FR2826754A1 - Analysis and indexing of parts of HTML or XML documents in which parts and attributes of a section of a document are memorized in memory zone - Google Patents

Abstract

Has a memory stage during which is memorized the number of parts of a section in a memory zone (ZM). During the memorization stage the number of attributes in the section in memory zone (ZM) is memorized. Method of a section of a document represented in a markup language. The section is delimited by opening and closing markers and eventually includes data. The method consists of getting the address of the opening marker (E1510) and eventually the address of data in the document; getting the size of the opening marker (1520) and eventually the size of the data ; memorizing in a zone (ZM) of; the addresses of the opening marker and data; the sizes of the opening marker and data Independent claims are also included for (1) document analysis and indexing systems; (2) a mobile phone (3) personal digital assistant with a document analysis system.

Description

<Desc / Clms Page number 1>

 The present invention is in the field of processing documents represented with a markup language.

 It will be recalled first of all that a markup language is a language for structuring documents with a view to their use on the Web. The invention relates in particular to the processing of HTML and XML documents, these languages being defined by the World Wide Web Consortium (W3C).

 More precisely, the invention relates to a method for analyzing and indexing a document represented with a markup language, this method being particularly adapted to be implemented in a digital data processing apparatus having little memory, such as a mobile phone or PDA (Personal Digital Assistant, "PDA").

 Indeed, and conventionally, the analysis and indexing of documents represented with a markup language require, especially during a parsing phase, a significant use of memory.

 In order to illustrate this problem, we will describe with reference to FIGS. 1A and 1B an example of a DOM tree ("Document Object Model") built by a classical XML parser for the representation of an XML document.

 Figure 1A is an example of an XML document 10.

 The XML document 10 has a part P1 delimited by a opening tag <mother> and a closing tag </ mother>.

 The part P1 comprises a daughter part P2 delimited by an opening tag <tag> and a closing tag </ tag>, this opening tag having an attribute determined by a name "test" and a value "yes".

 This part P2 has two daughter parts P3 and P4 respectively delimited by the opening tags <daughter 1> <daughter2> and closing </ daughter1> </ daughter2>.

<Desc / Clms Page number 2>

 The document part P3 delimited by the <daughter1> and </ daughter1> tags includes "My first daughter" data. The document portion P4 delimited by the <daughter2> and </ daughter2> tags includes "My second daughter" data.

 FIG. 1B represents a DOM tree 20 representing the XML document 10 of FIG. 1A.

This representation is known to those skilled in the art
Each opening tag of the XML document 10 is associated with a node of the DOM tree 20.

 In addition, each node associated with the opening tag of a given part has as its child the nodes associated with the opening tags of the child parts of this part.

 Thus, the vertex of the DOM tree 20 is a node N1 associated with the opening tag of the document part P1. This node N1 has a single child N2 associated with the document part P2.

 For simplification of language, it will also be said that node N2 has two child nodes N3 and N4 respectively associated with document parts P3 and P4.

 Node N2 also has a child N7 associated with its attribute (test = "yes") contained in the opening tag <test tag = "yes">.

 The node N3 has a son N5 representing the data "My first daughter" of the part P3; likewise, the node N4 has a son N6 representing the data "My second daughter" of the part P4.

 It thus appears, with reference to these FIGS. 1A and 1B, that all the information contained in the document 10 is found in the DOM tree 20 associated with it.

 In particular, the data "My first daughter" and "My second daughter" of the parts P3 and P4 are present in the DOM 20 tree.

 This representation has a disadvantage in the case of a digital data processing apparatus having little memory, because this parsing requires a duplication of all the

<Desc / Clms Page number 3>

 information contained in the document represented with a markup language, and in particular data.

 In order to solve this problem, the invention proposes a syntax analysis allowing a compact representation of a document represented with a markup language.

 More specifically, the invention provides a method for analyzing a portion of a document represented with a markup language, the portion being delimited by an opening tag and a closing tag and possibly including data.

 The analysis method comprises: a step of obtaining: the address of said opening tag and possibly the address of the data in the document; and - the size of the opening tag and possibly the size of the data; and a storage step in a memory zone of: the address of the opening tag and possibly the address of the data; and - the size of the opening tag and possibly the size of the data.

 Thus, the analysis method according to the invention does not copy the data contained in the document part but stores, instead of these data themselves, the address and the size of these data. This saves memory space and therefore has a significant advantage when this analysis method is implemented in a digital data processing apparatus with little memory.

 According to a particular characteristic of the analysis method according to the invention, during the storage step, the number of daughter parts of the document part and the number of attributes of its tag are memorized in the memory area. opens.

 This particular characteristic makes it possible to find directly, during a step of a method of using a part of a document

<Desc / Clms Page number 4>

 represented with a markup language according to the invention, the number of daughter parts and attributes of this part, without it being necessary to restart the parsing of this document.

 According to another particular characteristic of the analysis method according to the invention, the opening tag comprising at least one attribute determined by a name and a value, - during the obtaining step, one obtains: - the size and the address of the name of the attribute in the document; - the size and address of the value of the attribute in the document; and during the storage step, the memory area is memorized by: the size and the address of the name of the attribute in the document; and - the size and address of the value of the attribute in the document.

 This particular feature makes it possible to avoid the duplication of the name and the value of the attribute, and thus to save space in memory, by memorizing only the address and the size of this information.

 According to a particular characteristic of the analysis method according to the invention, the document comprising a sister part of the part, this sister part being consecutive to the said part in the document, - during the obtaining step, one obtains a value representative of the address of the sister party in the document; and during the storage step, the value representative of the address of the sister part is stored in the memory area.

 This particular characteristic makes it possible to find directly, during a step of a method of using a portion of a document represented with a markup language according to the invention, the sister part of this part, without which it is necessary to restart the parsing of the document.

<Desc / Clms Page number 5>

 According to a preferred embodiment of the analysis method according to the invention, the value representative of the address of the sister part is obtained from the difference between the address of the sister part and the address of the party. in the document.

 This particular feature makes it possible to reduce the size of the data stored in the memory area, particularly in the case of indexing an important document, by saving offsets, and not physical addresses.

 The invention also relates to a method for indexing a document represented with a markup language, the document comprising a first part, the indexing method comprising, successively and recursively: a step of analyzing the first part ; a step of analyzing each daughter part of said first part, according to a previously described analysis method.

