CN109933362A - A kind of platform feature call method, system and equipment - Google Patents
A kind of platform feature call method, system and equipment Download PDFInfo
- Publication number
- CN109933362A CN109933362A CN201910070074.5A CN201910070074A CN109933362A CN 109933362 A CN109933362 A CN 109933362A CN 201910070074 A CN201910070074 A CN 201910070074A CN 109933362 A CN109933362 A CN 109933362A
- Authority
- CN
- China
- Prior art keywords
- platform
- parameter
- functional
- module
- calling
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Stored Programmes (AREA)
Abstract
This application discloses a kind of platform feature call method, system and equipment, method includes: that the functional method of at least one platform is called using unified standardization method of calling, wherein, the unified standardization method of calling includes: to specify scheduled parameter using scheduled sequence.Standardized unified call mode, to simplify code realization, has reached so that caller does not need a point platform selecting differentiation method of calling and has facilitated calling, easy to maintain and extending transversely, project integrates quick effect.
Description
Technical field
The invention belongs to field of computer technology, in particular to a kind of platform feature call method, system and equipment.
Background technique
Increasingly developed with computer software technology, software systems are comprehensive and technology specialty is more and more stronger, software
Sharing out the work and help one another between platform is more and more important with integration.AI (Artificial Intelligence), artificial intelligence is to grind
Study carefully, develop for simulate, extend and extend people intelligence theory, method, technology and application system a new engineering department
It learns.In practical application, often there is comprehensive service process platform and need to carry out specially in specific processing links by AI technology
Industryization processing.And it is directed to different field, AI is highly professional as one, a set of system of self-contained technical solution is often packaged
For a professional platform, by such as SDK, (functional interface of big granularity has the API of correlation including a batch, can run
It is serviced for one, the API of server-side, the API of last server-side is requested to handle data simultaneously by the assembling data of client
Return the result) a series of relevant professional data process&analysis functions are provided.This SDK can be software system development person
The platform of independent research encapsulation, also can be used the platform of third party software quotient offer.
For same field, such as technical field of face recognition, it is understood that there may be the SDK that multiple software business mans provide, such as Baidu people
Face identifies SDK, Tencent recognition of face SDK and face++ recognition of face SDK.The SDK that different platform provides has respective characteristic
And advantage and disadvantage.One needs to generally require such as certain business platform in difference using the comprehensive business platform of face recognition technology
Business module calls multiple AI platforms using the SDK from different AI platforms or in some processing links of same business module
SDK, analysis selection is carried out to processing result.Due to the docking standard that the docking of each AI platform is not unified, result in each
The situation that multi-platform AI technology presents diversification is docked in a business side, docks the face identification system of different AI platforms, all
Personalisation process is done for different platform, cause workload and maintenance task very heavy.
Simultaneously as the closure and independence of each AI platform development, cause the SDK of docking platform to be not quite similar, and
Each platform has respective AI algorithm advantage again, if it is desired to carrying out the docking of all platforms, each platform will carry out independent right
It connects, independent maintenance, independent process, has to safeguard several platforms, task simultaneously so as to cause the personnel docked
Repeated and redundant, job costs are high.The processing such as identification of multiple platforms is carried out to picture in a conventional manner, needs operation system
Multiple platforms, the module of the independent each platform docking SDK of exploitation are docked simultaneously in backstage.Lack unified entrance, also results in certain
When the related content of a AI platform is updated and upgrades, the maintenance cost of developer, a project can be greatly increased
N number of place for needing to modify is had, M project has the place of N*M modification, also with regard to there is leakage to correct mistakes while maintenance
Risk, and then may cause correlation function failure.Also, need to increase newly all must for the platform SDK when newly-increased AI platform
It wants the encapsulation of level to dock code, cannot achieve simple extending transversely.
Summary of the invention
In view of the above-mentioned problems, the present invention provides the multi-platform SDK funcall method of unified standardization a kind of, system and sets
It is standby.
A kind of platform feature call method, comprising:
The functional method of at least one platform is called using unified standardization method of calling, wherein
The unified standardization method of calling includes: to specify scheduled parameter using scheduled sequence.
Further,
The scheduled parameter includes: at least one in dock door, functional module, functional method and method call parameter
It is a.
Further,
The dock door determines that the method obtains corresponding according to the NameSpace by specified NameSpace
The application file of platform, and initialize according to the application file application management module of the corresponding platform.
Further,
The application management is for loading at least one described functional module.
Further,
The scheduled parameter further includes verifying configuration parameter, and the verifying configuration includes the function of at least one platform
The access certificate parameter of energy module.
Further, the method also includes:
By parsing the verifying configuration before loading the functional module, institute corresponding with the functional module is obtained
Certificate parameter is stated to verify the availability of the functional module.
