with partial division is applied to those logical nodes. But node 4
In the collation processing for 35, the node 435
Nests the nodes 436, 437, 438, and the logical node 216 corresponding to the node 435 specifies the partial layout template 250 (see FIG. 3). Therefore, the allocation process for those nodes is performed for the nodes 439, 440. It is performed prior to the allocation process. At this time, the node 43
The template with partial division 250 is applied to the logical nodes corresponding to 6, 437, and 438.

Now, the processing for the node 434 is completed and the processing for the node 435, that is, the processing for partial division by the partial division attaching means 109 when performing frame insertion will be described.

First, the copy division adding means 109 determines that the node 4
35, the logical node 21 corresponding to the node 435.
From the restriction of the template with partial division 250 (see FIG. 3) designated by No. 6, the minimum structure with partial division 810 is generated as shown in FIG. Next, at the time of partial division processing for the logical node (basic logical object) 217 corresponding to the node 436 in which the node 435 is nested,
This content allocation means 1 is called by calling the content allocation means 104.
The block 521 generated by 04 is arranged in the lowest frame. This is because the generated block is
As shown in (b), it is meant to be poured into the lowest frame of the partial division structure 810. Similarly,
When the process with partial division is executed for the nodes 437 and 438 in which the node 435 is nested, the process result 910 with partial division as shown in FIG. 9A is obtained.

Since the template 240 with partial division is applied to the processing result 910 with partial division, the processing result 910 with partial division is the node 431 as shown in FIG. 9A.
To 434 are arranged in the lowest frame 921 (text frame) in the structure with partial division 920 obtained by applying the template with partial division 240. This is the processing result with partial division 910 (that is, a frame (small set frame))
, As shown in FIG. 9 (b), means that it is poured into the lowest frame.

In the conventional allocation process, the process of pouring a frame into the lowest frame as described above cannot be performed.

Then, the processing for the node 435 is completed, and the nodes 439 and 4 nested in the node 431 are terminated.
By performing the allocation process for 40, the entire logical structure shown in FIG. 3 is allocated, and the allocation result shown in FIG. 5 is obtained. In the allocation structure 510 shown in FIG. 5, the allocation structure 48 shown in FIG.
The plate frame at 00 is the text frame. This is because the body frame is treated as the lowest frame in the partial division process for the first layer logical node sequence, and the left frame and the right frame are treated as the lowest frame in the partial division process for the second layer logical node sequence. This is because they are treated.

As described above, the partial division attaching means 109 activated for the root of the logical structure and the partial division attaching means 1
The entire logical structure is allocated by means of the partial division assigning means (recursive call) 109 and the content allocating means 104 which are indirectly activated by 09.

The logical objects 212 and 222 (that is, "Chapter 1") in the logical structure shown in FIG.
For the logical object of “Chapter 2”, if the corresponding template with partial division is described, the logical node sequence is configured as shown in FIG. In the example shown in FIG. 10, the logical node 211 is the logical nodes 212, 216, and 2.
A logical node sequence consisting of 20 is nested, a logical node 212 is nesting a logical node sequence consisting of logical nodes 213, 214, 215, and a logical node 2
20 nests a logical node sequence composed of logical nodes 221 and 222, and further, logical node 212,
Reference numeral 220 designates a chapter division template (chapter frame template) 1000 for chapter frames. Note that 1000a is a layout image represented by the partial layout template 1000.

Next, a second embodiment of the present invention will be described with reference to FIGS. 11 to 17.

FIG. 11 is a functional block diagram showing the second embodiment of the document processing apparatus according to the present invention. This functional block diagram is the same as the functional block diagram of the first embodiment shown in FIG. 1, except that the section division attaching unit 109 is changed to the section division attaching unit 1100. In the figure, the same reference numerals are given to the parts that perform the same functions as the components shown in FIG.

The partial division attaching means 1100 determines the area of the partial division attaching structure to be top-down from the remaining area of the upper allocation object.
And a bottom-up part dividing means 1120 for deciding the area of the part dividing structure from the area obtained by combining the lower-level layout objects to bottom-up, and selecting one of the respective part dividing means based on allocation means selection information described later. And a selection unit 1130 with partial division.

Note that the template holding means 10 with partial division is used.
In addition to the functions described in the first embodiment, reference numeral 5 is an allocation means selection as to which of the top-down part division attaching means 1110 and the bottom-up part division attaching means 1120 is applied to the division attaching template. Information is stored, and in this embodiment, a value indicating "top down" and "bottom up" are used as the allocation means selection information.
Holds a value that indicates.

Here, the copy division selection means 1130 is
When the value is "top down", the top down part dividing means 1110 is selected, and when the value is "bottom up", the bottom up part dividing means 1120 is selected. ..

By the way, in the above-described first embodiment, in the partial division adding process by the partial division adding unit 109 (see the flowchart shown in FIG. 7), a frame (small subframe) is added from the area in which the lower layout objects are combined to the bottom-up. )
Has determined the size of. Therefore, as shown in FIG.
When the allocation processing is performed by applying 50, when the remaining area (empty area) of the lowest frame is not sufficient, that is, when there is no room for the small set frame to enter the remaining area of the body frame, FIG. The allocation result is as shown in a). Figure 1
2 (b) shows a layout image represented by the allocation result shown in FIG. 12 (a).

If the size of the frame (small set frame) is decided to be top-down instead of bottom-hoop, the allocation result as shown in FIG. 13 (a) is obtained. FIG. 13B shows a layout image represented by the allocation result shown in FIG.

That is, also in the frame pouring to the lowest frame according to the present invention, similarly to the block pouring to the lowest frame by the conventional allocation processing,
There are two methods, bottom-up and top-down.

Next, the allocation process of bottom-up allocation and top-down allocation when the frame is poured will be described with reference to FIGS. 14 to 16.

FIG. 14 is a diagram for explaining a bottom-up layout when the frame is poured, and FIGS. 15 and 16 are diagrams for explaining a top-down layout when the frame is poured.

[Bottom-Up Allocation] 1) A minimum partial division structure is generated. At this time, the image of the layout of the frame (subset frame) represented by the structure with partial division has the content shown in FIG.

2) Next, the block having the content portion "a" is poured. The layout image in the frame (small set frame) at the time when the content portion "a" is poured into the small set frame is shown in FIG.
The contents are as shown in (b). As shown in FIG. 14 (b), the content portion "a" is poured into the left frame and the right frame, but this should be poured so as to be balanced in the region of the left frame and the right frame. This is because

3) Subsequently, the block having the content portion "b" is poured. A layout image in the frame (small set frame) at the time when the content portion “b” is poured into the small set frame is shown in FIG.
The contents are as shown in (c).

4) Further, the block having the content portion "c" is poured. FIG. 14 shows a layout image in the frame (small set frame) at the time when the content portion “c” is poured into the small set frame.
The contents are as shown in (d).

However, at the time when the content portion “c” is poured, the area of the frame (subset frame) is the remaining area (empty area) of the body frame (bottom frame) of a page,
In other words, since the size exceeds the fillable size, the generated frame (subset frame) cannot be poured into the lowest frame, and as a result, the body frame of the newly generated page (lowest frame). Will be poured into.

This corresponds to a case where the sub-set frame is poured into the text frame of the second page in the layout image shown in FIG. 12B, but the free space is insufficient, and the small frame is poured into the third page. ..