 This indexing method advantageously makes it possible to index an XML document by using a compact representation of the original document.

It is particularly suitable for being implemented in a digital data processing apparatus having little memory.

 The invention also relates to a method of using a part of a document represented with a markup language, this part being identified by a name of an opening tag and the document being associated with a table obtained by a method of indexing as described above.

 The method of use comprises: a step of searching for the opening tag in the document from information contained in said table; optionally a step of reading at least one attribute of this opening tag from: an address in the document and a size of the name of this attribute, an address in the document and a the size of the value of this attribute, these addresses and these sizes being obtained from the table;

<Desc / Clms Page number 6>

 possibly a step of reading the data of the part from an address in the document and a size of these data, this address and this size being obtained from the table; and recursively, a step of using the daughter parts of the game.

 This method of use makes it possible to reconstruct a part of a document represented with a markup language from a table obtained by an indexing method as briefly described and from the document itself without having to start the analysis again. this document.

 This method of use is particularly advantageous when it is implemented in a data processing apparatus having only a small amount of memory. In this case indeed, the document itself can be stored on another device, only the parts of interest being downloaded into the device as and when they are used.

 The invention also relates to a device for analyzing a portion of a document represented with a markup language, the part being delimited by an opening tag and a closing tag and possibly including data.

 The analysis device comprises: means for obtaining: the address of the opening tag and possibly the address of the data in the document; and - the size of the opening tag and possibly the size of the data; and storage means in a memory area: the address of the opening tag and possibly the address of the data; and - the size of the opening tag and possibly the size of the data.

 The invention also relates to a device for indexing a document represented with a markup language, the document comprising a first part. The indexing device comprises:

<Desc / Clms Page number 7>

 a device for analyzing a part of a document represented with a markup language as described above, adapted to analyze, successively and recursively: the first part; - each girl part of the first part.

 The invention also relates to a device for using a portion of a document represented with a markup language, the part being identified by a name of an opening tag and the document being associated with a table obtained by a device. indexing as described above. The device of use comprises: means for searching the opening tag in the document from information contained in the table; means for reading an attribute of the opening tag from: an address in the document and a size of the name of this attribute; an address in the document and a size of the value of this attribute, these addresses and these sizes being obtained from the table; means for reading the data of the part from an address in the document and a size of these data, this address and this size being obtained from the table.

 The invention also relates to a computer system comprising an analysis device, an indexing device or a device for use as described above.

 This computer system may for example consist of a mobile phone, a personal digital assistant, or a computer.

 The invention also relates to an information carrier, possibly totally or partially removable, readable by a computer system, characterized in that it contains instructions of a computer program, allowing the implementation of a method of analysis, an indexing method or a method of use as described above, when this program is loaded and executed by a computer system.

<Desc / Clms Page number 8>

 This information medium may be a CD-ROM, or a magnetic medium, such as a hard disk or a floppy disk.

 In a variant, this support may be a transmissible medium, such as an electrical or optical signal.

 The invention also relates to a computer program stored on an information medium, the program comprising instructions for implementing an analysis method, an indexing method or a method of use as described above, when loaded and executed by a computer system.

 The advantages of these devices, mobile phone, personal digital assistant, computer system, information carrier, and computer program being identical to those of the processes as briefly outlined above, they will not be recalled here.

 Other aspects and advantages of the present invention will appear more clearly on reading the description of a particular embodiment which follows, this description being given by way of nonlimiting example and with reference to the accompanying drawings, on which: - Figure 1A already described is an example of an XML document; FIG. 1B represents a DOM tree of representation of the document of FIG. 1A; FIG. 2 represents a flowchart describing the main steps of a document indexing method according to the invention; FIG. 3 is a flowchart representing the main steps of a procedure for analyzing a header of an XML document according to the invention; FIG. 4 is a flowchart representing the main steps of a procedure for analyzing a part of a document according to the invention; FIG. 5 represents the main steps of an opening tag processing procedure according to the invention; FIG. 6 is a flowchart representing the main steps of a procedure for processing an attribute according to the invention;

<Desc / Clms Page number 9>

 FIG. 7 is a flowchart representing the main steps of a data processing procedure according to the invention; FIG. 8 represents the table obtained by the indexing method according to the invention for the document of FIG. 1A; FIG. 9 is a flowchart representing the main steps of a method of using a portion of a document represented with a markup language according to the invention; FIG. 10 is a flowchart representing the main steps of FIG. a search procedure of a part of a document according to the invention; FIG. 11 represents the main steps of a procedure for displaying a part of a document according to the invention;

 FIG. 12 is a flowchart representing the main steps of a procedure for displaying an opening tag according to the invention; FIG. 13 represents the main steps of a procedure for displaying the attributes of an opening tag according to the invention; FIG. 14 represents the main steps of a closing tag display procedure according to the invention; FIG. 15 represents a device for analyzing a portion of a document represented with a markup language according to the invention; FIG. 16 represents an indexing device of a document represented with a markup language according to the invention; FIG. 17 represents a device for using a part of a document according to the invention; and - Figure 18 shows a computer system according to the invention.

 FIG. 2 represents a flowchart describing the main steps E200 to E240 of a document indexing method according to the invention.

 During a first step E200, a table T is created. This table T is stored in a non-volatile memory 1810 described later with reference to FIG. 18.

<Desc / Clms Page number 10>

 Alternatively, the table T can be stored in a register of a volatile memory RAM 1820 also described with reference to FIG. 18.

 Step E200 is followed by a step E210 in which a variable i is initialized to the value 1.

 This variable i is stored in a register of the same name of a volatile memory RAM 1820 described later with reference to FIG. 18.

 Step E210 is followed by a step E220 in which document D is traversed until the first occurrence of the "<" character.

 Step E220 is followed by step E225 of calling a procedure for analyzing a header of document D which will now be described with reference to FIG.

 FIG. 3 is a flowchart representing the main steps E300 to E340 of a procedure for analyzing a header of an XML document according to the invention.

 In known manner the header of an XML document begins with the characters "<?" and ends with the characters "?>".

 Since the character "<" has been read during step E220 described above, during step E300, the character following the character "<" in document D is obtained.

 This character C is stored in a register of the same name of the volatile memory RAM 1820.