Further, the method also includes:
The functional method parameter is judged and obtains the functional method in the access path of corresponding platform.
Further, it is described judgement include:
Whether whether the corresponding platform supported in the functional method and/or at least one version path comprising the function
It can method.
Further, the method also includes:
Parameter setting is carried out using database access keyword.
Further,
It is described to include: using database access keyword progress parameter setting
The functional method is specified using select keyword, and/or
Using where keyword, the method call parameters are set.
Further, the method also includes:
The data-handling capacity of the platform is evaluated.
Further, described evaluate includes:
The result repeatedly handled is executed to same data to count.
Further, the method also includes:
According to the evaluation of multiple platforms as a result, the platform and/or the multiple platform of selection of selection processing data
The result of data processing.
A kind of platform feature calling system, comprising:
Input module is called, the platform function of unified standardization is determined for specifying scheduled parameter according to a predetermined sequence
It can method call input；
Execution module is called, parsing execution is carried out to calling input.
Further, the scheduled parameter includes: dock door, functional module, functional method and method call ginseng
At least one in number；
The calling execution module includes at least one unit corresponding with the predefined parameter below:
Dock door resolution unit, for parsing the dock door parameter and loading the application management mould of corresponding platform
Block；
Functional module resolution unit, for parsing the functional module parameter and loading corresponding functional module；
Functional method resolution unit, for parsing the access path of functional method parameter and the determining functional method.
Further, the dock door parameter is determined by specified NameSpace, the dock door resolution unit
The application file of corresponding platform is obtained according to the NameSpace, and the institute of corresponding platform is initialized according to the application file
State application management module；
The application management is for loading at least one described functional module.
Further, the parameter further includes verifying configuration parameter, and the verifying configuration includes at least one described platform
Functional module access certificate parameter；
The system also includes authentication unit, the authentication unit verifies the function for parsing the verifying configuration
The availability of module.
Further, the calling input uses the predetermined clause of unified standardization, includes with ShiShimonoseki in the predetermined clause
At least one of key word:
Select keyword, for specifying the functional method；
Where keyword, for specifying the method call parameters；
Get keyword, for being called execution to calling input and obtaining implementing result.
A kind of electronic equipment, comprising:
At least one processor；
And the memory being connect at least one described processor communication；
Wherein, the memory is stored with the instruction that can be executed by least one described processor, and described instruction is described
When at least one processor executes, at least one described processor is made to execute above-mentioned platform feature call method.
By reference software architecture and the characteristic of class, the abstract of profound level is carried out for the method for calling of each platform,
Encapsulation, polymerization, provides standardized unified call mode, caller needs not distinguish between the tune of platform and Selection and call differentiation
Achieve the effect that facilitate calling to simplify the realization of business function code with mode.Each platform is directed to unified call side
The parsing of formula can be reused largely, realized and be easy transplanting, so that system is easy to maintain, facilitated extending transversely, be conducive to fast
Speed is integrated, avoids the work of a large amount of repeated and redundant, greatly improves the docking efficiency of developer.Meanwhile the later period being facilitated to repair
Change maintenance, the correctness that item-function after upgrading is updated for platform provides certain guarantee.Using similar to database access
Mode is as method of calling, it can be readily appreciated that simplifying member's learning process of co-development.Meanwhile by multi-platform point
Analysis processing result is evaluated, and selects the platform or processing result of subsequent processing accordingly, effectively increases Data Analysis Services
Efficiency and accuracy rate.
Other features and advantages of the present invention will be illustrated in the following description, also, partly becomes from specification
It obtains it is clear that understand through the implementation of the invention.The objectives and other advantages of the invention can be by specification, right
Pointed structure is achieved and obtained in claim and attached drawing.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is this hair
Bright some embodiments for those of ordinary skill in the art without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 shows the execution flow chart of the platform feature call method using the embodiment of the present invention；
Fig. 2 shows the system architecture schematic diagrams of the embodiment of the present invention；
Fig. 3 shows a kind of parsing execution flow chart of platform feature call method of the embodiment of the present invention；
Fig. 4 shows a kind of platform feature calling system structural schematic diagram of the embodiment of the present invention；
Fig. 5 shows the structural schematic diagram of a kind of electronic equipment of the embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical solution in the embodiment of the present invention clearly and completely illustrated, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
The present invention provides a kind of platform feature call methods.As shown in Figure 1, execution process includes:
Scheduled parameter is specified using scheduled sequence；
Form unified standardization method of calling；
The functional method of access platform.
The present embodiment by it is a kind of using multiple AI platform SDK carry out Data Management Analysis comprehensive business platform for,
Platform feature call method disclosed in this invention is specifically described.
