JPH02297284A - Document processing system and version control system - Google Patents

Abstract

PURPOSE:To perform the version control of document data processed at high speed and with high memory efficiency by finding difference between the document data before and after correction, and regenerating the document data with an arbitrary version. CONSTITUTION:The difference between the document data taken out from a recorder 3 and the document data after edited at a document editing part 2 for a corresponding comparison unit is differentially detected 103 with the comparison unit levels of a logical structure unit, a graphic unit, a page unit, and an area unit, etc., based on the logical ID dictionary of a document buffer 5 setting either document as a reference document. And obtained differential data and reference grammatical data are found as data with new and old versions from a version managing device 105, and they are stored in a memory device 103. Thereby, it is possible to efficiently perform the version control by remarkably reducing the working capacity of a memory.

Description

Detailed Description of the Invention [Field of Industrial Application] The present invention relates to a document processing system and a version control method, and particularly relates to a document processing system and version control method suitable for efficiently managing a large amount of documents. .

[Prior Art] In conventional document processing represented by word processors, version control of documents is hardly performed.

On the other hand, version control technology is most developed in the field of program development. Where.

Utilizing the fact that a source program is expressed as a collection of lines, the difference between before and after modification of a source program is detected line by line. Instead of storing all versions of the source program, only the standard version of the source program is stored, and the differences from the other versions are stored, thereby significantly reducing the storage capacity.

In this way, in the case of a program, changes in one line do not change data in other lines, so differences can be easily detected simply by comparing character string patterns between corresponding lines.

However, in general document processing, this method does not work well.
When a part of a character string in a row is deleted, in many document processing systems, the character string is moved to fill in the deleted portion, so that the data in all the rows from the second row onwards changes. In other words, most rows are affected by just one editing operation, and even if differences are calculated depending on whether or not row data has changed, the amount of information will not be compressed.

In this way, in the case of document data, the method of calculating differences between corresponding lines cannot be used. However, as an alternative method, a method can be considered in which the entire text of the two document data before and after the correction is compared and the difference between them is detected.

That is, it is checked whether any part of one document data exists in the other document, and if the data is the same, it is regarded as the same, and if not, it is treated as a difference.

However, in general, when a document is edited, the corresponding portion may be moved to the beginning or end of the document. Therefore, during pattern matching, it is necessary to sequentially check the correspondence from the beginning to the end of the other document.

[Problems to be Solved by the Invention] In the above-mentioned conventional technology, when calculating the difference between two document files, it is necessary to perform a character string search over the entire text, and it takes a lot of time to calculate the difference between versions. The problem was that it required .

The present invention has been made in view of the drawbacks of the prior art, and its purpose is to be able to perform differential detection at high speed,
Moreover, the object is to provide a document processing system with high memory efficiency.

Another object of the present invention is to provide a version control system that functions similarly.

Another object of the present invention is to provide a document reproduction method that can easily reproduce other versions of document data using differences.

[Means for solving the problem] In order to achieve the above purpose, Honmeimei utilizes document components such as the logical structure of the document (chapters, sections, ..., paragraphs, etc.) for difference detection. It is something.

That is, for example, differences between two documents are investigated at the level of logical structure, and then document data belonging to different logical structures are compared with each other.

As a means for this purpose, the present invention creates and creates document data.
A document processing system comprising a document editing function for modifying and a storage device for storing the document data, the document processing system comprising: a document editing function for modifying a document; and a storage device for storing the document data; document as the reference document,
The present invention is characterized by comprising a difference detection means for determining the difference between both document data for each preset comparison unit, and a version storage management means for storing the difference data and the reference document data in the storage device as new and old version data. shall be.

The comparison unit of documents in the present invention can be set as appropriate depending on the structure of the target document. Preferably, this can be carried out using one or more of logical structural elements, one page of diagrams, an area, etc. that the target document includes as its original constituent elements.

Any document data can be used as a reference document for difference detection, such as the latest document data that has been edited, document data at the time of creation, and version document data located in between.

In addition, in order to achieve the above object, the present invention obtains the difference data each time the version of document data after editing is updated and registered, and the document data obtained by the editing as a new version of the document data.

To provide a version management system that can store the difference data in place of previous document data and store the difference data coexisting with the difference data at the time of previous version update.

