CN102819609B - A kind of perdurable data model modelling approach - Google Patents

Abstract

The present invention discloses a kind of perdurable data modeling method, comprises following step: step 1: the mode adopting XML, the data pattern file after definition modeling; Step 2: drive engine dynamically to generate business datum storehouse table corresponding to data model by data model; Step 3: data model drives engine automatically verify according to data model and preserve business datum. The database design method that the modeling of business datum is replaced off-line is realized by data modeling, will fast and flexible data model be changed. Data model is set up and will dynamically be changed corresponding business datum table with change, business datum carries out modeling and amendment fast, substantially reduces the construction cycle of model, and reduces owing to data model change causes the chaotic equivalent risk of data mistake. Business datum is automatically processed by data-droved engine according to data model, greatly improves the tempo of development of system, decreases the system problem caused by different data processing code.

Description

A kind of perdurable data model modelling approach

Technical field

The present invention relates to a kind of Data Modeling Method, belong to technical field of data processing.

Background technology

At present, the persistence medium of systemic software development is substantially all relevant database, and adopting maximum in system design is exactly to utilizing database design instrument to carry out the method for the modeling of business model after business demand analysis. The data model of the infosystem of major part all adopts the database design instruments such as similar PowerDesinger to design and develop, after data base administrator takes concrete business demand, by setting up the form of business datum table, business model is established in database.

The demand of user can often change, and in order to adapt to the change of business demand, data model also must change the change by directly causing database structure once data model changes. Current data model all adopts data modeling tool create and change, data base administration (DBM) personnel find former data model, Update Table model again is carried out according to the change of demand, such change amount is generally all bigger, and needs the people understood especially in original business datum storehouse just can make best change. Existing data modeling is all the mode adopting off-line, data model after having built also must be generated in actual data storehouse, this process be all generally manually or semi-automatic mode utilize database structure language these changes to be modified in actual data storehouse. Such process can produce many problems, first, change from data modeling to database needs the time grown very much, but also need to be proficient in very much original data base administration (DBM) personnel and modify maintenance, changes in demand frequently after inconvenient, the demand that can not adapt to crucial user fast changes, and business analyzes personnel to be needed to find the content changed in original business datum, after these contents being analyzed, original business model could be reconstructed. After reconstruct business model, in addition it is also necessary to these changes changed in the database used, this process will for a long time lastingly; Secondly, original data also likely can be had an impact by the database after changing, it is possible to part data can be caused to lose, and data mistake etc. are abnormal, directly cause infosystem normally not use.

Summary of the invention

In order to the structure of data model can be changed flexibly, do not cause again data with existing to lose the problems such as mistake, present invention employs the data modeling based on expandable mark language XML. Utilizing the modeling method based on XML, be kept in the structure of XML by the data structure of modeling, this kind of structure can tackle the change of demand fast.

The technical scheme of the present invention is to provide a kind of perdurable data modeling method, it is characterized in that, comprises following step:

Step 1: the mode adopting XML, the data pattern file after definition modeling;

Step 2: drive engine dynamically to generate business datum storehouse table corresponding to data model by data model;

Step 3: data model drives engine automatically verify according to data model and preserve business datum.

Data pattern file in step 1 adopts schema grammer in XML to set up.

In step 1, data pattern file is stored in data model Maintenance Table.

In step 2, data model drives engine first to set up in business some fields all needed, then data model drives engine meeting ergodic data model, for each business datum element, arrange according to the business datum in the title of data model and type automatically generated data storehouse table.

In step 2, unique mark according to data model generates business datum storehouse table as table name;

In the business datum storehouse table set up, comprising common data row and business datum row, drive engine just can be mapped with business datum storehouse table by data model by data model, data preserve and inquiry all utilizes data model to drive engine to process automatically.

The Schema that in step 3, business datum utilizes XML to carry before being saved in business datum storehouse table verifies.

The step preserving business datum in step 3 is:

Data model drives engine to utilize the business datum element of title Automatic Adaptation Data model, directly data model elements is assembled SQL statement as row name, and whether existed in the table of business datum storehouse according to data and determined to use INSERT or UPDATE operation;

Utilize the SQL statement after assembling, by database manipulation, namely business datum can be saved in the table of business datum storehouse.

