FR2833094A1 - Personal digital assistant for storing and management of relational databases, comprises microprocessor, and user memory which contains recordings of tables defining fixed and variable sized fields - Google Patents

Abstract

The personal digital assistant has a microprocessor (1), read only memory (2) and user memory (3a) which is adapted to contain recordings of tables defining fixed and variable sized fields in a preestablished order. For each table make a first block having the sizes of the fixed size fields and the references of the variable sized fields and a second block containing variable field references, field sizes and field values

Description

<Desc / Clms Page number 1>

PDA-TYPE COMPUTER DEVICE FOR STORAGE
AND THE MANAGEMENT OF RELATIONAL DATA BASES
The invention relates to a device for digital data processing of the PDA ("Personal Digital Assistant") type: handheld computer such as those marketed under the names Palmez Pocket PC (V, PSION (g) or EPOC &commat; as well as any other portable computing device (electronic organizers, communicating mobile telephones and other terminals) and / or of low processing and storage capacity (such as an on-board device for a motor vehicle intended for example for remote control, or a measuring device and meteorological data processing ...) Such a PDA device comprises: processing means comprising at least one microprocessor (equipped with at least one internal memory called register); means, said non-volatile storage memory, directly read-only access by the processing means, and consisting exclusively of one or more non-volatile electronic memory (s); means, called user memory, storage directly accessible in read and write by the processing means for storing programs and / or user data (programs and / or personal data specific to a user and / or a particular use of the device, data acquired, that is to say, entered by a user and / or measured by measuring means related to the device-meteorological data for example- and / or transmitted to the device, etc. ), and consist exclusively of one or more electronic memory (s); autonomous power supply means; means forming a user interface comprising at least user data acquisition means and means for restoring said data;

<Desc / Clms Page number 2>

 software means stored in the read-only memory and / or in the user memory adapted to manage at least one relational database for organizing the storage of user data in the form of table records each comprising a plurality of fields.

 A PDA device is in particular free of magnetic or optical recording medium mass memory requiring the intervention of a reading device using mechanical means: the user data and programs are in fact stored in an electronic memory, which can be, for example, a non-volatile electronic memory type "flash" (rewritable non-volatile electronic memory without specific equipment, direct access and relatively fast), or a volatile memory electronic memory type "RAM" or memory "SRAM" (RAM, direct access and very fast), the device before, in this case-which is the most common-, be constantly supplied with electrical energy by a battery operating off. These memories, particularly expensive, have low storage capacity, currently of the order of 8 megabytes.

 A relational database consists of tables, each formed of a plurality of records of the same nature (each corresponding to a row of the table) and a plurality of fields (each corresponding to a column of the table), each field receiving a specific value for each record. The fields can be of various type (character string, boolean, date, time, integer, decimal, binary ...), fixed size or variable size. The fields are ordered according to a predetermined order (order of the columns of the table).

 Known PDA devices have software means (database management system) adapted to manage databases defined only with fields of variable size, in order to limit the memory space occupied by the database. These devices are programmed to insert each record in the database in the following way: the data of a record acquired (in particular entered by the user),

<Desc / Clms Page number 3>

 corresponding to the values of the various fields of the table concerned, are recorded one after the other, in the predetermined order of said fields, said data being separated by a separator character (in general, the integer 9, corresponding to the input tab, comma or semicolon depending on the machines and data formats used).

 The search for the value of the third field of a recording therefore requires, once the reading means (microprocessor processing means) of the PDA "positioned" on the beginning of the recording, the reading of the first field up to the recognition of the separator character, then the reading of the second field until recognition of the separator character, ie four operations to make it possible to reach said third field from the first. The construction of the database by means of fields of variable size makes it possible, of course, to limit the memory space used for its storage, but induces subsequent processing of particularly slow data.

 This problem could be solved by using fields of fixed size, sized to allow the recording of the largest values that can be encountered for each field. The format of the records of each of the tables (and in particular the size of the fields of said table) would in this case be stored in a descriptor of the table, so that the logical address of each of the fields of a record, expressed relative to at the address of said registration, would be known. The positioning of the reading means on the beginning of the third field of the recording thus requires only one operation. Such a record structure allows for quick subsequent searches. In return, it uses a very important memory space, which only inexpensive memories of magnetic or optical disk type can offer. This method of building and storing databases is therefore totally unsuitable for PDAs, which have a low memory capacity (and are notably free of mass memory).

 The invention proposes to provide a new solution to the problem of recording data in a database (in particular

<Desc / Clms Page number 4>

 relational) stored in an electronic memory (direct access and low capacity) of a PDA-type computer device, which consumes little memory space and allows for subsequent data processing extremely fast.

 In particular, the invention aims to provide a PDA having speeds of performing single-criteria or multi-criteria searches, on one or more tables, ten to one hundred times higher than those obtained in prior PDA devices.

 The invention relates to a device for digital data processing comprising processing means comprising at least one microprocessor; means, said non-volatile memory read-only memory directly accessible read-only by the processing means, and consisting exclusively of one or more nonvolatile electronic memory (s); means, said user memory, storage directly accessible read and write by the processing means for storing programs and / or user data, and consisting exclusively of one or more electronic memory (s); autonomous power supply means; means forming a user interface comprising at least acquisition means (by input, measurement, remote transmission, etc.) of user data and means for rendering (visual, audible, by transmission to a remote device, etc.) said data ; software means stored in the ROM and / or in the user memory adapted to manage at least one relational database. The device according to the invention is characterized in that it is suitable for: storing, in the user memory, user data in the form of table records each comprising fields of fixed size and / or fields of variable size defined according to a predetermined order (by an operator during the construction of the database), each record of each table corresponding to a plurality of data constituting the respective values of the fields of the table for said record,

<Desc / Clms Page number 5>

 for each record of each acquired table (for example entered by a user): e determine (calculate) for each field of variable size of the table, the size of the value of the field of the acquired record, and calculate a reference referred to as reference of the field, said field references having fixed sizes preferably identical, e register successively (i.e. following each other) in the user memory, following the predetermined order of fields, the value of the field if the field is of fixed size or the reference of the field if it is of variable size, all of these data occupying a first compact block, called fixed block, having a fixed size, to record successively (at the following each other) in the user memory, for each field of variable size, the reference of the field, the size of its value and the said value, all of these data occupying a second compact memory block, said variable block, having a variable size, record at least one piece of information representative of a recording format (type and size) of the data of the recording (and in particular of the fixed block).