Similarly, the present invention obtains the difference data each time the version of document data after editing is updated and registered, and uses the difference data as a new version of document data to update the version of the previous version. To provide a version management method that can be stored together with difference data at the time of update.

Furthermore, in order to achieve the above object, the present invention reads the latest version of data managed by the version management method described above and a series of differential data from the version to the version to be restored from the storage device. , a document reproduction method is provided that combines both of them and reproduces a version of document data intended for restoration.

Similarly, the present invention reads out a series of difference data from the latest version of difference data managed by the above-described version management method to the original version, and reference document data from the storage device, and stores both of them. Synthesize and
To provide a document reproduction method for reproducing a version of document data with only one restoration.

Preferably, the version storage management means has a function of distinguishing which of the difference data and reference document data that constitute the old and new version data is new version data and storing it in the storage device.

Further, the document processing system includes logical structure elements, diagrams,
When document data that has one or more of pages or areas as document components is generated, deleted, or changed by editing operations, this indicates that the document data has been changed in those components. It is preferable to include version identification information management means for adding version identification information, and further, the difference detection means:
It is preferable to provide a function of determining a difference between two versions of document data of document components having different version identification information.

Here, the date and time when the document was created, modified, etc. can be used as the version identification information. Information representing this type of time (time stamp) can be obtained by using the calendar and clock provided in the system.
It can be easily applied. Further, this version identification information and an identifier to be described later may be provided all at once when updating the version of the document data that has been subjected to the editing operation.

As this type of identification information, not only the above-mentioned time stamp but also one variety of information can be used.

For example, the number of version updates can be used.

In addition, in the document processing system described above, means is provided for assigning a unique identifier to each constituent unit for which a difference is sought for document data, and the difference detecting means is provided with a means for assigning a unique identifier to each constituent unit for which a difference is sought, and the document data is It is preferable to have a function of comparing the two and detecting the difference.

Similarly, in the document processing system of the present invention, reference document data stored in the storage device and difference data indicating a difference between the reference document and a version of the document data for the purpose of restoration are transferred from the storage device. It is preferable to include a version reproduction means for reading out the document data, composing them, and reproducing the version of the document data to be restored.

[Function] In the present invention, at a comparison unit level such as a logical structure element, diagram unit, page unit, area unit, etc., for a corresponding comparison unit,
Detect differences from each other. Therefore, since it is not necessary to successively compare the entire document, high-speed comparison can be performed.

That is, if there is no corresponding comparison unit on one side, then
whether the comparison unit itself is newly created;
It will be immediately clear whether the item has been deleted or not. In this case, the comparison unit itself becomes difference data as a whole.

Further, if a comparison unit exists in both documents, a difference can be detected by comparing the constituent character strings, etc.

Here, if an identifier is assigned to the comparison unit of both documents, the corresponding comparison unit can be found using this identifier, and the documents can be compared efficiently.

Furthermore, if the version identification information described above is provided, the version identification information is first compared.

If there is no difference in the information, there is no difference between the units of comparison, and there is no need to compare the constituent character strings. Therefore, the processing time can be reduced accordingly. On the other hand, if there is a difference in version identification information, one character string or the like is compared for that comparison unit to detect difference data.

In the present invention, after obtaining the difference data as described above,
One of the compared documents is set as reference document data, and this reference document and the difference data are stored and retained. In this case, the latest document data may be used as the reference document, or the original document data may be used as the reference document. If the latest document is used as the reference document data, the difference data will be the previous version of the document. This is the data for restoring the data. −
On the other hand, when the original data is used as the reference document, the difference data becomes the restored data of the subsequent version of the document.

If there are multiple versions, differential data is created each time a version is updated and saved in a chain. Therefore,
Using the reference document data and the difference data, it is possible to restore the previous or subsequent version of the document data, and by using this document and the next difference data, it is possible to restore the next version of the document data. Further, it is also possible to synthesize differential data chained from a reference part to a target version, and to restore the target version of document data from this and the reference document data.

Such version management is especially important when documents, such as technical manuals, instruction manuals, specifications, etc., are often modified after creation and documents at each stage must be saved. , is suitable. Such documents generally have a large number of pages, so storing each page in a storage device each time it is modified reduces memory usage efficiency and is undesirable.6 In the present invention, this is not a standard. By storing the document data and the differential data linked thereto, the amount of memory used can be significantly reduced. Furthermore, by referring to the difference data, only the revised portion can be known.

