Method and apparatus for formatting documents.

BACKGROUND OF THE INVENTION

Field of the Invention

0001The present invention relates to a method and an apparatus for formatting a document in accordance with structures of the document.

Description of the Background Art

0002Recently, there has been remarkable progresses in so-called desk-top publishing as well as in Japanese word-processors, in addition to the advances in the output devices such as displays and printers which made it possible to produce documents in much more diverse and effective manners.

0003However, producing documents which are both neat-looking and easily-readable requires thorough understandings of operations, commands and formats of a document formatting apparatus to be used. which makes such a document formatting apparatus almost inaccessible to those without formal trainings.

0004As a solution to this situation, progresses has been made in developing an automatic document formatting system which utilizes an automatically extracted logical structure of a document, along with a document formatting system capable of arranging figures and articles according to automatically derived referential relationships between figures and articles.

0005Although such logical structures and referential relationships are generally sufficient to pro- videstructural information on the document necessary for effective formatting, ambiguities involved in human languages may lead to misapprehension of the logical structures and referential relationships. Furthermore, apart from this problem, outputs of these automatic document formatting apparatus may not satisfy personal or temporal demands of a user.

0006On the other hand, there are document formatting system which utilizes command codes indicating manners of formatting to be embedded in the document, such as 'Roff'. More recently, there appeared document formatting systems such as 'Scribe' or 'T_eX' which has document data and format data independently so that change in command codes can be made at once on the format data, without looking for every embedded command codes as in 'Roff'. Moreover, 'Scribe' and 'T_eX' are capable of performing more sophisticated formatting than others. But, in these document formatting systems, a thorough understanding of command codes is indispensable for skillful maneuvering. Furthermore, even in 'Scribe' and 'T_EX', embedding of command codes is necessary which can easily be tedious. Though automatic document formatting apparatuses mentioned above are free of such problems concerning command codes, they are, as described above, prone to the misapprehension of the logical structures and referential relationships due to the ambiguities in the human languages, and the outputs of these automatic document formatting apparatus may not satisfy personal or temporal demands of a user.

0007Thus, with conventional document formatting apparatuses, either a possibility of misapprehension resulting from automatic extraction of document structures or else difficulties in dealing with command codes which need to be embedded in the document and thoroughly mastered by the user, has to be tolerated.

SUMMARY OF THE INVENTION

0008It is therefore an object of the present invention to provide a method an an apparatus for formatting a document which are simply manipulatable as well as consistently correct.

0009According to one aspect of the present invention, there is provided an apparatus for formatting a document which contains command codes indicating prescribed commands concerning structural function, comprising: means for enterring the document to the apparatus; means for structurally analyzing the document and deriving structural information of the document; means for detecting the command codes and adjusting the analysis by the structurally analyzing means in accordance with the indications of the command codes; and means for carrying out formatting of the document in accordance with the structural information.

0010According to another aspect of the invention, there is provided a method of formatting a document which contains command codes indicating prescribed commands concerning structural function, comprising the steps of; structurally analyzing the document and deriving structural information of the document; detecting the command codes and adjusting the analysis at the structurally analyzing step in accordance with the indications of the command codes; and formatting the document in accordance with the structural information.

0011Other features and advantages of the present invention will become apparent from the following description taken in conjunction with the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

0012<ul id="ul0001" list-style="none"><li>Fig. 1 is a block diagram of a document formatting apparatus according to one embodiment of the present invention.</li><li>Fig. 2 is a tabulated illustration of examples of command codes to be utilized in the document formatting apparatus shown in Fig. 1.</li><li>Fig. 3 is another tabulated illustration of logical structures to be utilized in the document formatting apparatus shown in Fig. 1.</li><li>Fig. 4 is a flow chart for the operation of formatting by the document formatting apparatus shown in Fig. 1.</li><li>Fig. 5 is an illustration of a document being formatted by the document formatting apparatus shown in Fig. 1, showing manners in which the command codes are used in this embodiment.</li><li>Fig. 6(A) and (B) are tabulated illustrations of logical structures for the document shown in Fig. 5, obtained without and with the command codes.</li><li>Fig. 7 is another illustration of a document being formatted by the document formatting apparatus shown in Fig. 1, showing manners in which the command codes are used in this embodiment.</li></ul>

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

0013Referring now to Fig. 1, there is shown one embodiment of a document formatting apparatus according to the present invention.

0014This document formatting apparatus comprises an input unit 10 from which a document to be formatted with command codes embedded is entered, an original document memory 20 for storing the document with the command codes entered at the input unit 10, a format memory 30 for storing a format into which the document is to be formatted, an analyzing unit 40 comprising a command code analyzing unit 50 for analyzing contents of the command codes, and a document structure analyzing unit 60 for analyzing logical and referential structures of the document in accordance with the analyzed contents of the command codes, a document structure memory 70 for storing the analyzed structures of the document, a formatting processing unit 80 for carrying out formatting in accordance with the structures of the document stored in the document structure memory 70 and the format stored in the format memory 30, an output unit 90 for presenting the document as formatted by the formatting processing unit 80, and an administration unit 99 for administerring the operations by all these parts of this document formatting apparatus mentioned above.

