Documentation

Nitrogen API

AJAX Updates

TargetID can refer to the name of an element, or it can be a JQuery selector path. See Element Paths for more information.

For every one of these functions which offer a Priority argument, Priority can be either the atom eager, normal, or defer. eager will always execute before normal, which will always execute before defer.

Any wiring which does not explicitly specify a priority will wire at normal priority.

  • wf:set(TargetID, Value) -> ok

  • wf:set(Priority, TargetID, Value) -> ok

    Update a form element (textbox, dropdown, checkbox, etc) to set text value of TargetID. Can also be used to set the src attribute of an image tag.

  • wf:enable(Target) -> ok

  • wf:enable(Priority, Target) -> ok

    Enables the target form field or button.

  • wf:disable(Target) ->

  • wf:disable(Priority, Target) ->

    Disables the target form field or button.

  • wf:update(TargetID, Elements) -> ok

  • wf:update(Priority, TargetID, Elements) -> ok

    Update the contents of TargetID with a new set of Nitrogen Elements.

  • wf:replace(TargetID, Elements) -> ok

  • wf:replace(Priority, TargetID, Elements) -> ok

    Replace TargetID with a new set of Nitrogen Elements.

  • wf:remove(TargetID) -> ok

  • wf:remove(Priority, TargetID) -> ok

    Remove TargetID from the DOM.

  • wf:insert_top(TargetID, Elements) -> ok

  • wf:insert_top(Priority, TargetID, Elements) -> ok

    Insert Nitrogen Elements at the top of TargetID (within the contents of TargetID), shifting the existing contents downward.

  • wf:insert_bottom(TargetID, Elements) -> ok

  • wf:insert_bottom(Priority, TargetID, Elements) -> ok

    Insert Nitrogen Elements at the bottom of the TargetID (within the contents of TargetID), below the existing contents.

  • wf:insert_before(TargetID, Elements) -> ok

  • wf:insert_before(Priority, TargetID, Elements) -> ok

    Insert Nitrogen Elements before TargetID in the DOM, shifting the existing contents downward.

  • wf:insert_after(TargetID, Elements) -> ok

  • wf:insert_after(Priority, TargetID, Elements) -> ok

    Insert Nitrogen Elements after TargetID in the DOM, below the existing contents.

  • wf:flash(Elements) -> ok

    Insert the Nitrogen Elements as a new flash message.

Convenience Functions

  • wf:f(Format, Data) -> String or Binary

    Convenience function to format a string similar to io_lib:format(Format, Data). Returns a flattened list.

    Note, if Format is a binary, the return type will be a binary, and if Format is a string, the return type will be a string.

  • wf:coalesce([List]) -> Item

    Return the first element in the list that is not 'undefined'.

  • wf:is_string(Term) -> Bool

    Return true if the Term is an Erlang string. That is, a flat list of integers.

  • wf:to_list(Term) -> List

    Convert the supplied term to a flat list, if possible. Useful for turning Integers, Atoms, Binaries into Strings.

  • wf:to_atom(Term) -> Atom

    Convert the supplied term into an Atom, if possible. Useful for turning Integers, Binaries, and Strings into Atoms.

  • wf:to_existing_atom(Term) -> Atom

    Convert the supplied term to an Atom that already exists in the Erlang VM. Same as wf:to_atom/1 but safer when in doubt. (see the advanced section of the Erlang docs pertaining to atoms).

  • wf:to_binary(Term) -> Binary

    Convert the supplied term into a Binary, if possible. Useful for turning Integers, Atoms, and Strings into Binaries.

  • wf:to_integer(Term) -> Integer

    Convert the supplied term into an Integer, if possible. Useful for turning Atoms, Strings, and Binaries into Integers.

  • wf:html_encode(String) -> EncodedString

    HTML encode the supplied String, converting things like < and > into < and >.

  • wf:html_decode(String) -> DecodedString

    HTML decoding decode the supplied String, converting things like < and > into < and >.

  • wf:url_encode(String) -> EncodedString

    URL encode the supplied String, converting potentially URL-breaking characters into percent notation (%XX).

  • wf:url_decode(String) -> DecodedString

    URL decode the supplied String, converting a percent-encoded String into a normal String.