As shown in Fig. 2, being the system architecture schematic diagram of the embodiment of the present invention.The system architecture of the present embodiment is divided
1) application layer includes multiple application subsystems, is source terminal and the receiving end of platform service of data, system
The pending data of user's input is received by application layer, such as audits the identity card picture to be identified or user's meaning of subsystem
See user speech the input information or license plate photographic intelligence etc. of subsystem.
2) component layer is that AI platform calls conformable layer, and component layer is to function such as picture recognition, Text region, recognition of face
Deng progress modularized encapsulation.Component layer includes: the definition of unified standardization interface, parses interface and according to module granularity level pair
Connect the function of execution level AI platform.Wherein unified standardization interface calls, and further includes uniting to the certificate parameter of each platform
One configuration management.
3) execution level carries out the module of dissection process to input data based on the technologies such as sample database analysis and deep learning,
It is the specific layer for executing data specialization AI processing.The layer mainly includes multiple third party AI platform SDK in the present embodiment.
Below with reference to system architecture and practical application example, platform feature call method of the invention is done furtherly
It is bright.The present embodiments relate to a kind of business system, system receives user by application layer and inputs information, such as enters client enrollment
Information.Station subsystem (audit subsystem) needs to audit to entering client retail shop after the user information audit of service layer, examines
Nuclear process needs the identity card picture uploaded to user to match with the identity card name and identification card number filled out, i.e., real name is recognized
Card.System calls third-party platform SDK to realize the identification to photo by AI execution level.In order to improve the accuracy of identification, need
Baidu is comprehensively utilized, the picture recognition function of multiple platforms such as FACE++ carries out picture recognition.
Audit subsystem carries out parameter encapsulation to the information to be processed (including pictorial information) received, uses unified mark
Quasi-ization mode calls third-party platform picture recognition API.In the present embodiment, unified standardized method of calling refers to, for
Kinds of platform, multiple functional modules, multiple functional methods specify scheduled parameter, scheduled ginseng using unified scheduled sequence
Number includes: dock door, functional module, functional method and method call parameter.
It may is that in such a way that predetermined order specifies multiple predefined parameters and successively specify dock door, configuration parameter, function
Energy module, functional method, method call parameter, then execute and call and obtain result.Illustratively, above-mentioned calling process can be with
It indicates are as follows: entry file -> configuration -> functional module title -> interface -> parameter -> as a result, i.e. Entry (entrance):: Baidu
(Baidu platform) ($ this- > config (configuration)) -> face (face identification functions module) -> select ('
Getfacesets'(obtains recognition of face and interface is arranged)) -> get () (result).Wherein pass through " entry file -> platform name
Claim " it can determine dock door, such as Baidu's AI entrance；The functional module to be called is determined by functional module title；Pass through
The interface (API) of functional method determines the functional method to be called, and functional method is that platform realizes specific function (such as identification people
Face, license plate number identification etc.), the specific execution unit of data processing is carried out, externally provides call entry by providing access interface,
When called, specific data processing or analytic function are completed by algorithm logic code；Method call is determined by parameter
Parameter, such as photographic intelligence to be identified；By executing the knot that get method finally calls the method for AI platform and acquisition methods execute
Fruit.In addition, also being needed before loading functional module through permission account come the availability of authentication function module, in the present embodiment
The permission account integration of different console modules is stored in one or more configuration file, simultaneously by specified configuration file
It is parsed in conjunction with platform names and wants corresponding permission account and execute verifying, functional module is just loaded after being verified.
The present embodiment further, using database access keyword carries out parameter setting, comprising: crucial using select
Word specifies functional method, and each functional method interface has a multiple parameters again, when requesting method interface, passes through where transmitting encapsulation
Call parameters well obtain the result returned eventually by get.The service of system different application is calling SDK pairs of different function
When answering multiple interfaces, by the parameter of differentiation and the access mode of unified standardization, it can be used as access database
The method that simple line code calls AI platform, maintains easily extension and co-development, since select..where clause is
The basic clause of most of database platforms inquiry data of mainstream, common developer are easy to according to the calling clause at present
The meaning for understanding the thinking and parameter called, facilitates study to use.
