JP2005148970A - Method for converting data - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce the size of tree-structure data, thereby rendering the data convertible into readable character-string data and vice versa. <P>SOLUTION: A data conversion part 2 converts the tree-structure data 1 into character-string data by converting it into a data structure formed by positioning the names of tree elements at the front, then positioning the names and values of the trees' attributes, then positioning tree-element value identifiers, then positioning tree-element values, and finally positioning tree-element separators. Methods of conversion for cases where the tree-element attribute values are the same, where there are elements in master/slave relationship, where tree-structure data assumes a matrix structure, and where the tree structure comprises master, slave, and sub-slave, elements, are also included. A method of restoring the tree-structure data from the character-string data is also included. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present proposal relates to a data conversion method for converting data having a tree structure (hereinafter referred to as tree structure data) into readable character string data and vice versa.

  In recent years, XML (Extensible Markup Language) is often used as a data format. XML is a document description language in which SGML, which is an international standard for defining document formats, is adapted to be distributed over the Internet, and can express data in a tree structure. When expressed in a tree structure, it is easy to make data meaningful and easy to create, edit, and search data. In addition, since XML is expressed in text, it has readability (see, for example, Non-Patent Document 1).

An example in which the tree structure data is represented by a diagram is shown in FIG. 9, and an example in which the tree structure data is described in XML is shown in FIG.
Published by Technical Critics Company, “Revised Standard XML Complete Description (above), Mikitoshi Nakayama, Yasuhiro Okui”, first edition published in April 2001

  When tree structure data constructed in a database or the like is described as character string data by an XML method, there is a problem that the data size becomes large. If the data size is large, it takes time to send and receive data, and the cost of storing data is high.

  An object of the present invention is to provide a data conversion method capable of converting tree structure data into readable character string data by reducing the data size and vice versa.

  The present invention is characterized by the following method in order to solve the above problems.

(1) A data conversion method for converting tree structure data having a tree structure into character string data representing a hierarchy of the tree structure and being readable, and vice versa.
The string data is arranged with a tree element name at the head, followed by a tree attribute name and a tree attribute value, followed by a tree element value identifier, followed by a tree element value, Finally, it is converted into a data structure in which a tree element separator is arranged.

  (2) If the tree element attribute value is common even if there is no tree element attribute, the tree element value is displayed with one tree element name, followed by a tree element value identifier. The element value is converted into a data structure in which the element values are connected and arranged.

  (3) When the tree structure data includes a tree element B belonging to a child hierarchy of the tree element for a certain tree element A, the character string data is arranged in the order of “A, B”, and the tree elements of the child hierarchy If there is not, the data element is converted into a data structure in which an identifier indicating it is arranged after the tree element A.

(4) The tree structure data has a structure arranged as a matrix of m * n, with a row m indicating a hierarchy and a column n representing a tree element in one hierarchy,
The tree structure data is used to convert the data into the character string data.

  (5) In the character string data, a tree hierarchy head identifier is first placed, then a parent tree element is placed next, a child tree element of the parent tree element is placed thereafter, and the end of the child tree element And a data structure in which the identifier at the end of the tree hierarchy is arranged, and the data structure is converted into a data structure in which identifiers are arranged at the head and tail of all tree elements corresponding to each tree element and its child elements.

  (6) When the tree structure data includes a grandchild tree element that becomes a child tree element of the child tree element, the character string data includes a tree hierarchy head identifier again in front of the child tree element, followed by A child tree element is arranged, a grandchild tree element is arranged behind the child tree element, and a tree hierarchy end identifier is arranged at the end of the grandchild tree element.

(7) The restoration from the character string data converted by the above (3) to the tree structure data is as follows:
A first process for obtaining the first element from the character string data and storing it in the stack area as a root tree element;
A second process of acquiring the next element in the tree character string data and checking its type;
When the type of the element is a tree element, a third process in which the acquired element is set as a child tree element, one element put last from the stack area is taken as a parent tree element,
A fourth process of adding a child tree element to the parent tree element and storing it in a stack area;
A fifth process for storing the child tree element in a stack area and returning to the second process to obtain a next element;
In the second process, when the type of the element is a hierarchy identifier, a sixth process of taking out the last element put in from the stack area and returning to the second process;
Is repeated until there are no more elements in the character string data.

