I had a quiet long debate with my colleague about the proper HTTP verb to be used for one of our operation that changes the STATE of a resource.
Suppose we have a resource called WakeUpLan that tries to send event to a system connected in a network. This is kind of a Generic State Machine,
{
id: 1,
retries: {
idle: 5, // after 5 retries it went to FAILED state
wakeup: 0,
process: 0,
shutdown: 0
},
status: 'FAILED',
// other attributes
}`
IDLE --> WAKEUP ---> PROCESS ---> SHUTDOWN | ----> [FAILED]
Every state has a retry mechanism, i.e in IDLE case it tries for x times to transition to WAKEUP and after x retries it dies out and goes to FAILED state.
All the FAILED resource can be again manually restarted or retried one more time from some interface.
So, we have a confusion regarding which HTTP verb best suits in this case.
In my opinion, it is just a change in status and resetting retry count to 0, so that our retry mechanism can catch this and try in next iteration.
so it should be a pure PATCH request
PATCH retry/{id}
{state: 'IDLE'}
But my colleague opposes it to be a POST request as this is a pure action and should be treated as POST.
I am not convinced because we are not creating any new resource but just updating existing resource that our REST server already knows about it.
I would like to know and corrected if I am wrong here.
Any suggestions/advices are welcome.
Thanks in advance.
Any suggestions/advices are welcome.
The reference implementation of the REST architectural style is the world wide web. The world wide web is built on a foundation of URI, HTTP, and HTML -- and HTML form processing is limited to GET and POST.
So POST must be an acceptable answer. After all, the web was catastrophically successful.
PATCH, like PUT, allows you to communicate changes to a representation of a resource. The semantics are more specific than POST, which allows the client to better take advantage. So if all you are doing is creating a message that describes local edits to the representation of the resource, then PATCH is a fine choice.
Don't overlook the possibilities of PUT -- if the size of the complete representation of the resource is of roughly the same order as the representation of your PATCH document, then using PUT may be a better choice, because of the idempotent semantics.
I am not convinced because we are not creating any new resource but just updating existing resource that our REST server already knows about it.
POST is much more general than "create a new resource". Historically, there has been a lot of confusion around this point (the language in the early HTTP specifications didn't help).
HTTP Basics
PATCH
What is PATCH actually? PATCH is a HTTP method defined in RFC 5789 that is similar to patching code in software engineering, where a change to one or multiple sources should be applied in order to transform the target resource to a desired outcome. Thereby a client is calculating a set of instructions the target system has to apply fully in order to generate the requested outcome. These instruction are usually called "patch", in the words of RFC 5789 such a set of instructions is called "patch document".
RFC 5789 does not define in which representation such a patch document need to be transferred from one system to the other. For JSON-based representations application/json-patch+json (RFC 6902) can be used which contains certain instructions like add, replace, move, copy, ... that are more or less clear on what they are doing but the RFC also describes each of the available instructions further.
A further JSON-based, but totally different take on how to inform a system on how to change a resource (or document) is captured in application/merge-patch+json (RFC 7386). In contrast to json-patch, this media-type does define a set of default rules to apply on receiving a JSON based representation to the actual target resource. Here, a single JSON representation of the modified state is sent to the server that only contains fields and objects that should be changed by the server. Default rules define that fields to be removed from the target resource need to be nullified in the request while fields that should change need to contain the new value to apply. Fields that remain unchanged can be left out in the request.
If you read through RFC 5789, you will find merge-patch as more of a hack though. Compared to json-patch, a merge-patch representation lacks the control of the actual sequence the instructions are applied, which might not always be necessary, as well as the lack of changing multiple, different resources at once.
PATCH itself is not idempotent. For a json-patch patch document it is pretty clear that applying the same instructions multiple times may lead to different results, i.e. if you remove the first field. A merge-patch document here behaves similar to a "partial PUT" request that so many developers perform due to pragmatism, even though the actual operation still does not guarantee idempotency. In order to avoid applying the same patch to the same resource unintentionally multiple times, i.e. due to network errors while transmitting the patch document, it is recommended to use PATCH alongside conditional requests (RFC 7232). This guarantees that the changes are only applied to a specific version of the resource and if that resource had changed either through a previous request or by an external source, the request would be declined to prevent data loss. This is basically optimistic locking.
