CN1296855C - Key constrained automatic generation method for expandable mark language mode - Google Patents

Abstract

The present invention provides a method for generating the level description and the key constraint of an extensible markup language schema (XML Schema) based on a schema diagram of a relational database, which comprises the following steps that keys and external keys of the relational database can be utilized to construct the schema diagram of the database; in the process of scanning the schema diagram, the corresponding relation among a relational table name, a candidate code, a selector XPath and a field XPath of the key constraint in the XML Schema can be established, and the XML selector index of the key constraint of the relational database can be generated; the key constraint on the XML Schema can be automatically generated by the index; thereby, the automatic acquisition of the key constraint based on the process of XML data dissemination can be completed. The present invention can be widely used for each application system based on the process of XML data dissemination under the current network environment, and can satisfy the requirements of XML data dissemination and transformation among lots of application systems on the existing Internet. The present invention has wide application foreground.

Description

Key constraint towards expandable mark language mode generates method automatically

Technical field

The present invention relates to the generation method of database description, especially a kind of method of the automatic generation expandable mark language mode XML Schema key constraint based on relational data base schema figure Schema Diagram belongs to the computer data management technical field.

Background technology

XML (eXtensible Markup Language) has become the data representation on the Internet and the de facto standard of exchange, and increasing application system adopts XML to issue and swap data as standard format.But because the relational database technology maturation is occupied leading position in the market, how relational database being converted to XML document becomes the current problem that presses for solution.

Existed some systems and instrument to carry out data-switching at present to relational database and XML, XML Extender as IBM Corporation, the XML Enable and the XPERANTO of Oracle company, SilkRoute system etc., but because these systems only consider the data structure aspect, lack analysis to the data integrity constraint, and key and foreign key constraint are most important integrity constraints in the relational data model, obtain widely applying in the data base management system (DBMS) field, but in the prior art, also there is not to solve the key constraint transfer problem from the relational model to XML, the problem that has caused the semantic information in the data pattern transfer process to lose.Specific as follows:

At first, the mapping ruler from the relational database to the XML document has only been considered the data structure aspect at present, lacks the analysis to the data integrity constraint.

Secondly, though the mechanism of definition XML key constraint is provided among the XML Schema at present, in data exchange process, originally key in the relational database and external key semantic constraint can not be automatically converted to key constraint on the XML by exchange process.

For example by the XPERANTO system of IBM Almaden Research Center development, though this system utilizes the Xquery view to realize issue and conversion from relational database to the XML data, but all do not have the constraint of key by any XML document that this system generates, therefore lost semantic constraints such as key in original relational database and external key.

Therefore, a kind of method that generates XML data issue key constraint based on the key and the foreign key constraint of relational database that provides is provided the data issue application system based on XML of widespread use at present.

Summary of the invention

Technical matters to be solved by this invention is: provide a kind of key constraint towards expandable mark language mode to generate method automatically, generate hierarchy description and the key constraint of expandable mark language mode XML Schema based on the key constraint (comprising key and external key) of relational data base schema figure Schema Diagram, solve at present the problem that integrity constraint is lost when the relational database data are converted to Structure Conversion in the process that the XML data issue.

Technical solutions according to the invention comprise the steps:

The mode chart of step 1, the key that utilizes relational database and external key constructs database;

Step 2, scan pattern figure, the selector switch XPath of key constraint, the corresponding relation between the XPath of territory among opening relationships table name, Candidate key and the XML Schema in scan pattern figure process generate the XML selector switch index of relational database key constraint;

Step 3, the index that utilizes step 2 to obtain generate the key constraint on the XML Schema, thereby finish the automatic generation based on key constraint in the XML data issuing process.

The present invention is by relational data base schema figure, the XML selector switch index of opening relationships database key constraint, thereby generate the key constraint of XML Schema, solved at present problem based on integrity constraint is lost during Structure Conversion in the data issuing process of XML, the data issuing process based on XML that can be widely used in each application system under the current network environment, satisfied present Internet and gone up issue of XML data and the demand of changing between the extensive application system, had broad application prospects.

Description of drawings

Fig. 1 is the synoptic diagram of overall flow of the present invention;

Fig. 2 is the database schema figure of the banking system of a simplification;

Fig. 3 is in the example shown in Figure 2, sets up root node and has only out-degree and do not have the synoptic diagram of in-degree node;

Fig. 4 is in the example shown in Figure 2, the synoptic diagram after the branch dequeue;

Fig. 5 is in the example shown in Figure 2, the synoptic diagram after the account dequeue;

Fig. 6 is in the example shown in Figure 2, the synoptic diagram after the loan dequeue;

Fig. 7 is in the example shown in Figure 2, and customer goes into the synoptic diagram after the formation among the R;

Fig. 8 is in the example shown in Figure 2, the synoptic diagram after the customer dequeue.