[Top-Down Allocation 1] 1) A minimum partial division structure is generated. At this time, the structure with partial division is set to the maximum size that can be poured, as shown in FIG. The maximum size that can be poured is the uppermost lowest frame (text frame)
Corresponds to the remaining area (empty area).

2) Next, the block having the content portion "a" is poured. The layout image in the frame at the time when the content portion "a" is poured into the frame (small set frame) is shown in FIG.
The contents are as shown in (b). As shown in FIG. 15 (b), the content "a" is poured into the left frame and the right frame, but it is poured so as to be balanced in the region of the left frame and the right frame. This is because.

3) Then, the block having the content portion "b" is poured. When the content portion "b" is poured into the frame, and the content portion "b" cannot be poured into the frame, FIG.
As shown in (c), the frame is poured into the uppermost lowest frame (text frame). This corresponds to the case where the frame (subset frame) divided into the second page is poured in the layout image shown in FIG. That is, in this pouring, a part of the content part “b”,
This is the case where the entire content section "c" could not be poured.

However, in the case of top-down, the remaining contents of the small frame are allocated to the next page, so the allocation will be described below.

[Top-Down Allocation 2] 1) A minimum partial division structure is generated. At this time, the structure with partial division is set to the maximum size that can be poured, as shown in FIG. The maximum size that can be poured corresponds to the area (empty area) of the newly generated uppermost and lowermost frame.

2) Next, the above "top-down allocation 1"
The block having the content part "c" and the part of the content part "b" that was not flowed in is poured. The layout image in the frame (subset frame) at the time when a part of the content part “b” and the entire content part “c” are poured is
The contents are as shown in FIG.

3) Subsequently, the frame is set to the minimum size, and this frame is poured into the uppermost and lowermost frames as shown in FIG. 16 (c). This corresponds to the case where the divided frame (subset frame) is poured into the third page in the layout image shown in FIG.

By the way, in the conventional allocation process,
Since the object to be flowed was a block, it was possible to determine whether to select bottom-up or top-down allocation processing according to the type of content part, for example, top-down for text and bottom-up for charts. .. However, in the present invention, since the object to be poured is also applied to the frame, since the frame does not directly hold the content part, it is determined whether to make the bottom-up or the top-down based on the content part. You cannot do it.

Therefore, in the second embodiment, the subdivision template retains either the value indicating "top down" or the value indicating "bottom up" as the allocation means selection information. ing. Of course, when the value is not set, the value may be set in advance as the default value.

Then, the partial division selection means 1130 refers to the partial division template at the time of the allocation process, and further recognizes the value (or default value) specified in this partial division template, and as a result of this recognition, "top down". If it is a value indicating, the top-down section dividing means 1110 is selected, and if it is a value indicating “bottom-up”, the bottom-up section dividing means 1120 is selected. As a result, the top-down part dividing means 111
0 executes the top-down allocation process, while the bottom-up part dividing means 1120 executes the bottom-up allocation process.

As described above, for example, as shown in FIG. 2, by clearly indicating which process of the top-down layout process or the bottom-up layout process is applied to the partial layout template and performing the layout process based on this information. Even if the lower structure of the logical object 216 (related article) of the logical structure 210 is exactly the same, it is possible to explicitly select whether to perform the bottom-up allocation process or the top-down allocation process. .. Also, for example, when a table is represented by a combination of frames, the table is
When the remaining area of a certain page is not entered, it is possible to divide the table and allocate the table over two pages, or to allocate the entire table to the next page.

Next, details of the bottom-up and top-down allocation processing will be described. Since the allocation processing operation in the bottom-up processing is the same as the flow shown in FIG. 7 described in the first embodiment, Here, the description of the processing operation is omitted. It goes without saying that the processing is performed by the bottom-up part dividing / attaching means 1120.

Next, the top-down part dividing means 111
The top-down partial division processing by 0 will be described with reference to the flowchart in FIG.

Now, as shown in FIG. 17, the top-down part division attaching means 1110 generates the smallest part division attaching structure which satisfies the constraint of the part division attaching template designated by the logical node of the root of the frame (step 1701). ..
Here, the generated structure with partial division is referred to as "LayStruct".

Next, the size of the frame that is the root of "LayStruct" is set to the size of the remaining area of the upper allocation object of that frame, that is, the size of the remaining area of the uppermost lowest frame into which "LayStruct" should be poured. (Step 1702) After that, it is determined whether or not all the logical node sequences have been allocated (step 1703).

Here, if all are not assigned,
The first logical node whose allocation has not been completed is extracted from the logical node sequence (step 1704). The extracted logical node is called "LogNode".

Then, it is checked whether or not this "LogNode" is a basic logical object (step 1705), and if it is a basic logical object, the content allocating means 104 is called. The called content allocating means 104 generates an allocation object (block) having a size that can be poured into the lowest frame of the root frame (step 1706). On the other hand, if it is not a basic logical object in step 1705, the subdivision assigning means 1100 is called for the subordinate logical node sequence of "LogNode" and the subdivision template specified by "LogNode" to assign an allocation object (frame). It is generated (step 1707). Here, which of the bottom-up and top-down allocation means is selected (executed) is determined by the partial division selection means 1130 by "LogNode".
It is determined based on the selection information of the allocation template designated by ". The generated block and frame are each" LayObject ".

Incidentally, in the above step 1707, the top-down part division adding means 1110 recursively called.
Is the same as the processing in step 1701 above.
Will generate the smallest structure with partial divisions that satisfies the constraint of the template with partial divisions specified by.

Furthermore, the top-down part dividing means 111 that has completed step 1706 or step 1707.
In the case of 0, "LayObject" is poured into the lowest frame of "LayStruct" (step 1708). After that, “LogNod
It is checked whether or not all the lowest logical node sequences or contents of e "have been allocated (step 1709). If there is any allocation yet, step 1 above.
Returning to step 703, executing this step and subsequent steps, on the contrary, when all are allocated, the size of the frame that is the root of "LayStruct" is set to the minimum size that can hold all the allocation objects that make up "LayStruct". Now, return that "LayStruct" as a result (Step 1
710). In addition, in step 1703, when all the logical node sequences are allocated, the above step 1
Proceed to 710.

As described above, in the present embodiment, the top-down part division processing by the top-down part division means 1100 (see the flow of FIG. 17) is performed when a structure with a frame root is generated. Only applicable in processing. The bottom-up part dividing means 112 is the part dividing process corresponding to the root of the entire logical structure.
A bottom-up allocation process of 0 (see the flow of FIG. 7) is performed. Only the bottom-up part dividing unit 1120 takes charge of the function of adding the layout structure such as pages to generate the lowest frame. When the top-down part dividing / attaching means 1110 is used and all the target logical node sequences cannot be flowed into one lowest frame, the top-down portion / dividing means 1110 is used.
Passes the result (partitioning process result) to a higher-order parting process (actually, the higher-level logical object becomes the parting process by the parting means 1100), and ends the parting process. Therefore, the delivery of the processing result with partial division ends up in the bottom-up partial processing. A new lowest frame is generated by the bottom-up subdivision means 1110, and the subdivision means 1100 is activated again for a logical node for which allocation has been completed halfway. That is, with respect to the overflowed logical node, the partial division processing using the same partial division template is activated.

