luciano/reitit
1
0
Fork 0
mirror of https://github.com/metosin/reitit.git synced 2025-12-17 00:11:11 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Move documentation to GitBook

Browse source
This commit is contained in:
Miikka Koskinen 2017-09-14 16:33:36 +03:00
parent 9b5a9479f9
commit 60ec0e9e81
14 changed files with 738 additions and 707 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
.gitignore vendored
View file

@ -8,5 +8,7 @@ pom.xml.asc
/.lein-* /.lein-*
/.nrepl-port /.nrepl-port
/.nrepl-history /.nrepl-history
/doc
/gh-pages /gh-pages
/node_modules
/package-lock.json
/_book

708
README.md
View file

@ -17,713 +17,9 @@ Ships with example router for [Ring](#ring). See [Issues](https://github.com/met
[![Clojars Project](http://clojars.org/metosin/reitit/latest-version.svg)](http://clojars.org/metosin/reitit) [![Clojars Project](http://clojars.org/metosin/reitit/latest-version.svg)](http://clojars.org/metosin/reitit)
## Route Syntax ## Documentation
Routes are defined as vectors, which String path, optional (non-vector) route argument and optional child routes. Routes can be wrapped in vectors. [Check out the full documentation!](https://metosin.github.io/reitit/)
Simple route:
```clj
["/ping"]
```
Two routes:
```clj
[["/ping"]
["/pong"]]
```
Routes with meta-data:
```clj
[["/ping" ::ping]
["/pong" {:name ::pong}]]
```
Routes with path and catch-all parameters:
```clj
[["/users/:user-id"]
["/public/*path"]]
```
Nested routes with meta-data:
```clj
["/api"
["/admin" {:middleware [::admin]}
["/user" ::user]
["/db" ::db]
["/ping" ::ping]]
```
Same routes flattened:
```clj
[["/api/admin/user" {:middleware [::admin], :name ::user}
["/api/admin/db" {:middleware [::admin], :name ::db}
["/api/ping" ::ping]]
```
## Routing
For routing, a `Router` is needed. Reitit ships with several different router implementations: `:linear-router`, `:lookup-router` and `:mixed-router`, based on the awesome [Pedestal](https://github.com/pedestal/pedestal/tree/master/route) implementation.
`Router` is created with `reitit.core/router`, which takes routes and optional options map as arguments. The route tree gets expanded, optionally coerced and compiled. Actual `Router` implementation is selected automatically but can be defined with a `:router` option. `Router` support both path- and name-based lookups.
Creating a router:
```clj
(require '[reitit.core :as reitit])
(def router
(reitit/router
[["/api"
["/ping" ::ping]
["/user/:id" ::user]]]))
```
`:mixed-router` is created (both static & wild routes are found):
```clj
(reitit/router-name router)
; :mixed-router
```
The expanded routes:
```clj
(reitit/routes router)
; [["/api/ping" {:name :user/ping}]
; ["/api/user/:id" {:name :user/user}]]
```
Route names:
```clj
(reitit/route-names router)
; [:user/ping :user/user]
```
### Path-based routing
```clj
(reitit/match-by-path router "/hello")
; nil
(reitit/match-by-path router "/api/user/1")
; #Match{:template "/api/user/:id"
; :meta {:name :user/user}
; :path "/api/user/1"
; :result nil
; :params {:id "1"}}
```
### Name-based (reverse) routing
```clj
(reitit/match-by-name router ::user)
; #PartialMatch{:template "/api/user/:id",
; :meta {:name :user/user},
; :result nil,
; :params nil,
; :required #{:id}}
(reitit/partial-match? (reitit/match-by-name router ::user))
; true
```
Only a partial match. Let's provide the path-parameters:
```clj
(reitit/match-by-name router ::user {:id "1"})
; #Match{:template "/api/user/:id"
; :meta {:name :user/user}
; :path "/api/user/1"
; :result nil
; :params {:id "1"}}
```
There is also a exception throwing version:
```clj
(reitit/match-by-name! router ::user)
; ExceptionInfo missing path-params for route /api/user/:id: #{:id}
```
## Route meta-data
Routes can have arbitrary meta-data. For nested routes, the meta-data is accumulated from root towards leafs using [meta-merge](https://github.com/weavejester/meta-merge).
A router based on nested route tree:
```clj
(def router
(reitit/router
["/api" {:interceptors [::api]}
["/ping" ::ping]
["/admin" {:roles #{:admin}}
["/users" ::users]
["/db" {:interceptors [::db]
:roles ^:replace #{:db-admin}}
["/:db" {:parameters {:db String}}
["/drop" ::drop-db]
["/stats" ::db-stats]]]]]))
```
Resolved route tree:
```clj
(reitit/routes router)
; [["/api/ping" {:interceptors [::api]
; :name ::ping}]
; ["/api/admin/users" {:interceptors [::api]
; :roles #{:admin}
; :name ::users}]
; ["/api/admin/db/:db/drop" {:interceptors [::api ::db]
; :roles #{:db-admin}
; :parameters {:db String}
; :name ::drop-db}]
; ["/api/admin/db/:db/stats" {:interceptors [::api ::db]
; :roles #{:db-admin}
; :parameters {:db String}
; :name ::db-stats}]]
```
Path-based routing:
```clj
(reitit/match-by-path router "/api/admin/users")
; #Match{:template "/api/admin/users"
; :meta {:interceptors [::api]
; :roles #{:admin}
; :name ::users}
; :result nil
; :params {}
; :path "/api/admin/users"}
```
On match, route meta-data is returned and can interpreted by the application.
Routers also support meta-data compilation enabling things like fast [Ring](https://github.com/ring-clojure/ring) or [Pedestal](http://pedestal.io/) -style handlers. Compilation results are found under `:result` in the match. See [configuring routers](#configuring-routers) for details.
