gen_event(3erl)

📦

version 25.3.2.8 · 5 versions

💖 NAME 💖

okay bestie, welcome to gen_event(3erl) — the Generic event handling behavior 💅✨ this is the Erlang manual page for it, and we're gonna get through it together.

✨ DESCRIPTION ✨

So gen_event is this behavior module that gives you event handling functionality, and it's literally made of a generic event manager process plus any number of event handlers that get added and deleted dynamically 🦋

An event manager built with this module has a standard set of interface functions, includes tracing and error reporting functionality, and fits into an OTP supervision tree. For more info, see OTP Design Principles. 👑

Each event handler is implemented as a callback module that exports a predefined set of functions. The relationship between the behavior functions and the callback functions is giving:

gen_event module                   Callback module
----------------                   ---------------
gen_event:start
gen_event:start_monitor
gen_event:start_link       ----->  -

gen_event:add_handler
gen_event:add_sup_handler  ----->  Module:init/1

gen_event:notify
gen_event:sync_notify      ----->  Module:handle_event/2

gen_event:send_request
gen_event:call             ----->  Module:handle_call/2

-                          ----->  Module:handle_info/2

gen_event:delete_handler   ----->  Module:terminate/2

gen_event:swap_handler
gen_event:swap_sup_handler ----->  Module1:terminate/2
                                   Module2:init/1

gen_event:which_handlers   ----->  -

gen_event:stop             ----->  Module:terminate/2

-                          ----->  Module:code_change/3

Since each event handler is one callback module, an event manager has many callback modules that are added and deleted dynamically. That means gen_event is more tolerant of callback module errors than the other behaviors. If a callback function for an installed event handler fails with Reason, or returns a bad value Term, the event manager does not fail. It deletes the event handler by calling callback function Module:terminate/2, giving as argument {error,{'EXIT',Reason}} or {error,Term}, respectively. No other event handler is affected. 💀✨

A gen_event process handles system messages as described in sys(3erl). You can use the sys module for debugging an event manager.

Heads up bestie: an event manager does trap exit signals automatically. 👀

The gen_event process can go into hibernation (see erlang:hibernate/3) if a callback function in a handler module specifies hibernate in its return value. This can be useful if the server is expected to be idle for a long time. But use this feature with care, because hibernation implies at least two garbage collections (when hibernating and shortly after waking up) and is not something you want to do between each event handled by a busy event manager. 😭

Notice that when multiple event handlers are invoked, it is sufficient that one single event handler returns a hibernate request for the whole event manager to go into hibernation. 🧚

Unless otherwise stated, all functions in this module fail if the specified event manager does not exist or if bad arguments are specified. periodt.

🧚 DATA TYPES 🧚

okay, type time, stay with me 🎀

handler() = atom() | {atom(), term()}

handler_args() = term()

add_handler_ret() = ok | term() | {'EXIT', term()}

del_handler_ret() = ok | term() | {'EXIT', term()}

emgr_ref() =
    atom() |
    {atom(), atom()} |
    {global, term()} |
    {via, atom(), term()} |
    pid()

`request_id()`

An opaque request identifier. See send_request/3 for details. 🫶

`request_id_collection()`

An opaque collection of request identifiers (request_id()) where each request identifier can be associated with a label chosen by the user. For more information see reqids_new/0. ⭐

response_timeout() = timeout() | {abs, integer()}

Used to set a time limit on how long to wait for a response using either receive_response/2, receive_response/3, wait_response/2, or wait_response/3. The time unit used is millisecond. Currently valid values:

0..4294967295 — Timeout relative to current time in milliseconds.
infinity — Infinite timeout. That is, the operation will never time out.
{abs, Timeout} — An absolute Erlang monotonic time timeout in milliseconds. That is, the operation will time out when erlang:monotonic_time(millisecond) returns a value larger than or equal to Timeout. Timeout is not allowed to identify a time further into the future than 4294967295 milliseconds. Identifying the timeout using an absolute timeout value is especially handy when you have a deadline for responses corresponding to a complete collection of requests (request_id_collection()), since you do not have to recalculate the relative time until the deadline over and over again. 💅

format_status() =
    #{state => term(),
      message => term(),
      reason => term(),
      log => [sys:system_event()]}

A map that describes the gen_event process status. The keys are:

state — The internal state of the event handler.
message — The message that caused the event handler to terminate.
reason — The reason that caused the event handler to terminate.
log — The sys log of the server.

New associations may be added into the status map without prior notice. 👑

🚀 EXPORTS 🚀

`add_handler(EventMgrRef, Handler, Args) -> Result`

Types:

EventMgrRef = Name | {Name,Node} | {global,GlobalName} |
              {via,Module,ViaName} | pid()
 Name = Node = atom()
 GlobalName = ViaName = term()
Handler = Module | {Module,Id}
 Module = atom()
 Id = term()
Args = term()
Result = ok | {'EXIT',Reason} | term()
 Reason = term()

Adds a new event handler to event manager EventMgrRef. The event manager calls Module:init/1 to initiate the event handler and its internal state. 🌸

EventMgrRef can be any of the following:

The pid
Name, if the event manager is locally registered
{Name,Node}, if the event manager is locally registered at another node
{global,GlobalName}, if the event manager is globally registered
{via,Module,ViaName}, if the event manager is registered through an alternative process registry

Handler is the name of the callback module Module or a tuple {Module,Id}, where Id is any term. The {Module,Id} representation makes it possible to identify a specific event handler when many event handlers use the same callback module. 🎀

Args is any term that is passed as the argument to Module:init/1.