(The following is a blank space) [Example] An example of the present invention will be described using FIGS. 1 to 9.

FIG. 1 shows the configuration of an embodiment of the document processing system of the present invention.

In the figure, 1 is a document processing system, and the hardware includes general-purpose information processing devices such as workstations, word processors, personal computers, etc.
It can be constructed using an electronic editing device, a general-purpose computer, or the like. This document processing system 1 includes a document editing section 2
, a storage device 3, a display 4, a document buffer 5, and a document version management section 6.

The document editing section 2 has a function of creating a new document or modifying an already created document. What is its function?

This is well known and is not directly related to the gist of the present invention, so it will be omitted.6 The document data created by the document editing section 2 is stored in the document buffer 5.
The document data is stored in the 1-document editing section 2 and updated every time a correction or addition is made in the document editing section 2.

This document buffer 5 is populated with, for example, a RAM or the like, and stores document data and logical ■D# documents, as shown in FIG.

The storage device 3 is constituted by a large-capacity storage device using an optical disk, a magnetic disk, etc., and stores document data as a file. Furthermore, a program for constructing this processing system may be stored.

Normally, document data is divided into units represented by words such as ``chapter'', ``section'', and ``section'', and the entire document data is constituted as a collection of these units. Usually, a chapter has a hierarchical structure (tree structure), in which a chapter contains multiple sections, and each section further contains multiple sections or paragraphs. In this specification, a logical structure refers to a data structure introduced to manage document data by focusing on the above-mentioned structural characteristics of document data. Regarding the data structure of the document, IS
O (International 5 standard
Organization), OD A (Offi
ce Document Architecture)
There is also a definition in ``Logical Structure,'' which also uses the term logical structure in a roughly equivalent sense.

FIG. 2 shows an example of a document and its corresponding logical structure.

As shown in this example, the logical structure is six hierarchical levels deep, represented by a pointer tree.

Although it differs depending on the document, there is a paragraph P at the lowest level.Therefore, even if there is no clear logical structure such as "chapter" or "section 1", it is assumed that a logical structure in paragraph graph units exists. Also, the nodes of the pointer tree in Figure 2.

This embodiment will be referred to as a logical structure element.This embodiment is intended for documents containing the above-mentioned logical structure.

In the following, a basic algorithm for detecting a difference using this logical structure will be explained first, and then a concrete implementation method will be explained.

In this embodiment, each logical structure element has a unique identifier (logical I
D) and information (time stamp) indicating the date and time it was updated.

The logical ID is information assigned at the time the logical structure element is generated, and is permanently saved regardless of the version of the document. In other words, even if the document data belonging to a logical structure element is changed or deleted, the logical ID does not change.The logical ID is information unique to each logical structure element, and within the same document. It is never used twice or triple.

The time stamp has the function of distinguishing, based on the date and time, that the document of the logical structure has undergone some kind of update, from other versions of the same part. In other words, it functions as version identification information.

Figure 3 shows an example of the data structure of a logical structure element, where (a) is an explanatory diagram of the data record format that stores the data of the logical structure element, and (b) is an explanatory diagram of the data record format for storing the data of the logical structure element.
This shows an example to which the data structure in (a) is applied.

As shown in FIG. 3(a), a logical ID and a time stamp are stored in the record of the logical structure element. The timestamp is the last two digits of the Western calendar, the month, and 1 o'clock on the day.

Minutes and seconds are stored as 12-digit data. In addition, in order to express the hierarchical relationship of the logical structure, we also use a ``parent node pointer'' that indicates the connection with the upper layer, a ``child node pointer'' that indicates the connection with the lower layer, and ``pointer toward the child node'' that indicates the order relationship within the same hierarchy. A "pointer towards the older brother node" and a "pointer towards the th node" are stored in the record. The value of the pointer represents the record number in which the parent, child, older brother, and first node data of the corresponding node are stored.

As can be seen from the example in (b), in this embodiment, a link is established from the parent node (record 200) to the first node of the child node (record 100), and nodes at the same level are linked by "older brother" and " A tree-like logical structure is expressed by linking with the "th" pointer.

The algorithm that uses logical IDs and timestamps to detect differences between document data before and after editing is shown below.
i) First, I that is the same as the logical structure element of one document data
Check whether an element with D exists in the other document data. If it does not exist, document data under that logical structure element is detected as a difference.