(8) The restoration from the character string data converted by the above (4) to the tree structure data is as follows:
A first process of creating matrix data of m rows and n columns from tree character string data;
A second process for determining an element in the first row and the first column of the matrix data as a root element of the tree structure data;
While sequentially incrementing the column n and the row m, the data of the m-th row and the n-th column is added as a child tree element of an element that is smaller than the m-th row and has the maximum value in the (n-1) th column 3 processing,
Is repeated until there are no more elements in the matrix data.

(9) The restoration from the character string data converted by the above (5) or (6) to the tree structure data is as follows:
A first process in which the first tree element of the entire tree string data is a parent tree element (car element) and the other elements are child tree elements (cdr elements);
As long as there is the child tree element, a second process for taking out the child tree elements one by one from the top,
A third process for checking whether or not the extracted child tree element is composed of the car element and the cdr element;
If the child tree element is composed of the car element and the cdr element, a fourth process of adding the car element in the child tree element to the parent tree element;
A fifth process in which the car element in the child tree element is a parent tree element and the cdr element in the child tree element is a child tree element;
A sixth process of determining the parent tree element and the child tree element divided in the fifth process as restored elements and returning to the second process and entering a repetition process for the next element;
A seventh process of adding a child tree element to a parent tree element and returning to the second process when the child tree element is not composed of the car element and the cdr element in the third process;
Is repeated until there are no more elements in the character string data.

  As described above, according to the present invention, there is an effect that the tree structure data can be converted into readable character string data by converting the data size to be small and vice versa. Specifically, the following effects are obtained.

  (1) By using the method of claim 1, the size of the tree character string data representing the tree element value can be made smaller than XML.

  (2) By using the method of claim 2, when tree elements having the same attribute value continue, the data size can be made smaller than that of the first embodiment.

  (3) When there is a child tree element in the tree element, it is necessary to express the tree hierarchy in the tree character string data. In this case, by using the method of claim 3, the closing tag can be omitted, and the size of the tree character string data can be reduced.

  (4) Depending on the structure of the tree structure data, the size of the tree character string data can be made smaller than that of the method of claim 3 by using the method of claim 4.

  (5) By using the method of claim 5 or 6, depending on the structure of the tree structure data, the size of the tree character string data may be made smaller than that of the method of claim 3.

  (6) By using the method of claim 7 or 8, the tree structure data can be restored from the character string data.

  In order to solve the above problem, an embodiment of a method for converting tree structure data into character string data (hereinafter referred to as tree character string data) and its inverse conversion method will be described below.

  First, terms used in this embodiment are defined as follows.

  “Tree structure data”: Data having a tree structure. For example, the data of FIG.

  “Tree element”: An element of a tree constituting tree structure data. For example, the shaded portion in FIG.

  "Tree element name": Name of the tree element. For example, “Accounting Department” in FIG.

  “Tree element attributes”: attributes of tree elements. For example, “zip code” in FIG.

  “Tree element attribute value”: the value of the tree element attribute. For example, “271-0067” in FIG.

  "Tree element value": The value of the tree element. For example, “Fukuzawa Yukichi” in FIG.

  “Tree character string data”: Data obtained by converting tree structure data into character strings.

  "Tree element separator": A specific character that separates tree elements in tree string data.

  In the present invention, tree element data, tree attribute value data structures, and tree element parent-child data structures can be converted to tree character string data with a reduced data size that can describe the hierarchy of tree structure data. provide. The former data structure conversion method is shown in the first and second embodiments, and the latter data structure conversion method is shown in the third to fifth embodiments.

(Embodiment 1) Tree element value, tree element attribute, data structure of tree element attribute value and conversion method In XML, a tree element value is described between tags using tags at the beginning and end of the tree element. Since tags are used, the data size increases accordingly. Therefore, the present embodiment proposes a structure in which no tag is used at the end of a tree element as a data structure representing a tree element value describing a hierarchy of the tree structure data.