If Module:init/1 returns a correct value indicating successful completion, the event manager adds the event handler and this function returns ok. If Module:init/1 fails with Reason or returns {error,Reason}, the event handler is ignored and this function returns {'EXIT',Reason} or {error,Reason}, respectively. 😭

`add_sup_handler(EventMgrRef, Handler, Args) -> Result`

Types:

EventMgrRef = Name | {Name,Node} | {global,GlobalName} |
              {via,Module,ViaName} | pid()
 Name = Node = atom()
 GlobalName = ViaName = term()
Handler = Module | {Module,Id}
 Module = atom()
 Id = term()
Args = term()
Result = ok | {'EXIT',Reason} | term()
 Reason = term()

Adds a new event handler in the same way as add_handler/3, but also supervises the connection by linking the event handler and the calling process. 🫶

If the calling process later terminates with Reason, the event manager deletes any supervised event handlers by calling Module:terminate/2, then calls Module:handle_info/2 for each remaining handler.
If the event handler is deleted later, the event manager sends a message {gen_event_EXIT,Handler,Reason} to the calling process. Reason is one of the following:
- normal, if the event handler has been removed because of a call to delete_handler/3, or remove_handler has been returned by a callback function (see below).
- shutdown, if the event handler has been removed because the event manager is terminating.
- {swapped,NewHandler,Pid}, if the process Pid has replaced the event handler with another event handler NewHandler using a call to swap_handler/3 or swap_sup_handler/3.
- A term, if the event handler is removed because of an error. Which term depends on the error.

For a description of the arguments and return values, see add_handler/3. 💖

`call(EventMgrRef, Handler, Request) -> Result`

`call(EventMgrRef, Handler, Request, Timeout) -> Result`

Types:

EventMgrRef = Name | {Name,Node} | {global,GlobalName} |
              {via,Module,ViaName} | pid()
 Name = Node = atom()
 GlobalName = ViaName = term()
Handler = Module | {Module,Id}
 Module = atom()
 Id = term()
Request = term()
Timeout = int()>0 | infinity
Result = Reply | {error,Error}
 Reply = term()
 Error = bad_module | {'EXIT',Reason} | term()
 Reason = term()

Makes a synchronous call to event handler Handler installed in event manager EventMgrRef by sending a request and waiting until a reply arrives or a time-out occurs. The event manager calls Module:handle_call/2 to handle the request. ⭐

For a description of EventMgrRef and Handler, see add_handler/3.

Request is any term that is passed as one of the arguments to Module:handle_call/2.

Timeout is an integer greater than zero that specifies how many milliseconds to wait for a reply, or the atom infinity to wait indefinitely. Defaults to 5000. If no reply is received within the specified time, the function call fails. 🔥

Okay so the Reply you actually get back is whatever Module:handle_call/2 returned. 🫶 If the event handler you wanted isn’t installed, it’s giving {error,bad_module}. If the callback fails with Reason or returns some unexpected value Term, you get {error,{'EXIT',Reason}} or {error,Term} respectively. And when this call fails, it exits the calling process — real main-character consequences. The exit term is {Reason, Location} where Location = {gen_event,call,ArgList}. For what Reason can be, see gen_server:call/3 — it has the relevant values laid out.

💅 `check_response(Msg, ReqId) -> Result`

Types:

Msg = term()
ReqId = request_id()
Response =
    {reply, Reply :: term()} |
    {error, {Reason :: term(), emgr_ref()}}
Result = Response | no_reply

Check if Msg is a response corresponding to the request identifier ReqId. The request must have been made by send_request/3.

If Msg is a response corresponding to ReqId, you get that response back. Otherwise you get no_reply, and no cleanup is done — so the function has to be invoked repeatedly until a response is returned. Lowkey important, bestie.

If the specified event handler is not installed, the function returns {error,bad_module}. If the callback function fails with Reason or returns an unexpected value Term, this function returns {error,{'EXIT',Reason}} or {error,Term}, respectively. If the event manager dies before or during the request, this function returns {error,{Reason, EventMgrRef}}.

🎀 `check_response(Msg, ReqIdCollection, Delete) -> Result`

Types:

Msg = term()
ReqIdCollection = request_id_collection()
Delete = boolean()
Response =
    {reply, Reply :: term()} |
    {error, {Reason :: term(), emgr_ref()}}
Result =
    {Response,
     Label :: term(),
     NewReqIdCollection :: request_id_collection()} |
    no_request | no_reply

Check if Msg is a response corresponding to a request identifier saved in ReqIdCollection. All request identifiers of ReqIdCollection must correspond to requests that have been made using send_request/3 or send_request/5, and all those requests must have been made by the process calling this function.

The Label in the response equals the Label associated with the request identifier that the response corresponds to. A request identifier’s Label gets associated when saving the request id in a request identifier collection, or when sending the request using send_request/5.