(ii) If it exists, then check the timestamp. If the timestamps are the same.

It can be determined that there has been no change in the logical structure elements between versions.

On the other hand, (iii) if the timestamps are different, this means that there is a difference in the document data under that logical structure element. In this case, (iv) if a lower level logical structure exists under the logical structure element, the processes of (i) to (ni) are repeated.

(V) If the logical structure element is the lowest logical structure element (that is, a baragraph), direct comparison of the document data belonging to the corresponding logical structure element is performed between the two document data.

The above is the difference detection algorithm used in this embodiment.

FIG. 4 shows a system configuration in which the document version management section 6 of FIG. 1 is implemented according to the algorithm described above.

In FIG. 4, the same symbols as in FIG. 1 indicate the same items.

The document version management section 6 includes a logical ID management section 101, a time stamp management section 102, a difference detection section 103, a version creation section 104, and a version storage management section 105.

The logical ID management section 101 and the time stamp management section 102 are each connected to the document editing section.

The logical ID management unit 101 is also connected to the document buffer 5.

Further, the difference detection unit 103 and the version creation unit 104 are connected to the document buffer 5 and the storage device 3, respectively, although the data flows are opposite to each other.

In the document editing section 2, new sentences can be input,
When the editing process is performed, a new bar graph is generated.

Sometimes this baragraph is a chapter or a section. In this case, a new logical structure element is generated. At this time, the logical ID management unit 101 performs the following for the logical structure element:
Assign a unique logical ID.

In order to avoid duplicate use of logical IDs, each document data must have a logical ID dictionary as shown in FIG. Even if a document has multiple vergebonds, only one dictionary exists. All logical data used in all versions are registered in this dictionary. By checking the logical data in this dictionary, duplication of logical IDs can be avoided.

Note that the logical ID dictionary is read into the document buffer 5 together with the document data that is subject to version control, and the logical ID management unit 101 refers to this dictionary.

There are various ways to assign a logical ID, but here we will explain.

The simplest method is to use integers starting from 1. In this case, the maximum value currently being used is a logical ID that has been used before, and duplication can be avoided by setting the new logical ID as (maximum value + 1).

When a logical structure element is defined, the timestamp management unit 102 assigns date and time data to the logical structure element.

Furthermore, when the document data is changed thereafter, the timestamp of the logical structure element corresponding to the document data is updated.

For example, as shown in Figure 5, the logical ID belonging to Chapter 2
Regarding paragraphs 9 and 10, when the document data is changed, the "9th paragraph" and "10th paragraph" of "Chapter 2"
Update the timestamp of the logical structure element corresponding to paragraph J. Among the logical structure elements in FIG. 5, those indicated by hatched circles have new time stamps, and the others have the same time stamps as before modification.

The difference detection unit 103 has a function of detecting a difference between two document data before and after editing. The difference detection algorithm is as described above.

In the case of FIG. 5, by first comparing the logical structure elements, it can be seen that the timestamps of "Paragraph 9" and "Paragraph 10" belonging to "Chapter 2" are different. Next, the character strings of paragraph 9 and paragraph 10 are compared to detect a difference between the character strings.

As a result, as shown in FIG. 6, the difference before and after correction can be detected. In FIG. 6, node differences represent differences in logical structure elements, and text differences represent differences in character strings. All differences are based on the document data after correction, and indicate how the document before correction differs from it.

The node difference in FIG. 6(a) indicates that there is a difference between logical nodes 6.9 and 10 when the document data before modification and the document data after modification in FIG. 5 are compared. Further, logical node 6 is the highest node, and is also at the same level as logical node 1, indicating that it is a node following logical node 1. "*" in differential format
” Both documents possess this logical structure element.

This indicates that there are differences in the texts that belong to it.

On the other hand, the text difference data shown in FIG. 6(b) is.

A method for recovering a pre-conversion document from post-conversion document data;
That is, the 24th character of logical ID9 and logical ID10
This indicates that it is sufficient to perform differential processing on the 23rd character of the logical structure element. In particular, the difference format shown in FIG. 6(b) represents the content of the difference processing, and regarding the former,
This represents the process of removing ``by pressing the conversion key,'' and the process of replacing the latter with ``Into a sentence mixed with kana and kanji.''

The difference detection unit 103 retrieves a certain version of document data from the storage device 3 into the document buffer 5, corrects it in the document editing unit 2, compares the two document data before and after the correction, and calculates the difference. Discover data.