 The device according to the invention is therefore adapted to allow the definition and the management of tables constructed with fields of fixed size and fields of variable size, and to treat in a different way the data corresponding to fields of fixed size and those corresponding to to fields of variable size. The device is adapted to create, for each record, a fixed size memory block grouping the fixed size data and the field references of the variable size data, and a variable size memory block grouping the variable size data recognized by their respective field reference.

 Advantageously and according to the invention, for each table, the device is adapted to store, in the user memory, at least one

<Desc / Clms Page number 6>

 a descriptor defining a data recording format, as defined and entered by an operator during the construction of the database, common to at least a plurality of records of said table, and storing, at the beginning of each record, as information representative of a recording format of its data, a format reference for pointing to the descriptor of the record. A single descriptor is preferably defined for each table, said descriptor of said table, describing a recording format common to all the records of the table. It should be noted that the term descriptor refers to both the data describing the format of the records of the table and the memory block in which these data are stored.

 Said descriptor presents a series of information defining the format of the fixed block of the record, comprising, recorded in the pre-established order of the fields, the type and the size of the field if it is of fixed size, or representative information. a field reference (which is an integer of fixed size for example) if said field is of variable size.

 The addition of its format reference to the beginning of the fixed block of a recording allows the processing means to delimit the memory spaces occupied by the different data of the recording. It also allows an operator to easily modify the format of a table or possibly a plurality of records of a table, for example to add or delete a field or to change the type or size. In this case, it suffices to create and save in the user memory a new descriptor defining a modified recording format, and to establish a correspondence between the old and the new descriptor. The PDA according to the invention is then advantageously adapted to convert each record of the table to the new format, and modify the format reference stored at the head of its fixed block. In a variant, the PDA device according to the invention is adapted to convert to the new format only the recordings being read, as part of other management operations of said recordings.

<Desc / Clms Page number 7>

 Advantageously and according to the invention, for each fixed size field of each record of each table, the device is adapted to calculate a logical address of the value of said field in the fixed block of the record (by means of representative information of its recording format), and read said value by accessing it directly in the user memory by means of this address.

 Advantageously and according to the invention, for each of the variable size fields of each record, the device is adapted to calculate a logical address of the reference of said field, said desired field, in the fixed block of the record (by means of the information representative of its recording format), read the reference of said field in the fixed block (by accessing it directly by means of this address), calculate a logical address of the beginning of the variable block (by means of information representative of its recording format), read a reference of a field, said current field, in the variable block, compare it with the reference of the searched field, read the size of the value of the current field in the variable block, read the value of said current field in the variable block if its reference is equal to the reference of the searched field, or, if the reference of the current field is not equal to the reference of the searched field, calculate (at the means of said size of the current field) an address of a reference of a field, said next field, in the variable block and reiterate the operations previously performed by taking the next field as the current field.

 Thus, to access the value of a field of variable size of a recording, the reading means (microprocessor) of the PDA according to the invention read the reference of the searched field in the fixed block (including a logical address expressed from the address of the record is known by means of its descriptor), are positioned at the beginning of the variable block (whose logical address expressed from the address of the record is also known) and successively read the references of field and value size of the various fields registered in the variable block, until recognizing the reference of the searched field. Part of the database being defined by size fields

<Desc / Clms Page number 8>

 fixed, the number of operations necessary to access data corresponding to a variable size field is reduced compared to that required in a previous device (in which all the fields are defined as having a variable size). In addition, the logical address (expressed from the address of the record) of the value of a fixed size field of a record is known by means of the descriptor of the record. Such a value is therefore instantly accessible. The reading and the processing of the stored data are therefore much faster than in the previous devices.

 In one version of the invention, advantageously and according to the invention, for each record acquired, and for each variable size field of the unspecified record, the device is adapted to assign (calculate) to said field and store a reference of field representative of the fact that said field contains no value for the recording, that is to say that no value has been acquired or stored for said field (zero size value). Similarly, for each record acquired, and for each field of variable size of zero value (that is to say equal to 0, being an integer or a real number), the device is adapted for assigning to said field and storing a representative field reference because the acquired field value is zero.

 Thus, the processing means can memorize the result unknown value (respectively zero value) for a field-variable size-searched (zero value respectively), at the simple reading of its field reference in the fixed block, without have to go through the variable block of the record. In fact, for each field of variable size not indicated or of zero value of a record, the device is adapted to calculate a logical address of the reference of said field in the fixed block of the record, to read this reference in the fixed block , and storing information representative of the fact that the reference read is a reference of an empty field or of zero value, that is to say representative of the fact that said field contains no value or contains a zero value for the 'recording. This

<Desc / Clms Page number 9>

 feature of the device allows, moreover, to overcome the registration in the variable block of the reference and size of the fields having a zero value or for which no value is acquired (zero size value) and thus save storage memory.

 Advantageously and according to the invention, the device is adapted to record the fixed block and the variable block of each of the records after each other in the user memory. Each record thus occupies a single compact block (of variable size but of which a first part is of fixed size whatever the data of the recording), in order to optimize the physical organization of the storage memory (it is known in fact, a memory manager uses an average of 20 bytes per block for managing memory).