  • wf:to_qs(PropList) -> EncodedQueryString

    Encode a PropList (a list of {Key, Value} tuples) into a URL-encoded query string (returned as an IOList).

  • wf:hex_encode(String) -> EncodedString.

    Hex-encode the supplied String.

  • wf:hex_decode(String) -> DecodedString

    Convert a hex-encoded String to a normal String.

  • wf:temp_id() -> String

    Return a temp id. Useful for naming an Element so that you can refer to it during a postback later, without giving it a specific name.

  • wf:js_escape(String) -> EscapedString

    Convert a String to a JS-safe string by adding backslashes to quotes and newlines.

  • wf:json_encode(Data) -> JsonEncodedString

    Convert a data structure (Proplist, list, etc) to a JSON encoded string.

  • wf:json_decode(JsonEncodedString) ->

    Convert a JSON-encoded string back to an Erlang data structure.

  • wf:join([Terms],Delimiter) -> [Terms]

    Because Erlang doesn't provide a means to join (or intersperse) a list in a non-string fashion, this will join the Terms on the delimiter, regardless of the type of Delimiter.

   wf:join([Line1,Line2,Line2],#br{}).

Event Wiring

  • wf:wire(Actions) -> ok

    Wire actions to the page. The Actions are applied against the entire page unless a trigger or target are specified within the action itself.

    For example, show a Javascript alert:

   wf:wire(#alert { text="Hello, World!" })
  • wf:wire(TargetID, Actions) -> ok

    Wire actions to the page, targeted against supplied TargetID. For example, hide a Panel:

   wf:wire(PanelID, #hide {})
  • wf:wire(TriggerID, TargetID, Actions) -> ok

    Wire actions to the page, triggering on the supplied TriggerID and targeting against the supplied TargetID. This allows you to wire actions (such as #event) that listen to a click on one element and modify a different element.

    For example, when a button is clicked, hide a panel:

   wf:wire(ButtonID, PanelID, #event { type=click, actions=#hide {} })
  • wf:eager(Actions) -> ok

  • wf:eager(TargetID, Actions) -> ok

  • wf:eager(TriggerID, TargetID, Actions) -> ok

    Wire actions to the page, however, execute the actions before any normal or deferred priority wired actions.

  • wf:defer(Actions) -> ok

  • wf:defer(TargetID, Actions) -> ok

  • wf:defer(TriggerID, TargetID, Actions) -> ok

    Wire actions to the page, however, execute these actions after any normal or eager priority wired actions.

  • wf:continue(Tag, Function, IntervalInMS, TimeoutInMS) -> ok

    Spawn the provided function (arity 0) and tell the browser to poll for the results at the specified interval, with a timeout setting. so See continuations example for usage.

Asynchronous Page Updates (Comet, Continuations)

  • wf:comet(Function) -> {ok, Pid}

    Spawn a comet function, and tell the browser to open a COMET request to receive the results in real time. See example 1, example 2, and example 3 for usage.

  • wf:comet(Function, LocalPool) -> {ok, Pid}

    Spawn a function connected to the specified local pool.

  • wf:comet_global(Function, GlobalPool) -> {ok, Pid}

    Spawn a function connected to the specified global pool.

  • wf:send(LocalPool, Message)

    Send the specified message to all comet functions connected to the specified local pool.

  • wf:send_global(GlobalPool, Message)

    Send the specified message to all comet function connected to the specified GlobalPool.

  • wf:flush() -> ok

    Normally, the results of a comet function are sent to the browser when the function exits. flush/0 pushes results to the browser immediately, useful for a looping comet function.

  • wf:async_mode() -> comet | {poll, Interval}

    Return the current async mode, either comet or {poll, IntervalInMS}.

  • wf:async_mode(Mode)

    Set the current async mode, either comet or {poll, IntervalInMS}.

  • wf:switch_to_comet()

    Run all current and future async processes in comet mode. This uses more resources on the server, as HTTP connections stay open.

  • wf:switch_to_polling(IntervalInMS)

    Run all current and future async processes in polling mode. This uses more resources on the client, as the application must issue a request every IntervalInMS milliseconds.

  • wf:continue(Tag, Function) -> ok

    Spawn the provided function (arity 0) and tell the browser to poll for the results. See continuations example for usage.