In the above ODA, etc., "variable" is assigned to the allocation template in order to determine the size of the frame.
Alternatively, a “fixed” parameter is allowed, and the allocation process activated by this parameter value is switched. In the case of "Variable", the instruction is for top-down allocation processing, but in the case of "Fixed", it means that the size is fixed regardless of the lower-level allocation object, and the bottom-up allocation pointed out in this embodiment is used. It is completely different from the processing instruction. Also in the present invention, “fixed” can be added as the type of parameter.

Next, a third embodiment of the present invention will be described with reference to FIGS.

FIG. 18 shows a third example of the document processing apparatus according to the present invention.
Is a functional block diagram of this embodiment. This functional block diagram is the same as the logical / allocation correspondence holding unit 180 in the configuration of the functional block diagram of the first embodiment shown in FIG.
It has a configuration in which 0 is added. In the figure, the same reference numerals are given to the parts that perform the same functions as the components shown in FIG.

The logical / allocation correspondence holding means 1800 has a partial logical structure in the logical structure to which the partial division processing is applied,
The correspondence relation with the partial division structure corresponding to this partial logical structure is held. For example, as indicated by the dotted line in FIG. 19, the correspondence relation between “logical root” and “allocation root”, “related article”. It holds information indicating the corresponding relationship between each of the sub-frames and the “small frame”.

FIG. 19 shows the logical structure 210 shown in FIG.
And the allocation structure 510 shown in FIG. 5 as the allocation processing result for this logical structure 210 shares the content part, thereby showing the correspondence relationship between the logical structure 210 and the allocation structure 510. Of course, the logical node 211
(Logical root) is the template 240 with partial division shown in FIG.
Is designated, and the logical node 216 (related article) designates the template with partial division 250 shown in FIG. Also,
Each content part has the same content as each content part shown in FIG.

Therefore, the logical / allocation correspondence holding means 1800
For example, in FIG. 19, like the correspondence relationship between the logical node 211 (logical root) and the allocation node 511 (allocation root) and the correspondence relationship between the logical node 216 (related article) and the allocation node 519 (small group frame), It holds a correspondence relationship between a logical object (logical node) that specifies a template with copy division and a layout object corresponding to this logical object.

Here, paying attention to the logical structure 210 and the allocation structure 210 shown in FIG. 19, when allocation is performed, the structure under the logical node 216 (related article), that is, the partial logical structure 1920A is the structure of the main body. It can be seen that it is not directly related to the partial logical structure 1910A. In this case, the lower part of the related article on the logical structure (partial logical structure 1
920A), only the lower part of the related article (structure with partial division 1920A) should be partial divided. On the contrary, when only a part of the text (partial logical structure 1910A) that is not related to the related article is changed, it is not necessary to reallocate the logical nodes under the related article, and the related article is dealt with. It should be sufficient to move the layout structure (structure with partial division 1920B) below the small frame.

Therefore, the logical / allocation correspondence holding means 1800
Holds the correspondence relationship between the logical node 211 (logical root) and the allocation node 511 (allocation root), and the correspondence relationship between the logical node 216 (related article) and the allocation node 519 (small frame), respectively, and When the 1910A is changed, only the partial logical structure 1910A needs to be reallocated, and when the partial logical structure 1920A is changed, only the partial logical structure 1920A needs to be reallocated.

FIG. 20 shows the data structure of the contents shown in FIG. 19. This data structure has the partial logical structure and the partial division structure in the data structure of the first embodiment shown in FIG. It has a structure in which data (hereinafter, referred to as logical / allocation correspondence data) indicating the correspondence relation of is added.

In the figure, the logical / allocation correspondence relationship data 2 managed by the logical / allocation correspondence holding means 1800 is stored.
Reference numeral 010 is a pointer to the node 431 (node corresponding to the logical node 211) in the logical node sequence held in the logical node sequence management unit 107, and the allocation structure holding unit 1
Allocation node 511 in the allocation structure held in 03
The logical / allocation correspondence data 2020 is composed of a pointer to (allocation root), and the logical / allocation correspondence data 2020 is a node 435 (a node corresponding to the logical node 216) in the logical node sequence held in the logical node sequence management means 107. ) And a pointer to the allocation node 511 (small set frame) in the allocation structure held in the allocation structure holding means 103.

Next, with reference to the flow charts of FIGS. 21 and 22, a description will be given of the operation of the partial division processing in consideration of the reallocation by the partial division unit 109.

As shown in FIG. 21, the part division attaching means 109
Determines whether or not there is a structure with partial division corresponding to the higher logical node in the logical node sequence (step 2101).

Here, if there is a structure with partial division, the structure with partial division corresponding to the upper logical node is "LayStruct".
(Step 2102), on the other hand, if there is not, a minimum partial-partitioned structure that satisfies the constraint of the partial-partitioned template specified by the upper logical node is generated, and this generated partial-partitioned structure is defined as “LayStruct”. (Step 21
03).

Next, when step 2102 or step 2103 is completed, it is checked whether or not there is a logical node whose contents or the connection relation with the lower logical node is changed (step 2104). The first logical node in the changed logical nodes is referred to as "FirstNode".

If there is a changed logical node, the allocation object corresponding to the changed logical node is removed (step 2105), and then "FirstNod
Separates the allocation objects corresponding to the logical nodes existing in the logical node sequence of the same level or higher level from the logical node after e ”(the allocation object itself is not removed).
(Step 2106) Further, even if the lower layout objects are no longer connected and are deleted, all the layout objects that do not conflict with the restrictions of the partial layout template are removed (step 2107).

Next, when step 2107 is completed,
Alternatively, when there is no changed logical node in the above step 2104, it is checked whether or not all the logical node sequences have been allocated (step 2108). here,
If all logical node sequences are assigned, "LayStruct
Is returned as a result (step 2109).

If all are not allocated in the above step 2108, the subdivision assigning means 109, as shown in FIG. 22, does not have the first logical node whose allocation is not completed from the logical node sequence, that is, the corresponding allocation object. Or, the first logical node in which the corresponding allocation object is separated from the higher-level allocation object is taken out, and that logical node is designated as "LogNode" (step 2110), and whether or not that "LogNode" is the basic logical object. (Step 211)
1).

Here, if it is a basic logical object,
It is checked whether or not there is a block corresponding to "LogNode" (step 2112). If it exists, the corresponding block is set to LayObject (step 2113). On the other hand, if it does not exist, the content allocation unit 104 is called and the allocation object (block) generated by the content allocation unit 104 is set to "LayObject". (Step 21
14).

If it is not a basic logical object in step 2111, the self (partitioning means 109) is called (recursively) with respect to the logical node sequence subordinate to "LogNode" and the template with subdivision specified by the "LogNode". Call) to obtain an allocation object (frame) (step 2115). This obtained frame is "LayO
bject ”.

When any of step 2113, step 2114, or step 2115 is completed, "LayStruct
It is checked whether or not there is a sufficient area remaining for the LayObject to be filled in the lowest frame of "(step 211).
6).

[0171] Here, if there is any, "LayObjec
"t" is poured into the lowest frame of "LayStruct" (step 2117), while if there are no more "LayStruct", "LayS"
The structure allowed by the allocation template is added to "truct" to generate a new lowest frame (step 211).
8) and proceeds to step 2117 above. In addition, step 2
After 117, the process returns to step 2108 of FIG. 21 and the steps after this step are executed.

Next, by the above-mentioned part division attaching means 109,
Deletion and separation caused by changing the logical node (Fig. 2
23 to 25 show the states of steps 2105 to 2107) of step 1.