 Step E300 is followed by a step E310 in which it is verified that the character C read in step E300 is the "?" Character.

 If this is not the case, it means that the XML document is malformed, and the header analysis procedure ends with a step of generating an error message not shown.

 If the character C read in step E300 is the "?" Character, the result of the test E310 is positive.

 This test is then followed by a step E320 during which the document D is traversed up to the character "?" following.

<Desc / Clms Page number 11>

 Step E320 is followed by a step E330 in which the ">" character is read.

 This character marks the end of the XML document header.

cela signifie que le document XML est mal formé, et la procédure d'analyse de l'en-tête se termine par une étape de génération d'un message d'erreur. In the case where a "?" additional is present in the header, or if the character read in step E330 is different from the ">" character,

this means that the XML document is malformed, and the header parsing procedure ends with a step of generating an error message.

 Step E330 is followed by a step E340 in which any separators are traversed.

 These separators are constituted inter alia by the space characters "", tab "-" "and carriage return" J ".

 Step E340 is followed by step E230 of the document indexing method which will now be described in Figure 2.

 During this step E230, document D is scanned to the next character "<".

 This "<" character marks the beginning of a P part of document D.

 Step E230 is followed by step E240 of calling an analysis procedure of this part P which will now be described with reference to FIG. 4.

 Fig. 4 is a flowchart showing the main steps E400 to E446 of a procedure for analyzing a part of a document according to the invention.

 During the first step E400, a variable j is initialized with the value of the variable i. This variable j is stored in a register of the same name of the volatile memory RAM 1820.

 As known, part of the XML document is delimited by an opening tag whose format is of the form "<b>" and a closing tag whose format is of the form "</ b>".

 The step E400 is followed by the step E402 of calling an opening tag processing procedure delimiting the part P of the document.

 This opening beacon processing procedure will now be described with reference to FIG. 5.

<Desc / Clms Page number 12>

 Figure 5 shows the main steps E500 to E540 of the opening tag processing procedure.

 During step E500, the contents of index box i of table T are initialized to 0.

 This box is the first box C1 of the table T modified since the beginning of the analysis procedure of the part P.

 This first box C1 is provided for storing the index of the first box C1 of the table T corresponding to the first sister part PS of the part P. Its contents will eventually be modified later in the step E422 described later.

 The step E500 is followed by a step E510 in which one reads: on the one hand the address A of the current character C of the document D; and on the other hand the word M starting with this character C.

 In the preferred embodiment described here, the address A corresponds to the position, expressed in number of characters, of the current character C in the document D.

 According to the format of the tags described above, the current word starting with this character C is the name of the opening tag defining the part P of the document D.

l'analyse de la partie P1 du document 10, le mot M courant est le mot "mère", le caractère C courant est la première lettre du mot "mère" soit "m" et l'adresse A courante est 2 car"m"est le deuxième caractère du document 10. For example, with reference to FIG. 1A already described, during

the analysis of the part P1 of the document 10, the current word M is the word "mother", the current character C is the first letter of the word "mother" is "m" and the current address A is 2 because "m "is the second character of the document 10.

 The step E510 is followed by a step E520 during which: the address A read during the step E510 is stored in a box C2 of the table T, the box C2 having index i + 1; and - the box C3 of the table T is initialized with the number of characters of the current word M read in step E510.

<Desc / Clms Page number 13>

 Boxes C2 and C3 of table T are provided respectively for storing the address and the length of the name of the opening tag of parts P of document D.

 In the example of the part P1 of the document 10 of FIG. 1A, these values are respectively 2 as we have already seen and 4.4 being the number of characters of the word "mother".

 The step E520 is followed by a step E530 in which the boxes C4 and C5 of the table T respectively of index i + 3 and i + 4 are initialized to 0.

 These boxes are respectively provided to memorize the number of attributes of the opening tag of the part P being analyzed, and the number of girls parts of this part. The contents of these boxes will be modified later.

 Step E530 is followed by a step E540 in which the variable i is incremented by 5 units.

 This step E540 is the last step in the process of processing an opening tag. This is followed by step E404 of the part-of-document analysis procedure which will now be described back in FIG. 4.

 . During the step E404 the variable N is assigned to the value 0.

 This variable is stored in a register of the same name of the volatile memory RAM 1820.

 Step E404 is followed by a step E406 during which any separators placed behind the name of the tag are traversed.

 This step is similar to step E340 already described.

 The step E406 is followed by a step E408 in which the character C immediately following the possible separators read in the previous step E406 is obtained.

 Step E408 is followed by an E410 test in which it is checked whether the character C read in step E408 is the "/" character or the ">" character.

<Desc / Clms Page number 14>

 If this is the case, it means that the opening tag of the P part does not contain any attributes. The result of the E410 test is then positive and this test is followed by a step E414 which will be described later.

 On the other hand, if the character C read during the step E408 differs from the characters "/" and ">" the result of the test E410 is negative. This test is then followed by the call step E411 of an attribute processing procedure which will now be described with reference to FIG. 6.

 FIG. 6 is a flowchart of the main steps E600 to E614 of an attribute processing procedure according to the invention.

During step E600, we read: on the one hand, the address A of the current character C; - and on the other hand, the word M starting with this character C.

This step is identical to the step E510 already described.

The step E600 is followed by a step E602 in which the boxes A 1 and A2 of the table T are initialized with the values A and M obtained previously during the step E600. The indexes of these boxes A 1 and A2 are respectively i and i + 1.

 These boxes A 1 and A2 are provided for storing the address and the length of the name of an attribute of the opening tag of the part P of the document D.

 Step E602 is followed by a step E604 in which the characters "=" "are read, depending on the format of the attributes in an XML document, these characters immediately precede the value of the attribute whose name was read in during the step E600.

 Step E604 is followed by a step E608 in which it reads: on the one hand, the current A address of the character C current in the document D; - and on the other hand, the word M starting with this character C.

 This step is identical to the step E510 already described.

<Desc / Clms Page number 15>

 The step E608 is followed by a step E610 in which the boxes A3 and A4 of the table T of respective indices i + 2 and i + 3 are initialized with the address A and the length of the word M.

 These boxes A3 and A4 are indeed intended to store the address in the document D and the length of the value, an attribute of the opening tag of the part P of the document D.