0015The input unit 10 may take a form of a keyboard, a mouse, or a communication network. The output unit 90 may take a form of a CRT, a display, or a printer.

0016In the analyzing unit 40, when the command codes are present in the document, the command code analyzing unit 50 adjusts operation of the document structure analyzing unit 60 such that the contents of the command codes are reflected in a manner of analyzing the logical and referential structures of the document. Thus, in this embodiment a user can deliberately control the analysis of the logical and referential structures. Moreover, the command code analyzing unit 50 also deciphers those command codes which are directly concerned with layout of the document, so that the user can also has a control over the layout of the document.

0017An example of a set of command codes to be utilized in this embodiment are shown in Fig. 2. As given in the section (0) of Fig. 2, any command code begins with a symbol <img file="EP0337068A2_D0001.tif" /> in this embodiment, and what follows this symbol<img file="EP0337068A2_D0001.tif" /> designates a type of command code. Various different types of command code and their corresponding logical attributes are summarized in the section (1) of Fig. 2. For example, a command code <img file="EP0337068A2_D0001.tif" /> DATE indicates that what follows is the date, a command code <img file="EP0337068A2_D0001.tif" /> NODC indicates what follows is the document number, and so on. In addition, there are anlysis prohibition codes shown in the section (2) of Fig. 2 which indicate a portion of the document not to be structurally analyzed such as those containing mathematical formulae and those requiring special type of formatting. Furthermore, there are graphic reference codes as shown in the section (3) of Fig. 2 which indicate the presence of reference from the document to graphics, as well as the location of data on the graphics being referred and the layout of the graphics, i.e., how the graphics are to be incorporated into the final output of the document. There are also a compulsory return code shown in the section (4) of Fig. 2 which indicates forcible changing to the next line, regardless of the logical structure.

0018An example of the logical structure in the document structure memory 70 is shown in Fig. 3. The logical structure in the document structure memory 70 comprises a sentence number labelling each sentences of the document, an attribute for each sentence, a level given to each attribute, and a header pattern for relevant sentences. For example, in Fig. 3 a sentence labelled by the sentence number 4 is given an attribute of Paragraph End in response to the command code C&lt; J HEAD present in that sentence, whose level is given as 3 and which has no header pattern. Such a logical structure will be utilized along with the format data in the format memory 30 in carrying out formatting of the document.

0019Referring now to Fig. 4, the operation of document formatting by this document formatting apparatus will be described.

0020At the step 100, the document with the command codes is entered from the input unit 10 and stored in the original document memory 20.

0021Then, one sentence of the document are read out from the original document memory 20 and given to the analyzing unit 40 at the step 101.

0022Then at the step 102, whether there is any command code in this sentence is determined by the command code analyzing unit 50.

0023When there is no command code in the sentence, an ordinary automatic extraction of logical and referential structures of the document is carried out by the document structure analyzing unit 60 at the step 103 and the process proceeds to the step 113 to be explained below.

0024On the other hand, when there is a command code in the sentence, the step 104 will be taken in which whether the command code present in the sentence is one of the logical structure codes is determined by the command code analyzing unit 50.

0025When the command code is one of the logical structure codes, the logical structure of the sentence is extracted by the document structure analyzing unit 60 at the step 105 in accordance with what the command code indicates and the process proceeds to the step 113.

0026Otherwise, the step 106 will be taken in which whether the command code present in the sentence is one of the analysis prohibition codes is determined by the command code analyzing unit 50.

0027When the command code is one of the analysis prohibition codes, the structural analysis by the document structure analyzing unit 60 is controlled at the step 107 in accordance with what the command code indicates and the process proceeds to the step 113.

0028Otherwise, the step 108 will be taken in which whether the command code present in the sentence is one of the graphic reference codes is determined by the command code analyzing unit 50.

0029When the command code is one of the graphic reference codes, the referential structure of the sentence is extracted by the document structure analyzing unit 60 at the step 109 in accordance with what the command code indicates and the process proceeds to the step 113.

0030Otherwise, the step 110 will be taken in which whether the command code present in the sentence is the compulsory return code is determined by the command code analyzing unit 50.

0031When the command code is the compulsory return code, the information of the compulsory return is deciphered and extracted by the command code analyzing unit 50 at the step 111 and the process proceeds to the step 113.

0032Otherwise, the command code present in the sentence is in error since it is not any one of command codes given in Fig. 2, so at the step 112 the error in the command code is corrected, and the process proceeds to the step 113.

0033At the step 113, the result of the logical and referential structures obtained up to this point is stored in the document structure memory 70.

0034Then at the step 114, whether all the sentences in the document have been checked is determined. When all the sentences in the document have not been checked, the process returns to the step 101 and the steps following will be repeated.

