My application provides the following resource:
GET /user/:id/orders
As commonly used, this returns a list of all the user's orders.
Now, a client wants to check if a user has any orders at all without actually getting the complete list.
My current approach looks like this:
GET /user/:id/orders/exist
But it looks kind of odd to me.
Is there a more "standard" way of designing this? In the end, this resource only needs to return the information:
yes, user has orders
no, user doesn't have any orders
What you will see in some API is the notion of a resource that exists (204) or does not exist (404).
But I really don't recommend that: saving a few bytes in your representation of the resource doesn't help very much when you are already sending a response-line and a bunch of HTTP headers.
Your "resource model" can be anything you want.
The REST interface is designed to be efficient for large-grain hypermedia data transfer, optimizing for the common case of the Web, but resulting in an interface that is not optimal for other forms of architectural interaction -- Fielding, 2000
So you can create fine grained resources if you like; but there are can be consequences to that. "Tradeoffs" are a thing.
Resources are generalizations of "documents"; if it makes sense to have a report that is just a count of the number of orders, or a statement that the number of orders is greater than zero, or whatever, then that report can certainly be a resource in your resource model.
If you know what the report is, then you might be able to guess at a name for the report, and from there to a spelling for it.
There's no particular reason that the report identifier has to be part of the /user hierarchy; the machines don't care what spelling conventions you use.
/user/:id/orders/report
/user/:id/orders-report
/user/:id/report
/report/:id
/report?user=:id
/report/user=:id
Those are all fine; choose whichever variation is appropriate to your local conventions.
Note that you want to be aware of caching - when you have information in two different resources, it is easy for the client's locally cached copies to contradict each other (report says that there are orders, but the orders list is empty; or the other way around). As far as REST, and general purpose components are concerned, different resources are different, and vary independently of each other.
In the large grained world, you don't have that problem so often, because you throw the kitchen sink into a single resource; as long as its produced representations are internally consistent, the cached copies will be as well.
Related
Lets assume we have some main-resource and a related sub-resource with 1-n relation;
User of the API can:
list main-resources so GET /main-resources endpoint.
list sub-resources so GET /sub-resources endpoint.
list sub-resources of a main-resource so one or both of;
GET /main-resources/{main-id}/sub-resources
GET /sub-resouces?main={main-id}
create a sub-resource under a main-resource
POST /main-resource/{main-id}/sub-resouces: Which has the benefit of hierarchy, but in order to support this one needs to provide another set of endpoints(list, create, update, delete).
POST /sub-resouces?main={main-id}: Which has the benefit of having embedded id inside URL. A middleware can handle and inject provided values into request itself.
create a sub-resource with all parameters in body POST /sub-resources
Is providing a URI with main={main-id} query parameter embedded a good way to solve this or should I go with the route of hierarchical URI?
In a true REST environment the spelling of URIs is not of importance as long as the characters used in the URI adhere to the URI specification. While RFC 3986 states that
The path component contains data, usually organized in hierarchical form, that, along with data in the non-hierarchical query component (Section 3.4), serves to identify a resource within the scope of the URI's scheme and naming authority (if any). The path is terminated by the first question mark ("?") and number sign ("#") character, or by the end of the URI. (Source)
it does not state that a URI has to have a hierarchical structure assigned to it. A URI as a whole is a pointer to a resource and as such a combination of various URIs may give the impression of some hierarchy involved. The actual information of whether URIs have some hierarchical structure to it should though stem from link relations that are attached to URIs. These can be registered names like up, fist, last, next, prev and the like or Web linking extensions such as https://acme.org/rel/parent which acts more like a predicate in a Semantic Web relation basically stating that the URI at hand is a parent to the current resource. Don't confuse rel-URIs for real URIs though. Such rel-URIs do not necessarily need to point to an actual resource or even to a documentation. Such link relation extensions though my be defined by media-types or certain profiles.
In a perfect world the URI though is only used to send the request to the actual server. A client won't parse or try to extract some knowledge off an URI as it will use accompanying link relation names to determine whether the URI is of relevance to the task at hand or not. REST is full of such "indirection" mechanism in order to help decoupling clients from servers.
I.e. what is the difference between a URI like https://acme.org/api/users/1 and https://acme.org/api/3f067d90-8b55-4b60-befc-1ce124b4e080? Developers in the first case might be tempted to create a user object representing the data returned by the URI invoked. Over time the response format might break as stuff is renamed, removed and replaced by other stuff. This is what Fielding called typed resources which REST shouldn't have.