In addition, in the example shown in FIG. 23, in addition to the logical / allocation relation data 2010 and 2020 showing the correspondence between the logical object designating the partial layout template and the layout object corresponding to this logical object, Logical / allocation relation data 2310 to 2380 indicating the correspondence between the basic logical object and the block is provided, but it is clearly shown that the basic logical object and the block are associated by sharing the content part. It is represented as data. In other words, a logical node group (a logical object that specifies a template with partial division, a basic logical object) and an allocation node corresponding to each of these logical nodes are associated with each other.

The logical structure shown in FIG. 23 has the same contents as the logical structure of the first embodiment shown in FIG. 2, and the contents part held by each basic logical object is shown in FIG.
It is assumed that the contents are the same as the contents part shown in.

When the content part C linked to the basic logical object 215 is changed from the state shown in FIG. 23, the correspondence between the logical structure and the layout structure becomes the content shown in FIG. In FIG. 24, since a part of the text is changed, the structure with partial division 2400 following the related article is separated. Also, the allocation objects 516, 517,
525 and 526 (each layout object of “text frame” and “page”) are deleted because all the blocks under each layout object are deleted (actually, the blocks are separated. However, from the viewpoint of each layout object of "text frame" and "page",
Since the block that existed until now does not exist in itself, it becomes the same state as when it was deleted), and even if those allocation objects are deleted, the partial division template (partial division template 240 shown in FIG. , 25
This is because there is no contradiction with (see 0).

When the content part b linked to the basic logic object 218 is changed from the state shown in FIG. 23, the correspondence between the logical structure and the layout structure becomes the content shown in FIG. In FIG. 25, the allocation object 52
3 (right frame) is not deleted. If only the layout object 523 is deleted, the template 240
This is because the constraint (see FIG. 2) is deviated.

Since the partial correspondence relationship between the logical structure and the allocation structure can be clarified in this way, when a part of the logical structure is changed, only a part of the allocation structure that is affected can be reallocated. Therefore, the efficiency of allocation processing is improved.

However, in the conventional allocation process, one allocation program is applied to the whole,
It has been difficult for a specific range of a logical structure to retain the correspondence of a specific logical structure.

Next, a fourth embodiment of the present invention will be described with reference to FIGS. 26 to 34.

An apparatus for realizing this embodiment is shown in FIG.
The configuration is the same as that of the functional block diagram of the first embodiment shown in FIG.

However, since the information held in the logical structure holding means 101 and the template with partial division 105 is different from the contents explained in the first embodiment, it will be explained next.

The logical structure holding means 101 holds the logical structure 2610 of the document as shown in FIG. 26, and the category “note” is designated for the logical node representing the “related article” in the logical structure 2610 ( Granted). The fact that "note" is specified in the logical node representing the "related article" means that the entire portion of the "related article" is passed to the subsequent page. By the way, in the conventional allocation processing method according to the category, since only the block can be specified in the category and poured into the specific lowest frame, originally, though I want to send the whole related article to the subsequent page, For each basic logical object, we had to give each category.

Further, the template 105 with partial divisions is provided in the template 105 with partial divisions as shown in FIG.
30 is held, and the template with partial division 2620
In the "text frame" inside, there is a category called "main"
The “note frame” is given the category “note”, while the template with partial division 2630
The category "main" is assigned to each of the "left frame" and the "right frame" in the inside. In addition, in FIG.
2620a is a layout image represented by the template with partial division 2620, and 2630a is a layout image represented by the template with partial division 2630.

Here, when the layout processing is performed by applying the partial layout templates 2620 and 2630 to the logical structure 2610, the layout result as shown in FIG. 27A is obtained. FIG. 27B shows a layout image represented by the layout structure shown in FIG. As can be seen from FIG. 27, the “related article” portion is certainly assigned to the final page.

In the example shown in FIG. 27, the first-layer logical node string (logical nodes 2613, 2 shown in FIG. 26).
614, 2615, and 2620), there are two methods of changing the layout structure permitted in the partial layout template 2620, that is, increasing the “text page” and increasing the “note page”. That is, in order to obtain the allocation result shown in FIG. 27, it is necessary to increase the “text page” up to the second page and the “note page” on the third page. Thus, a function for properly selecting the two types of changes is required, and this function is realized by the backtrack mechanism.

Next, the process of adding a partial division according to the category instruction by the partial dividing unit 109 will be described with reference to FIGS. 28 to 31.

28 to 31 show the allocation process when the backtrack mechanism is used.

Now, as shown in FIG. 28A, with respect to the logical structure 2610 shown in FIG. 26, an allocation structure 2810 due to the restriction of the template with partial division 2620 (see FIG. 26).
And a template with partial division 2630 (see FIG. 26).
It is assumed that a frame 2820 is obtained as a result of the process of partial division due to the restriction (see the reference).

From this state, as shown in FIG. 28 (a), the section division attaching means 109 transfers the frame 2820 to the lowest frame 2810a (text frame) of the layout structure 2810.
Suppose you try to pour (small frame). However,
As shown in FIG. 28B, the lowest frame 2810a
The category “main” is specified in (text frame), and “not” is set in the logical node 2615 of “related article”.
Since the category "e" is specified, the frame 2820 (subset frame) corresponding to the logical node 2615 cannot be poured into the lowest frame 2810a (text frame).

Therefore, this time, the means for partial copy 109 causes the lowest frame 291 as shown in FIG. 29A due to the restriction of the template for partial copy 2620 (see FIG. 26).
0a (note frame) is regenerated to obtain the layout structure 2910. It is assumed that an attempt is made to pour the frame 2820 (small frame) into the lowest frame 2910a (note frame) from this state. In this case, as shown in FIG. 29B, the category “note” is specified in the lowest frame 2910a (note frame), and this category is
Since it matches with the category specified in the logical node 2615 (related article), the frame 2820 will be poured into the lowest frame 2810a.

By the above processing, the content parts "A", "B", "a", "b", and "c" shown in FIG. 26 have been assigned, and the remaining content parts "C" are next. Will be explained.

In this case, as shown in FIG. 30A, the basic logical object 301 holding the content portion "C"
A block 3020 corresponding to 0 is generated and an attempt is made to fill this block 3020 into the bottom frame 3030. However, as shown in FIG.
The remaining area of the lowest frame 3030 is the block 30.
Since it does not have the size necessary to pour 20,
Pouring fails. Therefore, the template with partial division 2
When the number of “text pages” is increased due to the restriction of 620, FIG.
The allocation result shown in (a) is obtained. A block 3020 to the lowest frame 3110 of the “text frame” of this allocation result
31B, since the category of both the lowest frame 3110 and the block 3020 is “main”, the filling succeeds, as shown in FIG.

When this process ends, the allocation result (final allocation result) shown in FIG. 27A is obtained.

According to this embodiment, it is possible to easily realize the contents described in the first national language and the contents described in the second national language in association with each other. Here, for example, a process of allocating a Japanese part and an English part to which a Japanese-English bilingual document is allocated in a corresponding manner will be described.

For example, as shown in FIG. 32, logical structure 3
210, templates with copy divisions 3220 and 3230, and content portion 3240 may be prepared.