A requirement that all patch documents have to fulfill is, that they need to be applied atomically. Either all the changes are applied or none at all. This puts some transaction burden onto the service provider.
POST
POST method is defined in RFC 7231 as:
requests that the target resource process the representation enclosed in the request according to the resource's own specific semantics.
This is basically a get-free-out-of-jail-card that lets you do anything you want or have to do here. You are free to define the syntax and structure to receive on a certain endpoint. Most of these so-called "REST APIs" consider POST as the C in CRUD, which it can be used for, but is just an oversimplification of what it actually can do for you. HTML basically only supports POST and GET operations so POST requests are used for sending all kinds of data to the server to start of backing processes, create new resources such as blog-posts, Q&A, videos, ... but also to delete or update stuff.
The rule of thumb here is, if a new resource is created as an outcome of triggering a POST request on a certain URI the response code should be 201 Created containing a HTTP response header Location with a URI as a value that points to the newly created resource. In any other case POST does not map to the C (create) of the CRUD stereotype.
REST-related
REST isn't a protocol but an architectural style. As Robert (Uncle Bob) C. Martin stated, architecture is about intent and REST intention is about decoupling clients from servers which allows the latter one to evolve freely by minimizing interoperability issues due to changes introduced by the server.
These are very strong benefits if your system should still work in decades to come. However, these benefits are unfortunately not obtained easily. As outlined in Fieldings dissertation to benefit from REST the mentioned constraints need to be followed strictly or otherwise couplings will remain increasing the likelihood of breaking clients due to changes. Fielding later on ranted about people that did either not read or understand his dissertation and clarified what a REST API has to do in a nutshell.
This rant can be summarized into the following points:
The API should adhere to and not violate the underlying protocol. Altough REST is used via HTTP most of the time, it is not restricted to this protocol.
Strong focus on resources and their presentation via media-types.
Clients should not have initial knowledge or assumptions on the available resources or their returned state ("typed" resource) in an API but learn them on the fly via issued requests and responses that teaches clients on what they can do next. This gives the server the freedom over its namespace and move around things it needs to without negatively impacting clients.
Based on this, REST is about using well-defined standards and adhering to the semantics of the protocols used as transportation facilities. Through the utilization of HATEOAS and stateless communication, the concepts that proved the Web to be scalable and evolution-friendly, the same interaction model that is used by humans in the Web is now used by applications in a REST architecture.
Common media-types provide the affordance on what a system might be able to do with data received for that payload while content-type negotiation guarantees that both, sender and receiver, are able to process and understand the payload correctly. The affordance may differ from media-type to media-type. A payload received for a image/png might be rendered and shown to the user while a application/vnd.acme-form+json might define a form where a server teaches a client on the elements of a request the server does support and a client can enter data and issue the request without having to actively know the method to use or target URI to send the data to as this is already given by the server. This not only removes the need for out-of-band (external) API documentation but also the need for a client to parse or interpret URIs as they are all provided by the server, accompanied by link-relations, that should be either standardized by IANA, follow common conventions such as existing rel values microformats or ontologies like Dublin Core, or represent extension types as defined in RFC 5988 (Web linking).
Question-related
With the introductory done, I hope that for a question like
But my colleague opposes it to be a POST request as this is a pure action and should be treated as POST. I am not convinced because we are not creating any new resource but just updating existing resource that our REST server already knows about it
it is clear that there is no definite yes or no answer to this quest but more of a it depends.
There are a couple of questions that could be asked, i.e. like
How many (different) clients will use the service? Are they all under your control? If so, you don't need REST, but you can still aim for it
How is the client taught or instructed on to perform the update? Will you provide an external API documentation? Will you support a media-type that supports forms, such as HTML, hal-forms, halo+json, Ion or Hydra
In general, if you have multiple clients, especially ones that are not under your control, you might not know which capabilities they support. Here content-type negotiation is an important part. If a client supports application/json-patch+json it might also be able to calculate a patch document containing the instructions to apply onto the target resource. The chances that it will also support PATCH are also very likely as RFC 6902 mentions it. In such a case it would make sense to provide a PATCH endpoint the client can send the request to.