The second URI doesn't give you a clue on what content it returns, and you might start questioning on what benefit it brings then. While you might not be aware of what actual content the service returns for such URIs, you know at least that your client is able to process the data somehow as otherwise the service would have responded with a 406 Not Acceptable response. So, content-type negotiation ensures that your client will with high certainty receive data it is able to process. Maintaining interoperability in a domain that is likely to change over time is one of RESTs strong benefits and selling points. Depending on the capabilities of your client and the service, you might receive a tailored response-format, which is only applicable to that particular service, or receive a more general-purpose one, like HTML i.e.. Your client basically needs a mapping to translate the received representation format into something your application then can use. As mentioned, REST is probably all about introducing indirections for the purpose of decoupling clients from servers. The benefit for going this indirection however is that once you have it working it will work with responses issued not only from that server but for any other service that also supports returning that media type format. And just think a minute what options your client has when it supports a couple of general-purpose formats. It then can basically communicate and interoperate with various other services in that ecosystem without a need for you touching it. This is how browsers operate on the Web for decades now.
This is exactly why I think that this phrase of Fielding is probably one of the most important ones but also the one that is ignored and or misinterpreted by most in the domain of REST:
A REST API should spend almost all of its descriptive effort in defining the media type(s) used for representing resources and driving application state, or in defining extended relation names and/or hypertext-enabled mark-up for existing standard media types. (Source)
So, in a true REST environment the form of the URI is unimportant as clients rely on other mechanisms to determine whether to use that URI or not. Even for so called "REST APIs" that do not really care about the true meaning of REST and treat it more like old-school RPC the question at hands is probably very opinionated and there probably isn't that one fits all solution. If your framework supports injecting stuff based on the presence of certain query parameters, use that. If you prefer the more hierarchical structure of URIs, go for those. There isn't a right or wrong in such cases.
According to the URI standard when you have a hierarchical relationship between resources, then better to add it to the path instead of the query. https://datatracker.ietf.org/doc/html/rfc3986#page-22 Sometimes it is better to describe the relation itself, not just the sub-resource, but that happens only if the sub-resource can belong to multiple main resources, which is n:m relationship.
Years ago I created a tiny web service that serves the same resource in two representations.
# returns a collection of Foos
GET /foo
# returns the same collection of Foos in a different JSON representation
GET /foo?projection=X with 'Accept: my-specific-media-type'
This works quite well in (Java) code as I can have two methods mapped to the same #Path both with different return types. One accepts a #QueryParam and #Consumes a specific media type while the other doesn't.
However, according to the (current) #ApiOperation Swagger annotation I opted for the wrong API design.
A combination of a HTTP method and a path creates a unique operation
Hence, after I upgraded my old project to current library versions the Swagger model only contains a single GET /foo operation - which one is random as it depends on runtime code introspection through Java reflections.
So, the question is this: is the Foo resource in a different representation effectively the "same" resource or is it a different resource? The Swagger annotation seems to hint at the latter (different resource -> different path).
Part of the problem that you are running into is a mix of REST concepts and Swagger/OpenAPI concepts.
Resource is a REST concept: "any concept that might be the target of an author's hypertext reference must fit within the definition of a resource"
Representation is a REST concept: "A representation is a sequence of bytes, plus representation metadata to describe those bytes."
Operations are an OpenAPI concept: "OpenAPI defines a unique operation as a combination of a path and an HTTP method."
There's a certain amount of tension here because the viewpoints aren't actually in alignment with each other.
For example, from the perspective of REST, there's no reason to document a "GET operation", because GET is part of the uniform interface - it has the same semantics no matter what value is used as the target-uri. That's a part of a key architectural constraint that makes the world wide web possible - consistent semantics means that we can use general purpose components (like web browsers) to interact with all of the different resources on the web.
is the Foo resource in a different representation effectively the "same" resource or is it a different resource?
"It depends".
A classic example of "one resource, different representations" would be a picture, where we might have a GIF, JPEG, PNG, BMP. Same picture (ish), but different sequences of bytes that need to be processed in different ways.
Similarly, you might have a web page (HTML), and also a text/plain representation, or a JSON representation, etc.