## Route conflicts
Route trees should not have multiple routes that match to a single (request) path. `router` checks the route tree at creation for conflicts and calls a registered `:conflicts` option callback with the found conflicts. Default implementation throws `ex-info` with a descriptive message.
```clj
(reitit/router
[["/ping"]
["/:user-id/orders"]
["/bulk/:bulk-id"]
["/public/*path"]
["/:version/status"]])
; CompilerException clojure.lang.ExceptionInfo: router contains conflicting routes:
;
; /:user-id/orders
; -> /public/*path
; -> /bulk/:bulk-id
;
; /bulk/:bulk-id
; -> /:version/status
;
; /public/*path
; -> /:version/status
;
```
## Ring
[Ring](https://github.com/ring-clojure/ring)-router adds support for ring concepts like [handlers](https://github.com/ring-clojure/ring/wiki/Concepts#handlers), [middleware](https://github.com/ring-clojure/ring/wiki/Concepts#middleware) and routing based on `:request-method`. Ring-router is created with `reitit.ring/router` function. It runs a custom route compiler, creating a optimized stucture for handling route matches, with compiled middleware chain & handlers for all request methods. It also ensures that all routes have a `:handler` defined.
Simple Ring app:
```clj
(require '[reitit.ring :as ring])
(defn handler [_]
{:status 200, :body "ok"})
(def app
(ring/ring-handler
(ring/router
["/ping" handler])))
```
Applying the handler:
```clj
(app {:request-method :get, :uri "/favicon.ico"})
; nil
(app {:request-method :get, :uri "/ping"})
; {:status 200, :body "ok"}
```
The expanded routes:
```clj
(-> app (ring/get-router) (reitit/routes))
; [["/ping"
; {:handler #object[...]}
; #Methods{:any #Endpoint{:meta {:handler #object[...]},
; :handler #object[...],
; :middleware []}}]]
```
Note that the compiled resuts as third element in the route vector.
### Request-method based routing
Handler are also looked under request-method keys: `:get`, `:head`, `:patch`, `:delete`, `:options`, `:post` or `:put`. Top-level handler is used if request-method based handler is not found.
```clj
(def app
(ring/ring-handler
(ring/router
["/ping" {:name ::ping
:get handler
:post handler}])))
(app {:request-method :get, :uri "/ping"})
; {:status 200, :body "ok"}
(app {:request-method :put, :uri "/ping"})
; nil
```
Reverse routing:
```clj
(-> app
(ring/get-router)
(reitit/match-by-name ::ping)
:path)
; "/ping"
```
### Middleware
Middleware can be added with a `:middleware` key, with a vector value of the following:
1. ring middleware function `handler -> request -> response`
2. vector of middleware function `handler ?args -> request -> response` and optinally it's args.
A middleware and a handler:
```clj
(defn wrap [handler id]
(fn [request]
(handler (update request ::acc (fnil conj []) id))))
(defn handler [{:keys [::acc]}]
{:status 200, :body (conj acc :handler)})
```
App with nested middleware:
```clj
(def app
(ring/ring-handler
(ring/router
["/api" {:middleware [#(wrap % :api)]}
["/ping" handler]
["/admin" {:middleware [[wrap :admin]]}
["/db" {:middleware [[wrap :db]]
:delete {:middleware [#(wrap % :delete)]
:handler handler}}]]])))
```
Middleware is applied correctly:
```clj
(app {:request-method :delete, :uri "/api/ping"})
; {:status 200, :body [:api :handler]}
```
```clj
(app {:request-method :delete, :uri "/api/admin/db"})
; {:status 200, :body [:api :admin :db :delete :handler]}
```
### Middleware Records
Reitit supports first-class data-driven middleware via `reitit.middleware/Middleware` records, created with `reitit.middleware/create` function. The following keys have special purpose:
| key | description |
| -----------|-------------|
| `:name` | Name of the middleware as qualified keyword (optional,recommended for libs)
| `:wrap` | The actual middleware function of `handler args? => request => response`
| `:gen` | Middleware compile function, see [compiling middleware](#compiling-middleware).
When routes are compiled, all middleware are expanded (and optionally compiled) into `Middleware` and stored in compilation results for later use (api-docs etc). For actual request processing, they are unwrapped into normal middleware functions producing zero runtime performance penalty. Middleware expansion is backed by `reitit.middleware/IntoMiddleware` protocol, enabling plain clojure(script) maps to be used.
A Record:
```clj
(require '[reitit.middleware :as middleware])
(def wrap2
(middleware/create
{:name ::wrap2
:description "a nice little mw, takes 1 arg."
:wrap wrap}))
```
As plain map:
```clj
;; plain map
(def wrap3
{:name ::wrap3
:description "a nice little mw, :api as arg"
:wrap (fn [handler]
(wrap handler :api))})
```
### Async Ring
All built-in middleware provide both 2 and 3-arity and are compiled for both Clojure & ClojureScript, so they work with [Async Ring](https://www.booleanknot.com/blog/2016/07/15/asynchronous-ring.html) and [Node.js](https://nodejs.org) too.
### Meta-data based extensions
`ring-handler` injects the `Match` into a request and it can be extracted at runtime with `reitit.ring/get-match`. This can be used to build dynamic extensions to the system.
Example middleware to guard routes based on user roles:
```clj
(require '[clojure.set :as set])
(defn wrap-enforce-roles [handler]
(fn [{:keys [::roles] :as request}]
(let [required (some-> request (ring/get-match) :meta ::roles)]
(if (and (seq required) (not (set/intersection required roles)))
{:status 403, :body "forbidden"}
(handler request)))))
```
Mounted to an app via router meta-data (effecting all routes):
```clj
(def handler (constantly {:status 200, :body "ok"}))
(def app
(ring/ring-handler
(ring/router
[["/api"
["/ping" handler]
["/admin" {::roles #{:admin}}
["/ping" handler]]]]
{:meta {:middleware [wrap-enforce-roles]}})))
```
Anonymous access to public route:
```clj
(app {:request-method :get, :uri "/api/ping"})
; {:status 200, :body "ok"}
```
Anonymous access to guarded route:
```clj
(app {:request-method :get, :uri "/api/admin/ping"})
; {:status 403, :body "forbidden"}
```
Authorized access to guarded route:
```clj
(app {:request-method :get, :uri "/api/admin/ping", ::roles #{:admin}})
; {:status 200, :body "ok"}
```
## Parameter coercion
Reitit provides pluggable parameter coercion via `reitit.coercion.protocol/Coercion` protocol, originally introduced in [compojure-api](https://clojars.org/metosin/compojure-api). Reitit ships with `reitit.coercion.spec/SpecCoercion` providing implemenation for [clojure.spec](https://clojure.org/about/spec) and [data-specs](https://github.com/metosin/spec-tools#data-specs).
**NOTE**: Before Clojure 1.9.0 is shipped, to use the spec-coercion, one needs to add the following dependencies manually to the project:
```clj
[org.clojure/clojure "1.9.0-alpha20"]
[org.clojure/spec.alpha "0.1.123"]
[metosin/spec-tools "0.3.3"]
```
### Ring request and response coercion
To use `Coercion` with Ring, one needs to do the following:
1. Define parameters and responses as data into route meta-data, in format adopted from [ring-swagger](https://github.com/metosin/ring-swagger#more-complete-example):
* `:parameters` map, with submaps for different parameters: `:query`, `:body`, `:form`, `:header` and `:path`. Parameters are defined in the format understood by the `Coercion`.
* `:responses` map, with response status codes as keys (or `:default` for "everything else") with maps with `:schema` and optionally `:description` as values.
2. Define a `Coercion` to route meta-data under `:coercion`
3. Mount request & response coercion middleware to the routes.
If the request coercion succeeds, the coerced parameters are injected into request under `:parameters`.
If either request or response coercion fails, an descriptive error is thrown.
#### Example with data-specs
```clj
(require '[reitit.ring :as ring])
(require '[reitit.coercion :as coercion])
(require '[reitit.coercion.spec :as spec])
(def app
(ring/ring-handler
(ring/router
["/api"
["/ping" {:parameters {:body {:x int?, :y int?}}
:responses {200 {:schema {:total pos-int?}}}
:get {:handler (fn [{{{:keys [x y]} :body} :parameters}]
{:status 200
:body {:total (+ x y)}})}}]]
{:meta {:middleware [coercion/gen-wrap-coerce-parameters
coercion/gen-wrap-coerce-response]
:coercion spec/coercion}})))
```
```clj
(app
{:request-method :get
:uri "/api/ping"
:body-params {:x 1, :y 2}})
; {:status 200, :body {:total 3}}
```
#### Example with specs
```clj
(require '[reitit.ring :as ring])
(require '[reitit.coercion :as coercion])
(require '[reitit.coercion.spec :as spec])
(require '[clojure.spec.alpha :as s])
(require '[spec-tools.core :as st])
(s/def ::x (st/spec int?))
(s/def ::y (st/spec int?))
(s/def ::total int?)
(s/def ::request (s/keys :req-un [::x ::y]))
(s/def ::response (s/keys :req-un [::total]))
(def app
(ring/ring-handler
(ring/router
["/api"
["/ping" {:parameters {:body ::request}
:responses {200 {:schema ::response}}
:get {:handler (fn [{{{:keys [x y]} :body} :parameters}]
{:status 200
:body {:total (+ x y)}})}}]]
{:meta {:middleware [coercion/gen-wrap-coerce-parameters
coercion/gen-wrap-coerce-response]
:coercion spec/coercion}})))
```
```clj
(app
{:request-method :get
:uri "/api/ping"
:body-params {:x 1, :y 2}})
; {:status 200, :body {:total 3}}
```
## Compiling Middleware
The [meta-data extensions](#meta-data-based-extensions) are a easy way to extend the system. Routes meta-data can be transformed into any shape (records, functions etc.) in route compilation, enabling fast access at request-time.
Still, we can do better. As we know the exact route that interceptor/middleware is linked to, we can pass the (compiled) route information into the interceptor/middleware at creation-time. It can extract and transform relevant data just for it and pass it into the actual request-handler via a closure - yielding faster runtime processing.
To do this we use [middleware records](#middleware-records) `:gen` hook instead of the normal `:wrap`. `:gen` expects a function of `route-meta router-opts => wrap`. Middleware can also return `nil`, which effective unmounts the middleware. Why mount a `wrap-enforce-roles` middleware for a route if there are no roles required for it?
To demonstrate the two approaches, below are response coercion middleware written as normal ring middleware function and as middleware record with `:gen`. These are the actual codes are from [`reitit.coercion`](https://github.com/metosin/reitit/blob/master/src/reitit/coercion.cljc):