If the client supports application/patch-merge+json one might assume that it supports PATCH as well, as it is primarily intended for use with the HTTP PATCH method, according to RFC 7386. Here the update from a client side perspective is rather trivial as the updated document is send as is to the server.
In any other case though, it is less clear in what representation formats the changes will be transmitted to the server. Here, POST is probably the way to go. From a REST stance, an update here has probably to be similar to an update done to data that is edited in a Web form in your browser with the current content being loaded into each form-element and the client modifies these form elements to its liking and then submits the changes back to the server in probably an application/x-www-form-urlencoded (or the like) structure. In such a case though, PUT would probably be more appropriate as in such a case you'd transmit the whole updated state of the resource back to the service and therefore perform a full update rather than a partial update on the target resource. The actual media-type the form will submit is probably defined in the media-type of the respective form. Note that this does not mean that you can't process json-patch or merge-patch documents in POST also.
The rule of thumb here would be, the more media-type formats and HTTP methods you support, the more likely different clients will be able to actually perform their task.
I would say you're in the right since you are not creating any new resource.
Highlight the part that says use put when you modify the entire existing resource while use patch when you are modifying one component of existing resource.
More here
https://restfulapi.net/rest-put-vs-post/
Related
I want to perform two POST requests to two different endpoints, is it possible to do this in a transactional way?
I want to perform two POST requests to two different endpoints, is it possible to do this in a transactional way?
No.
There's no rule that says that the effects of a POST will be scoped to precisely one resource. On the web, we can submit one HTML form that changes multiple pages. Implementations have a lot of freedom when processing a request (HTTP constrains message meaning, not handler implementations).
POST /replace/these/resources/with/empty/documents
Content-Type: text/plain
/A
/B/C
/D/E/F?g
A piece that is missing, however, is the ability to communicate to general purpose components which resources have been changed when handling the POST request. The mechanism described by RFC 7234 handles only a few simple cases, and does not (as far as I can tell) extend generally.
You get at most three
effective request uri
Location
Content-Location
and both Location and Content-Location mean something, so you may end up with a mess if you try to force them out of their normal semantics.
You can, of course, send back response that includes in the payload a list of all of the resources that were changed by your transaction (if that's practical), but you don't have any way to lift that information into the metadata where general purpose components can use it.
While designing rest API's I time to time have challenge to deal with batch operations (e.g. delete or update many entities at once) to reduce overhead of many tcp client connections. And in particular situation problem usually solves by adding custom api method for specific operation (e.g. POST /files/batchDelete which accepts ids at request body) which doesn't look pretty from point of view of rest api design principles but do the job.
But for me general solution for the problem still desirable. Recently I found Google Cloud Storage JSON API batching documentation which for me looks like pretty general solution. I mean similar format may be used for any http api, not just google cloud storage. So my question is - does anybody know kind of general standard (standard or it's draft, guideline, community effort or so) of making multiple API calls combined into one HTTP request?
I'm aware of capabilities of http/2 which include usage of single tcp connection for http requests but my question is addressed to application level. Which in my opinion still make sense because despite of ability to use http/2 taking that on application level seems like the only way to guarantee that for any client including http/1 which is currently the most used version of http.
TL;DR
REST nor HTTP are ideal for batch operations.
Usually caching, which is one of RESTs constraints, which is not optional but mandatory, prevents batch processing in some form.
It might be beneficial to not expose the data to update or remove in batch as own resources but as data elements within a single resource, like a data table in a HTML page. Here updating or removing all or parts of the entries should be straight forward.
If the system in general is write-intensive it is probably better to think of other solutions such as exposing the DB directly to those clients to spare a further level of indirection and complexity.
Utilization of caching may prevent a lot of workload on the server and even spare unnecessary connecctions
To start with, REST nor HTTP are ideal for batch operations. As Jim Webber pointed out the application domain of HTTP is the transfer of documents over the Web. This is what HTTP does and this is what it is good at. However, any business rules we conclude are just a side effect of the document management and we have to come up with solutions to turn this document management side effects to something useful.
