I'm writing a small Clojure application which has a lot of interaction with a MongoDB database with 2-3 different collections.
I come from a OOP/Ruby/ActiveRecord background where standard practice is to create one class per data model and give each one access to the database. I've started doing the same thing in my clojure project. I have one namespace per "data model" and each has its own database connection and CRUD functions. However, this doesn't feel very functional or clojure-like, and I was wondering if there is a more idiomatic way of doing it, such as having a data or database namespace with functions like get-post, and limiting access to the database to only that namespace.
This seems like it would have the benefit of isolating the database client dependency to just one namespace, and also of separating pure functions from those with side effects.
On the other hand, I would have one more namespace which I would need to reference from many different parts of my application, and having a namespace called "data" just seems odd to me.
Is there a conventional, idiomatic way of doing this in Clojure?
A nice and, arguably, the most idiomatic (scored 'adopt' on the Clojure radar) way to manage state in a Clojure app is that proposed by Stuart Sierra's great Component library. In a nutshell, the philosophy of Component is to store all the stateful resources in a single system map that explicitly defines their mutual relationship, and then to architect your code in such a way that your functions are merely passing the state to each other.
Connection / environment access
One part of your system will be to manage the 'machinery' of your application: start the web server, connect do data stores, retrieve configuration, etc. Put this part in a namespace separate namespace from your business logic (your business logic namespaces should not know about this namespace!). As #superkondukr said, Component is a battle-tested and well-documented way to do this.
The recommended way to communicate the database connection (and other environmental dependencies for that matter) to your business logic is via function arguments, not global Vars. This will make everything more testable, REPL-friendly, and explicit as to who depends on whom.
So your business logic functions will receive the connection as an argument and pass it along to other functions. But where does the connection come from in the first place? The way I do it is to attach it to events/requests when they enter the system. For instance, when you start your HTTP server, you attach the connection to each HTTP request coming in.
Namespace organization:
In an OO language, the conventional support for data is instances of classes representing database entities; in order to provide an idiomatic OO interface, business logic is then defined as methods of these classes. As Eric Normand put it in a recent newsletter, you define your model's 'names' as classes, and 'verbs' as methods.
Because Clojure puts emphasis on plain data structures for conveying information, you don't really have these incentives. You can still organize your namespaces by entity to mimick this, but I actually don't think it's optimal. You should also account for the fact that Clojure namespaces, unlike classes in most OO languages, don't allow for circular references.
My strategy is: organize your namespaces by use case.
For example, imagine your domain model has Users and Posts. You may have a myapp.user namespace for Users CRUD and core business logic; similarly you may have a myapp.post namespace. Maybe in your app the Users can like Posts, in which case you'll manage this in a myapp.like namespace which requires both myapp.user and myapp.posts. Maybe your Users can be friends in your app, which you'll manage in a myapp.friendship namespace. Maybe you have a small backoffice app with data visualization about all this: you may put this in a myapp.aggregations namespace for example.
Related
I'm trying to create a modular application in clojure.
Lets suppose that we have a blog engine, which consists of two modules, for example - database module, and article module (something that stores articles for blog), all with some configuration parameters.
So - article module depends on storage, And having two instances of article module and database module (with different parameters) allows us to host two different blogs in two different databases.
I tried to implement this creating new namespaces for each initialized module on-the-fly, and defining functions in this namespaces with partially applied parameters. But this approach is some sort of hacking, i think.
What is right way to do this?
A 'module' is a noun, as in the 'Kingdom of Nouns' by Steve Yegge.
Stick to non side-effecting or pure functions of their parameters (verbs) as much as possible except at the topmost levels of your abstractions. You can organize those functions however you like. At the topmost levels you will have some application state, there are many approaches to manage that, but the one I use the most is to hide these top-level services under a clojure protocol, then implement it in a clojure record (which may hold references to database connections or some-such).
This approach maximizes flexibility and prevents you from writing yourself into a corner. It's analagous to java's dependency injection. Stuart Sierra did a good talk recently on these topics at Clojure/West 2013, but the video is not yet available.