Here, in the logical structure 3210, the basic logical objects holding the content parts “A”, “I”, “U”, and “E” respectively have the category “J” meaning Japanese. Specify, and the content section "A", "B",
The category "E", which means English, is designated as the basic logical object holding "C" and "D". In addition, the category "J" meaning Japanese is designated as the lowest frame of "day frame" of the template 3130 with partial division, and the category "E" meaning English is designated as the lowest frame of "English frame". specify.

When the layout process is performed by applying the partial layout templates 3220 and 3230 and the content section 3240 to the logical structure 3210, the layout result as shown in FIG. 33A is obtained. The layout image represented by this allocation result is shown in FIG.

Further, as shown in FIG. 34, by managing the logical node sequence for each category, the third node
The re-allocation efficiency described in the embodiment can be performed for each category, and further improvement in efficiency can be expected.

In the third embodiment described above, it was shown that the reallocation can be made more efficient by the present invention, but the backtrack mechanism can be considered as repeating the reallocation by discarding the partial layout structure. Therefore, as described in the fourth embodiment, the backtracking can be performed efficiently by performing the backtracking in units of the partial division structure. For example, when the third page is discarded, it is not necessary to discard / reallocate the structures below the small frame obtained by the partial division process.

Next, a fifth embodiment of the present invention will be described with reference to FIGS. 35 to 37.

The apparatus for realizing this embodiment is basically the same as the configuration of the functional block diagram of the first embodiment shown in FIG.

Also in this embodiment, the means for partial division 1
09 is realized by a control unit such as a processor or a central processing unit executing software (program) for executing the partial division process. That is, the partial division unit 109 is composed of a partial division program and a control unit that executes the partial division program.

Note that there is only one copy division unit 109 in the above-described first to fourth embodiments.
By executing such a program repeatedly or by writing a program that calls itself (recursive call), the partitioning process for a plurality of logical nodes is executed. However, the partial division means 10 in this embodiment
Reference numeral 9 corresponds to a logical node and has a plurality of partial partitioning means for performing partial partitioning processing for the logical node. That is,
The apparatus of this embodiment is applied to a computer which allows a plurality of control means for executing the same program in parallel. The plurality of control units operating in parallel may be realized by hardware (for example, having a plurality of central processing units), or may be realized by software in a pseudo manner.

In each of the partial partitioning means in the partial partitioning means 109, the partial partitioning means is generated corresponding to the logical node, and when the logical node is deleted, the corresponding partial partitioning means also disappears. ing. That is, the partial dividing means is generated by the number of logical nodes.

As one of the theories for enabling such parallel processing, CSP (Communicating Sequential Processe)
s) is known. A parallel computer based on this theory (Tra
nsputer) and programming language (OCCUM) have already been developed and implemented. In this theory, a program that executes programs in parallel is called a process. This process can specify other processes running in parallel to pass information, and vice versa. According to this theory, when receiving information, it waits until the information is sent from the specified party. In order to control such information exchange by a program, a symbol "!" Is used when passing information in the program, and a symbol "?" Is used when receiving information.

Therefore, also in this embodiment, according to this theory, a process is used to execute the partial layout program. That is, the processes operating in parallel perform the allocation process by exchanging information with other processes. It should be noted that in this embodiment, formatting is performed.
A process that performs processing is defined as a formatter. One of the formatters is activated corresponding to a logical node, and allocates a lower logical node sequence of the corresponding logical node. Therefore, the formatter forms a hierarchical structure corresponding to the hierarchy of logical nodes.

Here, when the logical structure is the logical structure 210 shown in FIG. 2, the hierarchical structure of the formatter generated corresponding to the logical node in the logical structure is shown in FIG.

In FIG. 35, the formatter 3501 is the logical node 211, the formatter 3502 is the logical node 213, and the formatter 3503 is the logical node 214.
The formatter 3504 is to the logical node 215, the formatter 3505 is to the logical node 216, the formatter 3506 is to the logical node 217, and the formatter 3507.
To the logical node 218, the formatter 3508 to the logical node 219, and the formatter 3509 to the logical node 22.
1, the formatter 3510 corresponds to the logical node 222. These formatters run in parallel.

Next, the part dividing means 109 of this embodiment.
The overall allocation process by the above will be described with reference to the flowchart shown in FIG.

When there is an allocation processing request, the partial division means 109 activates the formatters corresponding to all the logical nodes and hierarchizes those formatters (step 3601). Next, when an allocation start instruction is issued, the allocation start instruction is given to the highest formatter of the hierarchy to execute the allocation processing (step 3602). By assigning information necessary for allocation to the lower formatters in order from the highest formatter, the allocation process by each formatter proceeds (step 3603). The details of the allocation process will be described later, but here is a brief description of the process. Each formatter hands over the allocatable area (hereinafter referred to as available area) to the lower formatter and sends it to the upper formatter. On the other hand, the allocation result and status (allocation processing status) are delivered and the allocation processing proceeds. Then, the partial dividing means 109
Sets the result of the highest formatter as the result of the entire allocation process (step 3604).

There are the following three states in which the formatter hands over to the upper formatter.

[FAIL (allocation failure)] This indicates that the logical node sequence could not be allocated at all in the available area delivered from the upper formatter. The formatter that has passed the state "FAIL" to the upper formatter waits until it receives a new available area from the upper formatter.

[PARTIALLY (partially allocated successfully)] This indicates that part of the logical node sequence could be allocated in the available area delivered from the upper formatter. State "PARTIAL
The formatter that has delivered “LY” to the upper formatter waits until it receives a new available area from the upper formatter.

[DONE (End of allocation)] This indicates that all the logical node strings could be allocated within the available area delivered from the upper formatter. The formatter that has passed the state "DONE" to the upper formatter finishes the allocation.

Next, the allocation processing by each format in step 3603 will be described with reference to FIG.

The formatter as a means for partial division is
Higher formatter to available area (area-for-s
Wait until elf) is received (step 3701).

When an available area (area-for-self) is passed, the formatter creates the first minimal subdivision structure in the available area that satisfies the constraints of the subdivision template (template). la
y-subtree "(step 3702). In addition, "temp
"late" is a partial layout template designated by the logical node corresponding to the formatter.

Next, the formatter sets the first frame of the least significant frames in "lay-elmnt" to "lay-elmnt" (step 3703), and thereafter determines whether or not all logical node sequences have been allocated. (Step 3704).

Here, if all are not assigned,
The first logical node whose allocation has not been completed is extracted from the logical node sequence, and this is designated as "log-elmnt" (step 3705). The formatter activated corresponding to the "log-elmnt" is set to "sub-formatter" (step 3706), and the available area is handed over to the "sub-formatter" (step 3707).

Then, the formatter uses the "sub-formatte
Waiting for response (state, partial result) from "r" (step 3708), the value of the state which is the response result, that is, "FAIL", "DONE", "PARTIALLY"
After executing the processing according to (step 3709), it is determined whether the state is "DONE" (step 37).
10).

Here, the state from "sub-formatter" =
In the case of “DONE”, the process returns to the above step 3707 to execute the steps after this step, while in the case of the state from “sub-formatter” ≠ “DONE”, the above step 3707
Return to and execute the steps after this step.

When all the logical node sequences are allocated in the above step 3704, "DONE" is delivered as the own status and "lay-subtree" is delivered as the allocation result to the upper format (step 371).
0).