  • wf:continue(Tag, Function, Interval) -> ok

    Spawn the provided function (arity 0) and tell the browser to poll for the results at the specified interval. See continuations example for usage.

Comet notes

The comets functionality considers some internal messages other than the ones delivered by wf:send/2 and wf:send_global/2. These messages are:

  • 'INIT'

    The init message is sent to the first process in a comet pool.

  • {'JOIN', Pid}

    This message is sent to already existing comets when a new process joins to the pool

  • {'LEAVE', Pid}

    This message is triggered when certain comet process terminates and it is delivered to all other processes in the pool

    Optionally you can detect when the user leaves the page with a comet by trapping its exit signal.


comet_function() ->
  process_flag(trap_exit, true),
  receive
    {'EXIT', _, Message} -> 
        ?PRINT(Message),
        io:format("The user has left the page.~n")
  end.

Redirect

  • wf:redirect(URL) -> ok

    Redirect to the provided URL.

  • wf:redirect_to_login(URL) -> ok

    See Below.

  • wf:redirect_to_login(URL, PostLoginURL)= -> ok

    Redirect to the provided URL, attaching a token on the end. The receiving page can subsequently call wf:redirect_from_login(DefaultURL) to send the user to PostLoginURL.

    If PostLoginURL is not provided, it will default to the current page's URL.

  • wf:redirect_from_login(DefaultURL) -> ok

    Redirect the user back to a page that called wf:redirect_to_login(URL) or to the PostLoginURL provided to wf:redirect_to_login(URL, PostLoginURL) If the user came to the page for some other reason, then the user is redirected to the provided DefaultURL.

Session State

  • wf:session(Key) -> Value or 'undefined'

    Retrieve the session value stored under the specified key. For example, retrieve the value of 'count' for the current user: Count wf:session(count)=

  • wf:session_default(Key, DefaultValue) -> Value.

    Retrieve the session value stored under a specific key. If not found, return the supplied default value.

  • wf:session(Key, Value) -> OldValue

    Store a session variable for the current user. Key and Value can be any Erlang term. For example, store a count: wf:session(count, Count)

    Returns the previous value associated with Key.

  • wf:clear_session() -> ok

    Clear the current user's session.

  • wf:logout() -> ok

    Clear session state, page state, identity, and roles.

Page State

  • wf:state(Key) -> Value

    Retrieve a page state value stored under the specified key. Page State is different from Session State in that Page State is scoped to a series of requests by one user to one Nitrogen Page.

  • wf:state_default(Key, DefaultValue) -> Value.

    Retrieve a page state value stored under the specified key. If the value is not set, then return the supplied default value.

  • wf:state(Key, Value) -> ok

    Store a page state variable for the current user. Page State is different from Session State in that Page State is scoped to a series of requests by one user to one Nitrogen Page.

  • wf:clear_state() -> ok

    Clear a user's page state.

Authentication and Authorization

  • wf:user() -> User or 'undefined'

    Return the user value that was previously set by wf:user(User)

  • wf:user(User) -> ok

    Set the user for the current session.

  • wf:clear_user() -> ok

    Same as wf:user(undefined).

  • wf:role(Role) -> boolean

    Check if the current user has a specified role.

  • wf:role(Role, IsInRole) -> ok

    Set whether the current user is in a specified role. IsInRole should be a boolean (true or false)

  • wf:clear_roles() -> ok

    Remove the user from all roles.

Web Request and Response

  • wf:in_request() -> boolean()

    Checks if the current running process is actually inside a Nitrogen request, that is to say, that it has a Nitrogen-initialized context. Returns true if the current process is a Nitrogen request, and false otherwise.

    This will return true in initial requests, postback requests, and comet processes.

  • wf:q(AtomKey) -> String | undefined

    Get all query string and POST values for the provided key. If more than one AtomKey matches, then this will throw an error, use wf:qs(AtomKey) instead. Returns the atom undefined if there's no matching value.

    Mnemonic: Think q as "Query".

  • wf:mq(ListOfAtomKeys) -> [ListOfStrings]

    Get the list of query string and POST values for the provided keys, and return the list of values for the keys. Syntactical sugar equivilant of:

   [wf:q(AtomKey) || AtomKey <- ListOfAtomKeys]

Mnemonic: Think mq as "Multi Query".