### Naive
* Extracts the compiled route information on every request.
```clj
(defn wrap-coerce-response
"Pluggable response coercion middleware.
Expects a :coercion of type `reitit.coercion.protocol/Coercion`
and :responses from route meta, otherwise does not mount."
[handler]
(fn
([request]
(let [response (handler request)
method (:request-method request)
match (ring/get-match request)
responses (-> match :result method :meta :responses)
coercion (-> match :meta :coercion)
opts (-> match :meta :opts)]
(if (and coercion responses)
(let [coercers (response-coercers coercion responses opts)
coerced (coerce-response coercers request response)]
(coerce-response coercers request (handler request)))
(handler request))))
([request respond raise]
(let [response (handler request)
method (:request-method request)
match (ring/get-match request)
responses (-> match :result method :meta :responses)
coercion (-> match :meta :coercion)
opts (-> match :meta :opts)]
(if (and coercion responses)
(let [coercers (response-coercers coercion responses opts)
coerced (coerce-response coercers request response)]
(handler request #(respond (coerce-response coercers request %))))
(handler request respond raise))))))
```
### Compiled
* Route information is provided via a closure
* Pre-compiled coercers
* Mounts only if `:coercion` and `:responses` are defined for the route
```clj
(def gen-wrap-coerce-response
"Generator for pluggable response coercion middleware.
Expects a :coercion of type `reitit.coercion.protocol/Coercion`
and :responses from route meta, otherwise does not mount."
(middleware/create
{:name ::coerce-response
:gen (fn [{:keys [responses coercion opts]} _]
(if (and coercion responses)
(let [coercers (response-coercers coercion responses opts)]
(fn [handler]
(fn
([request]
(coerce-response coercers request (handler request)))
([request respond raise]
(handler request #(respond (coerce-response coercers request %)) raise)))))))}))
```
The `:gen` -version has 50% less code, is easier to reason about and is 2-4x faster on basic perf tests.
## Merging route-trees
*TODO*
## Validating route-trees
Namespace `reitit.spec` contains [specs](https://clojure.org/about/spec) for routes, router and router options.
To enable spec-validation of `router` inputs & outputs at development time, one can do the following:
```clj
; add to dependencies:
; [expound "0.3.0"]
(require '[clojure.spec.test.alpha :as st])
(require '[expound.alpha :as expound])
(require '[clojure.spec.alpha :as s])
(require '[reitit.spec])
(st/instrument `reitit/router)
(set! s/*explain-out* expound/printer)
(reitit/router
["/api"
["/publuc"
["/ping"]
["pong"]]])
; -- Spec failed --------------------
;
; ["/api" ...]
; ^^^^^^
;
; should satisfy
;
; (clojure.spec.alpha/cat
; :path
; :reitit.spec/path
; :arg
; (clojure.spec.alpha/? :reitit.spec/arg)
; :childs
; (clojure.spec.alpha/* (clojure.spec.alpha/and :reitit.spec/raw-route)))
;
; -- Relevant specs -------
;
; :reitit.spec/raw-route:
; (clojure.spec.alpha/cat
; :path
; :reitit.spec/path
; :arg
; (clojure.spec.alpha/? :reitit.spec/arg)
; :childs
; (clojure.spec.alpha/* (clojure.spec.alpha/and :reitit.spec/raw-route)))
; :reitit.spec/raw-routes:
; (clojure.spec.alpha/or
; :route
; :reitit.spec/raw-route
; :routes
; (clojure.spec.alpha/coll-of :reitit.spec/raw-route :into []))
;
; -- Spec failed --------------------
;
; [... [... ... ["pong"]]]
; ^^^^^^
;
; should satisfy
;
; (fn [%] (clojure.string/starts-with? % "/"))
;
; -- Relevant specs -------
;
; :reitit.spec/path:
; (clojure.spec.alpha/and
; clojure.core/string?
; (clojure.core/fn [%] (clojure.string/starts-with? % "/")))
; :reitit.spec/raw-route:
; (clojure.spec.alpha/cat
; :path
; :reitit.spec/path
; :arg
; (clojure.spec.alpha/? :reitit.spec/arg)
; :childs
; (clojure.spec.alpha/* (clojure.spec.alpha/and :reitit.spec/raw-route)))
; :reitit.spec/raw-routes:
; (clojure.spec.alpha/or
; :route
; :reitit.spec/raw-route
; :routes
; (clojure.spec.alpha/coll-of :reitit.spec/raw-route :into []))
;
; -------------------------
; Detected 2 errors
```
### Validating meta-data
*TODO*
## Swagger & Openapi
*TODO*
## Interceptors
*TODO*
## Configuring Routers
Routers can be configured via options. Options allow things like [`clojure.spec`](https://clojure.org/about/spec) validation for meta-data and fast, compiled handlers. The following options are available for the `reitit.core/router`:
| key | description |
| -------------|-------------|
| `:path` | Base-path for routes
| `:routes` | Initial resolved routes (default `[]`)
| `:meta` | Initial expanded route-meta vector (default `[]`)
| `:expand` | Function of `arg opts => meta` to expand route arg to route meta-data (default `reitit.core/expand`)
| `:coerce` | Function of `route opts => route` to coerce resolved route, can throw or return `nil`
| `:compile` | Function of `route opts => result` to compile a route handler
| `:conflicts` | Function of `{route #{route}} => side-effect` to handle conflicting routes (default `reitit.core/throw-on-conflicts!`)
| `:router` | Function of `routes opts => router` to override the actual router implementation
## Special thanks ## Special thanks