For identification image operation, may is that according to the concrete form of database access mode calling platform AI
According to the process of execution, the platform that Baidu is called as this is specified in above-mentioned call statement；Pass through config
Document analysis Authority Verification account；The face functional module of the AI SDK of Baidu is loaded after being verified；A step of going forward side by side is used should
Acquisition tabulating method getlist in module；The call parameters for calling this method include two；The photograph obtained respectively from application layer
Piece face_detect.jpeg and unique identification uid；The knot for executing the corresponding api of Baidu platform is returned finally by get method
Fruit.The specific parsing and implementation procedure of each step of the call statement (each corresponding parameter) are described further below:
1) dock door class is instantiated.Entrance class is instantiated, entrance class is defined as Entry in the present embodiment, and different AI are flat
According to the script file of respective platform, (script file is php file to platform in the present embodiment, in specific implementation, does not limit files classes
Type) realize Entry class object corresponding with platform.Specifically, passing through NameSpace (namespace, for carrying out group to code
Knit and limit, by using NameSpace, delimit the section that variable, title etc. come into force) find corresponding application file
Application.php, as Baidu NameSpace be AI Baidu, by NameSpace find AI Baidu
Application.php.Entry class according to the Application.php be converted to application management class AI Baidu
Application.The Application can be understood as the management class of Baidu's platform AI module, for loading and managing AI phase
The functional module of pass.In the present embodiment, Application is also used to administration authority authentication module.Illustratively,
Application can load following modules and (but not need to instantiate and load completely all platform function before authentication
Energy module, only simple statement):
Wherein Face ServiceProvider::class be the functional module to be accessed in this example.
2) Authority Verification.Need to carry out module availability verification before using function, (such as Baidu is flat for third-party platform
Platform) Authority Verification entrance is provided with to AI functional module, its corresponding function just only can be used by verifying.The present embodiment
In, in order to facilitate the Authority Verification account of each platform of unified call, the multi-platform calling of more efficient realization, by Authority Verification
Token is stored in configuration file, such as ai_conf, in ai_conf file such as about the content of each platform picture recognition module
Under:
Configuration file ai_conf.php is introduced by above-mentioned $ config parameter, parsing configuration file is available to arrive Baidu
The token of (in above-mentioned configuration file ' the part baidu') picture recognition module.It is illustrative to introduce configuration file mode, it can be with
It is:
$this->load->config("ai_conf")；
$ this- > face=$ this- > config- > item (" faceconf ")；
After getting Authority Verification token, certificate parameter assembly is carried out by AccessToken module (class), due to every
The certificate parameter of a module difference can obtain unified format verification parameter by certificate parameter assembly, and then pass through tune
Related platform access qualification is obtained with the general public authentication module (Common/AccessToken) of platform.
Public authentication module incorporates the method that all platforms generate code keys, further comprises token method for refreshing, it is convenient after
Continuous extension increases new token and verifies relevant method.
3) corresponding functional module is loaded.After access authority verification passes through, corresponding functional module can be used.At this point,
Corresponding functional module can be instantiated, such as:
Entry::Baidu($config)->face.By under load Face module (file or catalogue)
ServiceProvider class (finds corresponding ServiceProvider.php and is initialized as ServiceProvider class
Object) and the Face class under Face module is registered in ServiceProvider, realize load Face functional module.Example
Property, Face class specifically:
The variable declaration of Face class: $ client indicates to call AI platform feature method (by accessing corresponding API service
Interface is called) when access server-side client connection；The access address that $ endpoint passes through finally calling API；$
VersionUrl_1/ $ versionUrl is a part in API Access path, for limiting the version of access path；$
VersionApi_1 indicates the group platform functional method that version 1 (corresponding $ versionUrl_1) is supported.In these Face classes
Variable initializer show and using in the following step further embody.
4) functional method to be called is obtained.Obtaining the method called includes verify to method parameter come described in obtaining
Access path of the method in corresponding platform.That is, whether verifying corresponding platform supports the functional method and at least one version
It whether include the functional method in path.Call method specifies method name by select, and this specify is not safety
, that is, it is possible that corresponding method is not present in the platform in the method name that method call person specifies, it is therefore desirable to method name
It is verified.And by being spliced with the version path of support ad hoc approach in module class (Face), one is obtained completely
Access the path of third party API.
Illustratively, select function is accomplished by
The access methods types file supported by association AI platform determines the access method that platform is supported, above-mentioned example
SupportType.php in code is access methods types file.Illustratively, file includes the following contents:
' detect ' // Face datection
' match ' // face alignment
The method that ' getlist ' // Groups List inquiry-the present embodiment will access
…
The method name stored in file is converted into array by require, is stored in $ allowOcrType, so
Judge that incoming access method parameter $ param whether in $ allowOcrType array, if it does not exist, then illustrates that called side passes afterwards
The access method title entered is illegal, program throw exception.Subsequent authentication is then carried out if it exists:
Judge that access method title (defines) in first or second version API list in above-mentioned module class Face, if
It is present in the first API list, then accesses version Path selection versionUrl_1 and spliced, is arranged if it exists in the 2nd API
In table, then accesses version Path selection versionUrl_2 and spliced, first or second API list, then select if it does not exist
Default the splicing of road strength.Splicing result is $ endpoint, is the platform method access being spliced by version path and method name
Path.