At step 3709, "sub-fo
When the value of the state from "rmatter" is "FAIL", a new available area is prepared using the partial division template. "Lay-subtree" structure that is being created
According to the status of "," the status of "own status (FAIL or PARTIALLY in this case)" and the allocation result are returned to the upper formatter. On the other hand, when the value of the status from "sub-formatter" is "DONE" or "PARTIALLY", the allocation result received from "sub-formatter" is
Connect to the current "lay-subtree". That is, pour (P
our) processing.

An algorithm showing an example of the partial division process executed by each formatter is as follows.

01 Procedure formatter 02 AVAILABLE-AREA [sup-formatter]? Area-for-self; 03 lay-subtree = generate-first-min-subtree (template, area-for-self); 04 lay-elmnt = get -first-lay-stream-elmnt (lay-subtree); 05 foreach log-elmnt in log-stream do 06 sub-formatter = formatter-for (log-elmnt); 07 loop 08 area-for-sub-formatter = compute -available-area (lay-elmnt, area-for-self); 09 AVAILABLE-AREA [sub-formatter]! area-for-sub-forematter; 10 LAYOUT-RESULT [sub-formatter]? (status ， generated-subtree ); 11 case status of 12 DONE: 13 pour (generated-subtree, lay-elmnt); 14 exit loop; 15 PARTIALLY: 16 pour (generated-subtree, lay-elmnt); 17 FAIL: 18 if lay-stream-has -more-elmnts (lay-subtree) then 19 lay-elmnt = get-next-lay-stream-elmnt (lay-subtree); 20 elseif can-add-branch (lay-subtree, template, area-for-self). then 21 add-branch (lay-subtree, template, area-for-self); 22 lay-elmnt = get-next-lay-stream-e lmnt (lay-subtree); 23 else 24 if something-was-poured (lay-subtree) then 25 LAYOUT-RESULT [sup-formatter]! (PARTIALLY, lay-subtree); 26 AVAILABLE-AREA [sup-formatter]? area-for-self; 27 lay-subtree = generate-next-min-subtree (template, area-for-self); 28 else 29 LAYOUT-RESULT [sup-formatter]! (FAIL, nil); 30 AVAILABLE-AREA [sup-formatter]? area-for-self; 31 lay-subtree = re-generate-current-min-subtree (template, area-for-self); 32 endif; 33 lay-elmnt = get-first-lay- stream-elmnt (lay-subtree); 34 endif; 35 end loop; 36 end foreach; 37 LAYOUT-RESULT [sup-formatter]! (DONE, lay-subtree); 38 end; In the above algorithm, "01 "~" 38 "
Means the step number corresponding to each character string, and "lay-stream" means the row of the lowest frame in the partial division structure.

Next, the partial division process executed by each formatter will be described with reference to the above-mentioned algorithm. In the following description of the processing, each step number corresponds to the step number in the above algorithm.

[Step 01] Program name (form
atter) and its beginning.

[Step 02] Wait until an available area (area-for-self) is received from the upper formatter.

[Step 03] In "area-for-self", generate the first minimum subdivision structure satisfying the constraints of the subdivision template (template) and call it "lay-
subtree ". template is a partial layout template designated by the logical node corresponding to the formatter.

[Step 04] The first frame of the least significant frames in "lay-subtree" is set as "lay-elmnt".

[Step 05] Steps 07 to 36 are executed for each logical node (log-elmnt) included in the logical node sequence held by the corresponding logical node.

[Step 06] The formatter activated corresponding to "log-elmnt" is referred to as "sub-formatter".

[Step 07] Repeat steps 08 to 35 until "exit" appears.

[Step 08] The available area to be transferred from "area-for-self" and "lay-elmnt" to "sub-formatter" is calculated, and the calculation result is calculated as "area-for-sub".
-formatter ".

[Step 09] The "area-for-sub-formatter" is transferred to the "sub-formatter".

[Step 10] Wait until the status (status) and the allocation result (generated-subtree) are received from the "sub-formatter".

[Step 11] Value of "status" (DON
E, PARTIALLY, FAIL), and executes the processing according to each "status".

[Step 12] If the value of "status" is "DO"
In the case of "NE", steps 13 and 14 are executed.

[Step 13] "gene" is added to "lay-elmnt"
"rated-subtree" is connected.

[Step 14] The repetition started from step 07 is completed, and the process proceeds to step 36.

[Step 15] If the value of "status" is "PA"
In the case of “RTIALLY”, the process proceeds to step 16.

[Step 16] "gene" is added to "lay-elmnt".
"rated-subtree" is connected, and then the process proceeds to step 35.

[Step 17] If the value of "status" is "FA"
If “IL”, steps 18 to 34 are executed.

[Step 18] It is determined whether or not another lowest frame remains in the "lay-subtree". If the lowest frame remains, the process proceeds to step 19. If not, the step 20 proceeds. Go to.

[Step 19] The next lowest frame in "lay-subtree" is set to "lay-elmnt", and then the process proceeds to step 35.

[Step 20] Within "area-for-self", the "lay-subtree" is satisfied so that the restriction of the subdivision template can be satisfied and a new lowest frame can be prepared.
It is determined whether the number of nodes can be increased. If the number can be increased, the process proceeds to step 21, while if the number cannot be increased, steps 23 and 2 are performed.
Execute 4.

[Step 21] Within "area-for-self", "lay-subtree" is set so that the restriction of the partial layout template can be satisfied and a new lowest frame can be prepared.
Add more nodes.

[Step 22] The next lowest frame in "lay-subtree" is set to "lay-elmnt", and then the process proceeds to step 34.

[Steps 23 and 24] It is determined whether or not the allocation processing has been performed at all (whether something has been connected to the lowest frame somewhere in "lay-subtree"), and the allocation processing has been performed. If so, steps 25 to 27 are executed, while if not, steps 28 to 31 are executed.

[Step 25] "PARTIALLY" is delivered as the status and "lay-subtree" is delivered as the allocation result to the upper formatter.

[Step 26] Wait until a new available area (area-for-self) is received from the upper formatter.

[Step 27] Within "area-for-self", a new minimum subdivision structure that satisfies the constraints of the subdivision template is generated, and this is designated as "lay-subtree". After that, the process proceeds to step 33.

[Steps 28 and 29] "FAIL" is passed as the status and nil (representing nothing) is passed as the allocation result to the upper formatter.

[Step 30] Wait until a new available area (area-for-self) is received from the upper formatter.

[Step 31] Within "area-for-self", a subdivision structure that satisfies the constraints of the subdivision template is newly generated, and this is set as "lay-subtree". After that, the process proceeds to step 33.

[Step 32] It shows the end of the conditional statement starting from step 24, and proceeds to the next step.

[Step 33] The first frame of the least significant frames in "lay-subtree" is set as "lay-elmnt".

[Step 34] Indicates the end of the conditional statement starting from step 17, and the process proceeds to the next step.

[Step 35] It shows the end of the range of repetition starting from step 07, and returns to step 07.

[Step 36] The end of the range of repetition starting from step 05 is shown, and the process returns to step 05.

[Step 37] "DONE" is passed as the state and "lay-subtree" is passed as the allocation result to the upper format.

[Step 38] Indicates the end of the program.

However, the above-mentioned algorithm requires some modification for the highest and lowest formatters. This is because the highest formatter cannot wait until it receives information from higher formatters. Similarly, for the lowest format, it cannot wait until it receives information from a lower formatter.