At this time, from the difference detection unit 103 to the version storage management unit 1
05 that the document has been modified, the version storage management unit 105 replaces the document data of the corresponding version in the storage device 3 with the difference data, and further updates the latest document data on the document buffer 5 as the latest version data. Remember.

Thereby, document data other than the latest version can be stored in the form of a difference with respect to the latest version, so that the storage capacity of one document data can be significantly reduced.

The version creation unit 104 has a function of creating an arbitrary version of document data from document data expressed in such a differential format. may decide to consider the latest version data and differential data.

As shown in Figure 6(b), if the difference data is given to delete "by pressing the conversion key" from the 24th character of a paragraph whose logical ID is 9, then the latest version The previous version can be recovered by performing deletion processing on the document data.

On the other hand, if the version one before the latest version can be recovered in this way, then the document data of another version before the latest version can also be recovered in the same way. In this way, by sequentially repeating the recovery process going back through versions, document data of any version can be recovered.

In the above description, when there is a difference between the logical structure elements before and after the update, the difference between the corresponding document data is determined by comparison. On the other hand, when there is a difference in the logical structure elements, the document data belonging to the logical structure elements may be used as the difference data. In this way, the effort of comparing document data can be saved, and the processing speed can be increased.

According to this embodiment, the difference data between the documents before and after editing can be determined in units of the logical structure of the document, and version control data can be created, so high-speed version control can be achieved.

Although the above embodiments have been explained using text as an example, the present invention can also be applied to figures and tables.

Version control can be performed on figures and tables created in a multimedia document, such as the one shown in FIG.

In this embodiment, when a new figure or table is created, an identification number and a time stamp are assigned to the figure or table.

- Update timestamps when data in figures or tables in a document is changed.

Make it possible to identify that there has been a change.

That is, as shown in Fig. 8, when the contents of rFig2J are updated and rFig3J is newly created,
An identification number and a timestamp are assigned to the chart. This identification number corresponds to the logical ID shown in FIG. 3 mentioned above. Therefore, from the identification number and time stamp information, it can be determined whether there is a difference between which figures or tables in the document data before and after the update.

If it is found that there is a difference, the data constituting one figure or table are compared to detect the difference. Version control is performed by storing the difference data and the document before the update, or the difference data and the data after the update, just as in the case of documents.

If there is a difference between the diagram or table before and after the update, the entire data of the diagram or table may be stored as the difference without taking the difference. By doing so, the effort of calculating differences between chart data can be saved, and processing can be speeded up.

According to this embodiment, version management can also be performed for document information other than text, such as figures and tables.

In all of the above embodiments, the content information of the document (Text 1 Figure 1
The idea was to cut out tables (tables, etc.) into certain units and perform unit version control. In addition to this, it is also possible to perform version control using document layout information as a clue.

That is, a document has units of layout information such as pages and areas, as shown in FIG. Therefore, when a document is created, a unique identification number (P
ID) and a time stamp as management information.

If the page information or area information of a document created twice is updated.

8 After updating a document whose timestamp is updated, when it is made into a new version of the document, data in the different page or area is stored only when its identification number and timestamp are different from those before the update.

In this embodiment, it is possible to check whether there is a difference in units of arrangement information, regardless of the content data of one document.

The processing is simple, and the system can be made smaller than the embodiments up to FIG. 8.

[Effects of the Invention] According to the present invention, it is possible to quickly determine the difference between before and after modification of one document data, and to quickly create an arbitrary version of document data, so that version management of document data can be performed at high speed and with high memory efficiency. It can be realized.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention, FIG. 2 is an explanatory diagram for explaining the logical structure of a document, and FIG. 3 is an explanatory diagram for explaining records of logical structure elements. FIG. 4 is a block diagram showing the detailed configuration of the document version management section in FIG. An explanatory diagram showing an example, Fig. 7 is an explanatory diagram showing an example of a multimedia document subject to version control, Fig. 8 is an explanatory diagram showing a diagram version control method, and Fig. 9 is a version in units of pages. It is an explanatory diagram of a management method. 1... Document processing system, 2... Document editing department, 3.
...Storage device, 4...Display, 5...Document buffer, 6...Document version management unit, 101...
Logical ID management unit, 102...time stamp management unit,
103... Difference detection section, 104... Version creation section, 105... Version storage management section.