13
book.json Normal file
View file

@ -0,0 +1,13 @@
{
"root": "doc",
"plugins": ["editlink", "github"],
"pluginsConfig": {
"editlink": {
"base": "https://github.com/metosin/reitit/tree/master/doc",
"label": "Edit This Page"
},
"github": {
"url": "https://github.com/metosin/reitit"
}
}
}

16
doc/README.md Normal file
View file

@ -0,0 +1,16 @@
# reitit
[reitit](https://github.com/metosin/reitit) is a friendly data-driven router for Clojure(Script).
* Simple data-driven [route syntax](./routing/route_syntax.md)
* First-class [route meta-data](./routing/route_metadata.md)
* Generic, not tied to HTTP
* [Route conflict resolution](./routing/route_conflicts.md)
* [Pluggable coercion](./parameter-coercion.md) ([clojure.spec](https://clojure.org/about/spec))
* both [Middleware](./ring.md#middleware) & Interceptors
* Extendable
* Fast
## Latest version
[![Clojars Project](http://clojars.org/metosin/reitit/latest-version.svg)](http://clojars.org/metosin/reitit)

16
doc/SUMMARY.md Normal file
View file

@ -0,0 +1,16 @@
# Summary
* [Introduction](README.md)
* Routing
* [Route syntax](routing/route_syntax.md)
* [Routers](routing/routers.md)
* [Route metadata](routing/route_metadata.md)
* [Route conflicts](routing/route_conflicts.md)
* [Ring support](ring.md)
* [Parameter coercion](parameter_coercion.md)
* [Compiling middleware](compiling_middleware.md)
* [Validating route-trees](validating.md)
* [Configuring routers](configuring_routers.md)
* TODO: Merging route-trees
* TODO: Swagger & OpenAPI
* TODO: Interceptors

72
doc/compiling_middleware.md Normal file
View file

@ -0,0 +1,72 @@
## Compiling Middleware
The [meta-data extensions](#meta-data-based-extensions) are a easy way to extend the system. Routes meta-data can be transformed into any shape (records, functions etc.) in route compilation, enabling fast access at request-time.
Still, we can do better. As we know the exact route that interceptor/middleware is linked to, we can pass the (compiled) route information into the interceptor/middleware at creation-time. It can extract and transform relevant data just for it and pass it into the actual request-handler via a closure - yielding faster runtime processing.
To do this we use [middleware records](#middleware-records) `:gen` hook instead of the normal `:wrap`. `:gen` expects a function of `route-meta router-opts => wrap`. Middleware can also return `nil`, which effective unmounts the middleware. Why mount a `wrap-enforce-roles` middleware for a route if there are no roles required for it?
To demonstrate the two approaches, below are response coercion middleware written as normal ring middleware function and as middleware record with `:gen`. These are the actual codes are from [`reitit.coercion`](https://github.com/metosin/reitit/blob/master/src/reitit/coercion.cljc):
### Naive
* Extracts the compiled route information on every request.
```clj
(defn wrap-coerce-response
"Pluggable response coercion middleware.
Expects a :coercion of type `reitit.coercion.protocol/Coercion`
and :responses from route meta, otherwise does not mount."
[handler]
(fn
([request]
(let [response (handler request)
method (:request-method request)
match (ring/get-match request)
responses (-> match :result method :meta :responses)
coercion (-> match :meta :coercion)
opts (-> match :meta :opts)]
(if (and coercion responses)
(let [coercers (response-coercers coercion responses opts)
coerced (coerce-response coercers request response)]
(coerce-response coercers request (handler request)))
(handler request))))
([request respond raise]
(let [response (handler request)
method (:request-method request)
match (ring/get-match request)
responses (-> match :result method :meta :responses)
coercion (-> match :meta :coercion)
opts (-> match :meta :opts)]
(if (and coercion responses)
(let [coercers (response-coercers coercion responses opts)
coerced (coerce-response coercers request response)]
(handler request #(respond (coerce-response coercers request %))))
(handler request respond raise))))))
```
### Compiled
* Route information is provided via a closure
* Pre-compiled coercers
* Mounts only if `:coercion` and `:responses` are defined for the route
```clj
(def gen-wrap-coerce-response
"Generator for pluggable response coercion middleware.
Expects a :coercion of type `reitit.coercion.protocol/Coercion`
and :responses from route meta, otherwise does not mount."
(middleware/create
{:name ::coerce-response
:gen (fn [{:keys [responses coercion opts]} _]
(if (and coercion responses)
(let [coercers (response-coercers coercion responses opts)]
(fn [handler]
(fn
([request]
(coerce-response coercers request (handler request)))
([request respond raise]
(handler request #(respond (coerce-response coercers request %)) raise)))))))}))
```
The `:gen` -version has 50% less code, is easier to reason about and is 2-4x faster on basic perf tests.

15
doc/configuring_routers.md Normal file
View file

@ -0,0 +1,15 @@
## Configuring Routers
Routers can be configured via options. Options allow things like [`clojure.spec`](https://clojure.org/about/spec) validation for meta-data and fast, compiled handlers. The following options are available for the `reitit.core/router`:
| key | description |
| -------------|-------------|
| `:path` | Base-path for routes
| `:routes` | Initial resolved routes (default `[]`)
| `:meta` | Initial expanded route-meta vector (default `[]`)
| `:expand` | Function of `arg opts => meta` to expand route arg to route meta-data (default `reitit.core/expand`)
| `:coerce` | Function of `route opts => route` to coerce resolved route, can throw or return `nil`
| `:compile` | Function of `route opts => result` to compile a route handler
| `:conflicts` | Function of `{route #{route}} => side-effect` to handle conflicting routes (default `reitit.core/throw-on-conflicts!`)
| `:router` | Function of `routes opts => router` to override the actual router implementation

