I'm evaluating a piece of software (uPortal), and it lists one of its features as "Pluggable APIs."
I know what an API is, but have not heard the pluggable term before. What does it mean for an API to be "pluggable"?
It means that you can replace the implementation. Like a provider model, like ODBC database drivers - you can plug in drivers for various databases, using the same API.
Related
Hi everyone I’m new to server-side technologies so maybe this is a bit of a dumb question but after reading dozens of articles and viewing dozens of videos I’m still very confused. This has to do with arquitecture principles of modern apps.
Relational model:
I know that a few years ago the model was to have a database (mostly relational) and a DBMS that enabled the connection between an app and the database.
Question 1: Since we are talking about a relational model some examples of DBMS’s are MySQL or PostgreSQL?
Question 2: What is the process of information exchange? The client-side uses a language like PHP to make a request to the server and then the DBMS transforms the request into SQL and accesses the database? Is the conversion of the PHP into SQL part of DBMS function or another server-side software is needed?
(If someone could provide me summary detailed explanation I would be very thankful)
Non-Relational Models:
Question 2: Nowadays with the rise of NoSQL models does the same concept of DBMS apply? Since these systems allow other querying language other than SQL there should be some piece of software that has this function?
Service Oriented Arquitecture:
Almost every app uses this type of arquitecture. I understand the concept of avoid the creation of too tight software relation between client and server side allowing for multiple use across several platforms. What I don’t understand is what parts constitute a system that is build this way.
Question 3: Does the DBMS provides the API’s that constitute the web services made available?
Web Frameworks:
Last but not least, where do frameworks like Django or Ruby on Rails land on?
Question 4: These are supposed to provide tools to develop everything between the front-end and the database of a SOA system right?
Question 5: I’ve seen a lot of buzz about REST arquitecture. Can you explain me of the querying process happens and what are the software entities involved.
Thank you in advance for any explanation that helps me understating these questions. Please provide some links or any diagrams that you find useful.
EDIT:
I'll tackle your questions individually:
Question 1: Since we are talking about a relational model some examples of DBMS’s are MySQL or PostgreSQL?
Correct. The Database Management System is the suite of software that lets you interact with a particular database technology. The examples you give are correct.
Question 2: What is the process of information exchange? The client-side uses a language like PHP to make a request to the server and then the DBMS transforms the request into SQL and accesses the database? Is the conversion of the PHP into SQL part of DBMS function or another server-side software is needed?
There are many different avenues for this. Typically the API for accessing a database is done via ODBC (Open Database Connectivity). ODBC drivers are available for most (if not all) Relational DB vendors and are all very similar.
A language like PHP could connect to the database via an ODBC connection library (eg http://php.net/manual/en/intro.uodbc.php) which would allow you to send CRUD operations to the DBMS to be executed on the database.
Since most DBMS's use a subset or superset of a SQL standard for querying the database you can either pass this code directly via ODBC or you may use another level of abstraction. A common method is called an ORM (Object Relational Mapper). ORMs (eg SQLAlcmehy for Python: http://www.sqlalchemy.org/) provides an abstraction layer so you're not reliant on writing SQL but instead write queries and database commands in a format more common of your language of choice.
Question 2: Nowadays with the rise of NoSQL models does the same concept of DBMS apply? Since these systems allow other querying language other than SQL there should be some piece of software that has this function?
Same general concept (in that there is a DB Driver that exposes an API that languages can hook into) but there are generally more and differing ways of interacting with DB's now since they have many differing structures. Most NoSQL DBs still have ODBC connectors (eg MongoDB and Hadoop) so the general programming practices still apply whilst connecting with them, but the things you expect the database to do (and their natural query languages) will differ.
This continues to be an evolving space as these technologies evolve.
Question 3: Does the DBMS provides the API’s that constitute the web services made available?
Not sure I'm understanding this question. ODBC and webservices are different. Often webservices sit on top of ODBC if you want to query a database via a web API, but it's one layer of abstraction more than connecting to the DB directly via ODBC.
Last but not least, where do frameworks like Django or Ruby on Rails land on?
Web Frameworks are a way of quickening the development of web applications by trying to stop some of the "reinvent the wheel" things that you commonly do with every web application. They give you the basics, and have a lot of extensions that allow you to implement other common elements of web apps (like a subscription/login system, session management, admin system etc).
These are supposed to provide tools to develop everything between the front-end and the database of a SOA system right?
Both Django and RoR aim to be end-to-end frameworks. They include all the common elements you'll need including an Object Relational Mapper. They don't prescribe which DBMS you have to use, their ORMs can interface with many so that choice is still up to you.
Yes, they are aimed to cover everything from the front end to the DB, including the interaction with and initialization of the structure of the database.
Question 5: I’ve seen a lot of buzz about REST arquitecture. Can you explain me of the querying process happens and what are the software entities involved.
REST stands for Representational State Transfer (quick Wikipedia article: https://en.wikipedia.org/wiki/Representational_state_transfer). In a nutshell the creation of a "RESTful" (REST compliant) web API would mean that you have GET, PUT, POST and DELETE methods to accomplish all your services. REST aligns closely with the HTTP protocol which is why it's been very appropriate for the ever growing concept of Web Apps, it's helped transform the thinking from a Web Page (or set of web pages) into Web Apps.
It's hard to sum up better than the Wikipedia article does, I'd suggest you dive into it.
Hope that's cleared up a few of the questions!
I am developing a business application and, provided that many companies look for integration, I would like to make it "compatible" with business systems like SAP or Navision. What mechanisms do these systems use for importing/exporting/syncing data with third-party applications?
There exist software tools known as EAI (Enterprise Application Integration) whose purpose is to act as middleware to enable the integration of applications across a company.