 Step E610 is followed by a step E612 during which the "" character is read, which character marks the end of the attribute being processed.

 Thus, with reference to part P2 of the document shown in FIG. 1A, the attribute "test =" yes "", this attribute being determined by its name "test" and its value "yes", would be treated by the procedure attribute processing as follows: - address of the name of the attribute in the document D: 17; - length of the name of the attribute: 4; - address of the value of the attribute: 23; - length of the attribute value: 3.

 Step E612 is followed by a step E614 in which the variable i is incremented by 4 units. This step E614 completes the attribute processing procedure according to the invention. This is followed by step E412 of the document part analysis procedure which will now be described back to FIG. 4.

 During step E412 the variable N is incremented by one unit, this variable N being provided for storing the number of attributes of the opening tag of the part P being analyzed.

 Step E412 is followed by steps E406, E408 and E410 already described.

 When all the attributes of the opening tag of the P part have been processed, the result of the E410 test is positive.

 This test E410 is then followed by a step E414 in which the box C4 of index j + 3 of the table T is set to the value N, this value N storing the number of attributes of the opening tag of the part P .

<Desc / Clms Page number 16>

 In this regard, it is recalled that, since the variable j having been initialized to the value i during the step E400, the index box j + 3 corresponds to the box C4 previously described, the contents of the box C4 having been temporarily initialized to the value 0 in step E530.

 Now that the number of attributes of the opening tag of the part P being analyzed is known, the contents of the box C4 of the table is modified during this step E414.

 Step E414 is followed by a step E416 during which the current character C read previously in step E408 is compared with the character "/".

When these characters are identical, it means that the part P being analyzed does not include any daughter PF or data.

Indeed, the character "/" is, according to the XML format, the second character of the closing tag associated with the opening tag treated previously in step E402.

 The test E416 is followed in this case by a first step E418 in which the character "/" is read, then by a step E420, during which the character ">" is read.

 This character marks the end of the closing tag delimiting the part P being analyzed.

 The step E420 is followed by a step E422 during which the index box j of the table T is set to the value i-j.

 It is recalled that: - the variable j stores, since step E400, the index of the first box C1 of the table T filled during the analysis procedure of the part P; and that the variable i stores the index in the table T of the box which will be used for the analysis of the part following the part P, that is to say the sister part PS of the part P.

 Thus, by assigning with the value i-j, the box C1 index j of the table T, is stored in this box C1, the shift to be traveled in the

<Desc / Clms Page number 17>

 table T to obtain the first box of the table T associated with the sister part PS following the part P.

 This allows, during the reading of the table T, to be positioned simply, without having to analyze again the document D, on the first box of the part of the table T associated with the sister part PS immediately following a given part P of document D.

 When during step E416, we find that the character C is different from the character "l", the result of this test is negative. This means that the part P of the document D being analyzed comprises data and / or a daughter part PF. This test is then followed by step E424.

 In step E424, the ">" character is read. This character closes the opening tag processed in step E402.

 Step E424 is followed by a step E426 in which any separators are traversed. This step is similar to step E406 previously described.

 The step E426 is followed by a step E428 during which a variable K is initialized to 0. This variable K is stored in a register of the same name of the volatile memory RAM 1820.

 Step E428 is followed by the calling step E429 of a data processing procedure which will now be described with reference to FIG. 7.

 FIG. 7 represents the main steps E700 to E730 of a data processing procedure according to the invention.

 During the step E700, the address A of the current character C is read in the document D.

 Step E700 is followed by a step E710 during which the characters of document D are read until the character "<" is found. In the rest of the description, it is considered that p characters were read during this step E710.

 Step E710 is followed by an E715 test in which it is tested whether the variable p is strictly positive.

<Desc / Clms Page number 18>

 If this is not the case, this means that part P does not contain data at address A in step E700, and the data processing procedure ends.

 On the contrary, if the variable p is strictly positive, the result of the test E715 is positive. This test is then followed by a step E720 during which the cells 01 is D2 of the table T, these boxes respectively having for index i and i + 1 are set to the values A and p.

 The boxes 01 and D2 of the table T are in fact provided for storing the address A of the data of the part P being analyzed, and the length of these data.

 In the example of the analysis of the part P3 of the document 10 of FIG. 1A, the processing of the data "My first daughter" of this part P3 gives: - A = 39 because the character "M", corresponding to the first character of these data, occupies the 39th position in the document 10; and - P = 17 because the data "My first daughter" has 17 characters.

 Step E720 is followed by a step E730 in which the variable i is incremented by 2 units.

 This step E730 completes the data processing procedure according to the invention. This is followed by step E430 of the part-of-document analysis procedure which will now be described in Figure 4.

 In step E430, the character "<" is read.

 This step E430 is followed by step E432 in which the character C is obtained following the character "<" read in step E430.

 This step E432 is followed by an E434 test in which the character C read in step E432 is compared with the character "/".

 When these characters are identical, it means, in accordance with the XML format, that the "<" character read in step E430 was the first character of the closing tag of part P of document D.

<Desc / Clms Page number 19>

 This test E434 is then followed by a step E440 which will be described later.

 In the case where the result of the test E434 is negative, it means that the character "<" read in step E430 is the first character of an opening tag delimiting a daughter part PF of the part P in progress. analysis.

 This test E434 is then followed by the steps E400 to E446 analysis of this daughter PF.

 Indeed, in accordance with the invention, the indexing of a document represented by a markup language is done, recursively, by analysis of the first part of the document D then, by analysis of each daughter part of this first part .

 At the end of the analysis of the daughter part PF above, during a step E436, the document is scanned to the next character "<".

 The step E436 is followed by step E438 during which the variable K is incremented by one unit, the variable K being used to memorize the number of data and daughter parts PF of the part P being analyzed.

 Step E438 is followed by the calling step E429 of the data processing procedure described above.

 This data processing procedure is, as previously described, followed by steps E430, E432 and E434.

 When all daughter PF parts of part P have been analyzed, the result of the E434 test is positive. This test is then followed by the step E440 during which the box C5 of the table T, this box having for index j + 4, is set to the value K.

 It should be remembered that this box C5 is provided for storing the number K of daughter parts PF and data of the part P being analyzed.

 This step E440 is followed by a step E442 during which the character "/" is read.

 This character is indeed the second character of the closing tag of the part being analyzed.