Note the difference from your approach. You need to separate the management and resolution of objects from their lifecycles. Tying them to namespaces is quick for access, but it means any functions you write as clients that use that code are now accessing global state. With protocols, you can separate the implementation detail of global state from the interface of access.
If you need a motivating example of why this is useful, consider, how would you intercept all access to a service that's globally accessible? Well, you would push the full implementation down and make the entry point a wrapper function, instead of pushing the relevant details closer to the client code. What if you wanted some behavior for some clients of the code and not others? Now you're stuck. This is just anticipating making those inevitable trade-offs preemptively and making your life easier.
I'm making a asmx web service and I use LinqToSQL to deal with database.
It's seems to be easy for me to use LinqToSQL generated classes as arguments or return values in web methods.
Like this:
[WebMethod]
public OperationResult Meet_ListMeets(string clientToken, out meet[] meets)
{
ServiceMeet s = new ServiceMeet(sqlCon, clientToken);
return s.ListMeets(out meets);
}
Where "meet" is a LinqToSQL class.
I found that meet class is exposed as WSDL complex type including all dependencies (such as other classes which is referent to meet by the foreign keys in database).
Main question is "is it a good practice to use classes that way?". What about a security?
Should I use wrapper-classes to hide my entity structure?
Not a good practice, and you'll most likely run into problems at some point. Not to mention the overhead of all that extra cruft down the wire.
What I've done in the past is make a "inbetween" model with just the fields I need to actually send across the wire, and the map them back to the real object when they come in. You can do the mapping manually or with one of the many mapping toolkits for .NET (look in NuGet).
I've been using some code that implements the phpBB DBAL for some time. Recently I had to implement a more full package around it and decided to use the DBAL throughout. In the main, it's been OK. But occassionally there are circumstances where I can't see the logic in using it. It seems to make the simple much more complicated.
What benefits does a DBAL offer rather then writing sql statements directly?
From wikipedia (http://en.wikipedia.org/wiki/Database_abstraction_layer) :
API level abstraction
Libraries like OpenDBX unify access to databases by providing a single low-level programming interface to the application developer. Their advantages are most often speed and flexibility because they are not tied to a specific query language (subset) and only have to implement a thin layer to reach their goal. The application developer can choose from all language features but has to provide configurable statements for querying or changing tables. Otherwise his application would also be tied to one database.
When cooking a dish, you do not want several chefs having access to the pot. They could all be adding spices unaware that another chef had already added a spice. Ideally, you want a single chef that would serve as a single point of access to avoid spoiling the soup.
The same with databases. A single point of access can avoid problems of multiple services accessing the data in different ways.
I'm working on the initial architecture for a solution for which an SOA approach has been recommended by a previous consultant. From reading the Erl book(s) and applying to previous work with services (and good design patterns in general), I can see the benefits of such an approach. However, this particular group does not currently have any traditional needs for implementing web services -- there are no external consumers, and no integration with other applications.
What I'm wondering is, are there any advantages to going with web services strictly to stick to SOA, that we couldn't get from just implementing objects that are "service ready"?
To explain, an example. Let's say you implement the entity "Person" as a service. You have to implement:
1. Business object/logic
2. Translator to service data structure
3. Translator from service data structure
4. WSDL
5. Service data structure (XML/JSON/etc)
6. Assertions
Now, on the other hand, if you don't go with a service, you only have to implement #1, and make sure the other code accesses it through a loose reference (using dependency injection, or a wrapper, etc). Then, if it later becomes apparent that a service is needed, you can just have the reference instead point to #2/#3 logic above in a wrapper object (so all caller objects do not need updating), and implement the same amount of objects without a penalty to the amount of development you have to do -- no extra objects or code have to be created as opposed to doing it all up front.
So, if the amount of work that has to be done is the same whether the service is implemented initially or as-needed, and there is no current need for external access through a service, is there any reason to initially implement it as a service just to stick to SOA?
Generally speaking you'd be better to wait.
You could design and implement a web service which was simply a technical facade that exposes the underlying functionality - the question is would you just do a straight one for one 'reflection' of that underlying functionality? If yes - did you design that underlying thing in such a way that it's fit for external callers? Does the API make sense, does it expose members that should be private, etc.