One of the important questions to ask: is a general purpose cache going to have the information necessary to return the "correct" representation for a request?
That said: given that your original design was using a query parameter to distinguish one projection from another, you should likely respect that instinct and continue to treat the different representations as belonging to different resources (meaning that general purpose caches will keep them completely separate).
Whether that means that you want to share the same path /foo (treating projection as an optional #ApiParam), or give each projection a different path (defining separate operations for each unique path) is less clear. In a brownfield project, my bias would be toward documenting what you already have, rather than making a bunch of breaking changes.
But it is certainly reasonable to treat "easy to document" as a design constraint.
So, the question is this: is the Foo resource in a different representation effectively the "same" resource or is it a different resource?
Fielding defined a resource as such:
The key abstraction of information in REST is a resource. Any information that can be named can be a resource: a document or image, a temporal service (e.g. "today's weather in Los Angeles"), a collection of other resources, a non-virtual object (e.g. a person), and so on. In other words, any concept that might be the target of an author's hypertext reference must fit within the definition of a resource. A resource is a conceptual mapping to a set of entities, not the entity that corresponds to the mapping at any particular point in time.
More precisely, a resource R is a temporally varying membership function MR(t), which for time t maps to a set of entities, or values, which are equivalent. The values in the set may be resource representations and/or resource identifiers. A resource can map to the empty set, which allows references to be made to a concept before any realization of that concept exists -- a notion that was foreign to most hypertext systems prior to the Web [61]. Some resources are static in the sense that, when examined at any time after their creation, they always correspond to the same value set. Others have a high degree of variance in their value over time. The only thing that is required to be static for a resource is the semantics of the mapping, since the semantics is what distinguishes one resource from another.
...
REST uses a resource identifier to identify the particular resource involved in an interaction between components. REST connectors provide a generic interface for accessing and manipulating the value set of a resource, regardless of how the membership function is defined or the type of software that is handling the request. The naming authority that assigned the resource identifier, making it possible to reference the resource, is responsible for maintaining the semantic validity of the mapping over time (i.e., ensuring that the membership function does not change). (Source)
In short, a resource is something that you give a name in order to reference it later on. This resource is a container for data. That data can be represented in plenty of ways. A representation is a concrete instance of the resource' data with respect to the media-type the representation was created for. The media-type itself defines the syntax and semantic of a concrete instance. I.e. HTML defines which attributes and elements are admissible within the payload and what these things express.
As REST shouldn't have typed "resources" meaningful to clients content type negotiation should be used. Here a client express its capabilities via the Accept header to the server and the server will chose a representation format that will suite the data the best. A well-behaved server will only chose among the suggested media types as it knows the client can handle the data. A non-well-behaved client will just ignore the header and send whatever it wants which eventually may prevent clients from being able to process the payload at all.
REST is all about decoupling of clients from servers and allowing the server side from evolving in future without breaking clients. This however is only possible if both use some kind of indirection. I.e. not the URI itself is the relevant thing in a payload but the link-relations that are attached to that URI. A link relation might be something like next, prev, first or last for a traversable collection or something like prefetch witch just states that the content of the annotated URI may be loaded once the client has loaded all other things and is currently IDLE as this content may be requested next with some likelihood. Such link relations are either standardized or should follow the extension mechanism defined in Web Linking.
In regards to your actual question. Think of an arbitrary product ABC1234. This product contains some properties such as its price, the current number of items in stock, some metadata describing the product and what not. These properties might be expressed in JSON, in XML or in HTML. Clients which are able to process these media-types will be able to create an "object" with the same properties with hardly any issues. The actual representation format used shouldn't have an influence on the actual data of the resource itself. After all, the representation format is just a mutually agreed way of exchanging the data between client and server in order to allow the recipient of the payload to process it in the same way the sender intended it initially.
As Fielding mentioned before, such a resource may be static or change over time. With the product example from above, the price may change over time, though this doesn't change the semantics of the actual product. Over time sometimes the same data that is present for a resource need to be made available as part of an other resource. This is totally fine and here things start to get a bit more interesting. As part of a company merger one of our clients needed to expose all of their items with different names. In their case they opted for providing both product names for a year simultaneously. By definition these would be two different resources to an arbitrary HTTP client, i.e ABC1234 and XYZ12345 even though they "represent" the data of the same real-live product. They could also have opted for using (permanent) redirection of clients to the "new" URI and therefore hint clients that the product is actually the same.