<Desc / Clms Page number 20>

 This step E442 is followed by a step E444 during which the current word is read, this word corresponding to the name of the tag processed in step E402.

 The step E444 is followed by a step E446 reading the character ">", this character being the last character of the closing tag of the part P being analyzed.

 Step E446 is followed by step E422 already described, this step E422 being the last step of the procedure for analyzing a part of a document according to the invention.

 Fig. 8 shows the table T obtained by the indexing method for document 10 of Fig. 1A.

 It is divided into four blocks, each block containing information associated with a part of a document, or, in other words, with an opening tag.

 We will describe in detail the block associated with the tag <fille1> associated with the part P3, the description for the other blocks not being repeated because identical.

- la case C1 contient la valeur 7, car l'index dans le tableau de la case C1 associée à la partie soeur P4 de P3 est obtenu par un décalage de 7 cases à partir de la case C 1 courante ; - la case C2 contient la valeur 32, cette valeur correspondant à l'adresse du nom de la balise ouvrante < fille 1 > de la partie P3, cette adresse étant comptée en nombre de caractères ; - la case C3 contient la valeur 6 correspondant au nombre de caractères de la chaîne"fiiiel" ; - la case C4 contient la valeur 0 car la partie P3 n'a pas d'attributs ; - la case C5 contient la valeur 1 car la partie P3 contient un bloc de données mais n'a pas de partie fille ; - le bloc associé à la partie P3 ne comprend pas de cases A 1 à A4 car cette partie n'a pas d'attributs ; Thus, for part P3:

- Box C1 contains the value 7, because the index in the table of the box C1 associated with the sister part P4 of P3 is obtained by a shift of 7 boxes from the current box C 1; the box C2 contains the value 32, this value corresponding to the address of the name of the opening tag <daughter 1> of the part P3, this address being counted in number of characters; - box C3 contains the value 6 corresponding to the number of characters of the string "fiiiel"; - Box C4 contains the value 0 because the part P3 has no attributes; the box C5 contains the value 1 because the part P3 contains a block of data but has no daughter part; the block associated with the part P3 does not include boxes A 1 to A4 because this part has no attributes;

<Desc / Clms Page number 21>

 the box D1 comprises the value 39, corresponding to the position of the first character of the data of the part P3; and - the box D2 comprises the value 17 corresponding to the size of the string of characters constituted by these data, namely "My first daughter".

 We will now describe with reference to FIG. 9 the main steps E900 to E920 of a method of using a part P of a document represented with a markup language according to the invention, this part P being identified by the name of its opening tag.

 In the embodiment described here, the part P is displayed. Of course, alternatively, the part P may undergo other treatments, such as for example the translation into another language.

 We will assume later that this document D has been indexed previously by an indexing method according to the invention.

 We will also assume that during this indexing process, a table T has been created in accordance with the indexing method described above. We will now say that this table T is associated with the document D.

 The method of using a part P of the document D, consists in finding in the document D, and in the embodiment described here, to display, for the part P and all the daughter parts PF of this part P, to from the information saved in table T: - the opening and closing tags of these parts; - the attributes of the opening tags of these parts; - the data of these parts.

 During a first step E900, the variable i is initialized to the value 1.

 Step E900 is followed by a step E905 of calling a search procedure of part P.

 This search procedure of the part P will now be described with reference to FIG.

<Desc / Clms Page number 22>

 Fig. 10 is a flowchart showing the main steps E1000 to E1070 of a procedure for searching a part P of a document D according to the invention.

 During the first step E1000, we begin by reading the contents of the box C2 of index i + 1 of the table T.

 This box contains the address in the document D of the opening tag of the first part P of the document D.

 During this same step E1000 is placed in the document D at the address read previously.

Then, in the document D, from this address, a number of characters are read, this number corresponding to the content of the box C3 of index i + 2 of the table T;
This string is, during this same step E1000, assigned to a variable S, the variable S being stored in a register of the same name of the volatile memory RAM 1820.

 At the end of this step E1000, the variable S thus contains the name of the opening tag of the first part P of the document D.

 The step E1000 is followed by a step E1010 in which it is checked whether the name of the tag obtained in step E1000 corresponds to the name of the opening tag identifying the part P.

 This test E1010 consists of comparing the string S and the name of the opening tag of the part P.

 If these strings are identical, the result of the E1010 test is positive and the search procedure for an opening tag is complete.

 Otherwise, the result of the E1010 test is negative.

 This test is then followed by a step E1020 during which the content of box C5 of index i + 4 of table T is assigned to variable K, the content of box C5 corresponding to the number of data and daughter parts. of the first part of document D, as described above with reference to step E440.

<Desc / Clms Page number 23>

 Step E1020 is followed by a step E1030 during which the content of box C4 of index i + 3 of table T is assigned to variable N, the content of box C4 corresponding to the number of attributes of the opening tag of the part P, as described above with reference to step E414.

 During this same step E1030, the variable i is set to the value i + 5 + (4xN). The variable i thus contains the index of the first box C1 of the part of the table T containing information relating to the daughter part PF of the first part of the document D.

 Step E1030 is followed by an E1040 test in which it is tested whether the variable K is strictly positive.

If this is not the case, the result of the E1040 test is negative and the opening tag search procedure ends
When K is strictly positive, the result of the E1040 test is positive. This test is then followed by a calling step E1050 of the search procedure of the daughter part PF of the part P.

 At the end of step E1050, during a step E1060, it is tested whether the desired opening tag has been found in the daughter part PF of the part P.

 This test can be done by reading an indicator updated during step E1010 of the opening beacon search procedure in the PF daughter part, this indicator not being shown here.

 If the desired opening tag has not been found, the result of the E1060 test is negative. This test is then followed by a step E1070 in which the variable K is decremented by one unit.

 Step E1070 is followed by the test E1040 already described.

 On the other hand, when the searched opening tag has been found, the result of the test E1060 is positive and the search procedure of the part P ends.

 At the end of the search procedure of the part P previously described, the variable i contains the index in the table T of the box C 1 of the part of the table T associated with the part P.

<Desc / Clms Page number 24>

 The search procedure of the part P is followed by the step E920 of calling a display procedure of the part P, this procedure will now be described with reference to FIG. 11.

 FIG. 11 represents the main steps E1100 to E1170 of a procedure for displaying a part P.