 Advantageously and according to the invention, the device is adapted to calculate, for each record, a number of fields of variable size referenced in the variable block (fields whose reference, size and value are recorded in the variable block), and to store said number at the beginning of the variable block of said record. To read the value (acquired and non-zero) of a variable size field of a record, the device is in this case adapted to read the number of referenced fields before reading a first current field reference; on each reading of a current field reference, incrementing a number of current field references read; interrupting the reading of the variable block and storing information representative of the fact that an error has occurred as soon as the number of current field references read is equal to the number of referenced fields and no current field reference read is equal to the reference of the searched field (that is to say, automatically interrupt the reading of the variable block after reading the last field reference if the searched field reference has not been recognized). This characteristic constitutes a security measure making it possible to systematically limit the reading to a restricted memory space, and to avoid a possible loop reading of the entire memory

<Desc / Clms Page number 10>

 by the reading means if one of the records has formal anomalies resulting from an error occurred during its insertion for example.

 Advantageously and according to the invention, the device is adapted to calculate and store, for each record of the database, an identifier for pointing to the location of the recording in the user memory. The identifier may be a fixed-size integer, also stored in an address table in correspondence with the physical address or a logical address at the beginning of the record, and thus making it possible to locate the record in the user memory.

 Advantageously and according to the invention, for a field of a table (previously selected by the operator during the construction of the database or any modification operation of the structure of said database), said indexed field, the device is further adapted to develop, update for each new record of said acquired table, and record in the user memory, an index of said indexed field, comprising the identifiers of the records of the table stored one after the other of such so that the values of the indexed field of said records are ordered, that is, ordered according to an order predefined by the operator (alphabetical order for a field of alphabetic character strings, ascending or descending order for a field of integers or decimals ...). The device according to the invention is preferably adapted to perform a dichotomy search on the index in order to insert the identifier of a new acquired record and thus update the index.

 Advantageously and according to the invention, the device is adapted to store, in response to a search made in the database based on search criteria entered by a user, the identifier of each of the records verifying the search criteria. The advantages of this feature are detailed below in an exemplary embodiment of the invention.

<Desc / Clms Page number 11>

 Advantageously and according to the invention, the user memory is chosen from a random volatile electronic memory of the RAM (Random Access Memory) type, a static volatile electronic memory of the SRAM (Static Random Access Memory) type, a non-volatile memory type electronic memory. flash> (memory "EEPROM": "Electrically Erasable Programmable Read Only Memory", erasable in blocks), a memory space of one of these memories dedicated to the storage of data and user programs. In all cases, it is a direct access memory by the processing means, that is to say an electronic memory.

 The invention extends to a method of recording data (called user data) in a digital data processing device comprising processing means comprising at least one microprocessor; means, said non-volatile memory read-only memory directly accessible read-only by the processing means, and consisting exclusively of one or more nonvolatile electronic memory (s); means, said user memory, storage directly accessible read and write by the processing means for storing programs and / or user data, and consisting exclusively of one or more electronic memory (s); autonomous power supply means; means forming a user interface comprising at least user data acquisition means and means for restoring said data; software means stored in the ROM or in the user memory adapted to manage at least one relational database built according to specificities defined by an operator. The method according to the invention is characterized in that: the processing means store, in the user memory, the user data in the form of table records each comprising fields of fixed size and / or fields of variable size defined by the operator according to a pre-established order, each record of each

<Desc / Clms Page number 12>

 table corresponding to a plurality of data constituting the respective values of the fields of the table for said record, for each record of each acquired table (in particular entered by a user), the processing means perform the following operations: e determination (calculation), for each field of variable size of the table, the size of the value of the field of the record, and calculation of a reference of fixed size, called reference of the field, e successive registration in the user memory, following the pre-established field order, the field value if the field is of fixed size or the field reference if it is of variable size, all of these data occupying a first compact block, called a fixed block, presenting a fixed size, e successive registration in the user memory, for each field of variable size, the reference of the field, the size of its value and of said value, the set of e these data occupying a second compact memory block, said variable block, having a variable size, recording at least one piece of information representative of a data format of the recording (and in particular of the fixed block).

 The method for storing data according to the invention advantageously comprises, for each record of the database, one or more of the following steps: for each table, storing, in the user memory, at least one descriptor defining a a data recording format, as defined and entered by the operator, common to a plurality of records of said table, and storing, at the beginning of each record, a format reference for pointing to the descriptor of the record ;

<Desc / Clms Page number 13>

 storing the fixed block and the variable block one after the other in the user memory; for each field of variable size not filled with a record, allocation (calculation) and storage in the fixed block of a field reference representative of the fact that no value has been acquired or memorized for said field; for each field of variable size of zero value of a record, allocation (calculation) and storage in the fixed block of a field reference representative of the fact that the value of said field acquired for said record is zero. determining (calculating) a number of variable size fields referenced in the variable block, and registering said number at the beginning of the variable block of the record; - calculating and storing an identifier for pointing to the location of the recording in the user memory.

 The method according to the invention furthermore comprises, for a field of a table, said indexed field, the preparation, updating for each new record of said acquired table and the recording in the user memory of a index of said indexed field, comprising the identifiers of the records of the table, stored one after the other so that the values of the indexed field of said records are ordered (classified according to an order predefined by the operator). The ranking of the value of the indexed field of a newly acquired record relative to the values of the other records of the table, with a view to updating the index, is preferentially carried out by dichotomy (on the index).

 The invention extends to a method of reading data in a relational database of a PDA device stored according to the recording method according to the invention, comprising the following steps: for reading a data item constituting the value of a fixed size field of a record, calculation of a logical address of the value

<Desc / Clms Page number 14>

 said field in the fixed block of the recording, and reading said value by accessing it directly in the user memory by means of this address; for reading a data item constituting the value of a field of variable size of a record, calculating a logical address of the reference of said field, said desired field, in the fixed block of the record; reading the reference of said field in the fixed block; if the field reference is not an empty field reference or a zero value field, calculating a logical address of the beginning of the variable block, if necessary, reading the number of referenced fields, reading a reference of a field, said current field, in the variable block, incrementation of a number of current field references read, comparison of said reference of the current field with the reference of the searched field, reading of the size of the value of the current field in the variable block, reading the value of said current field in the variable block if its reference is equal to the reference of the searched field, or, if the reference of the current field is not equal to the reference of the searched field, calculation ( by means of the previously read size of the value of the current field) of an address of a reference of a field, said next field, in the variable block and reiteration of the operations previously carried out by taking the e following field as current field if the number of current field references read is less than the number of referenced fields, or interruption of the reading of the variable block and storage of information representative of an error occurred if the number of field references current read is equal to the number of referenced fields. If the reference of the searched field read in the fixed block is an empty reference of field (respectively null value), memorization of information representative of the fact that no value was acquired for this field (respectively that its value is zero).