The resource per name (or URI) concept is also noticable if you take a look at how caching works in the HTTP ecosystem. Here the effective request URI is used as cache-key in order to look up whether for the requested URI already a stored response is present. Any unsafe operation performed on that URI will lead to an eviction of that stored response. This is i.e. one of the reasons why HTTP isn't meant for batch-operations as these may bypass the cache at all and lead to wrong and/or misleading results.
Years ago I created a tiny web service that serves the same resource in two representations.
GET /foo # returns a collection of Foos
GET /foo?projection=X # returns a collection of Foos in a different coordinate system i.e. different representation
According to how HTTP defines effective request URIs these two URIs would target two different resources actually, event though they express the same data just with different representations. A probably better approach would have been to expose just /foo and use either specialized media-types for the different coordinate systems or even better a media-type that supports profiles and hint the recipients processor via the profile attribute which "kind of" data it receives. Link relations, as mentioned above, also define a profile relation name that can be used to allow a client to chose between the URI returning "metric" or "imperial", "Kelvin", "Fahrenheit" or "Celsius" or similar measurement figures or the like.
So, long story short, loosely speeking the absolut URI, including matrix, query and path parameters, is what "names" a resource at an arbitrary client. The whole URI is the identifier of that resource after all. Slightly different names might result in local or intermediary cache misses and therefore indicate a different resource, even though the data expressed is the same as before. Instead of using two slighly different URIs redirection directives, content type negotiation or profiles on the same resource can be used to "get rid" of the sibling "resource" that only differ in different representation formats returned.
Variety of REST practises suggest (i.e. 1, 2, 3) to use plurals in your endpoints and the result is always a list of objects, unless it's filtered by a specific value, such as /users/123 Query parameters are used to filter the list, but still result in a list, nevertheless. I want to know if my case should 'abandon' those best practices.
Let's use cars for my example below.
I've got a database full of cars and each one has a BuildNumber ("Id"), but also a model and build year which combination is unique. If I then query for /cars/ and search for a specific model and year, for example /cars?model=golf&year=2018 I know, according to my previous sentence, my retrieve will always contain a single object, never multiple. My result, however, will still be a list, containing just one object, nevertheless.
In such case, what will be the best practise as the above would mean the object have to be extracted from the list, even though a single object could've been returned instead.
Stick to best practises and export a list
Make a second endpoind /car/ and use the query parameters ?model=golf&year=2018, which are primarily used for filtering in a list, and have the result be a single object, as the singular endpoint states
The reason that I'm asking this is simply for the cleanness of the action: I'm 100% sure my GET request will result in single object, but still have to perform actions to extract it from the list. These steps should've been unnecessary. Aside of that, In my case I don't know the unique identifier, so cars/123 for retrieving a specific car isn't an option. I know, however, filters that will result in one object and one specific object altogether. The additional steps simply feel redundant.
1: https://learn.microsoft.com/en-us/azure/architecture/best-practices/api-design
2: https://blog.mwaysolutions.com/2014/06/05/10-best-practices-for-better-restful-api/
3: https://medium.com/hashmapinc/rest-good-practices-for-api-design-881439796dc9
As you've specifically asked for best practices in regards to REST:
REST doesn't care how you specify your URIs or that semantically meaningful tokens are used inside the URI at all. Further, a client should never expect a certain URI to return a certain type but instead rely on content-type negotiation to tell the server all of the capabilities the client supports.
You should furthermore not think of REST in terms of object orientation but more in terms of affordance and statemachines where a client get served every information needed in order to make an educated decision on what to do next.
The best sample to give here is probably to take a close look at the Web and how it's done for HTML pages. How can you filter for a specific car and how it will be presented to you? The same concepts that are used in the Web also apply to REST as both use the same interaction model. In regards to your car sample, the API should initially return some control-structures that teach a client how a request needs to be formed and what options could be filtered for. In HTML this is done via forms. For non-HTML based REST APIs dedicated media-types should be defined that translate the same approach to non-HTML structures. On sending the request to the server, your client would include all of the supported media-types it supports in an Accept HTTP header, which informs the server about the capabilities of the client. Media-types are just human-readable specification on how to process payloads of such types. Such specifications may include hints on type information a link relation might return. In order to gain wide-usage of media-types they should be defined as generic as possible. Instead of defining a media-type specific for a car, which is possible, it probably would be more convenient to use an existing or define a new general data-container format (similar to HTML).