92
doc/parameter_coercion.md Normal file
View file

@ -0,0 +1,92 @@
# Parameter coercion
Reitit provides pluggable parameter coercion via `reitit.coercion.protocol/Coercion` protocol, originally introduced in [compojure-api](https://clojars.org/metosin/compojure-api). Reitit ships with `reitit.coercion.spec/SpecCoercion` providing implemenation for [clojure.spec](https://clojure.org/about/spec) and [data-specs](https://github.com/metosin/spec-tools#data-specs).
**NOTE**: Before Clojure 1.9.0 is shipped, to use the spec-coercion, one needs to add the following dependencies manually to the project:
```clj
[org.clojure/clojure "1.9.0-alpha20"]
[org.clojure/spec.alpha "0.1.123"]
[metosin/spec-tools "0.3.3"]
```
### Ring request and response coercion
To use `Coercion` with Ring, one needs to do the following:
1. Define parameters and responses as data into route meta-data, in format adopted from [ring-swagger](https://github.com/metosin/ring-swagger#more-complete-example):
* `:parameters` map, with submaps for different parameters: `:query`, `:body`, `:form`, `:header` and `:path`. Parameters are defined in the format understood by the `Coercion`.
* `:responses` map, with response status codes as keys (or `:default` for "everything else") with maps with `:schema` and optionally `:description` as values.
2. Define a `Coercion` to route meta-data under `:coercion`
3. Mount request & response coercion middleware to the routes.
If the request coercion succeeds, the coerced parameters are injected into request under `:parameters`.
If either request or response coercion fails, an descriptive error is thrown.
#### Example with data-specs
```clj
(require '[reitit.ring :as ring])
(require '[reitit.coercion :as coercion])
(require '[reitit.coercion.spec :as spec])
(def app
(ring/ring-handler
(ring/router
["/api"
["/ping" {:parameters {:body {:x int?, :y int?}}
:responses {200 {:schema {:total pos-int?}}}
:get {:handler (fn [{{{:keys [x y]} :body} :parameters}]
{:status 200
:body {:total (+ x y)}})}}]]
{:meta {:middleware [coercion/gen-wrap-coerce-parameters
coercion/gen-wrap-coerce-response]
:coercion spec/coercion}})))
```
```clj
(app
{:request-method :get
:uri "/api/ping"
:body-params {:x 1, :y 2}})
; {:status 200, :body {:total 3}}
```
#### Example with specs
```clj
(require '[reitit.ring :as ring])
(require '[reitit.coercion :as coercion])
(require '[reitit.coercion.spec :as spec])
(require '[clojure.spec.alpha :as s])
(require '[spec-tools.core :as st])
(s/def ::x (st/spec int?))
(s/def ::y (st/spec int?))
(s/def ::total int?)
(s/def ::request (s/keys :req-un [::x ::y]))
(s/def ::response (s/keys :req-un [::total]))
(def app
(ring/ring-handler
(ring/router
["/api"
["/ping" {:parameters {:body ::request}
:responses {200 {:schema ::response}}
:get {:handler (fn [{{{:keys [x y]} :body} :parameters}]
{:status 200
:body {:total (+ x y)}})}}]]
{:meta {:middleware [coercion/gen-wrap-coerce-parameters
coercion/gen-wrap-coerce-response]
:coercion spec/coercion}})))
```
```clj
(app
{:request-method :get
:uri "/api/ping"
:body-params {:x 1, :y 2}})
; {:status 200, :body {:total 3}}
```