As REST is just a generalization of the concepts used in the browsable Web, it is no miracle that the same concepts that apply to Web development also apply to REST development in some form. Thereby a question like how something should be done in REST usually resolves around answering how something should be done on the Web.
As mentioned before, HTTP isn't ideal in terms of batch processing actions. Sure, a GET request may retrieve multiple results, though in reality you obtain one response containing links to further resources. The creation of resources has, according to the HTTP specification, to be indicated with a Location header that points to the newly created resource. POST is defined as an all purpose method that allows to perform tasks according to server-specific semantics. So you could basically use it to create multiple resources at once. However, the HTTP spec clearly lacks support for indicating the creation of multiple resources at once as the Location header may only appear once per response as well as define only one URI in it. So how can a server indicate the creation of multiple resources to the server?
A further indication that HTTP isn't ideal for batch processing is that a URI must reference a single resource. That resource may change over time, though the URI can't ever point to multiple resources at once. The URI itself is, more or less, used as key by caches which store a cacheable response representation for that URI. As a URI may only ever reference one single resource, a cache will also only ever store the representation of one resource for that URI. A cache will invalidate a stored representation for a URI if an unsafe operation is performed on that URI. In case of a DELETE operation, which is by nature unsafe, the representation for the URI the DELETE is performed on will be removed. If you now "redirect" the DELETE operation to remove multiple backing resources at once, how should a cache take notice of that? It only operates on the URI invoked. Hence even when you delete multiple resources in one go via DELETE a cache might still serve clients with outdated information as it simply didn't take notice of the removal yet and its freshness value would still indicate a fresh-enough state. Unless you disable caching by default, which somehow violates one of REST's constraints, or reduce the time period a representation is considered fresh enough to a very low value, clients will probably get served with outdated information. You could of course perform an unsafe operation on each of these URIs then to "clear" the cache, though in that case you could have invoked the DELETE operation on each resource you wanted to batch delete itself to start with.
It gets a bit easier though if the batch of data you want to remove is not explicitly captured via their own resources but as data of a single resource. Think of a data-table on a Web page where you have certain form-elements, such as a checkbox you can click on to mark an entry as delete candidate and then after invoking the submit button send the respective selected elements to the server which performs the removal of these items. Here only the state of one resource is updated and thus a simple POST, PUT or even PATCH operation can be performed on that resource URI. This also goes well with caching as outlined before as only one resource has to be altered, which through the usage of unsafe operations on that URI will automatically lead to an invalidation of any stored representation for the given URI.
The above mentioned usage of form-elements to mark certain elements for removal depends however on the media-type issued. In the case of HTML its forms section specifies the available components and their affordances. An affordance is the knowledge what you can and should do with certain objects. I.e. a button or link may want to be pushed, a text field may expect numeric or alphanumeric input which further may be length limited and so on. Other media types, such as hal-forms, halform or ion, attempt to provide form representations and components for a JSON based notation, however, support for such media-types is still quite limited.
As one of your concerns are the number of client connections to your service, I assume you have a write-intensive scenario as in read-intensive cases caching would probably take away a good chunk of load from your server. I.e. BBC once reported that they could reduce the load on their servers drastically just by introducing a one minute caching interval for recently requested resources. This mainly affected their start page and the linked articles as people clicked on the latest news more often than on old news. On receiving a couple of thousands, if not hundred thousands, request per minute they could, as mentioned before, reduce the number of requests actually reaching the server significantly and therefore take away a huge load on their servers.
Write intensive use-cases however can't take benefit of caching as much as read-intensive cases as the cache would get invalidated quite often and the actual request being forward to the server for processing. If the API is more or less used to perform CRUD operations, as so many "REST" APIs do in reality, it is questionable if it wouldn't be preferable to expose the database directly to the clients. Almost all modern database vendors ship with sophisticated user-right management options and allow to create views that can be exposed to certain users. The "REST API" on top of it basically just adds a further level of indirection and complexity in such a case. By exposing the DB directly, performing batch updates or deletions shouldn't be an issue at all as through the respective query languages support for such operations should already be build into the DB layer.