 The invention also relates to a computer device of the PDA type, methods for storing and reading data in such a device, characterized in combination by all or some of the characteristics mentioned above and hereinafter.

<Desc / Clms Page number 15>

 Other objects, features and advantages of the invention will become apparent on reading the following description which refers to the appended figures (and the tables below) representing preferred embodiments of the invention given solely for the purposes of the invention. non-limiting examples. Figure 1 is a schematic view of a PDA according to the invention; Figure 2 is a diagram illustrating the organization of the various components constituting a PDA according to the invention; Tables 1 to 3 below are an example of tables of a relational database according to the invention; Table 4 shows a record of a PDA according to the invention.

 The cash desk of a video rental store includes a payment terminal and a portable computing device according to the invention for managing the inventory of the store. The latter, illustrated in FIGS. 1 and 2, comprises a microprocessor 1, of small size and cost, but also of low processing capacity (approximately ten times lower than that of a standard desktop computer).

The microprocessor 1 is connected to a non-volatile electronic memory 2 of the "ROM" type ("Read Only Memory"), in which are stored the main operating programs (and in particular an operating system of the type marketed under the names PalmOS &commons; PocketPC &commat;). It is also connected to a volatile electronic memory 3 of the "SRAM" type ("Static Random Access Memory"), part of which 3b, called working memory, is dedicated to performing operations related to the operation of the device and computing operations, and another part 3a, said user memory, is dedicated to the storage of the database and any applications (including the database management system) added by the user. Said data and user applications are thus instantly available as soon as the device is put into operation (unlike a desktop computer whose applications and data are stored in a hard disk and must be loaded into the working memory in order to be used), and instantly saved when they are modified (no copying operation on a hard disk is necessary ...). Since the storage memory of the device is volatile, it must constantly be powered by

<Desc / Clms Page number 16>

 electrical energy by a battery 13, preferably rechargeable. Note that, in a variant, the device may contain a non-volatile memory of the "EEPROM" type (such as a flash memory) for storing data and user programs. In order to avoid the loss of the data stored in the "SRAM" memory 3a in the event of failure or depletion of the battery 13, the PDA incorporates means of communication with external devices in order to allow the export of said data to these devices. external as well as all synchronization operations (compliance of databases stored respectively in the PDA and in external devices): a connection port 10 of the type "USB" port ("Universal Serial Bus") to allow the cable connection of the PDA according to the invention to a desktop computer 8 and / or a modem 9, infrared transmission means 12, and means 11 for transmitting waves for transmitting files to a desktop computer by means of intermediary of an internet server (by sending an e-mail). The PDA is furthermore equipped with a display screen 4, preferably a liquid crystal display, and data input means such as a keyboard 5, control buttons 7 programmed to trigger certain functionalities, and a pen 6b for the handwritten input of data on the touch screen 6a, the device comprising for this purpose software character recognition means stored in the "ROM" memory 2.

To describe the activity of the store, we have established the following tables:
Table 1: Customers

<Tb>
<tb> Identifier-Name <SEP> First Name <SEP> n CL <SEP> Street <SEP> City <SEP> Code <SEP> Date
<tb> cateur <SEP> Postal <SEP> Birth
<tb> 230 <SEP> DURAND <SEP> John <SEP> 025 <SEP> 10 <SEP><SEP> street of <SEP> flowers <SEP> Paris <SEP> 75013 <SEP> 02061953
<tb> 860 <SEP> MARTIN <SEP> Pierre <SEP> 473 <SEP> 20 <SEP> street <SEP> of <SEP> birds <SEP> Paris <SEP> 75018 <SEP> 2108 <SEP> 1969
<tb> 420 <SEP> PINSON <SEP> Cécile <SEP> 325 <SEP> 6 <SEP> place <SEP> of <SEP> the <SEP> station <SEP> Evry <SEP> 91000 <SEP> 09 <SEP > 11 <SEP> 1965
<tb> 370 <SEP> BROWN <SEP> John <SEP> 102 <SEP> 1 <SEP><SEP> Street of the <SEP> Center <SEP> Versailles <SEP> 78000 <SEP> 27041949
<Tb>

<Desc / Clms Page number 17>

 It should be noted that the identifier of the record, indicated in the Customers table, is a reference automatically assigned to the record (when inserted) by the database management system, in order to allow the location of the record in the user memory. It is not a field of said table to be filled in by the user.

Table 2: Movies

<Tb>
<tb> Title <SEP> Gender <SEP> LocationHT <SEP> Barcode <SEP> Quantity
<tb> Indiana <SEP> Jones <SEP> 1 <SEP> Adventure <SEP> 17.00 <SEP> 123456789 <SEP> 4
<tb> West <SEP> Side <SEP> Story <SEP> musical <SEP> 17.00 <SEP> 647599965 <SEP> 2
<tb> Pinocchio <SEP> child <SEP> 10.00 <SEP> 254789462 <SEP> 1
<tb> The <SEP> war <SEP> of <SEP> stars <SEP> science <SEP> fiction <SEP> 13, <SEP> 00 <SEP> 687954285 <SEP> 6
<Tb>
Table 3: Invoices