5) method call parameter is determined.Method call parameter is determined by where keyword, illustrative:
Wherein, $ condition is call method parameter, may include the combination of a parameter or multiple parameters, such as includes
Photographic intelligence and uid information.
6) it call method and returns the result.Implementing result is obtained eventually by get keyword calling platform method.This implementation
In example, get method sends post by service access client and requests to platform service end (BaiduFace service) in fact
Existing above-mentioned getlist method call.
Wherein, client (defining in above-mentioned Face class) needs carry out initialization building before use:
publicfunction_construct($app){
$ this- > client=newclient ($ app)；
}
Wherein, $ app is the example of the application management class of picture recognition module (Face module), i.e., above-mentioned application
Object, it comprises the basic informations that can be used for initializing access client client, such as link information.
In the present embodiment, returning the result as an array after step 6 is executed, when the result that method executes is successfully
When, array illustratively can be with are as follows:
Above-mentioned steps process is to access the example of Baidu's picture recognition module, each platform different business, for difference
AI platform feature accessed using above-mentioned unified standard mode, i.e., specify parameter: entry file-according to same predetermined order
> configuration -> function title -> interface -> parameter -> as a result, and calling clause: select- > where- > get using same.Tool
During body is implemented, according to different business, data type, platform, in implementation detail, such as the composition of method call parameter, configuration file
Content etc. can voluntarily be adjusted according to the true spirit of above-described embodiment, step is merely exemplary in examples detailed above,
Specific item code is implemented with no restrictions.Meanwhile system is each according to what is be related in above-mentioned steps to each console module
A parameter realizes that the same or similar standardization is parsed and docked, and the module script file (php) under each platform is all with above-mentioned
Mode is parsed and is encapsulated.
Method of calling described in the present embodiment is simply unified, and fault-tolerance is high, is easy to develop and extend.Each platform is for system
The parsing of one method of calling is to be carried out according to similar platform structure, script file, configuration file, therefore resolving can
It is a large amount of to reuse or replicate, the convenient extension in multiple platforms, transplanting.
Meanwhile in order to preferably with corporate business set, need to different AI platforms to different types of business datum
Processing performance is evaluated, and can preferentially be selected in subsequent business data processing at the business datum according to evaluation result
The strong platform of reason ability is as main processing platform.For example, the demand of the speech recognition for company, utilizes the knowledge of Baidu's platform
Other advantage, setting Baidu's platform speech recognition are Gao Quanchong, and the preferential analysis using the speech recognition of Baidu's platform is as a result, other are flat
Platform is analyzed supplemented by result.
Since platform calculated result is established on the basis of large sample analysis, so land identification function also can be wrong
Each all platform processes results are done a sample database, by analyzing result sample for reduction identification errors more as far as possible by difference
This library obtains the accuracy rate of each platform processes.Illustratively, the identification using tri- platforms of A, B, C to some license plate, three
Platform respectively executes n times identification to the license plate photographic intelligence respectively.Wherein, A land identification goes out N-1 identical result, and B platform is known
Not Chu N-2 identical result, C land identification goes out N number of identical result, then can determine the identification for the license plate, three platforms
Weight (recognition capability evaluation) are as follows: A- > N-1, B- > N-2, C- > N；C>A>B.In practical application, the weight of three platforms is by right
The multiple batches of identification of multiple objects is accumulative to obtain, such asWherein Si indicates the weight repeatedly identified to i-th of object.It is real
In the application of border, weight can be also possible to other numbers for embodying recognition capability difference to identify correct number in examples detailed above
Value, such as number of multiple platforms carry out scale: A- > (N-1)/3N, B- > (N-2)/3N.In the present embodiment, weight is that evaluation is flat
The parameter of platform processing capacity is assessed by a certain amount of recognition result and obtains platform for the processing capacity of different types of data
Afterwards, in subsequent applications, can preferred mainly preferred processing platform of the high platform of weight as respective type data, can also be with
Multi-platform processing is carried out to certain form of data, platform weighted value, right to choose are comprehensively considered to the processing result of multiple platforms
The optimal result of the sum of weight is as processing result.It, can also continuous updating in use for the weight of each platform.
Platform processes are used not by evaluating multi-platform analysis and processing result, and using evaluation weight as subsequent
With the foundation of data, Data Analysis Services efficiency and accuracy rate can effectively improve.
Based on identical inventive concept, the embodiment of the present disclosure additionally provides a kind of platform feature calling system, such as Fig. 4 institute
Show, system includes calling input module 400, dock door resolution unit 411.