Embodiment

The present invention be directed to the technical scheme that a kind of new integrity constraint that each application system under the current network environment proposes based on the demand of XML data issue generates, it is generated as hierarchical relationship and the key constraint that XMLSchema describes automatically based on keys important and a large amount of existence and foreign key constraint among the relational data base schema figure (Schema Diagram), has realized the XML issue of relation data database data.

The inventive method is at first utilized the key of relational database and the mode chart of external key constructs database; The corresponding relation between the selector switch XPath of key constraint (Selector XPath), the territory XPath (Field XPath) among opening relationships table name, Candidate key and the XML Schema in scan pattern figure process then generates the XML selector switch index of relational database key constraint; Utilize this index to generate key constraint on the XMLSchema automatically again, thereby realized automatic acquisition based on key constraint in the XML data issuing process.

Mode chart of the present invention be defined as G=(V, E), wherein: (key attribute non-keyattribute), promptly comprises primary attribute and nonprime attribute to V=; E is the reference relation between the description list, and for table s, its attribute list is X, table t, its attribute list is Y, concerns t[β for each foreign key reference] _ s[α], α _ X wherein, β _ Y, and α is the primary attribute of s table then has a limit the α attribute of being shown to s by the β attribute of t table.

For a database schema that contains major key and external key dependence, can represent that promptly described structure mode chart comprises the steps: with mode chart SchemaDiagram

Step 11, each relation table in the database is represented with a rectangle that the name of rectangle top mark relation is listed the attribute of relation table in the rectangle.

If step 12 has major key, the major key attribute is separated on square frame top with a horizontal line.

Step 13, to each external key in the database, generate external key attribute from reference relation to a directed edge the major key attribute that is referenced relation.

Illustrate how to make up the mode chart that has internal key and foreign key constraint relational database below by a concrete example:

The database schema of the banking system of a simplification comprises following list structure:

branch( branch_name，branch_city，assets)，

account( account_number，branch_name，balance)，

depositor(customer_name，account_number，last_date)，

customer(customer_name，customer_street，customer_city)，

loan( loan_number，branch_name，amount)

borrower( customer_name，loan_number，last_date)

The key of this table that underscore wherein marks.In addition, also have the foreign key constraint contact between each table, for example the branch_name attribute in the account table is an external key, and it is with reference to the major key branch_name in the branch table.

Be the database schema figure that obtains according to above-mentioned relation database schema figure construction step as shown in Figure 2.The mode of mode chart is basic identical with prior art in the acquisition relational database of the present invention, can be with reference to Paul Dorsey and Peter Koletzke, Designer/2000 Handbook, OraclePress.

Scan database mode chart of the present invention generates index, specifically comprises the steps:

Step 21, according to database schema figure, set up XML root node and child thereof;

Step 22, from the XML root node, mode chart is carried out reverse scan; Result according to scanning sets up hierarchical structure and concordance list.

Above-mentioned step 21 is determined XML root node V according to following steps _rWith and child, specifically comprise the steps:

Step 211: set up root node V _r

Step 212: scan pattern figure finds out the node set R that has only in-degree not have out-degree; And to each node among the R, corresponding generation is by root node V _rPoint to a limit of this node.

Step 213: if there is not such node, in the expression mode chart ring is arranged, a then optional node from ring, and generation root node V _rPoint to a limit of this node.

Described root node V _rBe an empty node, just, irrelevant with the physical relationship database as the beginning node of XML tree.Because the XML data are tree structures, thus found all have only in-degree not have the node set R of out-degree after, necessarily can reach all nodes from R, so in the step 212, finished this step.If in the mode chart be ring-type, then there is not the node that has only in-degree not have out-degree in the node of this ring, but can arrives all nodes the ring from each node, therefore can from ring, appoint and get a node, to start with.

Be according to the database schema figure among Fig. 2, its R={branch, customer}, the synoptic diagram behind the completing steps 21 as shown in Figure 3.