Compared to check_response/2, the returned result associated with a specific request identifier or an exception associated with a specific request identifier will be wrapped in a 3-tuple. The first element of this tuple equals the value that would have been produced by check_response/2, the second element equals the Label associated with the specific request identifier, and the third element NewReqIdCollection is a possibly modified request identifier collection.

If ReqIdCollection is empty, the atom no_request will be returned. If Msg does not correspond to any of the request identifiers in ReqIdCollection, the atom no_reply is returned.

If Delete equals true, the association with Label will have been deleted from ReqIdCollection in the resulting NewReqIdCollection. If Delete equals false, NewReqIdCollection will equal ReqIdCollection. Note that deleting an association is not for free, and a collection containing already handled requests can still be used by subsequent calls to check_response/3, receive_response/3, and wait_response/3. However, without deleting handled associations, those calls will not be able to detect when there are no more outstanding requests to handle, so you’ll have to keep track of this some other way instead of relying on a no_request return. Also note: if you pass a collection only containing associations of already handled or abandoned requests to check_response/3, it will always return no_reply.

🦋 `delete_handler(EventMgrRef, Handler, Args) -> Result`

Types:

EventMgrRef = Name | {Name,Node} | {global,GlobalName} |
{via,Module,ViaName} | pid()
 Name = Node = atom()
 GlobalName = ViaName = term()
Handler = Module | {Module,Id}
 Module = atom()
 Id = term()
Args = term()
Result = term() | {error,module_not_found} | {'EXIT',Reason}
 Reason = term()

Deletes an event handler from event manager EventMgrRef. The event manager calls Module:terminate/2 to terminate the event handler.

For a description of EventMgrRef and Handler, see add_handler/3.

Args is any term that is passed as one of the arguments to Module:terminate/2.

The return value is the return value of Module:terminate/2. If the specified event handler is not installed, the function returns {error,module_not_found}. If the callback function fails with Reason, the function returns {'EXIT',Reason}.

🌈 `notify(EventMgrRef, Event) -> ok` / `sync_notify(EventMgrRef, Event) -> ok`

Types:

EventMgrRef = Name | {Name,Node} | {global,GlobalName} |
{via,Module,ViaName} | pid()
 Name = Node = atom()
 GlobalName = ViaName = term()
Event = term()

Sends an event notification to event manager EventMgrRef. The event manager calls Module:handle_event/2 for each installed event handler to handle the event.

notify/2 is asynchronous and returns immediately after the event notification has been sent. sync_notify/2 is synchronous in the sense that it returns ok after the event has been handled by all event handlers.

For a description of EventMgrRef, see add_handler/3.

Event is any term that is passed as one of the arguments to Module:handle_event/2.

notify/1 does not fail even if the specified event manager does not exist, unless it is specified as Name.

💖 `receive_response(ReqId, Timeout) -> Result`

Types:

ReqId = request_id()
Timeout = response_timeout()
Response =
    {reply, Reply :: term()} |
    {error, {Reason :: term(), emgr_ref()}}
Result = Response | timeout

Receive a response corresponding to the request identifier ReqId. The request must have been made by send_request/3 to the gen_statem process. This function must be called from the same process from which send_request/3 was made.

Timeout specifies how long to wait for a response. If no response is received within the specified time, the function returns timeout. Assuming that the server executes on a node supporting aliases (introduced in OTP 24), the request will also be abandoned. That is, no response will be received after a timeout. Otherwise, a stray response might be received at a later time.

The return value Reply is defined in the return value of Module:handle_call/3.

The difference between wait_response/2 and receive_response/2 is that receive_response/2 abandons the request at timeout so that a potential future response is ignored, while wait_response/2 does not.

🍓 `receive_response(ReqIdCollection, Timeout, Delete) -> Result`

Types:

ReqIdCollection = request_id_collection()
Timeout = response_timeout()
Delete = boolean()
Response =
    {reply, Reply :: term()} |
    {error, {Reason :: term(), emgr_ref()}}
Result =
    {Response,
     Label :: term(),
     NewReqIdCollection :: request_id_collection()} |
    no_request | timeout

Receive a response corresponding to a request identifier saved in ReqIdCollection. All request identifiers of ReqIdCollection must correspond to requests that have been made using send_request/3 or send_request/5, and all those requests must have been made by the process calling this function.

The Label in the response equals the Label associated with the request identifier that the response corresponds to. A request identifier’s Label gets associated when adding the request id in a request identifier collection, or when sending the request using send_request/5.

Compared to receive_response/2, the returned result associated with a specific request identifier will be wrapped in a 3-tuple. The first element of this tuple equals the value that would have been produced by receive_response/2, the second element equals the Label associated with the specific request identifier, and the third element NewReqIdCollection is a possibly modified request identifier collection.

If ReqIdCollection is empty, the atom no_request will be returned.

Timeout specifies how long to wait for a response. If no response is received within the specified time, the function returns timeout. Assuming that the server executes on a node supporting aliases (introduced in OTP 24), all requests identified by ReqIdCollection will also be abandoned. That is, no responses will be received after a timeout. Otherwise, stray responses might be received at a later time.