  FIG. 1 shows an example of tree structure data and tree character string data structure converted according to this embodiment. Each tree element constituting the tree structure data 1 and stored in a database or the like has data such as a tree element name, a tree element attribute, a tree element attribute value, and a tree element value.

  The data conversion unit 2 converts the target tree element from the tree structure data 1 into tree character string data and outputs it. The data conversion unit 2 also has a function of inversely converting tree character string data into tree structure data.

  In the tree character string data 3 converted by the data converter 2, a tree element name is first arranged. Next, the tree attribute name and the tree attribute value are arranged in the same description method as XML. Next, a tree element value identifier is arranged. This is a mark that a tree element value is arranged next to this identifier, and a dedicated character is used. Then, tree element values are arranged. Finally, place a tree element separator to indicate the end of the tree element. These are schematically represented as follows.

Tree character string data structure = "tree element name + tree element attribute name and tree element attribute value + tree element value identifier + tree element value + tree element separator"
For example, with respect to the tree structure data shown in FIG. 9, an example of XML shown in FIG. 10 is as follows.

[1] Example of tree character string data by XML <member number = “1”> Yukichi Fukuzawa </ member>
<Member number = “2”> Inazo Nitobe <// member>
<Member No. = “3”> Soseki Natsume <// Member>
In the present embodiment, this is converted into the following data structure. Here, the symbol “%” is used for the tree element value identifier and the symbol “,” is used for the tree element separator.

[2] Example of tree character string data according to this embodiment Member number = "1"% Yukichi Fukuzawa,
Member number = "2"% Nitobe Inazo,
Member number = "3"% Natsume Soseki,
As described above, in this embodiment, a tag name is not used for the portion indicating the end of the tag. Therefore, the data size of the tree character string data is reduced. For example, in the above example of [1] XML tree character string data, it is 90 bytes, but in the above [2] example of this embodiment, it is 71 bytes, and the data size is reduced by about 21%. be able to.

(Embodiment 2) Relay Element Value Integration Method In this embodiment, the tree element attribute value is the data structure representing the tree element value describing the hierarchy of the tree structure data, even when there is no tree element attribute or when there is no tree element attribute. If they are common, one tree element name is arranged, and a tree element value identifier is used behind the tree element value to form a structure in which the tree element values are connected and arranged.

  As in the first embodiment, the tree element names are arranged first. Next, the tree attribute name and the tree attribute value are arranged in the same description method as XML. Next, a tree element value identifier is arranged. This is a mark that a tree element value is arranged next to this identifier, and a dedicated character is used. Then, tree element values are arranged. Here, when there is a tree element value having the same tree element name, tree element attribute name, and tree element attribute value, a tree element value identifier is added behind the tree element value, and the tree element value is arranged behind the tree element value identifier. Repeat this combination. These are schematically represented as follows.

Tree character string data structure = “tree element name + tree element attribute and tree element attribute value + tree element value identifier + tree element value + tree element value identifier + tree element value + tree element value identifier + tree element value + …… + Tree element separator "
For example, an XML example in which the tree element name is “membership year” is as follows.

[3] Example of tree character string data by XML <membership year = “2003”> Yukichi Fukuzawa </ member>
<Year of joining the club = “2003”> Inazo Nitobe // Member
<Participation year = “2003”> Soseki Natsume </ members>
In the present embodiment, this is converted into the following data structure. Here,% is used for the tree element value identifier, and the symbol “,” is used for the tree element separator.

[4] Example of tree character string data according to this embodiment Membership year = “2003”% Yukichi Fukuzawa% Inazo Nitobe% Soseki Natsume,
As described above, in the present embodiment, even when there are a plurality of tree element names, they can be described as one. Therefore, the data size of the tree character string data is reduced. For example, in the above XML example [3], 107 bytes is used, but in the example [4] 8 of this embodiment, it is 48 bytes, and the data size is reduced by about 55%. Can do.