Another factor to consider is do you really know what the callers of the service want or need? The risk you run with building a service is that (as you're basically only guessing) you might need to re-write it when the first customers / callers come along. This can could result in all sorts of work including test cases, backwards compatibility if it drives change down to the lower levels, and so on.
having said that the advantage of putting something out there is that it might help spark use of the service - get people thinking - a more agile principled approach.
If your application is an isolated Client type application (a UI that connects to a service just to get data out of the Database) implementing a SOA like architecture is usually overkill.
Nevertheless there could be security, maintainability, serviceability aspects where using web services is a must. e.g. some clients needs access to the data outside the firewall or you prefer to separate your business logic/data access from the UI and put it on 1 server so that you don’t need to re-deploy the app every time some bus. rules changes.
Entreprise applications require many components interacting with each other and many developers working on it. In this type of scénario using SOA type architecture is the way to go.
The main reason to adopt SOA is to reduce the dependencies.
Enterprise Applications usually depends on a lot of external components (logic or data) and you don’t want to integrate these components by sharing assemblies.
Imagine that you share a component that implements some specific calculation, would you deploy this component to all the dependent applications? What will happen if you want to change some calculation logic? Would you ask all teams to upgrade their references and recompile and redeploy their app?
I recently posted on my blog a story where the former Architect had also choosed not to use web services and thought that sharing assemblies was fine. The result was chaos. Read more here.
As I mentioned, it depends on your requirements. If it’s a monolithically application and you’re sure you’ll never integrate this app and that you’ll never reuse the bus. Logic/data access a 2 tier application (UI/DB) is good enough.
Nevertheless this is an Architectural decision and as most of the architectural decisions it’s costly to change. Of course you can still factor in a web service model later on but it’s not as easy as you could think. Refactoring an existing app to add a service layer is usually a difficult task to accomplish even when using a good design based on interfaces. Example of things that could go wrong: data structure that are not serializable, circular references in properties, constructor overloading, dependencies on some internal behaviors…
So, I have an API that I need to implement in to an existing framework. This API manages interactions with an external server. I've been charged with coming up with a way to create an easily repeatable "pattern," so that if people are working on new projects in the given framework they have a simple solution for integrating the API.
My first idea was to create a class for your "main" class of the framework to extend that, would provide all the virtual functions necessary to interact with the API. However, my boss vetoed this, since the existing framework is "inheritence heavy" and he wants to avoid adding to the madness. I obviously can't incapsulate my API, because that is what the API itself is supposed to be doing, and doing so might hide functionality.
Short of asking futures developers to copy and paste my example, what do I do?
If your boss is hostile to inheritance, try aggregation. (Has-a relationships rather than inheritance's is-a relationship.) Assuming you interface with the API in question via an object, maybe you can just keep that object in a property of your framework 'main' class, so you'd interact with it like main->whateverapi->doWhatever(). If the API isn't object-implemented or you need to load a lot of functionality specific to your environment onto it, that points toward making your own class that goes into that role and relates to the third party API however it needs to. Yeah, this basically means you're building an API to the API. Aggregation allows you to avoid the masking-functionality problem, though; even if you do have to do an intermediary layer, you can expose the original API as main->yourobject->originalapi and not have to worry about inheritance mucking things up.
Sounds to me like what your boss is having a problem with is the Framework part of this. There is an important distiction between Framework and API, in order to code to a framework you must have a good understanding of it and how it fits within your overall development, much more of a wholeistic view, adding to frameworks should never be taken lightly.
API's on the other hand are just an interface to your application / Framework and usually just a library of utility calls, I can't see that he would have a problem with inheritance or aggregation in a library, seems to me that the issue would be creating additional complexity in the framework itself, i.e. requiring developers to extend the main class of the framework is much more onerous than creating a stand alone API library that people can just call into (if they choose) I would be willing to bet that your boss would not care (in fact probably support) if the library itself contained inheritance.
Like the answer from chaos above, I was going to suggest aggregation as an alternative to inheritance. You can wrap the API and make it configurable either via properties or via dependency injection.
Also for a related topic see my answer to "How do the Proxy, Decorator, Adaptor, and Bridge Patterns differ?" for a run-down on other "wrapper" design patterns.