The difference between receive_response/3 and wait_response/3 is that receive_response/3 abandons the requests at timeout so that potential future responses are ignored, while wait_response/3 does not.

reqids_add/3 vibes check: if Delete equals true, the association with Label is gone from ReqIdCollection in the resulting NewReqIdCollection (bye bestie 👋). If Delete equals false, then NewReqIdCollection just equals ReqIdCollection — no changes, no drama. But heads up: deleting an association is not free, and a collection that contains already-handled requests can still be used by later calls to receive_response/3, check_response/3, and wait_response/3. However — if you never delete handled associations, those calls can't detect when there are no more outstanding requests to handle 😭, so you'll have to track that yourself some other way instead of relying on a no_request return. And the real tea ☕: if you pass a collection that only contains associations of already handled or abandoned requests into receive_response/3, it will always block until a timeout determined by Timeout is triggered. Periodt.

💌 `reqids_add/3`

reqids_add(ReqId :: request_id(),
           Label :: term(),
           ReqIdCollection :: request_id_collection()) ->
              NewReqIdCollection :: request_id_collection()

Saves ReqId and associates a Label with the request identifier by adding this info into ReqIdCollection, then hands you back the resulting request identifier collection. Cute and functional 🎀

reqids_new() -> NewReqIdCollection :: request_id_collection()

Returns a brand new empty request identifier collection, bestie ✨ A request identifier collection is what you use when you gotta handle multiple outstanding requests at once.

Request identifiers from requests made by send_request/3 can be saved into a request identifier collection using reqids_add/3. Later, that collection can be used to get one response corresponding to a request in the collection — just pass the collection as the argument to receive_response/3, wait_response/3, or check_response/3. Iconic.

And reqids_size/1 tells you how many request identifiers are chilling in a request identifier collection 🧚

reqids_size(ReqIdCollection :: request_id_collection()) ->
                integer() >= 0

Returns the amount of request identifiers saved in ReqIdCollection. That's it, that's the whole function 💅

reqids_to_list(ReqIdCollection :: request_id_collection()) ->
                   [{ReqId :: request_id(), Label :: term()}]

Returns a list of {ReqId, Label} tuples matching all the request identifiers and their associated labels present in the ReqIdCollection collection 🌸

📨 `send_request/3`

send_request(EventMgrRef :: emgr_ref(),
             Handler :: handler(),
             Request :: term()) ->
                ReqId :: request_id()

Sends an asynchronous call request Request to event handler Handler installed in the event manager identified by EventMgrRef, and returns a request identifier ReqId. You'll use that ReqId later with receive_response/2, wait_response/2, or check_response/2 to fetch the actual result of the request. And besides passing the request identifier straight into those functions, you can also save it in a request identifier collection with reqids_add/3 — such a collection can later be used to get one response corresponding to a request in the collection by passing the collection as argument to receive_response/3, wait_response/3, or check_response/3. If you're about to save the request identifier in a collection anyway, maybe consider send_request/5 instead 👀

The call

gen_event:receive_response(gen_event:send_request(EventMgrRef, Handler, Request), Timeout)

can be seen as equivalent to gen_event:call(EventMgrRef, Handler, Request, Timeout), ignoring the error handling. Ngl that's kind of a slay equivalence.

The event manager calls Module:handle_call/2 to handle the request.

Request is any term that is passed as one of the arguments to Module:handle_call/3.

📨📨 `send_request/5`

send_request(EventMgrRef :: emgr_ref(),
             Handler :: handler(),
             Request :: term(),
             Label :: term(),
             ReqIdCollection :: request_id_collection()) ->
                NewReqIdCollection :: request_id_collection()

Sends an asynchronous call request Request to event handler Handler installed in the event manager identified by EventMgrRef. The Label gets associated with the request identifier of the operation and added to the returned request identifier collection NewReqIdCollection. Later you can use that collection to get one response corresponding to a request in it — pass the collection as argument to receive_response/3, wait_response/3, or check_response/3 🦋

It's the same as calling

gen_event:reqids_add(gen_event:send_request(EventMgrRef, Handler, Request), Label, ReqIdCollection)

but calling send_request/5 is slightly more efficient. Efficiency queen behavior fr fr 👑

🌱 `start/0,1,2`

start() -> Result
start(EventMgrName | Options) -> Result
start(EventMgrName, Options) -> Result

Types:

EventMgrName = {local,Name} | {global,GlobalName} | {via,Module,ViaName}
Name = atom()
GlobalName = ViaName = term()
Options = [Option]
Option = {debug,Dbgs} | {timeout,Time} | {hibernate_after,HibernateAfterTimeout} | {spawn_opt,SOpts}
Dbgs = [Dbg]
Dbg = trace | log | statistics | {log_to_file,FileName} | {install,{Func,FuncState}}
SOpts = [term()]
Result = {ok,Pid} | {error,{already_started,Pid}}
Pid = pid()

Creates a stand-alone event manager process — meaning an event manager that is not part of a supervision tree and therefore has no supervisor. Independent main character energy 💫

For a description of the arguments and return values, see start_link/0,1.

🌱🔗 `start_link/0,1,2`

start_link() -> Result
start_link(EventMgrName | Options) -> Result
start_link(EventMgrName, Options) -> Result

Types:

EventMgrName = {local,Name} | {global,GlobalName} | {via,Module,ViaName}
Name = atom()
GlobalName = ViaName = term()
Options = [Option]
Option = {debug,Dbgs} | {timeout,Time} | {hibernate_after,HibernateAfterTimeout} | {spawn_opt,SOpts}
Dbgs = [Dbg]
Dbg = trace | log | statistics | {log_to_file,FileName} | {install,{Func,FuncState}}
SOpts = [term()]
Result = {ok,Pid} | {error,{already_started,Pid}}
Pid = pid()

Creates an event manager process as part of a supervision tree 🎀 This function is meant to be called, directly or indirectly, by the supervisor — for example, it makes sure the event manager is linked to the supervisor. Teamwork!