 During the first step EU 00, a procedure is called for displaying the opening tag of the part P which will now be described with reference to FIG. 12.

Fig. 12 is a flow chart showing the main steps E1200 to E1230 of a display procedure of an opening tag according to the invention.

11 <11 During the first step E1200, the character is displayed
Step E1200 is followed by step E1210 during which the contents of box C2 of index i + 1 of table T are read first.

 This box C2 stores the address in the document D of the name of the opening tag.

 During this same step E1210 is read in the document D, from the address obtained above, a number of characters, this number being contained in the box C3 index i + 2 Table T.

 These characters read in this way are then displayed. They correspond to the name of the opening tag of part P.

 Step E1210 is followed by a step E1220 of calling a procedure for displaying the attributes of the opening tag which will now be described with reference to FIG. 13.

 FIG. 13 represents the main steps E1300 to E1380 of a procedure for displaying the attributes of an opening tag according to the invention.

 During the first step E1300, the content of the box C4 of index i + 3 of table T is assigned to the variable N.

 This box C4 contains the number of attributes of the opening tag being displayed.

 The step E1300 is followed by a step E1305 in which the variable j is set to the value i + 5.

<Desc / Clms Page number 25>

 The variable j thus contains the index in the table T of the box A 1 corresponding to the address in the document D of the first attribute of the opening tag.

 Step E1305 is followed by an E1310 test in which it is tested whether the variable N is strictly positive.

 If this is not the case, it means that all the attributes of the opening tag have been displayed.

 In this case, the attribute display procedure ends.

 In the case where N is strictly positive, the result of the E1310 test is positive.

 This test is then followed by a step E1315 during which the "space" character is displayed in order to separate an attribute from the name of the opening tag or two consecutive attributes.

 The step E1315 is followed by a step E1320 during which is read in the document D, at the address contained in the box A 1 index j of the table T, a number of characters, this number being contained in box A2 of index j + 1 of table T.

 These characters are assigned to the variable S.

 Thus, at the end of step E1320, the variable S contains the name of the first attribute of the opening tag.

 The step E1320 is followed by step E1330 during which the contents of the variable S are displayed.

 Step E1330 is followed by a step E1340 in which the characters "=" "are displayed.

 Step E1340 is followed by a step E1350 during which a number of characters are read in document D at address A3 of index j + 2 of table T. This number is contained in box A4 d index j + 3 of this table T.

 These characters are assigned to variable S.

 Thus, at the end of step E1350, the variable S contains the value of the first attribute of the opening tag currently being displayed.

<Desc / Clms Page number 26>

 The step E1350 is followed by a step E1360 during which the contents of the variable S are displayed.

 Step E1360 is followed by a step E1370 in which the character "" is displayed.

 The step E1370 is followed by a step E1380 during which: the variable j is incremented by 4 units; and the variable N is decremented by 1 unit.

 At the end of this step E1380, the variable j contains the index in the table T of the box containing the address of the name of the next attribute of the tag being displayed; the variable N contains the number of attributes of the tag currently being displayed to be displayed.

 This step E1380 is followed by the test E1310 already described.

 When all the attributes of the opening tag being displayed have been displayed, the result of the E1310 test is negative, and the attribute display procedure terminates.

 The test E1310 is then followed by the step E1230 of the beacon display procedure which will now be described back to FIG. 12.

 During step E1230, the character ">" is displayed.

 This character is the last of the opening tag to display.

 This step E1230 is followed by the step EU 05 which will now be described back to FIG. 11.

 During step E1105, the content of box C5 of index i + 4 of table T is assigned to variable K.

 The step E1105 is followed by a step E1110 during which: on the one hand, the content of the box C4 of index i + 3 of the table T is assigned to the variable N; and on the other hand the variable j is set to the value i + 5 + (4xN).

 The step E1110 is followed by a step E1120, it is tested whether the K value is strictly positive.

<Desc / Clms Page number 27>

Si tel n'est pas la cas, le test E1120 est suivi par une étape E1125 d'appel à une procédure d'affichage d'une balise fermante de la partie P qui sera décrite ultérieurement en référence à la figure 14.

If this is not the case, the test E1120 is followed by a step E1125 of calling a procedure for displaying a closing tag of the part P which will be described later with reference to FIG.

 When K is strictly positive, the result of the E1120 test is positive. This test is then followed by an EU 30 test, in which one tests if the character preceding the current character C is the character "<".

* If this is the case, it means that the current C character corresponds to the first character of a daughter part PF of part P.

 In this case, the test E1130 is followed by a step E1180 of calling a display procedure of the daughter part PF.

 This EU step 80 is followed by an EU step 70 described later.

When the result of the E1130 test is negative, it means that the current C character is the first data character of the P part
In this case, the test E1130 is followed by a step E1140 during which the variable S is assigned a number of characters read at the address contained in the index box j of the table T, this number of characters corresponding to the index j + 1 box of the table T.

 At the end of this step E1140, the variable S contains the data of the part P.

 The step E1140 is followed by a step E1150 during which the contents of the variable S are displayed.

 The step E1150 is followed by a step E1160 in which the variable j is incremented by two units.

 The step E1160 is followed by a step E1170 in which the variable K is decremented by one unit. Step E1170 is followed by the test E1120 already described.

 When the result of the test E1120 becomes negative, this test is followed by a step E1125 of calling a procedure for displaying a closing tag of the part P which will now be described with reference to FIG.

 Fig. 14 shows the main steps E1400 to E1420 of a closing tag display procedure.

<Desc / Clms Page number 28>

 Remember that this procedure is called when the result of the EU 20 test is negative.

 During the first step E1400, the characters "</" are displayed.

 The step E1400 is followed by a step E1410 during which the number of characters, the number of characters, is read in the document D in the box C3 of index i + 2 of the table T. being contained in box C2 of index i + 1 of table T.

 This step E1410 is similar to the already described step EU 10.

 Step E1410 is followed by a step E1420 in which the character ">" is displayed.

 This step E1420 is the last step of a closing tag display procedure according to the invention.

 This is also the last step of the procedure for using a part of a document D according to the invention.

 FIG. 15 represents an analysis device 1500 of a part P of a document D represented with a markup language.

 The analysis device 1500 comprises a module 1510 for obtaining the address A of a character string.