0035Otherwise the process proceeds to the step 115, at which the formatting of the document is carried out by the formatting processing unit 80 in accordance with the logical and referential structures stored in the document structure memory 70 as well as with the format stored in the format memory 30, and as the resulting formatted document is outputted by the output unit 90 the process terminates.

0036One practical examples of the document and. the use of the command codes is shown in Fig. 5.

0037In. this example of Fig. 5 which is a report entitled 'Intelligent document processing system', the first line "INTELLIGENT DOCUMENT PROCESSING SYSTEM" can be identified as a title by the ordinary automatic logical structure extraction so that no command code is necessary in this first line. On the other hand, the second line "RACHI YOZAN" will most likely not identifiable as a name of the author as Rachi Yozan is a very rare name so that this name cannot be found in a name database. Thus, that "RACHI YOZAN" is a name of the author is indicated by placing the command code <img file="EP0337068A2_D0001.tif" /> AUTH at the top of this second line. Accordingly, the document structure analyzing unit 60 can construe "RACHI YOZAN" as a name of the author correctly, Likewise, "NICHIBEI SOFTWARE" in the third line can be identified correctly as a name of the organization to which the author belongs be placing the command code : SECT at the top of this third line. As for the fourth line, there is a compulsory return code at the top of this fourth line, as this fourth line is to be left blank. The fifth line "1_'. INTRODUCTION" can be identified as a section header with 'introduction' as a reserved word, by the ordinary automatic logical structure extraction so that no command code is necessary in this fifth line. Also, the thirty-third line " <img file="EP0337068A2_D0001.tif" /> S Fig. 10 <img file="EP0337068A2_D0001.tif" /> bunsho-1 <img file="EP0337068A2_D0001.tif" />ZP d <img file="EP0337068A2_D0001.tif" /> E ..." means that "Fig. 10 is in the file 'bunsho-1' and it is to be laid out in the lower part of the current page ...".

0038How the logical structure obtained by the document structure analyzing unit 60 is affected by the presence of the command codes in the example of Fig. 5 is shown in Fig. 6(A) and (B), where Fig. 6(A) shows the logical structure obtained from the document without the command codes and Fig. 6(B) shows the logical structure obtained from the document with the command codes. As can be seen from Fig. 6, without the command codes, i.e., by a completely automatic logical structure extraction, the second and the third lines are construed incorrectly as sub-titles in Fig. 6(A), whereas with the use of the command codes this misapprehension can be avoided in Fig. 6(B).

0039Another practical example of the document and the use of the command codes, in particular the use of the analysis prohibition codes, is shown in Fig. 1.

0040In this example of Fig. 7 which is a portion of an article containing mathematical formulae, the analysis prohibition start line code <img file="EP0337068A2_D0001.tif" /> X and the analysis prohibition end line code <img file="EP0337068A2_D0001.tif" /> Y are placed at a top and a bottom, respectively, of the mathematical formulae so that this portion will not be structurally analyzed, as the meaning of the mathematical formulae is not analyzable by the document structure analyzing unit 60. Apart from the mathematical formulae, the analysis prohibition codes can similarly be used for those portions which are written in special or personal manners. In addition, the compulsory return code <img file="EP0337068A2_D0001.tif" /> C can be placed in a middle of a line as shown in Fig. 7.

0041As described according to this embodiment it is possible to have a document formatting apparatus which is both simply manipulatable and consistently correct. It can be seen from the above description that this is due to the particular use of both the automatic logical and referential structures, and the command codes. Consequently, according to this embodiment the misapprehension inevitably accompanying the completely automatic structural analysis can be rectified by the use of the command codes. At the same time, as this embodiment requires the use of the command code in only those places which may causes misapprehension, and not elsewhere, the encumbrances associated with the conventional usage of the command codes can be lessened enormously.

0042It is to be noted that although in the above embodiment, the command codes are placed at the top of lines, this can easily be modified to allow placement of the command code anywhere in the lines.

0043Also, although in the above embodiment the onlycommand code which are directly concerned with layout of the document is the compulsory return code, it is possible to incorporate more complicated command codes such as font command codes and style command codes used in 'Scribe', and thereby enhancing the scope of possible formatting.

0044Also, although in the above embodiment the structural analysis is performed line by line, this can be performed in so-called Top-Down fashion, i.e., the entire document all at once.

0045Moreover, the analysis prohibition codes <img file="EP0337068A2_D0001.tif" /> X and <img file="EP0337068A2_D0001.tif" /> Y can be modified such that within a portion indicated by them the analysis is to be performed by a particular processing system such as T_eX which is known to be highly effective in dealing with mathematical formulae.

0046Furthermore, this embodiment can effectively employed not only for the document containing references to the graphics as in the examples in the foregoing description, but also for the document having references to a list of references or bibliography, and for relational data-bases.

0047Besides these, many modifications and variations of the above embodiment may be made without departing from the novel and advantageous features of the present invention. Accordingly, all such modifications and variations are intended to be included within the scope of the appended claims.