If EventMgrName={local,Name}, the event manager gets registered locally as Name using register/2.
If EventMgrName={global,GlobalName}, the event manager gets registered globally as GlobalName using global:register_name/2. If no name is provided, the event manager is not registered.
If EventMgrName={via,Module,ViaName}, the event manager registers with the registry represented by Module. The Module callback is supposed to export the functions register_name/2, unregister_name/1, whereis_name/1, and send/2, which are to behave like the corresponding functions in global. So {via,global,GlobalName} is a totally valid reference ✨
If the option {hibernate_after,HibernateAfterTimeout} is present, the gen_event process awaits any message for HibernateAfterTimeout milliseconds and, if no message is received, the process goes into hibernation automatically (by calling proc_lib:hibernate/3). Nap time 😴

If the event manager is successfully created, the function returns {ok,Pid}, where Pid is the pid of the event manager. If a process with the specified EventMgrName already exists, the function returns {error,{already_started,Pid}}, where Pid is the pid of that already-existing process. Awkward but okay 🙈

🌱👀 `start_monitor/0,1,2`

start_monitor() -> Result
start_monitor(EventMgrName | Options) -> Result
start_monitor(EventMgrName, Options) -> Result

Types:

EventMgrName = {local,Name} | {global,GlobalName} | {via,Module,ViaName}
Name = atom()
GlobalName = ViaName = term()
Options = [Option]
Option = {debug,Dbgs} | {timeout,Time} | {hibernate_after,HibernateAfterTimeout} | {spawn_opt,SOpts}
Dbgs = [Dbg]
Dbg = trace | log | statistics | {log_to_file,FileName} | {install,{Func,FuncState}}
SOpts = [term()]
Result = {ok,{Pid,Mon}} | {error,{already_started,Pid}}
Pid = pid()

Creates a stand-alone event manager process — so, not part of a supervision tree and thus no supervisor — and atomically sets up a monitor to the newly created process 🫶 Two things at once, multitasking icon.

For a description of the arguments and return values, see start_link/0,1. Just note that the return value on a successful start is different from start_link/3,4: start_monitor/3,4 returns {ok,{Pid,Mon}}, where Pid is the process identifier of the process and Mon is a reference to the monitor set up to monitor that process. If the start is not successful, the caller gets blocked until the DOWN message has been received and removed from the message queue. Not the fun kind of blocking tbh 💀

🛑 `stop/1,2`

stop(EventMgrRef) -> ok
stop(EventMgrRef, Reason, Timeout) -> ok

Types:

EventMgrRef = Name | {Name,Node} | {global,GlobalName} | {via,Module,ViaName} | pid()
Name = Node = atom()
GlobalName = ViaName = term()
Reason = term()
Timeout = int()>0 | infinity

Orders event manager EventMgrRef to exit with the specified Reason, and waits for it to terminate. Before terminating, gen_event calls Module:terminate(stop,...) for each installed event handler. Everyone gets a proper goodbye 🌈

The function returns ok if the event manager terminates with the expected reason. Any reason other than normal, shutdown, or {shutdown,Term} causes an error report to be issued using logger(3erl). The default Reason is normal.

Timeout is a positive integer, bestie — how many milliseconds to wait for the event manager to actually terminate 💅 — or the atom infinity if you'd rather wait forever. Default is infinity. And if the event manager hasn't terminated within that time? The call straight-up exits the calling process with reason timeout 😭. Rude but iconic.

If the process doesn't exist at all, the call exits the calling process with reason noproc, and if the connection to the remote Node where the server runs fails, you get {nodedown,Node}. That's the vibe check failing, fr.

For a description of EventMgrRef, see add_handler/3.

swap_handler(EventMgrRef, {Handler1,Args1}, {Handler2,Args2}) -> Result — swaps out an old event handler for a new one 🦋

Types:

EventMgrRef = Name | {Name,Node} | {global,GlobalName} |
{via,Module,ViaName} | pid()
 Name = Node = atom()
 GlobalName = ViaName = term()
Handler1 = Handler2 = Module | {Module,Id}
 Module = atom()
 Id = term()
Args1 = Args2 = term()
Result = ok | {error,Error}
 Error = {'EXIT',Reason} | term()
 Reason = term()

Replaces an old event handler with a new event handler in event manager EventMgrRef ✨

For a description of the arguments, see add_handler/3.

So here's the order of operations, and it's a whole thing 👑: first the old event handler Handler1 is deleted. The event manager calls Module1:terminate(Args1, ...), where Module1 is the callback module of Handler1, and collects the return value.

Then the new event handler Handler2 gets added and initiated by calling Module2:init({Args2,Term}), where Module2 is the callback module of Handler2 and Term is the return value of Module1:terminate/2. That's literally the whole point — it makes it possible to transfer information from Handler1 to Handler2 🫶.