Apache Camel is an example of such framework, but there exist many of them. You can find a comparison list here: http://en.wikipedia.org/wiki/Comparison_of_business_integration_software
As the user nmiranda pointed out, in the case of SAP, the framework used for data interchange is SAP PI (SAP NetWeaver Process Integration).
I think your question was actually aimed to find this "starting point", wasn't it? I faced the same question some years ago and I also wondered if there was any "standard" interface to integrate applications. In such case, I hope have helped you.
There are multiple ways to integrate with ERP data sources. You can do batch integration where you setup a query that pulls the data from the source ERPs on a scheduled bases. ETL tools like Informatica and Talend shines on this front.
If you want online data integration when you want live data in your business application then you need to look at Data Virtualization solutions like Denodo, VirtDB or Composite.
Prices, feature sets, performance and flexibility highly differ. One distinguishing factor in my practice is security. Solutions tend to extract data into file system, which makes a problem when sensitive data is extracted. In real projects, implementors usually start a long process replicating the source system security objects in the target application.
Both API and IDL act as an interface between two components of software and play the role of bridge between two components of software or between two software.
What is the difference between them?
API is a concept. It is any external programming interface that a piece of software exposes so that it can accept external input from some other software, run some logic, and provide output. Usually when we talk about API's, like Facebook's Graph API, or the Windows API, we are talking about the types and logic contained within those API's, and how they can be used.
IDL, as the tag says, is a language you can use to describe a API, in a manner that other software may understand. It is platform independent, so can be used to facilitate integration. More information about this language is vastly available if you search for it ;)
The OP in this question asks about using an WCF/OData as an internal data access layer.
Arguments of using WCF/OData as access layer instead of EF/L2S/nHibernate directly
The resounding reply seems to be don't do it. I'm in similar position to the OP, but have a concern not raised in the original question. I'm trying to develop (natively) for a lot of different platforms but want to keep as much of the data and business logic server side as possible. So I'll have iOS/Android/Web (MVC)/Desktop applications. Currently, I have a single WinForms applications with an ORM data access layer (LLBLGen Pro).
I'm envisioning moving most of my business / data access logic (possibly still with LLBLGen or other ORM) behind a WCF / OData interface. Then making all my different clients on the different platforms very thin (basically UI and WCF calls).
Is this also overengineered? Am I missing a simpler solution?
I cannot see any problem in your architecture or consider it overengeenered as a OData is a standard protocol and your concept conforms the DRY principle as well.
I change the question: Why would you implement the same business logic in each client to introduce more possible bugs and loose the possibility to fix the errors at one single and centralized place. Your idea makes you able to implement the security layer only once.
OData is a cross-platform standard and you can find a OData libraries for each development platform (MSDN, OData.org, JayData for JavaScript). Furthermore, you can use OData FunctionImports/Service methods and entity-level methods, which will simplify your queries.
If you are running multiplatform development, then you may find more practical to choose platform-agnostic communication protocol, such as HTTP, rather than bringing multiple drivers and ORMs to access your data Sources directly. In addition since OData is a REST protocol, you don't need much on the Client side: anything that can format OData HTTP requests and parse HTTP responses. There are however a few aspects to be aware of:
OData server is not a replacement for an SQL database. It supports batches but they are not the same as DB transactions (although in many cases can be used to model transactional operations). It supports parent-child relations but it does not support JOINs in classic SQL sense. So you have to plan what you expose as OData entity. It's too easy to build an OData server using WCF Data Services wrapping EF model. Too easy because People often expose low Level database content instead of building high level domain types.
As for today an OData multiplatorm clients are still under development, but they are coming. If I may suggest something I am personally working on, have a look at Simple.Data OData adapter (https://github.com/simplefx/Simple.OData, look at its Wiki pages for examples) - it has a NuGet package. While this a Client Library that only supports .NET 4.0, part of it is being extracted to be published as a portable class Library Simple.OData.Client to support .NET 4.x, Windows Store, Silverlight 5, Windows Phone 8, Android and iOS. In fact, if you check winrt branch of the Git repository, you will find a multiplatform PCL already, it's just not published on NuGet yet.
Consider the client application that should store its data on remote server. We do not want it to access this data "on-fly", but rather want it to have a copy of this data in local database. So we do not need connection with remote server to use application. Eventually we want to sync local database with remote server. The good example of what I am talking about is Evernote service. This type of applications is very common, for instance, in mobile development, where user is not guaranteed to have permanent Internet connection, bandwidth is limited and traffic can be expensive.
ORM (object relational mapping) solutions generally allows developer to define some intermediate "model" for his business logic data. And then work with it as some object hierarchy in his programming language, having the ability to store it relational database.
Why not to add to an ORM system a feature that will allow automatic synchronization of two databases (client and server), that share the same data model? This would simplify developing the applications of the type I described above. Is there any systems for any platform or language that have this or similar feature implemented?
this links may provide some useful information
http://gwtsandbox.com/?q=node/34
http://www.urielkatz.com/archive/detail/google-gears-orm-v01/
http://www.urielkatz.com/archive/detail/google-gears-orm/
AFAIK, there are no such ORM tools.
It was one of original goals of our team (I'm one of DataObjects.Net developers), but this feature is still not implemented. Likely, we'll start working on database sync this spring, but since almost nothing is done yet, there is no exact deadline for this.
There is at least one Open Source ORM matching your needs, but it is a Delphi ORM.
It is called mORMot, and use JSON in a stateless/RESTless architecture to communicate over GDI messages, named pipes or HTTP/1.1. It is able to connect to any database engine, and embed an optimized SQlite3 engine.
It is a true Client-Server ORM. That is, ORM is not used only for data persistence of objects (like in other implementations), but as part of a global n-Tier, Service Oriented Architecture. This really makes the difference.