Described step 22 comprises the steps:

Step 221, node set R is pressed into formation,, begins mode chart is carried out reverse breadth-first scanning according to the reverse direction on mode chart limit;

Step 222, in reverse breadth-first scanning process, the note present node is s (X), its attribute list is X; Scan new node note and be t (Y), its attribute list is Y, because new node is the node that existing in-degree has out-degree again, so, necessarily exists external key to connect FK:t[β] _ s[α], α _ X wherein, β _ Y,

Connect for this external key, generate corresponding hierarchical structure according to following steps:

At first, the attribute that will appear in t in the s node is deleted; This attribute that guarantees the s node only occurs once;

Then, with the child node of the remaining attribute of t as the s node; So just formed the hierarchical structure of the tree of XML;

Hierarchical structure of every generation is all set up corresponding index in XML Schema selector switch index, the structure of selector switch index is as follows:

Database table name Table Name

Candidate Key attribute Candidate Key

Context path Context Path

Selector switch XPath Selector XPath

Territory XPath Field XPath

The content of selector switch index generates according to following steps:

Step a, for root node V _r, the table name of this nodes records in the mode chart and major key attribute are inserted in preceding two row of selector switch concordance list, its selector switch XPath and territory XPath are NULL;

Step b, in scan pattern figure process, whenever scan a new node, just the table name of this nodes records in the mode chart and major key attribute are inserted in preceding two row of selector switch concordance list; Its selector switch XPath adds the current limit that scans for its father node selector switch XPath, and territory XPath adds the limit of this node to the Candidate Key attribute for the object set path.

Step 223, according to top method, recurrence is handled all nodes successively, if the node that scans was scanned, illustrates in the XML tree to have comprised this node, then generates reference and concerns index, quotes this node that is scanned the location; All foreign key constraint relations in handling relational data base schema figure.

Recurrence is handled all nodes in the step 223 of the present invention, has adopted the scan mode of breadth-first.Described breadth-first scanning is meant from each node v of set R, first access node v, and it is labeled as access node, then visit all adjacent node w1 of v successively, w2, ..., wn, and then visit and w1, w2, ..., the node that all of wn adjacency are not accessed, the rest may be inferred, till all nodes are all accessed.

Still adopt the example of simplification banking system above-mentioned below, further specify technical scheme of the present invention:

Steps A, have only out-degree not have the set R={branch of in-degree, branch enters in the formation among the customer}, utilizes formation that relational data base schema figure is handled, the beginning width scan;

Step B, branch dequeue, the attribute list of branch becomes the child node of branch, owing to, be connected to account and loan table by foreign key constraint, so the attribute list of account and loan table becomes child node separately; Simultaneously queue element (QE) becomes that { account, loan} generate the XML pattern as shown in Figure 4;

Step C, account dequeue, adopt handle with the same mode of step B after, queue element (QE) becomes that { loan, depositor} generate the XML pattern as shown in Figure 5;

Step D, loan dequeue, queue element (QE) become that { depositor, borrower} generate the XML pattern as shown in Figure 6;

Step e, depositor dequeue, formation becomes { borrower};

Step F, borrower dequeue, formation are empty;

Customer goes into formation among step G, the R, and formation becomes that { customer} generates the XML pattern as shown in Figure 7;

Step H, customer go out team, because depositor, borrower was scanned, so generate two adduction relationship reference1 (depositor-customer) and reference2 (borrower-customer), processing finishes, and generates the XML pattern as shown in Figure 8.

In above step, the list item in the corresponding concordance list generates simultaneously, is the concordance list that obtains in the above-mentioned example below:

Database table name Table Name	Candidate Key attribute Candidate Key	Context path Context Path	Selector switch XPath Selector XPath	Territory XPath Field XPath
					Branch	branch_name	/	Branch	branch_name
Account	Account_numbe r	/branch	Account	account_name
					Loan	loan_number	/branch	Loan	loan_name
Depositor	NULL	NULL	NULL	NULL
					Borrower	NULL	NULL	NULL	NULL
Customer	customer_name	/	Customer	customer_name

Reference relation (limit that reverse scan customer sends) is as follows: for

depositor(customer_name)_customer(customer_name)

borrower(customer_name)_customer(customer_name)

Obtain the Reference relation:

Database table name Table Name	The forerunner	Context path Context Path	Selector switch XPath Selector XPath	Referencing key path Keyref XPath
					depositor	customer	/branch/account/	depositor	ref1
borrower	customer	/branch/loan	borrower	ref2

By top step, the present invention utilizes the XML structure of set R for root, obtain having the concordance list that the expression structure and being convenient to of tree-shaped mode chart of the XML Schema of key constraint is retrieved key constraint, utilized this indexed set, just can generate the key constraint of XML Schema easily.