And here's the unhinged part, no cap: the new handler is added even if the specified old event handler is not installed — in which case Term=error — or if Module1:terminate/2 fails with Reason, in which case Term={'EXIT',Reason}. And the old handler is deleted even if Module2:init/1 fails. It's chaos, it's consistent 💀.

If there was a supervised connection between Handler1 and a process Pid, there's a supervised connection between Handler2 and Pid instead. The bond transfers 🎀.

If Module2:init/1 returns a correct value, this function returns ok. If Module2:init/1 fails with Reason or returns an unexpected value Term, this function returns {error,{'EXIT',Reason}} or {error,Term}, respectively.

swap_sup_handler(EventMgrRef, {Handler1,Args1}, {Handler2,Args2}) -> Result — same swap, but the connection to the calling process gets supervised 👑

Types:

EventMgrRef = Name | {Name,Node} | {global,GlobalName} |
{via,Module,ViaName} | pid()
 Name = Node = atom()
 GlobalName = ViaName = term()
Handler1 = Handler 2 = Module | {Module,Id}
 Module = atom()
 Id = term()
Args1 = Args2 = term()
Result = ok | {error,Error}
 Error = {'EXIT',Reason} | term()
 Reason = term()

Replaces an event handler in event manager EventMgrRef in the same way as swap_handler/3, but also supervises the connection between Handler2 and the calling process 🧚.

For a description of the arguments and return values, see swap_handler/3.

wait_response(ReqId, WaitTime) -> Result — sit and wait for the response that matches your request id, bestie ⏳

Types:

ReqId = request_id()
WaitTime = response_timeout()
Response =
    {reply, Reply :: term()} |
    {error, {Reason :: term(), emgr_ref()}}
Result = Response | timeout

Waits for a response corresponding to the request identifier ReqId. The request must have been made by send_request/3 to the gen_statem process. And — important — this function must be called from the same process from which send_request/3 was made. Same process or it doesn't count 🙈.

WaitTime specifies how long to wait for a response. If no response is received within the specified time, the function returns timeout and no cleanup is done — which lowkey slaps, because it means you can call this function repeatedly until a reply actually shows up 🔥.

The return value Reply is defined in the return value of Module:handle_call/3.

The difference between receive_response/2 and wait_response/2? receive_response/2 abandons the request at timeout so that a potential future response is ignored, while wait_response/2 does not. One is a ghosting queen, the other keeps the door open 💅.

wait_response(ReqIdCollection, WaitTime, Delete) -> Result — the group-chat version, multiple request ids at once 🫶

Types:

ReqIdCollection = request_id_collection()
WaitTime = response_timeout()
Delete = boolean()
Response =
    {reply, Reply :: term()} |
    {error, {Reason :: term(), emgr_ref()}}
Result =
    {Response,
     Label :: term(),
     NewReqIdCollection :: request_id_collection()} |
    no_request | timeout

Waits for a response corresponding to a request identifier saved in ReqIdCollection. All request identifiers of ReqIdCollection must correspond to requests that have been made using send_request/3 or send_request/5, and all requests must have been made by the process calling this function. No borrowing your friend's request ids, bestie 🚫.

The Label in the response equals the Label associated with the request identifier that the response corresponds to. The Label of a request identifier is associated when saving the request id in a request identifier collection, or when sending the request using send_request/5 ✨.

Compared to wait_response/2, the returned result associated with a specific request identifier — or an exception associated with a specific request identifier — will be wrapped in a 3-tuple. The first element of this tuple equals the value that would have been produced by wait_response/2, the second element equals the Label associated with the specific request identifier, and the third element NewReqIdCollection is a possibly modified request identifier collection 🎀.

If ReqIdCollection is empty, no_request will be returned. If no response is received before the WaitTime timeout has triggered, the atom timeout is returned. It is valid to continue waiting for a response as many times as needed up until a response has been received and completed by check_response(), receive_response(), or wait_response(). So keep looping, it's fine 🔁.

The difference between receive_response/3 and wait_response/3 is that receive_response/3 abandons requests at timeout so that a potential future responses are ignored, while wait_response/3 does not. Same energy as the two-arg version, just with more friends 💀.

⚠️ Okay this next bit is the part everyone ignores and then gets confused by — pay attention: deleting an association is not for free, and a collection containing already handled requests can still be used by subsequent calls to wait_response/3, check_response/3, and receive_response/3. However, without deleting handled associations, the above calls will not be able to detect when there are no more outstanding requests to handle, so you will have to keep track of this some other way than relying on a no_request return. And heads up — if you pass a collection only containing associations of already handled or abandoned requests to wait_response/3, it will always block until a timeout determined by WaitTime is triggered and then return no_reply 😭💀.

which_handlers(EventMgrRef) -> [Handler] — roll call, who's currently installed 👑

Types:

EventMgrRef = Name | {Name,Node} | {global,GlobalName} |
{via,Module,ViaName} | pid()
 Name = Node = atom()
 GlobalName = ViaName = term()
Handler = Module | {Module,Id}
 Module = atom()
 Id = term()

Returns a list of all event handlers installed in event manager EventMgrRef 🦋.

For a description of EventMgrRef and Handler, see add_handler/3.