All of the steps mentioned here should help you to design and implement an API that is free to evolve without having to risk to break clients, that furthermore is also scalable and minimizes interoperability concerns.
Unfortunately your question targets something totally different IMO, something more related to RPC. You basically invoke a generic method via HTTP on an endpoint, similar like SOAP, RMI or CORBA work. Whether you respect the semantics of HTTP operations or not is only of sub-interest here. Even if you'd reached level 3 of the Richardson Maturity Model (RMM) it does not mean that you are compliant to REST. Your client might still break if the server changes anything within the response. The RMM further doesn't even consider media-types at all, hence I consider it as rather useless.
However, regardless if you use a (true) REST or RPC/CRUD client, if retrieving single items is your preference instead of feeding them into a collection you should consider to include the URI of the items of interest instead of its data directly into the collection, as Evert also has suggested. While most people seem to be concerned on server performance and round-trip-times, it actually is very elegant in terms of caching. Further certain link-relation names such as prefetch may inform the client that it may fetch the targets payload early as it is highly possible that it's content will be requested next. Through caching a request might not even have to be triggered or sent to the server for processing, which is probably the best performance gain you can achieve.
1) If you use query like cars/where... - use CARS
2) If you whant CAR - make method GetCarById
You might not get a perfect answer to this, because all are going to be a bit subjective and often in a different way.
My general thought about this is that every item in my system will have its own unique url, for example /cars/1234. That case is always singular.
But this specific item might appear as a member in collections and search results. When /cars/1234 apears in these, they will always appear as a list with 1 item (or 0 or more depending on the query).
I feel that this is ultimately the most predictable.
In my case though, if a car appears as a member of a search or colletion, it's 'true url' will still be displayed.
If entity (e.g person) has data that has to be presented in different representations:
I have a big profile that has different representations but for a small example:
representations1:
GET /profiles/{id}/activity/projection1
returns:
{"actions":["add", "delete", "add"], "dates":[1499865456, 1499861544, 1499863655]}
representations2:
GET /profiles/{id}/activity/projection2
returns:
{add_at:[1499865456, 1499863655], delete_at:[1499861544]}
So the question: how to design such cases?
I have some ideas but don't know which one is better
GET /profiles/{id}/activity/projection1
GET /profiles/{id}/activity/projections/1
GET /profiles/{id}/activities/projection1
GET /profiles/{id}/activities/projections/1
GET /profiles/{id}/activity-actions and GET /profiles/{id}/activity-timestamps
I found only one same question Different RESTful representations of the same resource but it is about filtering data in response not about change model
A couple things to keep in mind
As far as REST is concerned, two things with different identifiers (URI) are different resources. The fact that they have the same source of truth is an implementation detail.
In designing a REST api, your resources are integration points. The representations of your resources will depend on the state of your model at any given time.
For instance, if I look up Clayton Kershaw at Baseball Reference, I am probably directed to this resource
http://www.baseball-reference.com/players/k/kershcl01.shtml
But if I ask to see his "2014 splits", then I'll be directed to this resource instead.
http://www.baseball-reference.com/players/split.fcgi?id=kershcl01&year=2014&t=p
There's no particular reason that every resource identifier related to kershcl01 has to be under the same root in the hierarchy.
You may want to review Cool URIs don't change; stable URI over time is a good goal for a design, in which case you'll want to make sure that temporary implementation details don't leak into the URI space.
Jim Webber's observation was that REST resources are part of your integration domain; "Resources adapt your domain model for the web."
So your design guidance should probably come from asking what properties are important to your consumers, and what constraints will ensure that those properties are present (outside in), rather than starting from your (current) implementation.
This question is about optimal REST API design and a problem I'm facing to choose between nested resources and root level collections.
To demonstrate the concept, suppose I have collections City, Business, and Employees. A typical API may be constructed as follows. Imagine that ABC, X7N and WWW are keys, e.g. guids:
GET Api/City/ABC/Businesses (returns all Businesses in City ABC)
GET Api/City/ABC/Businesses/X7N (returns business X7N)
GET Api/City/ABC/Businesses/X7N/Employees (returns all employees at business X7N)
PUT Api/City/ABC/Businesses/X7N/Employees/WWW (updates employee WWW)
This appears clean because it follows the original domain structure - business are in a city, and employees are at a business. Individual items are accessible via key under the collection (e.g. ../Businesses returns all businesses, while ../Businesses/X7N returns the individual business).