Therefore, the above algorithm is modified as follows. Regarding the highest formatter (1) In particular, step 02 is a step for waiting until there is an instruction to start allocation, and an infinite available area is always passed. (2) The state delivered in each step 25, 29, 37 and the allocation processing result are the allocation result of the entire logical structure,
Finish execution. Therefore, steps 26 and 27 and steps 30 and 31 are not executed.

Regarding the lowest formatter (1) Instead of the processing of steps 08, 09 and 10, "content allocation processing" is executed and the status is "status".
And the result is called "generated-subtree".

Next, an application example of this embodiment will be described.

In this embodiment, the formatter delivers the allocation result and the status to the upper formatter, and the allocation result received from the lower formatter is passed to the lowest frame during the processing of the formatter. You may do the following.

The allocation process can also be advanced by passing only the information of the available area consumed by the allocation result to the upper formatter instead of passing the allocation result to the upper formatter. in this case,
It is necessary to save the subdivision structure generated by each formatter and finally connect it to the entire layout structure. By adopting such an implementation method, the amount of information exchanged between the formatters can be suppressed to a minimum, so that the overall processing efficiency can be improved.

In this embodiment, the case where the allocation process is performed from the beginning has been described, but the present invention is also effective when the reallocation is performed after a part of the logical structure is changed.

This is because each formatter can judge whether or not re-execution is necessary from the information on the available area from the upper formatter and the information on the allocation result from the lower formatter, and only the formatters that need to be re-executed. This is because reassignment can be realized by re-executing. In such re-allocation, when the logical structure is changed, the formatter corresponding to the changed portion (logical node) is generated and deleted as necessary, and the hierarchy of the formatter is partially changed. After that, the allocation start instruction is given to the highest formatter in the changed portion in the hierarchy, and the result of the highest formatter may be used as the changed allocation processing result.

Since each formatter performs partial partitioning to generate a partial partitioning structure corresponding to a logical node, one formatter knows the correspondence relationship between the logical node and the partial partitioning structure ( You have information about the correspondence). Therefore, each formatter has the logic / assignment correspondence holding means 1 described in the third embodiment.
Performs 800 functions.

Next, the reallocation process will be specifically described.

When the previously allocated result can be placed in the available area received from the upper formatter, the lower formatter does not need to be executed again, and the previously obtained partial partitioning processing result only needs to be returned to the upper formatter. This is effective, for example, when adding a paragraph. Because the width of a normal paragraph is always the same,
Splits occur only in the relatively rare cases where paragraphs are on page boundaries.

When the allocation result received from the lower formatter has the same size as the allocation result received the previous time, the size of the available area for allocating the subsequent logical node is not affected, and therefore the subsequent logical node is not affected. It is not necessary to execute the formatter corresponding to the node. This is effective, for example, when a paragraph is changed to such an extent that the number of lines does not change.

By adopting each of the above-mentioned methods, the partial division can be realized very efficiently. Therefore, even if the allocation process is always executed for every operation of the document editing, the processing efficiency is high. It never drops. As a result, WYSIW does not require the complicated operation of allocating after editing.
It becomes possible to realize a YG (widwig) editor. Further, in this embodiment, each formatter waits until the result of the lower or upper formatter is received, but the parallel processing may be executed by predicting the information of the available area.

For example, since the width of paragraphs is usually constant,
The paragraph allocator formatter can pre-allocate in an available area of fixed width and infinite length. Then, after allocation, only when the result of parallel allocation cannot be placed in the available area given by the upper formatter, the allocation process is re-executed as in the case of re-execution of the allocation process described above. It will be.

As described in each of the above-mentioned embodiments, instead of the conventional layout template for the entire document, the partial layout template corresponding to each partial logical structure when the logical structure of the document is classified by a predetermined relationship is used. In addition to the preparation, the partial partitioning template corresponding to the partial logical structure is applied to a certain partial logical structure to obtain the partial partitioning structure. Can be applied to the frame.

Further, a template equivalent to the conventional layout template for the entire document can be obtained dynamically and virtually at the time of executing the layout process by a combination of smaller partial layout templates. Therefore, the partial layout template can be reused. it can.

Since the operation of the allocation program itself plays a simple role of partial allocation, it becomes easy to create the allocation program.

Since the partial logical structure is processed as a block, operations such as editing become easy.

Since the correspondence relationship between the partial logical structure and the partial layout structure is clarified, the processing efficiency of the reallocation process can be improved.

Further, since the partial allocation processing is executed in parallel to allocate the entire logical structure, the allocation processing speed is higher than that in the case where the partial partitioning processing is recursively executed to allocate the entire logical structure. Can be improved.

[0283]

As described above, according to the present invention, a partial division template is prepared corresponding to each partial logical structure when the logical structure of a document is classified according to a predetermined relationship, and a partial logical structure is created. On the other hand, since the partial-partitioned template corresponding to this partial logical structure is applied to generate the partial-partitioned structure, the already-created partial-partitioned template can be reused.

When changing the layout of an existing document, only the existing template with partial division corresponding to the layout to be changed is changed, and other existing templates with partial division can be reused. Can be easily created.

[Brief description of drawings]

FIG. 1 is a functional block diagram showing a first embodiment of a document processing apparatus according to the present invention.

FIG. 2 is a diagram for explaining a correspondence relationship between a logical structure of a document and a copy separation template according to the first embodiment.

FIG. 3 is a diagram for explaining a logical node sequence of the logical structure shown in FIG.

FIG. 4 is a diagram showing a data structure of a logical structure, a logical node sequence corresponding to the logical structure, a partial layout template, and a content part.

FIG. 5 is a diagram showing an allocation result obtained by applying the partial layout template shown in FIG. 2 to the logical node sequence shown in FIG.

6 is a diagram showing a data structure having a configuration in which data of the allocation structure shown in FIG. 5 is added to the data structure shown in FIG.

FIG. 7 is a flowchart showing the operation of the partial division process (bottom-up partial division process) by the partial division unit in the first embodiment.

FIG. 8 is a diagram for explaining a copy division process in the first embodiment.

9A and 9B are views for explaining a partial division process in the first embodiment.

FIG. 10 is a diagram for explaining a partial division process in the first embodiment.

FIG. 11 is a functional block diagram showing a second embodiment of the document processing apparatus according to the present invention.

FIG. 12 is a layout result obtained when bottom-up layout processing is performed by applying the partial layout template shown in FIG. 2 to the logical structure shown in FIG. 2 in the second embodiment. FIG.

FIG. 13 is a layout result obtained when top-down layout processing is performed by applying the partial layout template shown in FIG. 2 to the logical structure shown in FIG. 2 in the second embodiment. FIG.

FIG. 14 is a diagram for explaining the processing steps of bottom-up allocation processing in the second embodiment.

FIG. 15 is a diagram for explaining the processing steps of top-down allocation processing in the second embodiment.

FIG. 16 is a diagram for explaining a processing process of top-down allocation processing according to the second embodiment.

FIG. 17 is a flowchart showing the operation of the partial division processing (top-down partial division processing) by the partial division means in the second embodiment.

FIG. 18 is a functional block diagram showing a third embodiment of the document processing apparatus according to the present invention.

FIG. 19 is a diagram for explaining a correspondence relationship between a partial logical structure and a partial division structure according to the third embodiment.

FIG. 20 is a diagram showing a data structure of a logical structure, a logical node sequence, a partial layout template, a content part, a layout structure, and a logical / allocation correspondence relationship in the third embodiment.