<Tb>
<tb> InvoiceID <SEP> n CL <SEP> Amount <SEP> HT <SEP> VAT <SEP> Date
<tb> 1021 <SEP> 473 <SEP> 17, <SEP> 00 <SEP> 3, <SEP> 33 <SEP> 03041999
<tb> 3524 <SEP> 025 <SEP> 27.00 <SEP> 5.29 <SEP> 2804 <SEP> 1999
<tb> 10458 <SEP> 025 <SEP> 13.00 <SEP> 2.55 <SEP> 13 <SEP> 01 <SEP> 2000
<tb> 76854 <SEP> 102 <SEP> 34.00 <SEP> 6.66 <SEP> 11 <SEP> 122001
<Tb>
Table 4: Rental (invoice lines)

<Tb>
<tb> InvoiceID <SEP> BarCode <SEP> QuantityLoaded <SEP> Back
<tb> 1021 <SEP> 647599965 <SEP> 1 <SEP> 0404 <SEP> 1999
<tb> 3524 <SEP> 254789462 <SEP> 1 <SEP> 31041999
<tb> 3524 <SEP> 123456789 <SEP> 2 <SEP> 31041999
<tb> 10458 <SEP> 687954285 <SEP> 1 <SEP> 15 <SEP> 012000
<tb> 76854 <SEP> 123456789 <SEP> 1 <SEP> 15 <SEP> 122001
<tb> 76854 <SEP> 647599965 <SEP> 1 <SEP> 16 <SEP> 122001
<Tb>

The fields of the previous tables are defined as follows:
Clients ~ Name: string, fixed size 16 bytes

<Desc / Clms Page number 18>

Clients~DateNaissance : date, taille fixe 4 octets Films-Titre : chaîne de caractères, taille variable Films-Genre : chaîne de caractères, taille variable FilmsJLocationHT : nombre décimal, taille fixe 4 octets FilmsCodeBarres : chaîne de caractères, taille fixe 10 octets Films-Quantité : nombre entier, taille fixe 1 octet Factures~ID : nombre entier, taille fixe 4 octets FacturesMontantHT : nombre décimal, taille variable
Factures~TVA : nombre décimal, taille variable Factures~Date : date, taille fixe 4 octets

LocationFactureID : nombre entier, taille fixe 4 octets LocationCodeBarres : chaîne de caractères, taille fixe 10 octets LocationQuantitéLouée : nombre entier, taille fixe 1 octet Location-Retour : date, taille fixe 4 octets
Selon l'invention, le premier enregistrement de la table Clients est inséré dans la mémoire utilisateur comme illustré sur le tableau suivant (colonne de gauche) :
Tableau 4

Clients ~ First Name: string, variable size Clientsn CL and Facturesn CL: whole number, fixed size 4 bytes Clients-Street: character string, variable size Clients ~ City: character string, variable size ClientsPostalCode: string, size fixed 5 bytes

Customers ~ DateBed: date, fixed size 4 bytes Movies-Title: character string, variable size Movies-Genre: character string, variable size MoviesJLocationHT: decimal, fixed size 4 bytes MoviesCodeBars: character string, fixed size 10 bytes Movies -Quantity: integer, fixed size 1 byte Invoices ~ ID: integer, fixed size 4 bytes InvoicesMontantHT: decimal, variable size
Invoices ~ VAT: decimal number, variable size Invoices ~ Date: date, fixed size 4 bytes

LocationFactureID: integer, fixed size 4 bytes LocationCodeBars: string, fixed size 10 bytes LocationQuantityRun: integer, fixed size 1 byte Location-Return: date, fixed size 4 bytes
According to the invention, the first record of the Customers table is inserted in the user memory as shown in the following table (left column):
Table 4

<Tb>
<tb> Registration-Address <SEP> Observations
<tb> trement
<tb> 230 <SEP> Address <SEP> A <SEP> identifier <SEP> of <SEP> record <SEP> (size <SEP> fixed <SEP>:
<tb> 2 <SEP> bytes)
<tb> 010 <SEP> A + 2 <SEP> reference <SEP> format <SEP> record <SEP> (size <SEP> fixed <SEP>:
<tb> 2 <SEP> bytes)
<Tb>

<Desc / Clms Page number 19>

<Tb>
<tb> DURAND <SEP> A + 4 <SEP> value <SEP> of <SEP> field "Name"<SEP> (size <SEP> fixed <SEP>: <SEP> 16 <SEP> bytes)
<tb> 01 <SEP> A + 20 <SEP> reference <SEP> of <SEP> field "First Name"<SEP> (fixed <SEP> size <SEP>:
<tb> 2 <SEP> bytes)
<tb> 025 <SEP> A + 22 <SEP> value <SEP> of <SEP> field "n CL"<SEP> (size <SEP> fixed <SEP>: <SEP> 4 <SEP> bytes)
<tb> 02 <SEP> A + 26 <SEP> reference <SEP> field "Street"<SEP> (size <SEP> fixed <SEP>: <SEP> 2 <SEP> bytes)
<tb> 03 <SEP> A + 28 <SEP> reference <SEP> of <SEP> field "City"<SEP> (fixed <SEP> size <SEP>: <SEP> 2 <SEP> bytes)
<tb> 75015 <SEP> A + 30 <SEP> value <SEP> of <SEP> field "Code <SEP>postal"<SEP> (fixed <SEP> size <SEP>):
<tb> 5 <SEP> bytes)
<tb> 02061953 <SEP> A + 36 <SEP> value <SEP> field "BirthTime"<SEP> (fixed <SEP> size <SEP>):
<tb> 4 <SEP> bytes)
<tb> A + 40-end <SEP> of the <SEP> block <SEP> fixed <SEP> memory,
<tb> start <SEP> of <SEP> block <SEP> variable-
<tb> 3 <SEP> A + 40 <SEP> number <SEP> fields <SEP> of <SEP> size <SEP> variable <SEP> (size <SEP> fixed <SEP>:
<tb> 2 <SEP> bytes)
<tb> 01 <SEP> A + 42 <SEP> reference <SEP> of <SEP> field "First Name"<SEP> (size <SEP> fixed <SEP>:
<tb> 2 <SEP> bytes)
<tb> 04 <SEP> A + 44 <SEP> size <SEP> value <SEP> of <SEP> field "First Name"<SEP> (size <SEP> fixed <SEP>:
<tb> 2 <SEP> bytes)
<tb> John <SEP> A + 46 <SEP> value <SEP> field "FirstName"<SEP> (size <SEP> variable <SEP>: <SEP> here,
<tb> 4 <SEP> bytes)
<tb> 02 <SEP> A <SEP> +46+ <SEP> T <SEP> -first name <SEP> reference <SEP> field "Street"<SEP> (fixed <SEP> size <SEP>: <SEP> 2 <SEP> bytes)
<tb> 17 <SEP> A + 48 + T ~ first name <SEP> size <SEP> of <SEP> field "Street"<SEP> (fixed <SEP> size <SEP>: <SEP> 2 <SEP> bytes )
<tb> 10 <SEP> street <SEP> of <SEP> A <SEP> +50+ <SEP> T <SEP> -first name <SEP> value <SEP> field "Street"<SEP> (size <SEP> variable <SEP>: <SEP> here,
<tb> flowers <SEP> 17 <SEP> bytes)
<tb> 03 <SEP> A + 50 + Tphename + True <SEP> reference <SEP> of <SEP> field "City"<SEP> (size <SEP> fixed <SEP>:
<tb> 2 <SEP> bytes)
<tb> 5 <SEP> A + 52 + Tfirst name + True <SEP> size <SEP> value <SEP> of <SEP> field "City"<SEP> (size <SEP> fixed <SEP>:
<tb> 2 <SEP> bytes)
<tb> Paris <SEP> A + 54 + Tfirst name + True <SEP> value <SEP> of <SEP> field "City"<SEP> (size <SEP> variable <SEP>: <SEP> here,
<tb> 5 <SEP> bytes)
<tb> A + 54 + Tfirst name + True + T <SEP> city-end <SEP> of <SEP> block <SEP> variable <SEP> and <SEP> of <SEP> record-
<Tb>