Input module 400 is called, the flat of unified standardization is determined for specifying scheduled parameter according to a predetermined sequence
Platform functional method calls input, and predefined parameter includes: dock door, verifying configuration, functional module, functional method and method tune
Use parameter.In the present embodiment, above-mentioned predefined parameter is specified simultaneously by the predetermined clause select- > where- > get of unified standardization
Calling function method obtains result.
Select keyword, for specifying the functional method；
Where keyword, for specifying the method call parameters；
Get keyword, for being called execution to calling input and obtaining implementing result.
It illustratively may is that using predetermined clause call method
It is consistent in its meaning of parameters and above-mentioned platform feature call method embodiment, it repeats no more.
Execution module 410 is called, for carrying out parsing execution to calling input.Calling execution module 410 includes: that platform enters
Mouth resolution unit 411, authentication unit 412, functional module resolution unit 413, functional method resolution unit 414.
Dock door resolution unit 411 for analyzing platform suction parameter and loads the application management module of corresponding platform；
Dock door parameter determines that dock door resolution unit obtains corresponding platform according to NameSpace by specified NameSpace
Application file, and initialize according to application file the application management module of corresponding platform；Application management for load one or
Multiple functional modules." 1) dock door is instantiated in the specific work process and above-described embodiment of dock door resolution unit 411
The process of class " is consistent, repeats no more.
Functional module resolution unit 413, for parsing functional module parameter and loading corresponding functional module.Its specific work
It is consistent with " 3) load corresponding functional module " process in above-described embodiment to make process, repeats no more.
Functional method resolution unit 414, for parsing the access path of functional method parameter and the determining functional method.
Process is consistent with " 4) functional method to be called is obtained " in above-described embodiment for its specific work process, repeats no more.
Authentication unit 412 verifies the availability of the functional module for parsing the verifying configuration.In the present embodiment,
Before functional module resolution unit 413 loads corresponding functional module, availability verification is carried out.Specifically: it is defeated to obtain calling
Enter the configuration parameter in module 400, configuration file is obtained according to configuration parameter, parses the function mould of multiple platforms in configuration file
The access certificate parameter of block carries out availability verification according to certificate parameter.Certificate parameter is a series of permission accounts, executes verifying
When, the relevant service for checking credentials of access platform transmits permission account parameter, determines whether to verify according to the feedback result of platform service
Pass through.Its specific work process is consistent with " 2) Authority Verification " process in above-described embodiment, repeats no more.
Calling execution module 410 further includes processing result assessment unit, is used for different AI platforms to different types of industry
The processing performance of business data is evaluated, and using evaluation result as platform processes ability weight, in subsequent processing according to flat
The high platform of the difference of platform weight, preferably weight is as main processing platform.
In the present embodiment, the standardized packages of the method call parameter of the system and transmitting, method call execution (pass through
Get sends post request to Platform Server), processing result assessment is all by calling execution module 410 to execute.It is specific to execute
Process discloses in the embodiment of above-mentioned platform feature call method, repeats no more.
Based on identical inventive concept, the embodiment of the present disclosure additionally provides a kind of electronic equipment, structure as shown in figure 5,
The electronic equipment includes:
In at least one processor (processor) 501, Fig. 4 by taking a processor 500 as an example；And memory
(memory) 501, it can also include communication interface (Communication Interface) 502 and bus 503.Wherein, it handles
Device 500, communication interface 502, memory 501 can complete mutual communication by data connection 503.Communication interface 502 can
To be transmitted for information.Processor 500 can call the logical order in memory 501, to execute the platform of above-described embodiment
Funcall method.
In addition, the logical order in above-mentioned memory 501 can be realized by way of SFU software functional unit and conduct
Independent product when selling or using, can store in a computer readable storage medium.
Memory 501 is used as a kind of computer readable storage medium, can be used for storing software program, journey can be performed in computer
Sequence, such as the corresponding program instruction/module of the method in the embodiment of the present invention.Processor 500 is stored in memory 501 by operation
In software program, instruction and module, thereby executing functional application and data processing, i.e., in realization above method embodiment
Realization platform feature call method.
Memory 501 may include storing program area and storage data area, wherein storing program area can storage program area,
Application program needed at least one function；Storage data area, which can be stored, uses created data etc. according to terminal device.
In addition, memory 501 may include high-speed random access memory, it can also include nonvolatile memory.
Although the present invention is described in detail referring to the foregoing embodiments, those skilled in the art should manage
Solution: it is still possible to modify the technical solutions described in the foregoing embodiments, or to part of technical characteristic into
Row equivalent replacement；And these are modified or replaceed, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The spirit and scope of scheme.
Claims (15)
1. a kind of platform feature call method, which is characterized in that
The functional method of at least one platform is called using unified standardization method of calling, wherein
The unified standardization method of calling includes: to specify scheduled parameter using scheduled sequence.