🌸 CALLBACK FUNCTIONS 🌸

The following functions are to be exported from a gen_event callback module. These are the ones you write, bestie — the event manager calls them, you just gotta have them ready 💅

✨ EXPORTS ✨

Module:code_change(OldVsn, State, Extra) -> {ok, NewState} — the glow-up-your-internal-state function 🦋

Types:

OldVsn = Vsn | {down, Vsn}
 Vsn = term()
State = NewState = term()
Extra = term()

📌 Note: This callback is optional, so callback modules need not export it. If a release upgrade/downgrade with Change={advanced,Extra} specified in the .appup file is made when code_change/3 isn't implemented, the event handler will crash with an undef error reason 💀. Not a vibe.

This function is called for an installed event handler that is to update its internal state during a release upgrade/downgrade — that is, when the instruction {update,Module,Change,...}, where Change={advanced,Extra}, is specified in the .appup file. For more information, see OTP Design Principles.

For an upgrade, OldVsn is Vsn, and for a downgrade, OldVsn is {down,Vsn} ⬆️⬇️. Vsn is defined by the vsn attribute(s) of the old version of the callback module Module. If no such attribute is defined, the version is the checksum of the Beam file ✨.

State is the internal state of the event handler.

Extra is passed "as is" from the {advanced,Extra} part of the update instruction. No touch-ups, no filtering, literally as-is 🎀.

The function is to return the updated internal state. Give me the new you, bestie 👑

Module:format_status(Status) -> NewStatus — pretty-printing yourself for the logs 🌈

Types:

Status = format_status()
NewStatus = format_status()

📌 Note: This callback is optional, so event handler modules need not export it. If a handler does not export this function, the gen_event module uses the handler state directly for the purposes described below.

If this callback is exported but fails, to hide possibly sensitive data, the default function will instead return the fact that format_status/1 has crashed 🙈. Privacy queen behavior, honestly.

Okay so this callback — the format_status/1 one — is called by a gen_event process in these situations, bestie:

One of sys:get_status/1,2 is invoked to get the gen_event status.
The event handler terminates abnormally and gen_event logs an error.

This callback is literally there to limit the status of the event handler returned by sys:get_status/1,2 or sent to logger. 💅

The callback gets a map Status describing the current status and shall return a map NewStatus with the same keys, but it may transform some values. Same keys, bestie — do not come for that map structure. 🫶

Two possible use cases for this callback: remove sensitive information from the state so it doesn’t get printed in log files, or compact large irrelevant status items that would only clutter the logs. Iconic cleanup behavior, honestly. ✨

format_status(Status) ->
  maps:map(
    fun(state,State) ->
            maps:remove(private_key, State);
       (message,{password, _Pass}) ->
            {password, removed};
       (_,Value) ->
            Value
    end, Status).

🎀 `Module:format_status(Opt, [PDict, State]) -> Status`

Types:

                 Opt = normal | terminate
                 PDict = [{Key, Value}]
                 State = term()
                 Status = term()

Warning: This callback is deprecated, in new code use format_status/1. If a format_status/1 callback exists, then this function will never be called. No cap, it’s fully sidelined. 💀

Note: This callback is optional, so event handler modules need not export it. If a handler does not export this function, the gen_event module uses the handler state directly for the purposes described below. This function is called by a gen_event process in the following situations: - One of sys:get_status/1,2 is invoked to get the gen_event status. Opt is set to the atom normal for this case. - The event handler terminates abnormally and gen_event logs an error. Opt is set to the atom terminate for this case.

This function is useful for changing the form and appearance of the event handler state for these cases. An event handler callback module wishing to change the sys:get_status/1,2 return value as well as how its state appears in termination error logs, exports an instance of format_status/2 that returns a term describing the current state of the event handler. 🦋

PDict is the current value of the process dictionary of gen_event.

State is the internal state of the event handler.

The function is to return Status, a term that changes the details of the current state of the event handler. Any term is allowed for Status. The gen_event module uses Status as follows:

When sys:get_status/1,2 is called, gen_event ensures that its return value contains Status in place of the state term of the event handler.
When an event handler terminates abnormally, gen_event logs Status in place of the state term of the event handler.

One use for this function is to return compact alternative state representations to avoid that large state terms are printed in log files. Compact and cute, we love to see it. 🌸

🦋 `Module:handle_call(Request, State) -> Result`

Types:

                 Request = term()
                 State = term()
                 Result = {ok,Reply,NewState} | {ok,Reply,NewState,hibernate}
                  | {swap_handler,Reply,Args1,NewState,Handler2,Args2}
                  | {remove_handler, Reply}
                 Reply = term()
                 NewState = term()
                 Args1 = Args2 = term()
                 Handler2 = Module2 | {Module2,Id}
                 Module2 = atom()
                 Id = term()

Whenever an event manager receives a request sent using call/3,4, this function is called for the specified event handler to handle the request.

Request is the Request argument of call/3,4.

State is the internal state of the event handler.

The return values are the same as for Module:handle_event/2 except that they also contain a term Reply, which is the reply to the client as the return value of call/3,4. So it’s giving handle_event but with a little reply moment. 💌

🌈 `Module:handle_event(Event, State) -> Result`

Types:

                 Event = term()
                 State = term()
                 Result = {ok,NewState} | {ok,NewState,hibernate}
                  | {swap_handler,Args1,NewState,Handler2,Args2} |
                 remove_handler
                  NewState = term()
                  Args1 = Args2 = term()
                  Handler2 = Module2 | {Module2,Id}
                  Module2 = atom()
                  Id = term()

Whenever an event manager receives an event sent using notify/2 or sync_notify/2, this function is called for each installed event handler to handle the event.