Claims (1)

[Claims] 1. A document processing system comprising a document editing function for creating and modifying document data, and a storage device for storing the document data, the document processing system comprising: document data retrieved from the storage device; Difference detection means for determining the difference between the two document data for each preset comparison unit, using one of the documents as a reference document regarding the document data after the editing operation; A document processing system comprising: version storage management means for storing version data in the storage device. 2. A document processing system equipped with a document editing function for creating and modifying document data, and a storage device for storing the document data, wherein the document data retrieved from the storage device and after being subjected to editing operations are regarding the document data, a difference detection means for determining the difference between both document data in units of logical structure elements using one of the documents as a reference document, and the storage device which uses the difference data and the reference document data as new and old version data. A document processing system comprising: a version storage management means for storing a version in a document processing system; 3. A document processing system equipped with a document editing function for creating and modifying document data, and a storage device for storing the document data, wherein the document data retrieved from the storage device and after being subjected to editing operations are With regard to the document data, one of the documents is taken as the reference document, and the diagram data included in these documents is
A document characterized by comprising: a difference detection means for determining the difference between both document data on a chart-by-table basis; and a version storage management means for storing the difference data and the reference document data in the storage device as new and old version data. processing system. 4. A document processing system equipped with a document editing function for creating and modifying document data, and a storage device for storing the document data, wherein the document data retrieved from the storage device and after being subjected to editing operations are regarding the document data, a difference detection means for calculating the difference between the two document data, using one of the documents as a reference document, in units of logical structure elements for sentences, and in units of figures and tables for diagram data; 1. A document processing system comprising: version storage management means for storing document data as new and old version data in the storage device. 5. A document processing system equipped with a document editing function for creating and modifying document data, and a storage device for storing the document data, wherein the document data retrieved from the storage device and after being subjected to editing operations are regarding the document data, a difference detection means for determining the difference between both document data in page or area units using one of the documents as a reference document, and the storage device which uses the difference data and the reference document data as new and old version data. A document processing system comprising: a version storage management means for storing a version in a document processing system; 6. A version control method in a document processing system according to claim 2, 3, 4 or 5, wherein the difference data is obtained each time a version of document data after editing is updated and registered; and storing the document data obtained by the editing operation as a new version of document data in place of the previous document data, and storing the difference data coexisting with the difference data at the time of updating the previous version. A version control method characterized by: 7. A version control method in a document processing system according to claim 2, 3, 4 or 5, wherein the difference data is obtained every time a version of document data after editing is updated and registered; A version management method characterized in that the difference data is stored as a new version of document data coexisting with the difference data at the time of previous version update. 8. Reading the latest version of data managed by the version management method according to claim 6 and a series of difference data from the version to the version to be restored from the storage device, and composing both, A document playback method characterized by playing back a version of document data for the purpose of restoration. 9. A series of difference data from the latest version of difference data to the version to be restored managed by the version management method according to claim 7 and the reference document data are read from the storage device, and the two are combined. A document reproduction method characterized in that a version of document data for the purpose of restoration is reproduced. 10. The document processing system according to claim 1, 2, 3, 4, or 5, wherein the version storage management means determines which of the difference data and reference document data that constitute the old and new version data is new version data. A document processing system having a function of distinguishing between different types and storing them in the storage device. 11. The document processing system according to claim 1, 2, 3, 4, or 5, wherein by an editing operation on document data having one or more of logical structure elements, diagrams, pages, or areas as document constituent elements, comprising version identification information management means for adding version identification information indicating that the document data has been changed to those constituent elements when the document data is generated, deleted or changed; and the difference detection means includes two A document processing system comprising a function of determining a difference between document data of document components whose versions have different version identification information. 12. The document processing system according to claim 1, 2, 3, 4, 5, 10, or 11, further comprising means for assigning a unique identifier to each constituent unit for which a difference is to be found in the document data, and the difference detecting means. is a document processing system that has a function of comparing document data for each constituent unit having the same identifier and detecting the differences. 13.Claim 1, 2, 3, 4, 5, 10, 11 or 1
2. The document processing system according to item 2, wherein reference document data stored in the storage device and difference data indicating a difference between the reference document and a version of the document data for the purpose of restoration are read from the storage device. , a document processing system comprising a version reproduction means for synthesizing these and reproducing a version of document data to be restored.