2. the method according to claim 1, wherein
The scheduled parameter includes: at least one in dock door, functional module, functional method and method call parameter.
3. according to the method described in claim 2, it is characterized in that,
The dock door determines that the method obtains corresponding platform according to the NameSpace by specified NameSpace
Application file, and initialize according to the application file application management module of the corresponding platform.
4. method according to any one of claim 1-3, which is characterized in that
The scheduled parameter further includes verifying configuration parameter, and the verifying configuration includes the function mould of at least one platform
The access certificate parameter of block.
5. according to the method described in claim 4, it is characterized in that, the method also includes:
By parsing the verifying configuration before loading the functional module, obtain it is corresponding with the functional module described in test
Parameter is demonstrate,proved to verify the availability of the functional module.
6. according to the method described in claim 2, it is characterized in that, the method also includes:
The functional method parameter is judged and obtains the functional method in the access path of corresponding platform.
7. method according to claim 1 or 2, which is characterized in that the method also includes:
Parameter setting is carried out using database access keyword, comprising:
The functional method is specified using select keyword, and/or
Using where keyword, the method call parameters are set.
8. method according to claim 1 or 2, which is characterized in that the method also includes:
The data-handling capacity of the platform is evaluated.
9. method described according to claim 1 or 2 or 8, which is characterized in that the method also includes:
According to the evaluation of multiple platforms as a result, the platform and/or the multiple platform data of selection of selection processing data
The result of processing.
10. a kind of platform feature calling system characterized by comprising
Input module is called, determines the platform feature side of unified standardization for specifying scheduled parameter according to a predetermined sequence
Method calls input；
Execution module is called, parsing execution is carried out to calling input.
11. system according to claim 10, which is characterized in that
The scheduled parameter includes: at least one in dock door, functional module, functional method and method call parameter；
The calling execution module includes at least one unit corresponding with the predefined parameter below:
Dock door resolution unit, for parsing the dock door parameter and loading the application management module of corresponding platform；
Functional module resolution unit, for parsing the functional module parameter and loading corresponding functional module；
Functional method resolution unit, for parsing the access path of functional method parameter and the determining functional method.
12. system according to claim 11, which is characterized in that
The dock door parameter determines that the dock door resolution unit is empty according to the name by specified NameSpace
Between obtain the application file of corresponding platform, and the application management mould of corresponding platform is initialized according to the application file
Block；
The application management is for loading at least one described functional module.
13. system described in any one of 0-12 according to claim 1, which is characterized in that
The parameter further includes verifying configuration parameter, and the verifying configuration includes the visit of the functional module of at least one platform
Ask certificate parameter；
The system also includes authentication unit, the authentication unit verifies the functional module for parsing the verifying configuration
Availability.
14. system described in 0 or 11 according to claim 1, which is characterized in that
The calling input uses the predetermined clause of unified standardization, includes at least one in following keyword in the predetermined clause
It is a:
Select keyword, for specifying the functional method；
Where keyword, for specifying the method call parameters；
Get keyword, for being called execution to calling input and obtaining implementing result.
15. a kind of electronic equipment, comprising:
At least one processor；
And the memory being connect at least one described processor communication；
Wherein, the memory be stored with can by least one described processor execute instruction, described instruction by it is described at least
When one processor executes, at least one described processor perform claim is made to require method described in any one of 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910070074.5A CN109933362A (en) | 2019-01-24 | 2019-01-24 | A kind of platform feature call method, system and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910070074.5A CN109933362A (en) | 2019-01-24 | 2019-01-24 | A kind of platform feature call method, system and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109933362A true CN109933362A (en) | 2019-06-25 |
Family
ID=66985217