T e Prénom désigne la taille (variable) de la valeur du champ Clients-Prénom ; T-rue désigne la taille de la valeur du champ Clients~Rue ;

First Name is the size (variable) of the value of the Customer-First Name field; T-street is the size of the value of the Customer field ~ Street;

<Desc / Clms Page number 20>

 Tville is the size of the Customer-City field value. The address column of the preceding table indicates the logical address of each of the data of the record, expired with respect to the physical address (or a logical address), denoted A, of the record. The value of the field Customers ~ Birthdate is preferably stored at the address A + 36 (and not the odd address A + 35) to facilitate its subsequent operation.

champ de taille fixe est instantané : la valeur du champ ClientsDateNaissance d'un enregistrement donné d'adresse A de la table Clients se trouve à l'adresse A + 36 octets. Access to data is extremely fast. Access to the value of a field of variable size of a recording of the device according to the invention requires a number of operations comparable to that necessary in the case of a recording of an earlier device. To access the value of the Customer-Street field, the microprecessor 1 reads the reference of the said field to the address A + 26 bytes, is "positioned" at the beginning of the variable block at the address A + 40 bytes to read the number of referenced fields, reads, following, the first field reference, not being the searched reference, reads the size of said first field to be able to position itself at the address A + 46 + T-First name and read the second reference field, with respect to the searched reference, reads the size of the value of the field then said value at the address A + 50 + T-First name. It should be noted that this number of operations is in fact generally less than that required for an identical search carried out on a known PDA device, in which the fields are usually all defined as variable size fields in order to reduce the memory space used by the database. Moreover, access to the value of a

fixed size field is instantaneous: the value of the CustomerDateBirth field of a given address A record of the Customers table is at address A + 36 bytes.

 The device according to the invention displays significantly improved utilization performance (in terms of speed of execution of the database processing operations) compared to the previous devices, while using a limited memory space for storing the database. of data.

 Other records in the Customers table are inserted similarly, in a format that can be defined, and stored in a

<Desc / Clms Page number 21>

descripteur Format-Clients dans lequel sont stockés, pour chaque champ, le type du champ et la position, dans le bloc fixe, de sa valeur ou de sa référence.

Format-Clients descriptor in which are stored, for each field, the type of the field and the position, in the fixed block, of its value or its reference.

1

<Tb>
<tb> Format <SEP> Observations
<tb> Customers
<tb> NF2,2 <SEP> NF, <SEP> CF, <SEP> CV, <SEP> D, <SEP> designate <SEP> the <SEP> type <SEP> of <SEP> stored <SEP> field <SEP>:
<tb> CF16, <SEP> 4 <SEP> NF <SEP>: <SEP> number <SEP> of <SEP> fixed <SEP> size
<tb> CV <SEP> (2), <SEP> 20 <SEP> CF <SEP>: <SEP><SEP> string of <SEP><SEP> character of <SEP> fixed <SEP> size
<tb> NF4,22 <SEP> CV <SEP>: <SEP> string <SEP> of <SEP> character <SEP> of <SEP> size <SEP> variable
<tb> CV <SEP> (2), <SEP> 26 <SEP> D <SEP>: <SEP> date, <SEP> of <SEP> Fixed <SEP> size
<tb> CV <SEP> (2), <SEP> 28 <SEP> The <SEP><SEP> digit following <SEP> said <SEP><SEP> letters indicate <SEP><SEP> size <SEP> of the <SEP> field <SEP> if <SEP> is <SEP> of a
<tb> CF5.30 <SEP> field <SEP> of <SEP> fixed <SEP> size, <SEP> or <SEP><SEP> size <SEP> of <SEP> its <SEP> reference <SEP> (number <SEP> between <SEP> parentheses) <SEP> if
<tb> D4,36 <SEP> is <SEP> of a <SEP><SEP> field of <SEP> size <SEP> variable.
<Tb>

The <SEP> number <SEP> specified <SEP> after <SEP> the <SEP> comma <SEP> indicates <SEP> the <SEP> position <SEP> of the <SEP> field <SEP> or <SEP> of <SEP>SEP> sa
<tb> reference <SEP> in <SEP> the <SEP> block <SEP> fixed <SEP> of <SEP> the record.
<Tb>