Here is what the API consumer needs to be able to do:
Get businesses in a city (GET Api/City/ABC/Businesses)
Get all employees at a business (GET Api/City/ABC/Businesses/X7N/Employees)
Update individual employee information (PUT Api/City/ABC/Businesses/X7N/Employees/WWW)
That second and third call, while appearing to be in the right place, use a lot of parameters that are actually unnecessary.
To get employees at a business, the only parameter needed is the key of the business (X7N).
To update an individual employee, the only parameter needed it the key of the employee (WWW)
Nothing in the backend code requires non-key information to look up the business or update the employee. So, instead, the following endpoints appear better:
GET Api/City/ABC/Businesses (returns all Businesses in City ABC)
GET Api/Businesses/X7N (returns business X7N)
GET Api/Businesses/X7N/Employees (returns all employees at business X7N)
PUT Api/Employees/WWW (updates employee WWW)
As you can see, I've created a new root for businesses and employees, even though from a domain perspective they are a sub/sub-sub-collection.
Neither solution appears very clean to me.
The first example asks for unnecessary information, but is structured in a way that appears "natural" to the consumer (individual items from a collection are retrieved via lower leafs)
The second example only asks for necessary information, but isn't structured in a "natural" way - subcollections are accessible via roots
The individual employee root would not work when adding a new employee, as we need to know which business to add the employee to, which means that call would at least have to reside under the Business root, such as POST Api/Businesses/X7N7/Employees, which makes everything even more confusing.
Is there a cleaner, third way that I'm not thinking of?
I don't see how REST adds a constraint that two resources could not have the same value. The resourceType/ID is just an example of the easiest use case rather than the best way to go from a RESTful point of view.
If you read paragraph 5.2.1.1 of Roy Fielding's dissertation carefully, you will notice that Fielding makes the disctinction between a value and a resource. Now a resource should have a unique URI, that's true. But nothing prevents two resources from having the same value:
For example, the "authors' preferred version" of an academic paper is a mapping whose value changes over time, whereas a mapping to "the paper published in the proceedings of conference X" is static. These are two distinct resources, even if they both map to the same value at some point in time. The distinction is necessary so that both resources can be identified and referenced independently. A similar example from software engineering is the separate identification of a version-controlled source code file when referring to the "latest revision", "revision number 1.2.7", or "revision included with the Orange release."
So nothing prevents you from, as you say, changing the root. In your example, a Business is a value not a resource. It is perfectly RESTful to create a resource which is a list of "every business located in a city" (just like Roy's example, "revisions included with the Orange release"), while having a "business which ID is x" resource as well (like "revision number x").
For Employees, I would keep API/Businesses/X7N/Employees as the relation between a business and its employees is a composition relationship, and thus as you say, Employees can and should only be accessed through the Businesses class root. But this is not a REST requirement, and the other alternative is perfectly RESTful as well.
Note that this goes in pair with the application of the HATEAOS principle. In your API, the list of Businesses located in a city could (and perhaps should from a theoretical point of view) be just a list of links to the API/Businesses. But this would mean that the clients would have to do one round-trip to the server for each of the items in the list. This is not efficient and, to stay pragmatic, what I do is embed the representation of the business in the list along with the self link to the URI that would be in this example API/Businesses.
You should not confuse REST with the application of a specific URI naming convention.
HOW the resources are named is entirely secondary. You are trying to use HTTP resource naming conventions - this has nothing to do with REST. Roy Fielding himself states so repeatedly in the documents quoted above by others. REST is not a protocol, it is an architectural style.
In fact, Roy Fielding states in his 2008 blog comment (http://roy.gbiv.com/untangled/2008/rest-apis-must-be-hypertext-driven 6/20/2012):
"A REST API must not define fixed resource names or hierarchies (an obvious coupling of
client and server). Servers must have the freedom to control their own namespace. Instead,
allow servers to instruct clients on how to construct appropriate URIs, such as is done in
HTML forms and URI templates, by defining those instructions within media types and link relations."
So in essence:
The problem you describe is not actually a problem of REST - conceptually, it is a problem of HIERARCHY STRUCTURES versus RELATIONAL STRUCTURES.