 This module for obtaining the address 1510 of a string of characters is adapted to obtain the address of the opening tag of the part P of the document D.

 This module for obtaining the address 1510 of a character string is also adapted to obtain the address of any data of the part P.

 The analysis device 1500 also comprises a module 1520 for obtaining the size of a string of characters.

 This obtaining module 1520 the size of a character string is adapted to obtain the size of the opening tag of the part P and possibly the size of its data.

<Desc / Clms Page number 29>

ZM. The analysis device 1500 includes a storage module 1530 adapted to store different information in a memory area

ZM.

 This storage module 1530 is particularly adapted to store in the memory zone ZM the address of the opening tag of the part P and possibly the address of the data of the part P, these addresses having been obtained by the module D obtaining 1510 the address of a string of characters.

 This storage module 1530 is also adapted to store in the memory area ZM the size of the opening tag and possibly the size of the data of the part P, these sizes having been obtained by the obtaining module 1520 of the size of a string of characters.

 FIG. 16 represents an indexing device 1600 of a document D represented with a markup language.

 The indexing device 1600 comprises an analysis device 1500 as described above.

 This analysis device 1500 is particularly adapted to successively and recursively analyze a first portion P of the document D, each daughter part PF of the first part P, and each sister part PS of the first part P.

 The indexing device 1600 furthermore comprises a creation module 1610 of a table T, and a backup module 1620 adapted to progressively save in the table T the contents of the memory areas ZM obtained by the analysis device 1500. .

 The indexing device 1600 can be used to index a document in XML.

 The indexing device 1600 can be incorporated in a mobile phone, a PDA or a computer system.

 FIG. 17 represents a device 1700 for using a part P of a document D, the part P being identified by the name of its opening tag.

<Desc / Clms Page number 30>

 The document D is associated with a table T obtained by an indexing device 1600 as described above.

 The user device 1700 comprises a module 1710 for searching an opening tag in the document D from information contained in the table T.

 This module 1710 seeking an opening tag is particularly adapted to read the contents of boxes C2 and C3 of Table T, these boxes respectively containing the address and the size of an opening tag in the document D.

 The search module 1710 of an opening tag is also adapted to compare the name of the opening tag identifying the part P with a string of characters.

 The user device 1700 also comprises a reading module 1720: an attribute of the opening tag identifying the part P; - and data from part P.

 This reading module 1720 is particularly adapted to read the contents of boxes A 1, A2, A3, A4, 01 and D2 of Table T.

 The various means described with reference to FIGS. 15, 16 and 17 are adapted to be incorporated in a computer system, which will now be described with reference to FIG. 18.

 According to the embodiment chosen and represented in FIG. 18, a computer system implementing the invention is for example a microcomputer 1800.

 The computer system 1800 comprises a communication bus 1802 to which are connected: a central unit 1803 (microprocessor); a read only memory 1804, comprising a program "Progr", making it possible to implement an analysis method, an indexing method and a method of use according to the invention; a non-volatile flash memory 1810, in particular comprising a register for storing the table T;

<Desc / Clms Page number 31>

 RAM 1820, comprising registers adapted to record modified variables during the execution of the aforementioned program; a screen 1808 making it possible to display a document represented with a markup language, for example during a display procedure as previously described; - a keyboard 1817; a hard disk 1812; a floppy disk drive 1814 adapted to receive a floppy disk 1816; a communication interface 1818 with a communication network 1807 making it possible to transmit the digital document to a client.

 The communication bus allows communication between the various elements included in the computer system 1800 or connected to it. The representation of the bus is not limiting and, in particular, the central unit is capable of communicating instructions to any element of the microcomputer 1800 directly or through another element of the microcomputer 1800.

 The program noted "Progr" allowing the computer system 1800 to implement the invention, can be stored for example in ROM 1804 (called ROM in the drawing) as shown in Figure 18.

 According to one variant, the floppy disk 1816, just like the hard disk 1812, may contain the program "Progr".

 In the second variant, the program "Progr" can be received to be stored in a manner identical to that described previously via the communication network 1807.

 Floppies can be replaced by any information medium such as, for example, a CD-ROM or a memory card.

 In general, a means of storing information, readable by a computer or by a microprocessor, integrated or not into the device,

<Desc / Clms Page number 32>

 optionally removable, stores a program implementing the methods of analysis, indexing and use according to the invention.

 More generally, the program can be loaded into one of the storage means of the device 1800 before being executed.

 The central unit 1803 will execute the instructions relating to the implementation of the invention, instructions stored in the ROM 1804 or in the other storage elements. Upon power-up, the computer programs described briefly above, which are stored in a non-volatile memory, for example the ROM 1804, are transferred into RAM 1820 which will then contain the executable code of the computer. invention, as well as registers for storing the variables necessary for the implementation of the invention.

 The random access memory RAM 1820 comprises in particular the registers i, j, N, C, K, S and possibly T to store the variables of the same names already described.

Claims (31)