Thus, <SEP> CF <SEP> 16, <SEP> 4 <SEP> is <SEP> a <SEP> representative <SEP> representative <SEP> of <SEP> makes <SEP> that <SEP> the <SEP> stored <SEP> data
<tb> to <SEP> the address <SEP> A + 4, <SEP> which <SEP> is <SEP> by <SEP> elsewhere <SEP> known <SEP> by <SEP> the <SEP> means <SEP> from <SEP> treatment
<tb> As <SEP> Corresponding <SEP> In <SEP> Field <SEP> Name, <SEP> Is <SEP> A <SEP><SEP> String Of <SEP><SEP> Character Of <SEP> 16 <SEP> bytes
<tb> and <SEP> that it <SEP> matches <SEP> to <SEP> the <SEP> value <SEP> of the said <SEP> field <SEP> Name.
<Tb>

CV <SEP> (2), <SEP> 20 <SEP> is <SEP> a <SEP> representative <SEP> representative <SEP> of <SEP> makes <SEP> that <SEP> the <SEP> given <SEP > stored <SEP> at
<tb> the address <SEP> A + 20 <SEP> is <SEP> the <SEP> reference, <SEP> on <SEP> 2 <SEP> bytes, <SEP> a <SEP> string <SEP > of <SEP> character <SEP> of <SEP> size
<tb> variable <SEP> stored <SEP> in <SEP> the <SEP> block <SEP> variable <SEP> corresponding <SEP> to the <SEP> field <SEP><SEP> First <SEP>.
<Tb>

The <SEP> end <SEP> of the <SEP> block <SEP> fixed <SEP> is <SEP> to <SEP> the address
<tb> A + 36 + 4 <SEP> = <SEP> A + 40.
<Tb>

The <SEP> data <SEP> actually <SEP> stored <SEP> for <SEP> describe <SEP> the <SEP> type <SEP> and <SEP> the <SEP> size <SEP> of the <SEP> field
<tb> in <SEP> the <SEP> descriptor, <SEP> noted <SEP> NF2, <SEP> CF16, <SEP> CV <SEP> (2), <SEP> NF4, <SEP> CF5 <SEP> and <SEP> D4, <SEP> are <SEP> in <SEP> makes <SEP> of
<tb> integer <SEP> (variant <SEP> from <SEP> 1 <SEP> to <SEP> 100).
<Tb>

 Fields that can be searched frequently and for which it is interesting to sort the values are preferably indexed. An index is created defining the corresponding table and indexed field, and listing the identifiers of the concerned records, classified in the order of the different values of said field. For example, the IndexClientsName index corresponding to the Customer-Name field of the Customers table, sorted in alphabetical order, will include the following ordered items: 370,230, 860,420.

 The updating of the index, that is to say the insertion in the index of the identifier of any newly acquired record, is done by dichotomy: to classify in index Index ~ Clients ~ Name a new record whose value of the Customer-Name field is LANIER, the processing means are adapted to read the identifier located in the middle of the list (in

<Desc / Clms Page number 22>

 instance 860 for example (it can also be 230)) which provides access to the corresponding record (physical address B); read the reference of the recording format; in the case of a classification concerning the Customers-Name field, read the name of the registration (MARTIN, this data being located, according to the descriptor's teaching, at the address B + 4), compare the first letter of the name read (M) with that of the name of the record to be inserted (L), L being "lower" than M, read the identifier (230) in the middle of the list delimited by the first identifier of the index and the identifier just read, which allows access to the corresponding record (physical address A); read the reference of its recording format then the name at address A + 4 (DURAND); compare the first letter of the name read (D) with that of the name of the record to be inserted (L), and D being less than L, insert the identifier of the new record between 230 and 860.

 A single-criteria search performed on a single table (example: search, in the Clients table, for clients whose name begins with the letter D) can be performed as follows: - the user specifies at least the following four search criteria : the field and the table concerned by the search, an operator (in the example cited and a reference value (the letter D), if the field concerned is not indexed, that is to say if the values of said field for the different records of the table are not sorted, the processing means 1 read the value of said field for each of the records of the table, as explained above: if it is a size field fixed, the logical address of said field expressed from the address of the record is known, if it is a field of variable size, the processing means read the reference of said field in the fixed block- his address is known-then read successively the number of ch referenced amps and the first field reference of the variable block, compare this first reference to the reference of the searched field previously read in the fixed block, read the size of the field (next data in the variable block) and its value if the two references are equal, or read the size of the field (to delimit the memory space occupied by its value)

<Desc / Clms Page number 23>

 and the reference of the next field in the variable block if the two references being compared are different, etc. if the field is indexed, the processing means 1 performs a dichotomic search on the corresponding index (in the example: IndexClientsName).

 Advantageously and according to the invention, the result of a search consists of the identifiers of the records whose data verify the search criterion (s) (ie, in the previous example: 230). The device according to the invention thus provides the software management means of the database, as a result of a search, means for accessing the searched records. This feature is particularly advantageous in view of the fact that the database is stored on a fast and direct access memory. It results in a considerable reduction in the time and memory space consumed for each search. This feature of the invention also makes it possible to execute queries based on the results of a previous request, in an extremely fast manner, which is impossible in previous devices equipped with search engines of the so-called SQL type ("Structured Query Language ").

 Advantageously and according to the invention, a multi-criteria search carried out on a single table (example: search for clients whose name begins with a D and who live in Paris) is decomposed by the processing means into a plurality of single-criteria, linked searches. by Boolean operators. Said single-criteria searches are performed independently of one another. The identifiers of the records verifying the first criterion (name beginning with a D) are noted in a first "table bit" (bit string comprising a number of bits greater than or equal to the largest identifier used for the table concerned, said bits initially having the value 0): the 230th bit of the "table bit" takes the value 1. The results of the second single-criteria search (Customers-City = Paris) are noted in a second "table bit": the 230th and 860th bits of said table take the value 1. The results of the

<Desc / Clms Page number 24>

 single-criteria searches are then compiled by means of a table, called the truth table: if the Boolean operator is "and" (as is the case in the example described), only the bits having the value 1 in the two "bit tables" keep the value 1 in the truth table (in this case, the 230th bit). The results (record identifiers) of the multi-criteria search are given by the bits of the truth table having the value 1.