In regards to the number of connections clients create: HTTP from 1.0 on allows the reusage of connections via the Connection: keep-alive header directive. In HTTP/1.1 persistent connections are used by default if not explicitly requested to close via the respective Connection: close header directive. HTTP/2 introduced full-duplex connections that allow many channels and therefore requests to reuse the same connections at the same time. This is more or less a fix for the connection limitation suggested in RFC 2626 which plenty of Web developers avoided by using CDN and similar stuff. Currently most implementations use a maximum limit of 100 channels and therefore simultaneous downloads via a single connections AFAIK.
Usually opening and closing a connection takes a bit of time and server resources and the more open connections a server has to deal with the more a system may suffer. Though open connections with hardly any traffic aren't a big issue for most servers. While the connection creation was usually considered to be the costly part, through the usage of persistent connections that factor moved now towards the number of requests issued, hence the request for sending out batch-requests, which HTTP is not really made for. Again, as mentioned throughout the post, through the smart utilization of caching plenty of requests may never reach the server at all, if possible. This is probably one of the best optimization strategies to reduce the number of simultaneous requests, as probably plenty of requests might never reach the server at all. Probably the best advice to give is in such a case to have a look at what kind of resources are requested frequently, which requests take up a lot of processing capacity and which ones can easily get responded with by utilizing caching options.
reduce overhead of many tcp client connections
If this is the crux of the issue, the easiest way to solve this is to switch to HTTP/2
In a way, HTTP/2 does exactly what you want. You open 1 connection, and using that collection you can send many HTTP requests in parallel. Unlike batching in a single HTTP request, it's mostly transparent for clients and response and requests can be processed out of order.
Ultimately batching multiple operations in a single HTTP request is always a network hack.
HTTP/2 is widely available. If HTTP/1.1 is still the most used version (this might be true, but gap is closing), this has more to do with servers not yet being set up for it, not clients.
If we create the multiple resource of update request using POST method in REST.what will be the impact at server side if number of resource created .
I Know using put request ,we can achieve fault tolerance due to idempotence.if we use post instead put,what will happen?
If we created number of resource using post for update , is there any performance issue ?if we created number of resource then what is impact on server ?
In post and put if we call same request n times ,we are going to hit the server n time then creating new resource and same resource should not impact on server.can please confirm this statement right or wrong .
If we create the multiple resource of update request using POST method in REST.what will be the impact at server side if number of resource created .
First of all, HTTP, the de-facto transport layer of REST, is an application protocol for transferring documents over a network and not just YOUR application domain you can run your business rules on. Any business rules you infer from sending data over the network are just a side-effect of the actual documentation management you perform via HTTP. While certain thins might map well from the documentation management to your business layer, certain things might not. I.e. HTTP isn't designed to support larger kinds of batch processing.
By that, even though HTTP itself defines a set of methods you can use, with IANA administering additional ones, the actual implementation depends on the server itself. It should follow the semantics outline in the RFC, though it might not. It may harm interoperability with other clients though in such a case, that is why it is recommended to follow the spec.
What implications or impact a request may have on the server depends on a couple of factors such as the kind of the server, the data that needs to be processed and whether work can be offloaded, i.e. by a cache, as well as the internal infrastructure you use. If you have a server that supports a couple of hundred cores and terabytes of address space a request to be processed might have less of an impact on the server than if you have a server with only a single CPU core and just a gigabyte of RAM, which has to fit in a couple of other applications as well as the OS itself. In general though the actual impact a request has on the server isn't tide to the operation you invoke as at its core HTTP is just a remote document management protocol, as explained before. Certain methods, such as PATCH, may be an exception to this rule though as it clearly demands transaction support as either all or none of the operations defined in the patch document need to be applied.
I Know using put request ,we can achieve fault tolerance due to idempotence.if we use post instead put,what will happen?
RFC 7231 includes a hint on the difference between POST and PUT:
The fundamental difference between the POST and PUT methods is highlighted by the different intent for the enclosed representation. The target resource in a POST request is intended to handle the enclosed representation according to the resource's own semantics, whereas the enclosed representation in a PUT request is defined as replacing the state of the target resource. Hence, the intent of PUT is idempotent and visible to intermediaries, even though the exact effect is only known by the origin server.