(Embodiment 3) Depth priority method In tree structure data, a parent-child relationship (tree hierarchy) exists in each tree element. In XML, a tree hierarchy is expressed by using a tag at the beginning and end of a tree element and writing a child tag between them (see FIG. 10). The present embodiment proposes a method in which the tag at the end of the tree element is not required to represent the tree hierarchy.

(1) Tree Hierarchy Relationship Data Conversion Method In this embodiment, as a data structure representing the hierarchy of tree structure data, a tree element (here A) is a tree element belonging to a child hierarchy of the tree element (here A) In this case, the tree character string data is arranged in the order of “A, B”. If there is no tree element in the child hierarchy, an identifier indicating the element is arranged. The tree element value, the tree element attribute, and the tree element attribute value are arranged by the method shown in the first and second embodiments.

  A data conversion procedure by the data conversion unit 2 in FIG. 1 is shown below.

  (S1) First, a root tree element is acquired.

  (S2) A tree element for which data conversion is designated is added to the tree character string data.

  (S3) If there is no child tree element for the tree element for which data conversion is designated, a unique identifier is added to the tree character string data. This identifier is called a tree hierarchy identifier.

  (S4) If there is a child tree element, it is designated and the processing (S2) and (S3) are recursively repeated. When there are a plurality of child tree elements, (S2) and (S3) are recursively repeated one by one.

  (S5) When processing for all child tree elements is completed, a tree hierarchy identifier is added to the tree character string data.

  For example, as shown in FIG. 2A, when there are a tree element B and a tree element C in a hierarchy one level below a certain tree element A, the symbol “@” is used as the tree hierarchy identifier, and the tree element When the symbol “,” is used for the separator, it becomes as shown in FIG.

  Similarly, the data conversion result of FIG. 9 has the following description.

[5] Data conversion example of tree element having tree hierarchy according to this embodiment Company organization, Tokyo head office, accounting department address = “7th floor” ZIP code = “271-0067”, department number = “1”% Fukuzawa Yukichi, @, Department number = "1"% Nitobe Inazo, @, department number = "1"% Natsume Soseki, @, @, @, Numazu factory, engineering department, @, testing department, @, @, Kansai branch office, sales department , @, @, @,
(2) Method for Restoring Tree Character String Data to Tree Structure Data The data conversion unit 2 has a function for restoring tree character string data to tree structure data. This process is shown in FIG. The

  First, the first element is acquired from the tree character string data, and is used as the root tree element. The element is stored in the stack area (S11). Next, the next element in the tree character string data is acquired (S12), and the type of the element is checked (S13). If the element is a tree element, the acquired element is set as a child tree element, and one element put last from the stack area is taken out and set as a parent tree element (S14). The child tree element is added to the parent tree element (S15), the parent tree element is stored in the stack area (S16), the child tree element is stored in the stack area, the process returns to step S2, and the next An element is acquired (S17).

  If the type of the acquired element is a hierarchy identifier, the last element entered is taken out from the stack area, and the process returns to step S12 to acquire the next element (S18).

  The above processes (S12) to (S18) are repeated until there are no more elements, and the process ends when there are no more elements.

(3) Features of the present embodiment In the present embodiment, each element is arranged once in the tree character string data. Therefore, the degree of increase in the tree character string data size when the number of tree elements is increased as compared to XML. Can be lowered. In addition, when used in combination with the first and second embodiments, tree elements can have element values and attribute values, so that reverse conversion to tree structure data similar to XML can be performed.

(4) Data amount in this embodiment In this embodiment, the data amount in the depth priority method is as follows.

Data amount = total of each element size + tree hierarchy identifier size * number of tree hierarchy identifiers + tree element separator size * number of tree element separators The data amount d of the above tree character string data is x for the total number of elements and e for the element size. When the tree hierarchy identifier size is 1 and the tree element separator size is 1, it can be obtained from the following equation.

d = (e + 3) x-1
This is because d = element size * total number of elements + tree hierarchy identifier size * number of tree hierarchy identifiers + tree element separator size * number of tree element separators. In this expression, element size = e, tree hierarchy identifier size = 1, tree Hierarchy identifier number = total element number = x, tree element separator size = 1, tree element separator number = total element number + tree hierarchy identifier number−1 = 2x−1.
d = ex + x + 2x-1 = (e + 3) x-1
become.

