Does each Route include its Request and the Response returned by the Server?

For each dir.. , it is a Route which takes in a Request and returns a Response created by the Server. I am wondering if the Request and the Response included in each line of the code. Thanks in advance!

simpleHTTP serverConf $ do
decodeBody (defaultBodyPolicy "/tmp/" 4096 4096 4096)
msum [ do
      nullDir
      seeOther "graph" (toResponse "Redirecting to /graph"),
      dir "grid" gridResponse,
      dir "graph" graphResponse,
      dir "image" graphImageResponse,
      dir "timetable-image" $ look "courses" >>= \x -> look "session" >>= timetableImageResponse x,
      dir "graph-fb" $ seeOther redirectUrlGraphEmail $ toResponse "",
      dir "post-fb" $ seeOther redirectUrlGraphPost $ toResponse "",
      dir "test" $ look "code" >>= getEmail,
      dir "test-post" $ look "code" >>= postToFacebook,
      dir "post" postResponse,
      dir "draw" drawResponse,
      dir "about" $ aboutResponse aboutContents,
      dir "privacy" $ privacyResponse privacyContents,
      dir "static" $ serveDirectory DisableBrowsing [] staticDir,
      dir "course" $ look "name" >>= retrieveCourse,
      dir "all-courses" $ liftIO allCourses,
      dir "graphs" $ liftIO queryGraphs,
      dir "course-info" $ look "dept" >>= courseInfo,
      dir "depts" $ liftIO deptList,
      dir "timesearch" searchResponse,
      dir "calendar" $ lookCookieValue "selected-lectures" >>= calendarResponse,
      dir "get-json-data" $ look "graphName" >>= \graphName -> liftIO $ getGraphJSON graphName,
      dir "loading" $ look "size" >>= loadingResponse,
      dir "save-json" $ look "jsonData" >>= \jsonStr -> look "nameData" >>= \nameStr -> liftIO $ saveGraphJSON jsonStr nameStr,
      notFoundResponse
]

Solution

Each line is a handler. Think of a handler as a function which takes a request and returns a response.

graphReponse is a handler which handles the request in a certain way.

dir modifies the handler so that it doesn't get called unless the request url begins a certain way.

There is nothing special about having the word Response in the name of the handlers. graphResponse is just the name of a function - you can use any names for your handler functions.

What you have is a list of handlers, e.g.:

[ handler1,
  handler2,
  ...,
  notFoundResponse
]

When a request comes in a check is made to see if the handler1 will handle the request. If not, handler2 is checked and so on. If no handler accepts the request the notFoundHandler is called which will probably generated a 404 page.

That's why the dir ... part is important - it prevents a handler from responding unless the url begins a certain way.