POST does not give a client any promises on what happens in case of a network error i.e. You might not know whether a request reached the server and only the response got lost or if the actual request didn't make it to the server at all. Jim Webber gave an example why idempotency is important, especially when you deal with money and currencies.
HTTP is rather specific to inform a client when a resource was created by including a HTTP Location header in the response that contains a URI to the created resource. This works on POST as well as PUT and PATCH. This premise can now be utilized to "safely" upload data. A client can send POST requests to the server until it receives a response with a Location header pointing to the created resource which is then used in the next step to perform a PUT update on that resource with the actual content. This pattern is called the POST-PUT creation pattern and it is especially useful if you have either a large payload to send or have to guarantee that the state only triggers a business rule once, i.e. in case of an online purchase.
Note that with the help of conditional requests some form of optimistic locking could be used as well, though this would require to at least know the state of the current resource beforehand as here a certain value, that is unique to the current state, is included in the request that acts as distributed lock which, if different to the state the server currently has, as there might have been an update by an other client in the meantime, will result in a rejection of the request at the server side.
If we created number of resource using post for update , is there any performance issue ?if we created number of resource then what is impact on server ?
I'm not fully sure what you mean by created a number of resources using post for update. Do you want to create or update a resource via POST? These methods just differ in the semantics they promise. How you map the event of the document modification to trigger certain business rules in your backend is completely up to you. In general though, as mentioned before, HTTP isn't ideal in terms of batch processing.
In post and put if we call same request n times ,we are going to hit the server n time then creating new resource and same resource should not impact on server.can please confirm this statement right or wrong
If you send n requests via POST to the server, the server will perform the logic that should perform on a POST request n times (if all of the requests reached the server). Whether a new resource is created or not depends on the implementation. A POST request might only start a backing process, some kind of calculation or actually doing nothing. If one was created though the response should contain a Location header with the URI that points to the location of the new resource.
In terms of sending n requests via PUT, if the same URI is used for all of these requests, the server in general should apply the payload as the new state of the targeted resource. If it internally results in a DB update or not is an implementation detail that may very from project to project. In general a PUT request does not reflect in the creation of a new resource unless the resource the target URI pointed to didn't exist before, though it also may create further resources as a side-effect. Imagine if you design some kind of version control system. PUT is allowed to have side effects. Such a side effect may be that you perform an update on the HEAD trunk, which applies the new state to the HEAD, though as a side-effect a new resource is created for that commit in the commit history.
So in summary, you can't deduce the impact a request has on a server solely based on the HTTP operation you use as at its heart HTTP is just an application protocol that transfers documents over a network. The actual business rules that get triggered are just a side effect of the actual document management. What impact a request has on the server depends on multiple factors, such as the type of the server you use but also on the length of the request and what you do with it on the server. Each of the available methods has its own semantics and you shouldn't compare them by the impact they might have on the server, but on the premise they give to a client. Certain things like anything related to a balance or money should be done via PUT due to the idempotent property of that method.
by considering the application of REST principles in the web.
i am doing a case study on REST and I have some doubt mostly on Uniform interface.
I assumes that Uniform Interface has only one single PROCESS instead of HTTP verbs (e.g. get, post, put, delete, head, ...). Is there any potential consequences of this kind of process with conventional HTTP verbs?
Is there any potential consequences of this kind of process with conventional HTTP verbs?
There are a few.
One consideration is safety. In the RFC-7231, safe is defined this way.
Request methods are considered "safe" if their defined semantics are essentially read-only; i.e., the client does not request, and does not expect, any state change on the origin server as a result of applying a safe method to a target resource.
So if PROCESS were a safe verb, like GET, you would have an analog of the read-only web. The HTTP spec also defines HEAD and OPTIONS (which are optimized reads) and TRACE (a debugging tool); given that HTML has been an extremely successful hypermedia format without including support for these methods suggests that they aren't particularly critical of themselves.
A safe specification of PROCESS preserves all of the scaling benefits of REST. But it's utility is limited - clients can consume content, but they can't produce any.