  • wf:q_pl(ListOfAtomKeys) -> [{Key,Values},...]

    Takes a list of keys returns a proplist of keys and respective values from query string and POST values.

    Mnemonic: Think q_pl as "Query Proplist"

Example:

  wf:q_pl([favorite_robot,favorite_dinosaur,favorite_hobbit])

Returns something like:

  [
      {favorite_robot,"Optimus Prime"},
      {favorite_dinosaur,"Velociraptor"},
      {favorite_hobbit,"Samwise"}
  ].
  • wf:qs(AtomKey) -> [String]

    Get a list of query string and POST values for the provided key. (This acts like wf:q(AtomKey) in Nitrogen 1.0.)

    Mnemonic: Think qs as "Query Plural"

  • wf:mqs(ListOfAtomKeys) -> [ListOfStrings]

    Get a list of query string and POST values for the provided list of keys. Syntactical sugar equivilant of:

   [wf:qs(AtomKey) || AtomKey <- ListOfAtomKeys]

Mnemonic: Think mqs as "Multi Query Plural"

  • wf:qs_pl(ListOfAtomKeys) -> [{Key,ListOfValues},...]

    Takes a list of keys and returns a proplist of keys and respective list of values from the query string and POST values.

    Mnemonic: Think qs_pl as "Query Plurals into a Proplist"

Example:

  wf:qs_pl([fruit,veggie,meat])

Returns something like:

  [
     {fruit,["Apple","Peach"]},
     {veggie,["Broccoli"]},
     {meat,["Pork","Beef","Venison"]}
  ].
  • wf:request_body() -> String

    Return the complete text of the request body to the server. Note, this value will use the context of the current request. For example, the result of calling this during the page's initial request will be different than calling it within a postback event.

  • wf:encoding(Encoding) -> ok

    Set the encoding to be applied just before sending to the client. Valid values are:

    • none :: No encoding. Send the response as an iolist of bytes. This is what you would want to use if you were returning binary data (like if you were generating an image file on the fly).
    • unicode :: This runs the return value through Erlang's unicocde:characters_to_binary/1 before sending to the client.
    • Function/1 :: Run the response through this particular function before sending, such as if you have a custom encoding you wish to use.
    • {Module, Function} :: Run this through Module:Function/1. This is merely an alternative to the previous option, but one that is configuration-file friendly.
    • auto (Default) :: Reads the specified response header to attempt to determine which encoding should be provided. It's quite naive, however, in the name of speed. If the header starts with "text/" or is set to "application/json" or "application/javascript", it will use unicode, otherwise, it will use none. For example, if your page you had set wf:content_type("image/png"), it Encoding would apply as none.
  • wf:encoding() -> Encoding

    Return the current encoding that will be applied to the response. It's worth noting that this will return auto if it's currently set to such.

  • wf:status_code(IntegerCode) -> ok

    Set the HTTP response code. Default is 200.

  • wf:content_type(ContentType) -> ok

    Set the HTTP content type. Defaults is "text/html". This can be used to return text images or other files to the browser, rather than returning HTML.

  • wf:download_as(Filename) -> ok

    Set the HTTP Content-Disposition header in such a way that the browser will treat the page as a download with the specified Filename.

  • wf:request_method() -> atom()

    Return the HTTP request method used (POST, GET, PUT, etc), as a lower-case atom (example: post, get, and put).

  • wf:url() -> String

    Return the complete requested URL for the page. For exaple, if the request is for http://myapp.com/some/file?some_qs, the return is ="http://myapp.com/some/file?some_qs".

  • wf:uri() -> String

    Return the complete requested URI for the page. Note, this does not include the protocol (http/https) or the hostname. For example, if the request is for http://myapp.com/some/file?some_qs, the return is "/some/file?some_qs").

  • wf:path() -> String

    Return the string of the requested path. This excludes any query_string information. For example, if the request is for http://myapp.com/some/file?some_qs the return value is "/some/file"

  • wf:path_info() -> String

    Return only the path info for the requested page module. In other words, if the module web_my_page is requsted with the path "/web/my/page/some/extra/stuff then wf:path_info() would return "some/extra/stuff".