205
doc/ring.md Normal file
View file

@ -0,0 +1,205 @@
# Ring support
[Ring](https://github.com/ring-clojure/ring)-router adds support for ring concepts like [handlers](https://github.com/ring-clojure/ring/wiki/Concepts#handlers), [middleware](https://github.com/ring-clojure/ring/wiki/Concepts#middleware) and routing based on `:request-method`. Ring-router is created with `reitit.ring/router` function. It runs a custom route compiler, creating a optimized stucture for handling route matches, with compiled middleware chain & handlers for all request methods. It also ensures that all routes have a `:handler` defined.
Simple Ring app:
```clj
(require '[reitit.ring :as ring])
(defn handler [_]
{:status 200, :body "ok"})
(def app
(ring/ring-handler
(ring/router
["/ping" handler])))
```
Applying the handler:
```clj
(app {:request-method :get, :uri "/favicon.ico"})
; nil
(app {:request-method :get, :uri "/ping"})
; {:status 200, :body "ok"}
```
The expanded routes:
```clj
(-> app (ring/get-router) (reitit/routes))
; [["/ping"
; {:handler #object[...]}
; #Methods{:any #Endpoint{:meta {:handler #object[...]},
; :handler #object[...],
; :middleware []}}]]
```
Note that the compiled resuts as third element in the route vector.
### Request-method based routing
Handler are also looked under request-method keys: `:get`, `:head`, `:patch`, `:delete`, `:options`, `:post` or `:put`. Top-level handler is used if request-method based handler is not found.
```clj
(def app
(ring/ring-handler
(ring/router
["/ping" {:name ::ping
:get handler
:post handler}])))
(app {:request-method :get, :uri "/ping"})
; {:status 200, :body "ok"}
(app {:request-method :put, :uri "/ping"})
; nil
```
Reverse routing:
```clj
(-> app
(ring/get-router)
(reitit/match-by-name ::ping)
:path)
; "/ping"
```
### Middleware
Middleware can be added with a `:middleware` key, with a vector value of the following:
1. ring middleware function `handler -> request -> response`
2. vector of middleware function `handler ?args -> request -> response` and optinally it's args.
A middleware and a handler:
```clj
(defn wrap [handler id]
(fn [request]
(handler (update request ::acc (fnil conj []) id))))
(defn handler [{:keys [::acc]}]
{:status 200, :body (conj acc :handler)})
```
App with nested middleware:
```clj
(def app
(ring/ring-handler
(ring/router
["/api" {:middleware [#(wrap % :api)]}
["/ping" handler]
["/admin" {:middleware [[wrap :admin]]}
["/db" {:middleware [[wrap :db]]
:delete {:middleware [#(wrap % :delete)]
:handler handler}}]]])))
```
Middleware is applied correctly:
```clj
(app {:request-method :delete, :uri "/api/ping"})
; {:status 200, :body [:api :handler]}
```
```clj
(app {:request-method :delete, :uri "/api/admin/db"})
; {:status 200, :body [:api :admin :db :delete :handler]}
```
### Middleware Records
Reitit supports first-class data-driven middleware via `reitit.middleware/Middleware` records, created with `reitit.middleware/create` function. The following keys have special purpose:
| key | description |
| -----------|-------------|
| `:name` | Name of the middleware as qualified keyword (optional,recommended for libs)
| `:wrap` | The actual middleware function of `handler args? => request => response`
| `:gen` | Middleware compile function, see [compiling middleware](#compiling-middleware).
When routes are compiled, all middleware are expanded (and optionally compiled) into `Middleware` and stored in compilation results for later use (api-docs etc). For actual request processing, they are unwrapped into normal middleware functions producing zero runtime performance penalty. Middleware expansion is backed by `reitit.middleware/IntoMiddleware` protocol, enabling plain clojure(script) maps to be used.
A Record:
```clj
(require '[reitit.middleware :as middleware])
(def wrap2
(middleware/create
{:name ::wrap2
:description "a nice little mw, takes 1 arg."
:wrap wrap}))
```
As plain map:
```clj
;; plain map
(def wrap3
{:name ::wrap3
:description "a nice little mw, :api as arg"
:wrap (fn [handler]
(wrap handler :api))})
```
### Async Ring
All built-in middleware provide both 2 and 3-arity and are compiled for both Clojure & ClojureScript, so they work with [Async Ring](https://www.booleanknot.com/blog/2016/07/15/asynchronous-ring.html) and [Node.js](https://nodejs.org) too.
### Meta-data based extensions
`ring-handler` injects the `Match` into a request and it can be extracted at runtime with `reitit.ring/get-match`. This can be used to build dynamic extensions to the system.
Example middleware to guard routes based on user roles:
```clj
(require '[clojure.set :as set])
(defn wrap-enforce-roles [handler]
(fn [{:keys [::roles] :as request}]
(let [required (some-> request (ring/get-match) :meta ::roles)]
(if (and (seq required) (not (set/intersection required roles)))
{:status 403, :body "forbidden"}
(handler request)))))
```
Mounted to an app via router meta-data (effecting all routes):
```clj
(def handler (constantly {:status 200, :body "ok"}))
(def app
(ring/ring-handler
(ring/router
[["/api"
["/ping" handler]
["/admin" {::roles #{:admin}}
["/ping" handler]]]]
{:meta {:middleware [wrap-enforce-roles]}})))
```
Anonymous access to public route:
```clj
(app {:request-method :get, :uri "/api/ping"})
; {:status 200, :body "ok"}
```
Anonymous access to guarded route:
```clj
(app {:request-method :get, :uri "/api/admin/ping"})
; {:status 403, :body "forbidden"}
```
Authorized access to guarded route:
```clj
(app {:request-method :get, :uri "/api/admin/ping", ::roles #{:admin}})
; {:status 200, :body "ok"}
```

24
doc/routing/route_conflicts.md Normal file
View file

@ -0,0 +1,24 @@
## Route conflicts
Route trees should not have multiple routes that match to a single (request) path. `router` checks the route tree at creation for conflicts and calls a registered `:conflicts` option callback with the found conflicts. Default implementation throws `ex-info` with a descriptive message.
```clj
(reitit/router
[["/ping"]
["/:user-id/orders"]
["/bulk/:bulk-id"]
["/public/*path"]
["/:version/status"]])
; CompilerException clojure.lang.ExceptionInfo: router contains conflicting routes:
;
; /:user-id/orders
; -> /public/*path
; -> /bulk/:bulk-id
;
; /bulk/:bulk-id
; -> /:version/status
;
; /public/*path
; -> /:version/status
;
```

55
doc/routing/route_metadata.md Normal file
View file

@ -0,0 +1,55 @@
# Route meta-data
Routes can have arbitrary meta-data. For nested routes, the meta-data is accumulated from root towards leafs using [meta-merge](https://github.com/weavejester/meta-merge).
A router based on nested route tree:
```clj
(def router
(reitit/router
["/api" {:interceptors [::api]}
["/ping" ::ping]
["/admin" {:roles #{:admin}}
["/users" ::users]
["/db" {:interceptors [::db]
:roles ^:replace #{:db-admin}}
["/:db" {:parameters {:db String}}
["/drop" ::drop-db]
["/stats" ::db-stats]]]]]))
```
Resolved route tree:
```clj
(reitit/routes router)
; [["/api/ping" {:interceptors [::api]
; :name ::ping}]
; ["/api/admin/users" {:interceptors [::api]
; :roles #{:admin}
; :name ::users}]
; ["/api/admin/db/:db/drop" {:interceptors [::api ::db]
; :roles #{:db-admin}
; :parameters {:db String}
; :name ::drop-db}]
; ["/api/admin/db/:db/stats" {:interceptors [::api ::db]
; :roles #{:db-admin}
; :parameters {:db String}
; :name ::db-stats}]]
```
Path-based routing:
```clj
(reitit/match-by-path router "/api/admin/users")
; #Match{:template "/api/admin/users"
; :meta {:interceptors [::api]
; :roles #{:admin}
; :name ::users}
; :result nil
; :params {}
; :path "/api/admin/users"}
```
On match, route meta-data is returned and can interpreted by the application.
Routers also support meta-data compilation enabling things like fast [Ring](https://github.com/ring-clojure/ring) or [Pedestal](http://pedestal.io/) -style handlers. Compilation results are found under `:result` in the match. See [configuring routers](#configuring-routers) for details.