Also comprise the step to business datum inquiry:

Data model drives engine automatically to assemble the database statement of inquiry according to data model definitions, obtains corresponding business datum by performing inquiry statement.

The useful effect that the present invention reaches:

The database design method of the replacement off-line of the modeling to business datum is realized by data modeling, will fast and flexible data model be changed, once data model changes, only need the data model that dynamically amendment is set up in systems in which, and each model changed content will be preserved.

Data model is set up and will dynamically be changed corresponding business datum table with change, business datum carries out modeling and amendment fast, substantially reduces the construction cycle of model, and reduces owing to data model change causes the chaotic equivalent risk of data mistake.

Business datum automatically can be processed by data-droved engine according to data model, developer without the need to carrying out the code development of data processing again to each business, greatly improve the tempo of development of system, decrease the system problem caused by different data processing code.

Embodiment

The invention will be further described below. Following examples are only for clearly illustrating the technical scheme of the present invention, and can not limit the scope of the invention with this.

The present invention adopts the thinking of model-driven, utilizes expandable mark language XML that data model is carried out modeling. The present invention mainly comprises following settlement steps to deal:

Step one, adopt XML(extensible markup language) mode, definition modeling after data pattern file;

Step 2, by data model drive engine dynamically generate business datum storehouse table corresponding to data model.

Step 3, data model drive engine automatically verify according to data model and preserve business datum.

Detailed data modeling step process is as follows:

1, data pattern file

Present method utilizes XML to preserve the data pattern file after modeling, utilizes the characteristic that XML can expand, and data model can change according to the carrying out that the change of business demand continues. Data pattern file adopts schema grammer in XML to set up, and the specific form of model file is as follows:

All configurations are all that the grammer based on XMLSchema configures, and the maximum advantage of XMLSchema is exactly can expand arbitrarily for the demand in future.

Xs:schema is the root node of data model, and this node mainly is used for defining some name spaces of XMLSchema.

What the xs:element under xs:schema was used for definition is the title of data model and affiliated name space, and name attribute representation's is the title of data model, will uniquely identify this data model with this name after definition. Schema model is the name space formulated and change data model. By schema and name two attributes, it is possible to automatically generating corresponding database table in business datum storehouse, wherein schema is the pattern name of table, and name is table name.

Xs:complexType element representation starts business element concrete in definition of data model.

The order of the business element below xs:all element assignment can arrange arbitrarily.

Xs:element below xs:all has formulated each business element in business data model, and final business datum is all to be retained in business element corresponding to these elements. Name attribute representation's is unique mark of business element, and what text represented is the title of this business element, and what type represented is the type of this business element, defines following several data type in XMLSchema:

1) xs:string character string type

2) xs:decimal float

3) xs:integer integer type

4) xs:boolean boolean Value Types

5) xs:date date type

6) xs:time time type

Specification and element introduction according to xs:schema just can formulate a concrete business model, and xml content below is a simple business data model example.

2, data pattern file stores

Data model after configuration will be saved in system, has the data model after the next special stored configuration of a business data model Maintenance Table in system.

Following table is the database structure storing data pattern file:

Wherein, code is that data model uniquely identifies, and name is the title of data model, and model saves the XML structure of data model.

Consider that the data model of different times has specific meaning, so needing to be saved in system by the data model of each version, the version field references version number of data model, up-to-date data model version number adds 1 on original maximum version basis, the time that this version of version_time field references is set up, it is possible to be used for reviewing the time of concrete data model amendment.

3, business datum storehouse table is dynamically generated

The data model definitions data structure of business, these data structures corresponding have a concrete business datum. Business datum is kept in the table of business datum storehouse by present method, and these database tables are generated by data model engine automatically when data model is set up.

Data model drives engine first can some fields all needed in business be set up, and then model-driven engine meeting ergodic data model, arranges according to the business in its title and type automatically generated data Kuku table for each business model element.

Following table is the content of business datum storehouse table:

First, unique mark according to data model generates business datum storehouse table as table name.