  • wf:page_module() -> Atom

    Return the requested page module. Useful information to know when writing a custom element or action.

  • wf:page_module(Module) -> ok

    Changes the current page module for the rendering engine. This will likely only need to be set when you need to set an otherwise valid page that happens to be to, for some reason, need to return a specific error. For example, maybe you have a view_profile module that expects a userid on the query-string, but the provided userid is invalid, so you want the page to throw a 404 instead.

  • wf:protocol() -> http | https

    Return whether the request to the server is made by HTTP or HTTPS. Note, that if the request is made behind a proxy, this will only return whether the request from the proxy to the server is in HTTP or HTTPS, so the protocol seen by Nitrogen might not be the same protocol used by the client.

  • wf:peer_ip() -> IPAddress (4-tuple or 8-tuple)

    Return the IP address of the client. Note that Erlang IP Addresses are 4-tuples (IPv4) or 8-tuples (IPv6). For example: {127,0,0,} or {8193,3512,51966,47806,0,0,0,1}

  • wf:peer_ip(ListOfProxies) -> IPAddress (4-tuple or 8-tuple)

    Shortcut method for wf:peer_ip(ListOfProxies, x_forwarded_for)

  • wf:peer_ip(ListOfProxies, ForwardedHeader) -> IPAddress (4-tuple or 8-tuple)

    This will compare the peer_ip address against the provided list of proxies, and if any of them match the connected IP, then return the IP address from the ForwardedHeader.

Cookies

  • wf:cookies() -> [{AtomKey, StringValue}, ...].

    Return a proplist of all cookies.

  • wf:cookie(Key) -> String

    Get the value of a cookie.

  • wf:cookie_default(Key, Default) -> String.

    Get the value of a cookie, if it doesn't exist, return the default.

  • wf:cookie(Key, Value) -> ok

    Tell Nitrogen to set a cookie on the browser. Uses "/" for the Path, and Nitrogen's session timeout setting for the MinutesToLive value.

  • wf:cookie(Key, Value, Path, MinutesToLive) -> ok

    Tell Nitrogen to set a cookie on the browser under the specified Path that is valid for a certain number of minutes.

  • wf:delete_cookie(Key) -> ok

    Tell Nitrogen to set the cookie to expire immediately, effectively deleting it from the browser. Is a shortcut for wf:cookie(Key,"","/",0).

HTTP Headers

  • wf:headers() -> [{AtomKey, StringValue}, ...]

    Return a proplist of all HTTP headers.

  • wf:header(AtomKey) -> Value

    Get the value of an HTTP header.

  • wf:header_default(AtomKey, Default) -> Value.

    Get the value of an HTTP header, if it doesn't exist, return the default.

  • wf:header(StringKey, HeaderValue) -> ok

    Set an HTTP header during the next response.

Serialization

  • wf:pickle(Term) -> PickledBinary

    Serialize a term into a web-safe, checksummed, and AES-encrypted hex string.

  • wf:depickle(PickledBinary) -> Term

    Turn a pickled binary back into the original term.

  • wf:depickle(PickledBinary, SecondsToLive) -> Term or 'undefined'

    Turn a pickled binary back into the original term, checking to see if the term was pickled more than SecondsToLive second ago. Returns the Term if it is still 'fresh' or the atom 'undefined'.

Logging

  • wf:info(String)

    Log an informational message.

  • wf:info(Format, Args)

    Log an informational message.

  • wf:warning(String)

    Log a warning message.

  • wf:warning(Format, Args)

    Log a warning message.

  • wf:error(String)

    Log an error message.

  • wf:error(Format, Args)

    Log and error message.

  • wf:console_log(Terms)

    Send a console.log() to the browser with Terms as the value. (See Console Log Action)

Configuration

  • wf:config(Key) -> Term

    Get the Nitrogen configuration setting under the specified Key.

  • wf:config_default(Key, DefaultValue) -> Term

    Get the Nitrogen configuration setting under the specified Key. If not set, then return DefaultValue.



View Source or Submit Corrections for this Documentation Page
Copyright © 2008-2024 Nitrogen Web Framework. Released under the MIT License.