While a business is "in" a city and so can be considered to be part of the city "hierarchy" - what about international companies which have offices in 75 cities. Then the city suddenly becomes the junior element in a hierarchy with the business name at the senior level of the structure.
The point is, you can view data from various angles, and depending on the viewpoint you take, it may be simplest to see it as a hierarchy. But the same data can be seen as a hierarchy with different levels. When you are using HTTP type resource names, then you have entered a hierarchy structure defined by HTTP. This is a constraint, yes, but it's not a REST constraint, it's a HTTP constraint.
From that angle, you can chose the solution which fits better to your scenario. If your customer cannot supply the city name when he supplies the company name (he may not know), then it would be better to have the key with only city name. As I said, it's up to you, and REST won't stand in your way ...
More to the point:
The only real REST constraints you have, if you have already decided to use HTTP with GET
PUT and so on, are:
Thou shalt not presumeth any prior ("out of band") knowledge between client and servers. *
Look at your proposal #1 above in that light. You assume that customers know the keys for the cities which are contained in your system? Wrong - that's not restful. So the server has to give the list of cities as a list of choices in some way. So are you going to list every city in the world here?
I guess not, but then you'll have to do some work on how you are planning to do this, which brings us to:
A REST API should spend almost all of its descriptive effort in defining the media type(s) used for representing resources and driving application state ...
I think, reading the mentioned Roy Fielding blog will help you out considerably.
In a RESTful-API URL design should be quite unimportant - or at least a side issue since the discoverability is encoded in the hypertext and not in the URL path. Have a look at the resources linked in the REST tag wiki here on StackOverflow.
But if you want to design human readable URLs for your UC, I would suggest the following:
Use the resource type you are creating/updating/querying as the first part of the URL (after your API prefix). So when somebody sees the URL he immediately knows to which resources this URL points. GET /Api/Employees... is the only only way to receive Employee resources from the API.
Use Unique IDs for each resource independent of the relations they are in. So GET /Api/<CollectionType>/UniqueKey should return a valid resource representation. Nobody should have to worry where the Employee is located. (But the returned Employee should have the links to the Business (and for convenience sake City) he belongs to.) GET /Api/Employees/Z6W returns the Employee with this ID no matter where is is located.
If you want to get a specific resource: Put your query parameter at the end (instead in the hierarchical order described in the question). You can use the URL query string (GET /Api/Employees?City=X7N) or a matrix parameter expression (GET /Api/Employees;City=X7N;Business=A4X,A5Y). This will allow you to easily express a collection of all Employees in a specific City - independent of the Business they are in.
Side node:
In my experience an initial hierarchical domain data model seldom survives additional requirements that come up during a project. In your case: Consider a business located in two Cities. You could create a workaround by modelling it as two separate businesses but what about the employee who works half his time in one place and the other half at the other location? Or even worse: It's only clear for which business he works but it's undefined, in which city?
The third way that I see is to make Businesses and Employees root resources and use query parameters to filter collections:
GET Api/Businesses?city=ABC (returns all Businesses in City ABC)
GET Api/Businesses/X7N (returns business X7N)
GET Api/Employees?businesses=X7N (returns all employees at business X7N)
PUT Api/Employees/WWW (updates employee WWW)
Your both solutions use concept of REST sub-resources which requires that subresource is included in parent resource so:
GET Api/City/ABC/Businesses
in response should also return data provided by:
GET Api/City/ABC/Businesses/X7N
GET Api/City/ABC/Businesses/X7N/Employees
similar for:
GET Api/Businesses/X7N
which should return data provided by:
GET Api/Businesses/X7N/Employees
It will make size of the response huge and time required to generate will increase.
To make REST API clean each resource should have only one bounded URI which fallow below patterns:
GET /resources
GET /resources/{id}
POST /resources
PUT /resources/{id}
If you need to make links between resources use HATEOAS
Go with example 1. I wouldn't worry about unnecessary information from the point of view of the server. A URL should clearly identify a resource in a unique fashion from the point of view of the client. If the client would not know what /Employee/12 means without first knowing that it is actually /Businesses/X7N/Employees/12 then the first URL seems redundant.
The client should be dealing with URLs rather than the individual parameters that make up the URLs, so there is nothing wrong with long URLs. To the client they are just strings. The server should be telling the client the URL to do what it needs to do, not the individual parameters that then require the client to construct the URL.