In the business datum storehouse table set up, id, item and time are three fixing row, and wherein id is the major key representing business datum, and item be then that master data corresponding to this business datum encodes, and time represents the time dimension degree that business datum relates to. These fields are the public field that business datum exists, it is possible to use these service fields carry out the operations such as data filter.

It is all concrete business datum row after common data arranges, these data rows are that the business datum element according to definition in data model generates, wherein row name is specified by business element name, data type is then specified by type, and engine changes into, according to the type that XMLSchema defines, the database field type that database can understand automatically.

Like this, driving engine just can be mapped with business datum storehouse table by data model by data model, data preserve and inquiry can utilize data model to drive engine to process automatically.

4, data processing

Data model drives engine automatically business datum can be operated according to the data model of definition. The operation of data is mainly divided into the following aspects:

1) data verification

Business datum has to pass through the verification of careful data before being saved in business datum storehouse table, and present method utilizes the verification of the Schema that XML carries to complete.

Business datum can being preserved into by data model the content of xml structure, utilize the schema of the xml after assembling and data model to verify, what verification mainly comprised is data type, data length etc. If verifying unsuccessful, then these business datums can not be saved in the table of business datum storehouse.

The similar following content of the data xml assembled by data model:

2) data preserve

Data, after successfully verification, will be saved in concrete traffic table. Data model drives engine still to assemble the statement of database insertion or renewal according to data model.

Data are after importing engine into, engine can utilize the business datum element of title Automatic Adaptation Data model, because the row name of business datum table is consistent with the business element of data model, so directly data model elements, as row name, SQL statement can be assembled, and whether exist in traffic table according to data and determined to use INSERT or UPDATE operation.

Utilize the SQL statement after assembling, by database manipulation, namely data can be saved in the table of business datum storehouse.

3) data query

Data model drives engine automatically to assemble the database statement of inquiry according to data model definitions, will obtain corresponding business datum by performing inquiry statement.

After obtaining business datum, engine is corresponding with traffic data field row according to data model elements, and the data after these correspondences will return to the function needing to use with unified data object.

The above is only the preferred embodiment of the present invention; it is noted that for those skilled in the art, under the prerequisite not departing from the technology of the present invention principle; can also making some improvement and distortion, these improve and distortion also should be considered as protection scope of the present invention.

Claims (6)

1. a perdurable data modeling method, is characterized in that, comprises following step:

Step 1: the mode adopting XML, the data pattern file after definition modeling;

Step 2: drive engine dynamically to generate business datum storehouse table corresponding to data model in data pattern file by data model;

Step 3: data model drives engine automatically verify according to the data model in data pattern file and preserve business datum;

The Schema that in step 3, business datum utilizes XML to carry before being saved in business datum storehouse table verifies;

The step preserving business datum in step 3 is:

Data model driving engine utilizes the business datum element of the title Automatic Adaptation Data model of data model, directly the business datum element of data model is assembled SQL statement as row name, and whether existed in the table of business datum storehouse according to data and determined to use INSERT or UPDATE operation;

Utilize the SQL statement after assembling, by database manipulation, namely business datum can be saved in the table of business datum storehouse.

2. perdurable data modeling method according to claim 1, is characterized in that, the data pattern file in step 1 adopts schema grammer in XML to set up.

3. perdurable data modeling method according to claim 1, is characterized in that, in step 1, data pattern file is stored in data model Maintenance Table.

4. perdurable data modeling method according to claim 1, it is characterized in that, in step 2, data model drives engine first to set up in business some fields all needed, then data model drives engine meeting ergodic data model, for each business datum element, arrange according to the business datum in the title of data model and type automatically generated data storehouse table.

5. perdurable data modeling method according to claim 1, is characterized in that, in step 2, unique mark according to data model generates business datum storehouse table as table name;

In the business datum storehouse table set up, comprising common data row and business datum row, drive engine just can be mapped with business datum storehouse table by data model by data model, data preserve and inquiry all utilizes data model to drive engine to process automatically.

6. perdurable data modeling method according to claim 1, is characterized in that, also comprises the step to business datum inquiry:

Data model drives engine automatically to assemble the database statement of inquiry according to data model definitions, obtains corresponding business datum by performing the database statement of inquiry.