48
doc/routing/route_syntax.md Normal file
View file

@ -0,0 +1,48 @@
# Route Syntax
Routes are defined as vectors, which String path, optional (non-vector) route argument and optional child routes. Routes can be wrapped in vectors.
Simple route:
```clj
["/ping"]
```
Two routes:
```clj
[["/ping"]
["/pong"]]
```
Routes with meta-data:
```clj
[["/ping" ::ping]
["/pong" {:name ::pong}]]
```
Routes with path and catch-all parameters:
```clj
[["/users/:user-id"]
["/public/*path"]]
```
Nested routes with meta-data:
```clj
["/api"
["/admin" {:middleware [::admin]}
["/user" ::user]
["/db" ::db]
["/ping" ::ping]]
```
Same routes flattened:
```clj
[["/api/admin/user" {:middleware [::admin], :name ::user}
["/api/admin/db" {:middleware [::admin], :name ::db}
["/api/ping" ::ping]]
```

85
doc/routing/routers.md Normal file
View file

@ -0,0 +1,85 @@
# Routers
For routing, a `Router` is needed. Reitit ships with several different router implementations: `:linear-router`, `:lookup-router` and `:mixed-router`, based on the awesome [Pedestal](https://github.com/pedestal/pedestal/tree/master/route) implementation.
`Router` is created with `reitit.core/router`, which takes routes and optional options map as arguments. The route tree gets expanded, optionally coerced and compiled. Actual `Router` implementation is selected automatically but can be defined with a `:router` option. `Router` support both path- and name-based lookups.
Creating a router:
```clj
(require '[reitit.core :as reitit])
(def router
(reitit/router
[["/api"
["/ping" ::ping]
["/user/:id" ::user]]]))
```
`:mixed-router` is created (both static & wild routes are found):
```clj
(reitit/router-name router)
; :mixed-router
```
The expanded routes:
```clj
(reitit/routes router)
; [["/api/ping" {:name :user/ping}]
; ["/api/user/:id" {:name :user/user}]]
```
Route names:
```clj
(reitit/route-names router)
; [:user/ping :user/user]
```
### Path-based routing
```clj
(reitit/match-by-path router "/hello")
; nil
(reitit/match-by-path router "/api/user/1")
; #Match{:template "/api/user/:id"
; :meta {:name :user/user}
; :path "/api/user/1"
; :result nil
; :params {:id "1"}}
```
### Name-based (reverse) routing
```clj
(reitit/match-by-name router ::user)
; #PartialMatch{:template "/api/user/:id",
; :meta {:name :user/user},
; :result nil,
; :params nil,
; :required #{:id}}
(reitit/partial-match? (reitit/match-by-name router ::user))
; true
```
Only a partial match. Let's provide the path-parameters:
```clj
(reitit/match-by-name router ::user {:id "1"})
; #Match{:template "/api/user/:id"
; :meta {:name :user/user}
; :path "/api/user/1"
; :result nil
; :params {:id "1"}}
```
There is also a exception throwing version:
```clj
(reitit/match-by-name! router ::user)
; ExceptionInfo missing path-params for route /api/user/:id: #{:id}
```

92
doc/validating.md Normal file
View file

@ -0,0 +1,92 @@
# Validating route-trees
Namespace `reitit.spec` contains [specs](https://clojure.org/about/spec) for routes, router and router options.
To enable spec-validation of `router` inputs & outputs at development time, one can do the following:
```clj
; add to dependencies:
; [expound "0.3.0"]
(require '[clojure.spec.test.alpha :as st])
(require '[expound.alpha :as expound])
(require '[clojure.spec.alpha :as s])
(require '[reitit.spec])
(st/instrument `reitit/router)
(set! s/*explain-out* expound/printer)
(reitit/router
["/api"
["/publuc"
["/ping"]
["pong"]]])
; -- Spec failed --------------------
;
; ["/api" ...]
; ^^^^^^
;
; should satisfy
;
; (clojure.spec.alpha/cat
; :path
; :reitit.spec/path
; :arg
; (clojure.spec.alpha/? :reitit.spec/arg)
; :childs
; (clojure.spec.alpha/* (clojure.spec.alpha/and :reitit.spec/raw-route)))
;
; -- Relevant specs -------
;
; :reitit.spec/raw-route:
; (clojure.spec.alpha/cat
; :path
; :reitit.spec/path
; :arg
; (clojure.spec.alpha/? :reitit.spec/arg)
; :childs
; (clojure.spec.alpha/* (clojure.spec.alpha/and :reitit.spec/raw-route)))
; :reitit.spec/raw-routes:
; (clojure.spec.alpha/or
; :route
; :reitit.spec/raw-route
; :routes
; (clojure.spec.alpha/coll-of :reitit.spec/raw-route :into []))
;
; -- Spec failed --------------------
;
; [... [... ... ["pong"]]]
; ^^^^^^
;
; should satisfy
;
; (fn [%] (clojure.string/starts-with? % "/"))
;
; -- Relevant specs -------
;
; :reitit.spec/path:
; (clojure.spec.alpha/and
; clojure.core/string?
; (clojure.core/fn [%] (clojure.string/starts-with? % "/")))
; :reitit.spec/raw-route:
; (clojure.spec.alpha/cat
; :path
; :reitit.spec/path
; :arg
; (clojure.spec.alpha/? :reitit.spec/arg)
; :childs
; (clojure.spec.alpha/* (clojure.spec.alpha/and :reitit.spec/raw-route)))
; :reitit.spec/raw-routes:
; (clojure.spec.alpha/or
; :route
; :reitit.spec/raw-route
; :routes
; (clojure.spec.alpha/coll-of :reitit.spec/raw-route :into []))
;
; -------------------------
; Detected 2 errors
```
### Validating meta-data
*TODO*