On the other hand, if PROCESS isn't safe, then a bunch of use cases can no longer be supported. Prefetching content is no longer an option, because the components can no longer assume that invoking PROCESS has no side effects on the server. Crawling is no longer an option, for the same reason.
It's probably worth noticing the mechanics of methods in the web -- it's the hypermedia format that describes which methods are appropriate for which links. So you could potentially work around some of the issues by defining the restrictions on the method within the hypermedia format itself. It's a different way of communicating the same information to any component that can consume the media type in question.
But there are at least two additional considerations. First, that the information in the links can only be understood by components that know that media type. On the web, most components are media type agnostic -- caching html looks exactly like caching json looks like caching jpeg.
Second, that the information in the links is only available on the outbound side. REST means stateless - all of the information needed to process the request is contained within the request. That implies that the request must include within it all of the information needed to handle a communication failure.
For instance, the http spec defines idempotent.
A request method is considered "idempotent" if the intended effect on the server of multiple identical requests with that method is the same as the effect for a single such request.
This property is important for intermediary components when they forward a request along an unreliable, and receive no response from the server. We've got no way to know if a message is lost, or just slow, and we've got no way to distinguish the request message being lost from the response message being lost.
If the request includes the information that it is idempotent, then the intermediaries know that they can resend the message to the server, rather than reporting the error to the client.
Contrast this with correct handling of POST in http; since the POST request does not have an idempotent marker on it, the components do not know that a resending the message is going to have the desired effect (which is why web browsers typically display a warning if you try to POST a form twice).
Locking yourself into a single method give you a choice; do you want to support error recovery by intermediaries? or do you want the flexibility to support not-idempotent writes?
What is the value of RESTful “methods” (ie. GET, POST, PUT, DELETE, PATCH, etc.)?
Why not just make every client use the “GET” method w/ any/all relevant params, headers, requestbodies, JSON,etc. etc.?
On the server side, the response to each method is custom & independently coded!
For example, what difference does it make to issue a database query via GET instead of POST?
I understand that GET is for queries that don’t change the DB (or anything else?).
And POST is for calls that do make changes.
But, near as I can tell, the RESTful standard doesn’t prevent one to code up a server response to GET and issue a stored procedure call that indeed DOES change the DB.
Vice versa… the RESTful standard doesn’t prevent one to code up a server response to POST and issue a stored procedure call that indeed does NOT change the ANYTHING!
I’m not arguing that a midtier (HTTP) “RESTlike” layer is necessary. It clearly is.
Let's say I'm wrong (and I may be). Isn't it still likely that there are numerous REST servers violating the proper use of these protocols suffering ZERO repercussions?
The following do not directly address my questions but merely dance uncomfortably around it like an acidhead stoner at a Dead concert:
Different Models for RESTful GET and POST
RESTful - GET or POST - what to do?
GET vs POST in REST Web Service
PUT vs POST in REST
I just spent ~80 hours trying to communicate a PATCH to my REST server (older Android Java doesn't recognize the newer PATCH so I had to issue a stupid kluge HTTP-OVERIDE-METHOD in the header). A POST would have worked fine but the sysop wouldn't budge because he respects REST.
I just don’t understand why to bother with each individual method. They don't seem to have much impact on Idempotence. They seem to be mere guidelines. And if you "violate" these "guidelines" they give someone else a chance to point a feckless finger at you. But so what?
Aren't these guidelines more trouble than they're worth?
I'm just confused. Please excuse the stridency of my post.
Aren’t REST GET/POST/etc. methods superfluous?
What is the value of RESTful “methods” (ie. GET, POST, PUT, DELETE, PATCH, etc.)?
First, a clarification. Those aren't RESTful methods; those are HTTP methods. The web is a reference implementation (for the most part) of the REST architectural style.
Which means that the authoritative answers to your questions are documented in the HTTP specification.
But, near as I can tell, the RESTful standard doesn’t prevent one to code up a server response to GET and issue a stored procedure call that indeed DOES change the DB.
The HTTP specification designates certain methods as being safe. Casually, this designates that a method is read only; the client is not responsible for any side effects that may occur on the server.