(Embodiment 4) Matrix (1) Data conversion method related to tree hierarchy In this embodiment, a tree structure data is arranged as an m * n matrix as a data structure representing a hierarchy of tree structure data. Columns indicate hierarchies and rows may assign tree elements in one hierarchy, or rows and columns may be reversed. Place the number of rows and columns at the top of the matrix. In addition, a tree element separator is used to classify tree elements. The tree element value, the tree element attribute, and the tree element attribute value are arranged by the method shown in the first and second embodiments.

  For example, when there is a tree element B and a tree element C in the hierarchy immediately below a certain tree element A, it is described as a matrix having 2 rows and 2 ranks, and schematically shown in FIG. It becomes like this.

  When this is described by tree character string data, the symbol “,” is used for the tree element separator as follows.

  [6] Example of tree character string data according to the present embodiment.

2, 2, A, B, C
Similarly, the data in FIG. 9 is as shown in FIG. The numbers in the figure indicate row numbers and column numbers for convenience, and are not included in actual data. This is shown as a character string as follows.

  [7] Example of tree character string data according to the present embodiment.

6,4, company organization, Tokyo head office, accounting department address = "7th floor" zip code = "271-0067", department number = "1"% Yukichi Fukuzawa, ..., department number = "2"% Nitobe Inazo, ,, Employee ID = "3"% Natsume Soseki, Numazu Factory, Engineering Department, Testing Department, Kansai Branch, Sales Department,
(2) Restoration method from tree character string data to tree structure data The data converter 2 has a restoration function from tree character string data to tree structure data. This process is shown in FIG. The

  (S21) Create matrix data of m rows and n columns as shown in FIG. 4B from the tree character string data.

  (S22) The element in the first row and first column of the matrix data is determined as the root element of the tree structure data.

  (S23) Initial setting is made such that m = 1 and n = 2.

  (S24) An element in the m-th row and the n-th column of the matrix data is acquired and added as a child tree element in the m-th row (n-1) column.

  (S25) Increment n (+1).

  (S26) It is checked whether or not n exceeds the maximum value max. If not, the process returns to S24 to add the element in the next m-th row and n-th column.

  (S27) If n exceeds the maximum value max, m is incremented and returned to n = 2, and the element in the second column of the next m row is set as an addition target.

  (S28) It is checked whether or not the increment result of m exceeds the maximum row max. If not, the process returns to S24 to add the next m-th row and n-th element.

  As described above, to restore from the tree character string data to the tree structure data, the matrix data of m rows and n columns is created from the tree character string data, and the element of the first row and the first column of the matrix data is changed to the tree structure data. Determine as root element. After that, the column n and the row m are sequentially incremented, and the data in the m-th row and the n-th column are added as child tree elements of an element that is smaller than the m-th row in the n-1th column and has the maximum value. This process is realized by repeating the process until there are no more matrix data elements.

(3) Features of the present embodiment In the matrix method, it is not necessary to specify the tree hierarchy identifier shown in the third embodiment, and the amount of tree character string data is reduced accordingly. If the matrix method is used, the amount of data of the tree character string data may be smaller than the depth priority method depending on the structure of the tree structure data. By using in combination with the first and second embodiments, tree elements can have element values and attribute values, so that tree structure data similar to XML can be described. The disadvantage is that the data size varies greatly depending on the data structure.

(4) Data volume The data volume in the matrix method is as follows (excluding the number of rows and columns).