Described generation has the XML Schema of key constraint, comprises the steps:

Step 30, read in the hierarchical structure that step 2 generates, set up the pattern framework that element among the XML Schema and daughter element constitute;

Step 31, read in the selector switch concordance list line by line, with database table name, Candidate Key attribute, context path, selector switch XPath and territory XPath are set at T, K, C, S, F respectively;

Step 32, in pattern framework, search the element that element term is T, increase following key constraint according to context path C:

＜xsd:key name=key name 〉

<xsd:selector xpath=S/>

<xsd:field xpath=F/>

</xsd:key>。

Below have the XML Schema segment of the text description of key constraint in the above-mentioned example, the part of italic is represented the key constraint that generates:

<xsd:element name="customer">

<xsd:complexType>

<xsd:sequence>

<xsd:element name="customer_name"type="xsd:string"/>

<xsd:element name="customer_street"type="xsd:string"/>

<xsd:element name="customer_city"type="xsd:string"/>

</xsd:sequence>

</xsd:complexTyp>

<xsd:key name="PK_c">

<xsd:selector xpath="customer"/>

<xsd:field xpath="customer_name"/>

</xsd:key>

</xsd:element>

<xsd:element name="branch">

<xsd:complexType>

<xsd:sequence>

<xsd:element name="branch_name"type="string"/>

<xsd:element name="branch_city"type="string"/><xsd:element name="assets"type="integer"/>

<xsd:element name="account"minOccurs=0 maxOccurs="unbounded">

<xsd:complexType>

<xsd:sequence>

<xsd:element name="account_name"type="string"/>

<xsd:element name="balance"type="integer"/>

<xsd:element name="depositor"minOccurs=0 maxOccurs="unbounded">

<xsd:complexType>

<xsd:sequence>

<xsd:element name="last_date"type="xsd:date"/>

<xsd:element name="ref1"type="xsd:string"/>

</xsd:sequence>

</xsd:complexType>

<xsd:keyref name="ref1-customer"refer="PK_c">

<xsd:selector xpath="branch/account/depositor"/>

<xsd:field xpath="ref1"/>

</xsd:keyref>

</xsd:element>

</xsd:sequence>

</xsd:complexType>

<xsd:key name="PK_a">

<xsd:selector xpath="branch/account"/>

<xsd:field xpath="account_name"/>

</xsd:key>

</xsd:element>

<xsd:element name="loan"minOccurs=0 maxOccurs="unbounded">

<xsd:compl exType>

<xsd:element name="borrower"minOccurs=0 maxOccurs="unbounded">

<xsd:complexType>

<xsd:sequence>

<xsd:element name="last_date"type="xsd:date"/>

<xsd:element name="ref2"type="xsd:string"/>

</xsd:sequence>

</xsd:complexType>

<xsd:keyref name="ref2_customer"refer="PK_c">

<xsd:selector xpath="branch/loan/borrower"/>

<xsd:field xpath="ref2"/>

</xsd:keyref>

</xsd:element>

</xsd:complexType>

<xsd:key name="PK_1">

<xsd:selector xpath="branch/loan"/>

<xsd:field xpath="loan-number"/>

</xsd:key>

</xsd:element>

</xsd:sequence>

</xsd:complexType>

<xsd:key name="PK_b">

<xsd:selector xpath="branch"/>

<xsd:field xpath="branch_name"/>

</xsd:key>

</xsd:element>

It should be noted last that: above embodiment is the unrestricted technical scheme of the present invention in order to explanation only, although the present invention is had been described in detail with reference to the foregoing description, those of ordinary skill in the art is to be understood that: still can make amendment or be equal to replacement the present invention, and not breaking away from any modification or partial replacement of the spirit and scope of the present invention, it all should be encompassed in the middle of the claim scope of the present invention.

Claims (8)

1, a kind of key constraint towards expandable mark language mode generates method automatically, it is characterized in that: comprise the steps:

The mode chart of step 1, the key that utilizes relational database and external key constructs database;

Step 2, scan pattern figure, the selector switch XPath of key constraint, the corresponding relation between the XPath of territory among opening relationships table name, Candidate key and the XML Schema in scan pattern figure process generate the XML selector switch index of relational database key constraint;

Step 3, the index that utilizes step 2 to obtain generate the key constraint on the XML Schema, thereby finish the automatic generation based on key constraint in the XML data issuing process.