 1. A method for analyzing a part (P) of a document (D) shown with a markup language, said part (P) being delimited by an opening tag and a closing tag and optionally comprising data, characterized in that it comprises: a step of obtaining: the address of said opening tag (E510) and possibly the address of said data (E700) in said document (D); and - the size of said opening tag (E520) and possibly the size of said data (E710); and a step of storage in a memory area (ZM) of: said address of said opening tag (E520) and possibly said address of said data (E720); and - said size of said opening tag (E520) and possibly said size of said data (E720).  2. Analysis method according to claim 1, characterized in that during said storage step, the number of daughter parts (PF) of said part (P) in said memory area is also stored (E440) ( ZM).  3. Analysis method according to any one of claims 1 or 2, characterized in that during said storing step, one stores (E412) the number of attributes of said portion (P) in said memory area (ZM).  4. Analysis method according to claim 3, said opening tag having at least one attribute determined by a name and a value, characterized in that; during said obtaining step, the following is obtained: the size and address of the name of said at least one attribute in said document (E600); <Desc / Clms Page number 34>  the size and the address of the value of said at least one attribute in said document (E608); and in that: during said storage step, said memory zone (ZM) stores: the size and address of the name of said at least one attribute in said document (E602); and the size and address of the value of said at least one attribute in said document (E610).  5. Analysis method according to any one of claims 1 to 4, said document comprising a sister portion (PS) of said portion (P), said sister portion (PS) being consecutive to said portion (P), characterized in that in that: during said obtaining step, a value representative of the address of said sister part (PS) in said document (D) is obtained (E422); and in that: during said storage step, said value representative of said address of said sister part (PS) in said memory area (ZM) is memorized (E422).  6. Analysis method according to claim 5, characterized in that said value representative of said address of said sister part is obtained from the difference between: a storage index in said memory area (ZM) of the address of the opening tag of said sister part (PS); and an index for storing in said memory zone (ZM) the address of the opening tag of said part (P).  7. A method of indexing a document (D) represented with a markup language, said document (D) comprising a first part (P), characterized in that it comprises, successively and recursively: - a step (E420) analyzing said first portion (P); and - an analysis step (E435) of each daughter part (PF) of said first part (P), according to an analysis method according to any one of claims 1 to 6. <Desc / Clms Page number 35>  8. An indexing method according to claim 7, characterized in that it comprises a preliminary step (E200) for creating an array (T), and in that one saves, as and when said steps analyzing, and progressively in said table (T), the content of said memory areas (ZM) obtained during said analysis steps.  9. An indexing method according to any one of claims 7 or 8 for indexing a document in XML language.  10. A method of using a part (P) of a document (D) represented with a markup language, said part (P) being identified by a name of an opening tag and said document (D) being associated to a table (T) obtained by an indexing method according to claim 8, characterized in that it comprises: a search step (E1 01 0) of said opening tag in said document (D) from information contained in said table (T); possibly a reading step (E1320, E1350) of at least one attribute of said opening tag from: an address in said document and a size of the name of said at least one attribute, an address in said document and a size of the value of said at least one attribute, these addresses and these sizes being obtained from said table; optionally, a step of reading (E1140) the data of said part (P) from an address in said document (D) and a size of said data, this address and this size being obtained from said table (T ); and recursively, a step (E1180) of using the daughter parts (PF) of said part (P).  11. Device for analyzing (1500) a part (P) of a document (D) represented with a markup language, said part (P) being delimited by an opening tag and a closing tag and possibly including data characterized in that it comprises: means for obtaining: <Desc / Clms Page number 36>  the address of said opening tag and possibly the address of said data in said document (1510); and - the size of said opening tag and optionally the size of said data (1520); and means (1530) for storing in a memory area (ZM): - said address of said opening tag and possibly said address of said data; and - said size of said opening tag and possibly said size of said data.  12. Device for analyzing (1500) a part (P) of a document (D) represented with a markup language, characterized in that it comprises means adapted to implement a corresponding analysis method. in any one of claims 1 to 6.  13. Device for indexing (1600) a document (D) represented with a markup language, said document (D) comprising a first part (P), characterized in that it comprises: - an analysis device (1500) of a part of a document (D) shown with a markup language according to any one of claims 11 or 12, adapted to analyze, successively and recursively: - said first part (P); each daughter part (PF) of said first part (P).  14. An indexing device (1600) according to claim 13, characterized in that it further comprises means (1610) for creating an array (T), and means (1620) backup adapted to save, progressively in said table (T), the contents of said memory areas (ZM) obtained by said analysis device (1500).  15. Use of the indexing device (1600) according to any one of claims 13 or 14 for indexing a document (D) in XML. <Desc / Clms Page number 37>  16. A device (1700) for using a part (P) of a document (D) represented with a markup language, said part (P) being identified by a name of an opening tag and said document (D). ) being associated with a table (T) obtained by an indexing device according to claim 14, characterized in that it comprises: - means (1710) for searching said opening tag in said document (D) from information contained in said table (T); means (1720) for reading at least one attribute of said opening tag from: an address in said document (D) and a size of the name of said at least one attribute; an address in said document (D) and a size of the value of said at least one attribute, these addresses and these sizes being obtained from said table (T); means (1720) for reading the data of said part from an address in said document (D) and a size of said data, this address and this size being obtained from said table (T).  17. Mobile phone characterized in that it comprises an analysis device (1500) according to claim 11 or 12.  18. Personal digital assistant characterized in that it comprises an analysis device (1500) according to claims 11 or 12.  19. Computer system (1800) characterized in that it comprises an analysis device (1500) according to claims 11 or 12.  20. Mobile phone characterized in that it comprises an indexing device (1600) according to claim 13 or 14.  21. Personal digital assistant characterized in that it comprises an indexing device (1600) according to claim 13 or 14.  22. Computer system (1800) characterized in that it comprises an indexing device (1600) according to claim 13 or 14.  23. Mobile telephone characterized in that it comprises a device of use (1700) according to claim 16. <Desc / Clms Page number 38>  24. Personal digital assistant characterized in that it comprises a device of use (1700) according to claim 16.  25. Computer system (1800) characterized in that it comprises a device of use (1700) according to claim 16.  26. Information medium readable by a computer system, possibly totally or partially removable, in particular a CD-ROM or magnetic medium, such as a hard disk or a diskette, or a transmissible medium, such as an electrical or optical signal, characterized in that it contains instructions of a computer program, allowing the implementation of an analysis method according to any one of claims 1 to 6, when the program is loaded and executed by a computer system.  27. Information medium readable by a computer system, possibly totally or partially removable, in particular a CD-ROM or magnetic medium, such as a hard disk or a diskette, or a transmissible medium, such as an electrical or optical signal, characterized in that it contains instructions of a computer program, allowing the implementation of an indexing method according to any one of claims 7 to 9, when the program is loaded and executed by a computer system.  28. Information carrier readable by a computer system, possibly totally or partially removable, in particular a CD-ROM or magnetic medium, such as a hard disk or a diskette, or a transmissible medium, such as an electrical or optical signal, characterized in that it contains instructions of a computer program, allowing the implementation of a method of use according to claim 10, when this program is loaded and executed by a computer system.  29. Computer program stored on an information carrier, said program comprising instructions for carrying out the analysis method according to any one of claims 1 to 6, when it is loaded and executed by a system. computer.  30. Computer program stored on an information carrier, said program comprising instructions for implementing the <Desc / Clms Page number 39>  indexing method according to any one of claims 7 to 9, when loaded and executed by a computer system.  31. Computer program stored on an information carrier, said program comprising instructions for implementing the method of use according to claim 10, when it is loaded and executed by a computer system.