Data amount = total of each element size + tree element separator size * number of tree element separators Data amount d can be expressed as the following equation where x is the total number of elements, e is the element size, and 1 is the tree element separator size.

d = (1/4) * x ^ 2 + (e + 1/2) * x- (3/4) (at maximum)
d = (e + 1) × −1 (at minimum)
(Note) The minimum time is when there are no blank elements in the matrix. The maximum time is a case of an m * n matrix having elements only in one row and n columns and the others being blank elements.

because,
d = element size * total number of elements + tree element separator size * number of tree element separators
The number of tree element separators = the total number of elements becomes 1, 1 d = e * x + x−1 = (e + 1) x−1
At maximum,
Number of tree element separators = ((total number of elements + 1) / 2) ^ 2-1
It becomes as follows.

d = e * x + ((x + 1) / 2) ^ 2-1 = (1/4) * x ^ 2 + (e + 1/2) * x- (3/4)
(Embodiment 5) List (1) Data conversion method of tree hierarchy relation In this embodiment, as a data structure representing the hierarchy of tree structure data, the beginning and end of a tree element and all tree elements corresponding to its child elements are identified. The structure is separated by Here, the first identifier is called a tree hierarchy first identifier, the last identifier is called a tree hierarchy last identifier, and is collectively called a tree hierarchy identifier.

  The tree hierarchy head identifier is placed first, the parent tree element is placed next, the child tree element of the parent tree element is placed thereafter, and the tree hierarchy tail identifier is placed at the end of the child tree element. . Tree element separators are used to classify tree elements and tree hierarchy identifiers.

  If there are more child tree elements (called grandchild tree elements) in the child tree element, the tree hierarchy head identifier is again placed before the child tree element, the child tree element is placed behind it, and the grandchild behind it. A tree element is arranged, and a tree hierarchy tail identifier is arranged at the end of the grandchild tree element. Tree element separators are also used to distinguish between tree elements and tree hierarchy identifiers.

  By doing so, the same character as the tree hierarchy identifier can be described in the tree element name.

  The first tree element surrounded by the tree hierarchy identifier is called a car element, and the other tree elements are called cdr elements.

  The tree element value, the tree element attribute, and the tree element attribute value are arranged by the method shown in the first and second embodiments.

  For example, when there are a tree element B and a tree element C in the hierarchy immediately below a certain tree element A, the symbol “(” is used as the tree hierarchy head identifier and the symbol “)” is used as the tree hierarchy tail identifier. When the symbol “,” is used for the tree element separator, the result is as shown in FIG. The car element is the tree element A, and the cdr element is the tree elements B and C.

  Similarly, when there is a tree element B in the hierarchy immediately below a certain tree element A and the tree element C is below it, the result is as shown in FIG. The car element is the tree element A, and the cdr element is (, B, C,).

  Similarly, when there are tree elements B and E in the hierarchy one level below a certain tree element A, tree element C is one level below tree element B, and tree element D is one level below, It becomes like (c) of 6. The car element is the tree element A, and the cdr element is (, B, (, C, D,),) and E.

  Similarly, the data conversion result of FIG. 9 has the following description.

  [8] Example of tree character string data according to the present embodiment.

(, Company organization, (, Tokyo head office, (, accounting department address = "7th floor" zip code = "271-0067", department number = "1"% Fukuzawa Yukichi, department number = "2"% Nitobe Inzo, staff) Number = "2"% Natsume Soseki,),), (, Numazu Factory, Engineering Department, Testing Department,), (, Kansai Branch, Sales Department,),)
(2) Restoration method from tree character string data to tree structure data The data conversion unit 2 is provided with a restoration function from tree character string data to tree structure data. This process is shown in FIG. The

  The first tree element (car element) of the entire tree character string data is set as a parent tree element, and the cdr element is set as a child tree element (S31). The parent tree element and child tree element are passed to the process A and executed (S32). As long as there is a designated child tree element at the start of the process A, child tree elements are taken out one by one from the top (S33). “One by one” means that, for example, if the child tree element is B, (C, C,) E, the first is B, the second is (, C, D,), and the third is E. It is.

  Next, it is checked whether or not the extracted child tree element is composed of a car element and a cdr element (S34). If the child tree element is composed of a car element and a cdr element, the car element in the child tree element is added to the parent tree element (S35). Further, the car element in the child tree element is set as a parent tree element, and the cdr element in the child tree element is set as a child tree element (S36). The parent tree element and the child tree element divided in S36 are determined as restored elements, and the process returns to the process S33 to enter the repetition process for the next element (S37). If the child tree element is not composed of the car element and the cdr element in the process S34, the child tree element is added to the parent tree element, and the process returns to the process S33 to repeat the process for the next element (S38).