2, the key constraint towards expandable mark language mode according to claim 1 generates method automatically, it is characterized in that: described database schema figure is G=[V, E], wherein: V=[key attribute, non-key attribute], promptly comprise primary attribute and nonprime attribute; E is the reference relation between the description list, and for table s, its attribute list is X, table t, its attribute list is Y, concerns t[β for each foreign key reference] _ s[α], α _ X wherein, β _ Y, and α is the primary attribute of s table then has a limit the α attribute of being shown to s by the β attribute of t table.

3, the key constraint towards expandable mark language mode according to claim 1 generates method automatically, and it is characterized in that: described step 1 comprises the steps:

Step 11, each relation table in the database is represented with a rectangle that the name of rectangle top mark relation is listed the attribute of relation table in the rectangle;

If step 12 has major key, the major key attribute is separated on square frame top with a horizontal line;

Step 13, to each external key in the database, generate external key attribute from reference relation to a directed edge the major key attribute that is referenced relation.

4, the key constraint towards expandable mark language mode according to claim 1 generates method automatically, and it is characterized in that: described step 2 comprises:

Step 21, according to database schema figure, determine XML root node and child thereof;

Step 22, from the XML root node, mode chart is carried out reverse scan; Result according to scanning sets up hierarchical structure and XML Schema selector switch concordance list.

5, the key constraint towards expandable mark language mode according to claim 4 generates method automatically, and it is characterized in that: described step 21 comprises:

Step 211: set up root node V _r

Step 212: scan pattern figure finds out the node set R that has only in-degree and do not have out-degree, and to each node among the R, corresponding generation is by root node V _rPoint to a limit of this node;

Step 213: if there is not such node, in the expression mode chart ring is arranged, a then optional node from ring, and generation root node V _rPoint to a limit of this node.

6, the key constraint towards expandable mark language mode according to claim 4 generates method automatically, and it is characterized in that: described step 22 comprises the steps:

Step 221, node set R is pressed into formation,, begins mode chart is carried out reverse breadth-first scanning according to the reverse direction on mode chart limit;

Step 222, note present node are s (X), and its attribute list is X; Scan new node note and be t (Y), its attribute list is Y, because new node is the node that existing in-degree has out-degree again, so, necessarily exists external key to connect FK:t[β] _ s[α], α _ X wherein, β _ Y,

Connect for this external key, generate corresponding hierarchical structure according to following steps:

At first, the attribute that will appear in t in the s node is deleted; This attribute that guarantees the s node only occurs once;

Then, with the child node of the remaining attribute of t as the s node; So just formed the hierarchical structure of the tree of XML;

Hierarchical structure of every generation is all set up corresponding index in the selector switch index, the structure of selector switch index is as follows: The database table name The Candidate Key attribute Context path Selector switch XPath Territory XPath Table Name Candidate Key Context Path Selector XPath Field XPath

Step 223, according to the described method of step 222, by reverse priority scan order, handle all nodes successively, if the node that scans was scanned, then generate reference and concern index, quote this node that is scanned the location, all the foreign key constraint relations in handling relational data base schema figure.

7, the key constraint towards expandable mark language mode according to claim 6 generates method automatically, it is characterized in that: in the step 222, the content of selector switch index generates according to following steps:

Step a, for root node V _r, the table name of this nodes records in the mode chart and major key attribute are inserted in preceding two row of selector switch concordance list, its selector switch XPath and territory XPath are NULL;

Step b, in scan pattern figure process, whenever scan a new node, just the table name of this nodes records in the mode chart and major key attribute are inserted in preceding two row of selector switch concordance list; Its selector switch XPath adds the current limit that scans for its father node selector switch XPath, and territory XPath adds the limit of this node to the Candidate Key attribute for the object set path.

8, generate method automatically according to claim 1 or 6 described key constraints towards expandable mark language mode, it is characterized in that: described step 3 comprises the steps:

Step 30, read in the hierarchical structure that step 2 generates, set up the pattern framework that element among the XML Schema and daughter element constitute;

Step 31, read in the selector switch concordance list line by line, with database table name, Candidate Key attribute, context path, selector switch XPath and territory XPath are set at T, K, C, S, F respectively;

Step 32, in pattern framework, search the element that element term is T, increase following key constraint according to context path C:

＜xsd:key name=key name 〉

<xsd：selector xpath＝S/>

<xsd：field xpath＝F/>

</xsd：key>。

Images