FIG. 21 is a partial-partitioning process by the partial-partitioning unit in the third embodiment (partial-partitioning process in consideration of reallocation).
3 is a flowchart showing the operation of the above.

FIG. 22 is a partial-partitioning process (partial-partitioning process in consideration of reallocation) by a partial-partitioning unit in the third embodiment.
3 is a flowchart showing the operation of the above.

FIG. 23 is a diagram showing a logical structure / allocation correspondence relationship between a logical structure of a document after allocation processing and a layout structure in the third embodiment.

24 is a diagram for explaining reallocation processing when the logical structure of the document shown in FIG. 23 is changed.

FIG. 25 is a diagram for explaining reallocation processing when the logical structure of the document shown in FIG. 23 is changed.

FIG. 26 is a diagram showing a correspondence relationship between a logical structure of a document and a partial layout template in the fourth embodiment of the document processing apparatus according to the present invention.

27 is a diagram showing an allocation result obtained by applying the partial layout template shown in FIG. 26 to the logical structure shown in FIG.

FIG. 28 is a diagram for explaining a partial division process (a partial division process according to a category) according to the fourth embodiment.

FIG. 29 is a diagram for explaining a partial division process (a partial division process according to a category) according to the fourth embodiment.

FIG. 30 is a diagram for explaining a partial division process (a partial division process according to a category) according to the fourth embodiment.

FIG. 31 is a diagram for explaining a partial division process (a partial division process according to a category) according to the fourth embodiment.

FIG. 32 is a diagram for explaining a partial division process (a partial division process according to a category) according to the fourth embodiment.

FIG. 33 is a diagram for explaining a partial division process (a partial division process according to a category) according to the fourth embodiment.

FIG. 34 is a diagram for explaining a partial division process (a partial division process according to a category) according to the fourth embodiment.

FIG. 35 is a diagram for explaining a formatter in the fifth embodiment of the document processing apparatus according to the present invention.

FIG. 36 is a flowchart showing the operation of allocation processing in the fifth embodiment.

FIG. 37 is a flowchart showing the operation of partial division processing by each formatter in the fifth embodiment.

FIG. 38 is a diagram showing a conventional relationship between a logical structure of a document and a layout template.

39 is a diagram showing an allocation result obtained by applying the allocation template shown in FIG. 38 to the logical structure shown in FIG. 38.

FIG. 40 is a diagram showing a correspondence relationship between the logical structure shown in FIG. 38 and the allocation structure shown in FIG. 39.

FIG. 41 is a flowchart showing the operation of a conventional allocation process.

FIG. 42 is a view for explaining a processing process of conventional allocation processing.

FIG. 43 is a diagram for explaining a processing process of conventional allocation processing.

FIG. 44 is a diagram for explaining a processing process of conventional allocation processing.

FIG. 45 is a diagram for explaining a processing process of conventional allocation processing.

FIG. 46 is a view for explaining a processing process of conventional allocation processing.

FIG. 47 is a diagram showing a conventional correspondence relationship between a logical structure of a document and an allocation template expressing a rule for laying out a small set in two columns.

48 is a diagram showing an allocation result obtained by applying the allocation template shown in FIG. 47 to the logical structure shown in FIG. 47.

FIG. 49 is a diagram for explaining a process of generating the layout structure shown in FIG. 48.

FIG. 50 is a diagram for explaining a process of generating the layout structure shown in FIG. 48.

FIG. 51 is a diagram showing an allocation result obtained by performing conventional top-down allocation processing.

FIG. 52 is a view showing an allocation result obtained by performing conventional bottom-up allocation processing.

FIG. 53 is a diagram showing a conventional correspondence relationship between a logical structure of a document and an allocation template for executing an allocation process in accordance with a category instruction.

FIG. 54 is a diagram showing a conventional allocation template expressing rules for laying out small sets in three columns.

FIG. 55 is a view showing a conventional layout template that expresses rules for laying out small sets in two columns or three columns.

[Explanation of symbols]

101 ... Logical structure holding means, 102 ... Content holding means, 103 ... Allocation structure holding means, 104 ... Content allocating means, 105 ... Partial division template, 106 ... Logical node sequence extracting means, 107 ... Logical node sequence managing means, 108 ... Partial logical structure designating means, 109, 1100 ... Division dividing means, 240, 250, 2620, 2630 ... Division dividing template 1110 ... Top-down division dividing means, 1120 ... Bottom-up division dividing means, 1130 ... Division division selecting means, 1800 ... Logical / allocation management means.

Claims (6)

[Claims] 1. A partial division template holding unit that holds a partial division template showing a rule for allocating the partial logical structure corresponding to each partial logical structure when the logical structure of a document is classified according to a predetermined relationship. Means, a partial logical structure designating means for designating a predetermined partial logical structure in the logical structure, and the partial allocation corresponding to the partial logical structure to the partial logical structure designated by the partial logical structure designating means. A document processing apparatus, comprising: a section dividing unit that applies a template to perform a section dividing process. 2. A partial division template holding unit that holds a partial division template indicating a rule for allocating the partial logical structure corresponding to each partial logical structure when the logical structure of a document is classified by a predetermined relationship. Means, a partial logical structure designating means for designating a predetermined partial logical structure in the logical structure, and the partial division corresponding to the partial logical structure to the partial logical structure designated by the partial logical structure designating means. And a section dividing unit that performs a section dividing process by applying a template, wherein the layout process of the entire logical structure is performed by repeatedly or recursively executing the allocation process by the section dividing unit. Document processing device. 3. The partial logical structure designating means extracts a logical node sequence composed of a basic logical object holding document contents and a logical object representing a lower logical node sequence from the logical structure. 3. The document processing apparatus according to claim 2, further comprising a column extracting unit and a logical node sequence managing unit managing the logical node sequence extracted by the logical node sequence extracting unit. 4. The top-down section dividing means for determining the area of the side-partitioning structure obtained as a result of the side-by-side division from the remaining area of the higher-order layout object to top-down, and the means for obtaining the part-side division. The area of the subdivision structure is applied bottom-up subdivision means for determining bottom-up from the area in which lower-level layout objects are combined, and either the top-down subdivision means or the bottom-up subdivision means is applied. 3. The document processing apparatus according to claim 2, further comprising: a partial division selection unit that selects the partial division unit based on the allocation unit selection information regarding the. 5. A partial layout template holding unit that holds a partial layout template indicating a rule for allocating the partial logical structure corresponding to each partial logical structure when the logical structure of a document is classified by a predetermined relationship. Means, a partial logical structure designating means for designating a predetermined partial logical structure in the logical structure, and the partial division corresponding to the partial logical structure to the partial logical structure designated by the partial logical structure designating means. Partial division means for applying a template to perform partial division processing; logical / allocation management means for managing the partial logical structure and the partial division structure as a result of partial division processing by the partial division processing means in association with each other; A document processing device comprising: 6. A partial division template holding unit that holds a partial division template indicating a rule for allocating the partial logical structure corresponding to each partial logical structure when the logical structure of a document is classified by a predetermined relationship. Means, a partial logical structure designating means for designating a predetermined partial logical structure in the logical structure, and the partial division corresponding to the partial logical structure with respect to the partial logical structure designated by the partial logical structure designating means. Document processing characterized by comprising: a plurality of partial division means for applying a template to perform partial division processing, and executing the processing by the respective partial division means in parallel to perform the allocation processing of the entire logical structure. apparatus.