(3) Features of the present embodiment In the present embodiment, the size of the tree character string data may be smaller than that of the third embodiment. However, the data size changes depending on the structure of the tree structure data to be converted. The minimum degree of change is smaller than that in the third embodiment and is almost the same as the minimum value in the fourth embodiment. Even the maximum case is smaller than the maximum case of the fourth embodiment. Comparison of data sizes will be described later.

  Further, in this embodiment, by using together with the first and second embodiments, tree elements can have element values and attribute values, so that tree structure data similar to XML can be obtained.

  Disadvantages include: when converting tree string data to tree structure data, it is necessary not only to process the tree string data from the beginning of the tree string data but also to find the tree hierarchy tail identifier corresponding to a certain tree hierarchy head identifier. , Processing becomes complicated. Therefore, the processing load at the time of conversion increases.

(4) Data amount The data amount in this embodiment is as follows.

Data amount = total of each element size + tree hierarchy identifier size * number of tree hierarchy identifiers + tree element separator size * number of tree element separators The data amount d is x for all elements, e for element size, 1 for identifier size, tree If the element separator size is 1, it can be expressed as:

d = (e + 5) × −5 (at maximum)
d = (e + 1) x + 3 (minimum)
(Note) The minimum time is a case where there is one parent tree element and all other tree elements are child tree elements. The maximum time is when there is one parent tree element, one child element tree element, one child element tree in the child element tree, and so on.

because,
At the minimum, the number of identifiers = 2,
d = ex + 2 + (x + 2-1)
= (E + 1) x + 3
At the maximum, the identification index = (x 1 1) * 2
d = ex + (x-1) * 2 + {x + (x-1) * 2-1}
= (E + 5) x-5
The XML data size is the case where the abbreviation tag “/” is used, and the data size of the portion excluding the header and attributes is as follows.

Minimum: (e + 3) x-1
Maximum time: 2x + 7e-2
(Data amount comparison of each embodiment)
Embodiments 3 to 5 have been shown as methods for expressing a tree hierarchy, and the following table shows a comparison of the size of tree character string data when the tree hierarchy is expressed in XML.

  The tree structure data used for the comparison is the data shown in FIG. From this, it can be seen that the data size is 20% or more smaller than XML in any case.

  FIG. 8 is a graph showing the relationship between the number of tree elements and the size of tree character string data in the third to fifth embodiments and XML. Here, the size of each tree element name is 24, the tree element separator size is 1, and the end identifier size is 1. Further, since the data amounts of the fourth and fifth embodiments and the XML method differ depending on the tree structure data, the number of data is minimum (the matrix is most dense) and the number of data is the maximum (matrix). In this example, the tree structure data is binary tree (one tree element has two child tree elements).

  FIG. 8 shows that the data size is smaller in the third embodiment (depth priority) than in XML. In the fourth embodiment (matrix) and the fifth embodiment (list), it can be seen that the data size may be smaller than that in the third embodiment.

5 shows an example of tree structure data and tree character string data structure after data conversion in Embodiment 1 of the present invention. 14 is a data conversion example of a tree hierarchy relationship according to the third embodiment. The conversion processing procedure from the tree character string data to tree structure data in Embodiment 3. 10 is a description example of a matrix structure and a matrix method according to the fourth embodiment. The restoration processing procedure from the tree character string data to the tree structure data in the fourth embodiment. FIG. 10 is a hierarchical structure and character string data example in Embodiment 5. FIG. The conversion process procedure from the tree character string data to tree structure data in Embodiment 5. An example of the number of elements and tree character string data size in XML and the embodiment. Example of tree structure data. Description example of XML.