 The advantages of the device according to the invention, whose software means take into account the general specifics of the PDAs, are therefore multiple: fast access to the data of the database, compared to a previous device in which the data are defined as the field values of variable size, significantly reduced processing time for database management operations (adding, deleting, modifying data), speed of search execution and easy use of results, which are in the form of means of access to data, memory space occupied by the relatively small database and compatible with the limited storage capacity of PDAs. PDAs according to the invention are more efficient in their use than known devices.

 The various functions according to the invention, described above, can easily be performed by simple programming using traditional tools, using in particular the C ++ language.

 It goes without saying that the invention may be subject to numerous variants with respect to the embodiments previously described and shown in the figures.

Claims (13)

1 / -Device for digital data processing comprising processing means (1) comprising at least one microprocessor; means (2), said non-volatile memory read-only memory directly accessible read-only by the processing means, and consist exclusively of one or more non-volatile electronic memory (s); means (3a), said user memory, storage directly accessible read and write by the processing means for storing programs and / or user data, and consisting exclusively of one or more electronic memory (s) ( s); autonomous power supply means (13); means forming a user interface comprising at least means (5,6a, 6b, 7) for acquiring user data and means (4) for restoring said data; software means stored in the read-only memory and / or in the user memory adapted to manage at least one relational database,

 said device being characterized in that it is adapted to: - store user data in the user memory (3a) in the form of table records, each comprising fields of fixed size and / or fields of variable size defined according to an order preset by an operator, each record of each table corresponding to a plurality of data constituting the respective values of the fields of the table for said record, for each record of each acquired table: 'determine, for each variable size field of the table, the size of the value of the field of the acquired record, and compute a reference of fixed size, called reference of the field, * to register successively in the user memory (3a), following the pre-established order of fields, the value of the field if the field is of fixed size or the reference of the field if it is of variable size, <Desc / Clms Page number 26>  all of these data occupying a first compact block, called a fixed block, having a fixed size, * successively registering in the user memory (3a), for each field of variable size, the reference of the field, the size of its value and said value, all of these data occupying a second compact block, said variable block, having a variable size, to record at least one piece of information representative of a format of the data of the recording.  2 / -Dispositif according to claim 1, characterized in that it is adapted to store in the user memory, for each table, a descriptor defining a data recording format, as defined and entered by an operator, common to at least a plurality of records of said table, and storing, at the beginning of each record, a format reference for pointing to the descriptor of the record.  3 / -Dispositif according to one of claims 1 or 2, characterized in that, for each field of fixed size of each record of each table, the device is adapted to calculate a logical address of the value of said field in the fixed block of the recording, and read said value by accessing it directly by means of this address.  4 / -Dispositif according to one of claims 1 to 3, characterized in that, for each field of variable size of each record of each table, the device is adapted to calculate a logical address of the reference of said field, said desired field in the fixed block of the record, read the reference of said field in the fixed block, calculate a logical address of the beginning of the variable block, read a reference of a field, said current field, in the variable block, compare it with the reference of the searched field, read the size of the value of the current field in the variable block, read the value of said current field in the variable block if its reference is equal to the reference of the searched field, or, if the reference of the current field is not equal to the reference of the searched field, to calculate an address of a reference of a field, said field <Desc / Clms Page number 27>  next, in the variable block and repeat the operations previously performed by taking the next field as the current field.  5 / -Dispositif according to one of claims 1 to 4, characterized in that, for each record acquired, and for each field of variable size unspecified or, respectively, of zero value of said record, the device is adapted to calculate and storing a field reference representative of the fact that the field is unspecified or, respectively, of zero value.  6 / -Dispositif according to claim 5, characterized in that, for each field of variable size unspecified or, respectively, of zero value of a record, the device is adapted to calculate a logical address of the reference of said field in the fixed block of the record, read the reference of said field in the fixed block, and store information representative that said field contains no value or, respectively, contains a zero value for the record.  7 / -Dispositif according to one of claims 1 to 6, characterized in that it is adapted to record the fixed block and the variable block of each of the records after each other in the user memory.  8 / -Dispositif according to one of claims 1 to 7, characterized in that it is adapted to calculate, for each record, a number of variable size fields referenced in the variable block, and store said number at the beginning of the block variable of said record.  9 / -Dispositif according to claims 4,6 and 8, characterized in that it is adapted to read the number of referenced fields before reading a first current field reference, each reading of a current field reference, increment a number of current field references read, interrupt the reading of the variable block and store information representative of an error occurring as soon as the number of current field references read is equal to the number of referenced fields and no current field reference read is not equal to the reference of the searched field. <Desc / Clms Page number 28>  10 / -Dispositif according to one of claims 1 to 9, characterized in that it is adapted to calculate and store, for each record, an identifier for pointing to the location of the recording in the user memory.  11 / -Dispositif according to claim 10, characterized in that, for a field of a table, said indexed field, the device is programmed to develop, update for each new record of said acquired table and save in the user memory an index of said indexed field, comprising the identifiers of the records of the table, stored one after the other so that the values of the indexed field of said records are ordered.  12 / -Dispositif according to one of claims 10 or 11, characterized in that it is adapted to store, in response to a search made on the database from search criteria entered by a user, the identifier each record checking the search criteria.  13 / -Device according to one of claims 1 to 12, characterized in that the user memory is selected from a memory

 dynamic volatile electronic RAM type, a static volatile electronic memory of the SRAM type, a non-volatile electronic memory of the flash memory type, a memory space of one of these memories dedicated to the storage of data and user programs.