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910070074.5A Pending CN109933362A (en) | 2019-01-24 | 2019-01-24 | A kind of platform feature call method, system and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109933362A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110336840A (en) * | 2019-08-12 | 2019-10-15 | 苏州思必驰信息科技有限公司 | Third party's account register method and system for voice dialogue platform |
| CN110554968A (en) * | 2019-09-10 | 2019-12-10 | 联想(北京)有限公司 | interface calling method and device |
| CN110806932A (en) * | 2019-10-30 | 2020-02-18 | 广东博智林机器人有限公司 | Algorithm scheduling method and device |
| CN112395098A (en) * | 2019-08-19 | 2021-02-23 | 网易（杭州）网络有限公司 | Application program interface calling method and device, storage medium and electronic equipment |
| TWI765447B (en) * | 2019-12-05 | 2022-05-21 | 台達電子工業股份有限公司 | Decoupling method and system for decomposing services |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101827302A (en) * | 2010-04-19 | 2010-09-08 | 中兴通讯股份有限公司 | Multi-service unified processing method and unified service platform |
| CN102833357A (en) * | 2011-06-17 | 2012-12-19 | 上海晨阑数据技术有限公司 | Interface platform communicating with third party |
| CN108846586A (en) * | 2018-06-26 | 2018-11-20 | 南通森雅商务服务有限公司 | A kind of interface platform |
-
2019
- 2019-01-24 CN CN201910070074.5A patent/CN109933362A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101827302A (en) * | 2010-04-19 | 2010-09-08 | 中兴通讯股份有限公司 | Multi-service unified processing method and unified service platform |
| CN102833357A (en) * | 2011-06-17 | 2012-12-19 | 上海晨阑数据技术有限公司 | Interface platform communicating with third party |
| CN108846586A (en) * | 2018-06-26 | 2018-11-20 | 南通森雅商务服务有限公司 | A kind of interface platform |
Non-Patent Citations (3)
| Title |
|---|
| 柴林香: "基于Android的移动应用开发中间件技术研究与实现", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
| 申屠青春: "当中间件遇上区块链：为上层应用提供统一的API接口，进行私有化部署", 《HTTPS://WWW.01CAIJING.COM/FINDS/DETAILS/16221.HTM》 * |
| 进击中的PARK哥: "手游发行四大兵器之首---统一SDK的打造法则", 《HTTPS://BLOG.CSDN.NET/A_ASINCEO/ARTICLE/DETAILS/48106511》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110336840A (en) * | 2019-08-12 | 2019-10-15 | 苏州思必驰信息科技有限公司 | Third party's account register method and system for voice dialogue platform |
| CN110336840B (en) * | 2019-08-12 | 2022-05-13 | 思必驰科技股份有限公司 | Third party account registration method and system for voice conversation platform |
| CN112395098A (en) * | 2019-08-19 | 2021-02-23 | 网易（杭州）网络有限公司 | Application program interface calling method and device, storage medium and electronic equipment |
| CN112395098B (en) * | 2019-08-19 | 2024-03-15 | 网易（杭州）网络有限公司 | Application program interface calling method and device, storage medium and electronic equipment |
| CN110554968A (en) * | 2019-09-10 | 2019-12-10 | 联想(北京)有限公司 | interface calling method and device |
| CN110554968B (en) * | 2019-09-10 | 2021-05-18 | 联想(北京)有限公司 | Interface calling method and device |
| CN110806932A (en) * | 2019-10-30 | 2020-02-18 | 广东博智林机器人有限公司 | Algorithm scheduling method and device |
| TWI765447B (en) * | 2019-12-05 | 2022-05-21 | 台達電子工業股份有限公司 | Decoupling method and system for decomposing services |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109933362A (en) | A kind of platform feature call method, system and equipment | |
| CN108595157B (en) | Block chain data processing method, device, equipment and storage medium | |
| US20240087009A1 (en) | Data reconciliation based on computer analysis of data | |
| CN105095970B (en) | The execution method and system of third-party application | |
| US11250436B2 (en) | Index of usability for a replacement payment card | |
| US20080010561A1 (en) | Method for validating the proper operation of a transactional management system | |
| CN108492005B (en) | Project data processing method and device, computer equipment and storage medium | |
| CN109634996A (en) | Customer information table generating method, device, equipment and computer readable storage medium | |
| CN111625809B (en) | Data authorization method and device, electronic equipment and storage medium | |
| CN111290742A (en) | Parameter verification method and device, electronic equipment and readable storage medium | |
| CN113268336B (en) | Service acquisition method, device, equipment and readable medium | |
| CN107862425B (en) | Wind control data acquisition method, device and system and readable storage medium | |
| US11720825B2 (en) | Framework for multi-tenant data science experiments at-scale | |
| CN106487603A (en) | A kind of response test method and device | |
| US20040015499A1 (en) | System and method for transaction access control | |
| CN113726890A (en) | Block chain data service-oriented federal prediction method and system | |
| CN112711640A (en) | Method and device for configuring business handling process | |
| US9495367B2 (en) | System and method for performing a software comparison | |
| CN107341106A (en) | Application compatibility detection method, exploitation terminal and storage medium | |
| CN113938481A (en) | Receipt processing method, processing device, electronic equipment and readable storage medium | |
| US11615330B2 (en) | Virtual subject matter expert provisioning | |
| CN111402069B (en) | Accumulation fund data acquisition method and device, computer equipment and storage medium | |
| CN112487453A (en) | Data security sharing method and device based on central coordinator | |
| CN110086826A (en) | Information processing method | |
| CN114924783A (en) | Interface calling method, device, equipment and storage medium of application program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190625 |