Explanation of symbols

1 Tree structure data 2 Data converter 3 Tree character string data

Claims (9)

A data conversion method for converting tree structure data having a tree structure into character string data that represents the hierarchy of the tree structure and is readable, and vice versa.
The string data is arranged with a tree element name at the head, followed by a tree attribute name and a tree attribute value, followed by a tree element value identifier, followed by a tree element value, A data conversion method characterized by converting to a data structure in which a tree element separator is arranged at the end.   Even if there is no tree element attribute or the tree element attribute value is common, the tree element value displays one tree element name, followed by the tree element value identifier using the tree element value identifier. 2. The data conversion method according to claim 1, wherein the data is converted into a concatenated data structure.   When the tree structure data includes a tree element B belonging to a child hierarchy of the tree element for a certain tree element A, the character string data is arranged in the order of “A, B”, and there is no tree element of the child hierarchy 2. The data conversion method according to claim 1, wherein the data is converted into a data structure in which an identifier indicating the tree element A is arranged after the tree element A. The tree structure data has a structure arranged as a matrix of m * n, with a row m indicating a hierarchy and a column n representing a tree element in one hierarchy,
2. The data conversion method according to claim 1, wherein the tree structure data is used to convert the data into the character string data.   In the character string data, a tree hierarchy top identifier is first arranged, then a parent tree element is arranged, child tree elements of the parent tree element are arranged thereafter, and the tree hierarchy is arranged at the end of the child tree element. 2. A data conversion method according to claim 1, wherein the data structure is a data structure in which the last identifier is arranged, and is converted into a data structure in which identifiers are arranged at the head and tail of all tree elements corresponding to each tree element and its child elements. .   When the tree structure data includes a grandchild tree element that becomes a child tree element of the child tree element, the character string data includes a tree hierarchy head identifier again before the child tree element, and a child tree element after that. 2. The data conversion method according to claim 1, wherein a grandchild tree element is arranged behind the grandchild tree element, and a tree hierarchy tail identifier is arranged at the end of the grandchild tree element. The restoration from the character string data converted by the data conversion method according to claim 3 to tree structure data is as follows:
A first process for obtaining the first element from the character string data and storing it in the stack area as a root tree element;
A second process of acquiring the next element in the tree character string data and checking its type;
When the type of the element is a tree element, a third process in which the acquired element is a child tree element, one element that is finally inserted from the stack area, and is used as a parent tree element;
A fourth process of adding a child tree element to the parent tree element and storing it in a stack area;
A fifth process for storing the child tree element in a stack area and returning to the second process to obtain a next element;
In the second process, when the type of the element is a hierarchy identifier, a sixth process of taking out the last element put in from the stack area and returning to the second process;
4. The data conversion method according to claim 3, wherein the process is repeated until there are no more elements in the character string data. The restoration from the character string data converted by the data conversion method of claim 4 to the tree structure data is as follows:
A first process of creating matrix data of m rows and n columns from tree character string data;
A second process for determining an element in the first row and the first column of the matrix data as a root element of the tree structure data;
While sequentially incrementing the column n and the row m, the data of the m-th row and the n-th column is added as a child tree element of an element that is smaller than the m-th row and has the maximum value in the (n-1) th column 3 processing,
5. The data conversion method according to claim 4, wherein the process is repeated until there are no more elements in the matrix data. The restoration from the character string data converted by the data conversion method according to claim 5 to the tree structure data is as follows:
A first process in which the first tree element of the entire tree string data is a parent tree element (car element) and the other elements are child tree elements (cdr elements);
As long as there is the child tree element, a second process for taking out the child tree elements one by one from the top,
A third process for checking whether or not the extracted child tree element is composed of the car element and the cdr element;
If the child tree element is composed of the car element and the cdr element, a fourth process of adding the car element in the child tree element to the parent tree element;
A fifth process in which the car element in the child tree element is a parent tree element and the cdr element in the child tree element is a child tree element;
A sixth process of determining the parent tree element and the child tree element divided in the fifth process as restored elements and returning to the second process and entering a repetition process for the next element;
A seventh process of adding a child tree element to a parent tree element and returning to the second process when the child tree element is not composed of the car element and the cdr element in the third process;
The data conversion method according to claim 5, wherein the process is repeated until no character string data element is left.

Images