The purpose of distinguishing between safe and unsafe methods is to allow automated retrieval processes (spiders) and cache performance optimization (pre-fetching) to work without fear of causing harm.
But you are right, the HTTP standard doesn't prevent you from changing your database in response to a GET request. In fact, it even calls out specifically a case where you may choose to do that:
a safe request initiated by selecting an advertisement on the Web will often have the side effect of charging an advertising account.
The HTTP specification also designates certain methods as being idempotent
Of the request methods defined by this specification, PUT, DELETE, and safe request methods are idempotent.
The motivation for having idempotent methods? Unreliable networks
Idempotent methods are distinguished because the request can be repeated automatically if a communication failure occurs before the client is able to read the server's response.
Note that the client here might not be the user agent, but an intermediary component (like a reverse proxy) participating in the conversation.
Thus, if I'm writing a user agent, or a component, that needs to talk to your server, and your server conforms to the definition of methods in the HTTP specification, then I don't need to know anything about your application protocol to know how to correctly handle lost messages when the method is GET, PUT, or DELETE.
On the other hand, POST doesn't tell me anything, and since the unacknowledged message may still be on its way to you, it is dangerous to send a duplicate copy of the message.
Isn't it still likely that there are numerous REST servers violating the proper use of these protocols suffering ZERO repercussions?
Absolutely -- remember, the reference implementation of hypermedia is HTML, and HTML doesn't include support PUT or DELETE. If you want to afford a hypermedia control that invokes an unsafe operation, while still conforming to the HTTP and HTML standards, the POST is your only option.
Aren't these guidelines more trouble than they're worth?
Not really? They offer real value in reliability, and the extra complexity they add to the mix is pretty minimal.
I just don’t understand why to bother with each individual method. They don't seem to have much impact on idempotence.
They don't impact it, they communicate it.
The server already knows which of its resources are idempotent receivers. It's the client and the intermediary components that need that information. The HTTP specification gives you the ability to communicate that information for free to any other compliant component.
Using the maximally appropriate method for each request means that you can deploy your solution into a topology of commodity components, and it just works.
Alternatively, you can give up reliable messaging. Or you can write a bunch of custom code in your components to tell them explicitly which of your endpoints are idempotent receivers.
POST vs PATCH
Same song, different verse. If a resource supports OPTIONS, GET, and PATCH, then I can discover everything I need to know to execute a partial update, and I can do so using the same commodity implementation I use everywhere else.
Achieving the same result with POST is a whole lot more work. For instance, you need some mechanism for communicating to the client that POST has partial update semantics, and what media-types are accepted when patching a specific resource.
What do I lose by making each call on the client GET and the server honoring such just by paying attention to the request and not the method?
Conforming user-agents are allowed to assume that GET is safe. If you have side effects (writes) on endpoints accessible via GET, then the agent is allowed to pre-fetch the endpoint as an optimization -- the side effects start firing even though nobody expects it.
If the endpoint isn't an idempotent receiver, then you have to consider that the GET calls can happen more than once.
Furthermore, the user agent and intermediary components are allowed to make assumptions about caching -- requests that you expect to get all the way through to the server don't, because conforming components along the way are permitted to server replies out of their own cache.
To ice the cake, you are introducing another additional risk; undefined behavior.
A payload within a GET request message has no defined semantics; sending a payload body on a GET request might cause some existing implementations to reject the request.
Where I believe you are coming from, though I'm not certain, is more of an RPC point of view. Client sends a message, server responds; so long as both participants in the conversation have a common understanding of the semantics of the message, does it matter if the text in the message says "GET" or "POST" or "PATCH"? Of course not.
RPC is a fantastic choice when it fits the problem you are trying to solve.
But...
RPC at web scale is hard. Can your team deliver that? can your team deliver with cost effectiveness?
On the other hand, HTTP at scale is comparatively simple; there's an enormous ecosystem of goodies, using scalable architectures, that are stable, tested, well understood, and inexpensive. The tires are well and truly kicked.
You and your team hardly have to do anything; a bit of block and tackle to comply with the HTTP standards, and from that point on you can concentrate on delivering business value while you fall into the pit of success.