Event is the Event argument of notify/2/sync_notify/2.

State is the internal state of the event handler.

If {ok,NewState} or {ok,NewState,hibernate} is returned, the event handler remains in the event manager with the possible updated internal state NewState.
If {ok,NewState,hibernate} is returned, the event manager also goes into hibernation (by calling proc_lib:hibernate/3), waiting for the next event to occur. It is sufficient that one of the event handlers return {ok,NewState,hibernate} for the whole event manager process to hibernate. One handler said nap time, and the whole manager said periodt. 😴
If {swap_handler,Args1,NewState,Handler2,Args2} is returned, the event handler is replaced by Handler2 by first calling Module:terminate(Args1,NewState) and then Module2:init({Args2,Term}), where Term is the return value of Module:terminate/2. For more information, see swap_handler/3.
If remove_handler is returned, the event handler is deleted by calling Module:terminate(remove_handler,State).

🧚 `Module:handle_info(Info, State) -> Result`

Types:

                 Info = term()
                 State = term()
                 Result = {ok,NewState} | {ok,NewState,hibernate}
                  | {swap_handler,Args1,NewState,Handler2,Args2} |
                 remove_handler
                  NewState = term()
                  Args1 = Args2 = term()
                  Handler2 = Module2 | {Module2,Id}
                  Module2 = atom()
                  Id = term()

Note: This callback is optional, so callback modules need not export it. The gen_event module provides a default implementation of this function that logs about the unexpected Info message, drops it and returns {ok, State}. Default behavior said “not my problem,” logged it, and moved on. 🙈

This function is called for each installed event handler when an event manager receives any other message than an event or a synchronous request (or a system message).

Info is the received message.

In particular, this callback will be made when a process terminated after calling add_sup_handler/3. Any event handler attached to an event manager which in turn has a supervised handler should expect callbacks of the shape Module:handle_info({'EXIT', Pid, Reason}, State).

For a description of State and possible return values, see Module:handle_event/2.

💫 `Module:init(InitArgs) -> {ok,State} | {ok,State,hibernate} | {error,Reason}`

Types:

                 InitArgs = Args | {Args,Term}
                  Args = Term = term()
                 State = term()
                 Reason = term()

Whenever a new event handler is added to an event manager, this function is called to initialize the event handler. Fresh handler, who dis? 👑

If the event handler is added because of a call to add_handler/3 or add_sup_handler/3, InitArgs is the Args argument of these functions.

If the event handler replaces another event handler because of a call to swap_handler/3 or swap_sup_handler/3, or because of a swap return tuple from one of the other callback functions, InitArgs is a tuple {Args,Term}, where Args is the argument provided in the function call/return tuple and Term is the result of terminating the old event handler, see swap_handler/3.

If successful, the function returns {ok,State} or {ok,State,hibernate}, where State is the initial internal state of the event handler.

If {ok,State,hibernate} is returned, the event manager goes into hibernation (by calling proc_lib:hibernate/3), waiting for the next event to occur. Sleepy era activated. 🎀

🍓 `Module:terminate(Arg, State) -> term()`

Types:

                 Arg = Args | {stop,Reason} | stop | remove_handler
                  | {error,{'EXIT',Reason}} | {error,Term}
                 Args = Reason = Term = term()

Note: This callback is optional, so callback modules need not export it. The gen_event module provides a default implementation without cleanup. Default says bye with zero extra tasks. 💅

Whenever an event handler is deleted from an event manager, this function is called. It is to be the opposite of Module:init/1 and do any necessary cleaning up.

If the event handler is deleted because of a call to delete_handler/3, swap_handler/3, or swap_sup_handler/3, Arg is the Args argument of this function call.

Arg={stop,Reason} if the event handler has a supervised connection to a process that has terminated with reason Reason.

Arg=stop if the event handler is deleted because the event manager is terminating.

The event manager terminates if it is part of a supervision tree and it is ordered by its supervisor to terminate. Even if it is not part of a supervision tree, it terminates if it receives an 'EXIT' message from its parent.

Arg=remove_handler if the event handler is deleted because another callback function has returned remove_handler or {remove_handler,Reply}.

Arg={error,Term} if the event handler is deleted because a callback function returned an unexpected value Term, or Arg={error,{'EXIT',Reason}} if a callback function failed.

State is the internal state of the event handler, periodt. 🫶

Okay so the return value babes — this function can return any term at all, no rules, go off 💅. But like, what actually happens to it depends on the vibe check:

If the event handler gets deleted because of a call to gen_event:delete_handler/3, then the return value of that function becomes the return value of this function. Chain reaction, iykyk ✨.
If the event handler is about to be replaced with another event handler because of a swap, the return value gets passed straight to the init function of the new event handler 🫶. Passing the baton, iconic.
Otherwise? The return value is just ignored. Literally nobody's looking at it 💀. Not me returning a whole dramatic term for nothing.

✨ SEE ALSO ✨

supervisor(3erl) — the supervising bestie 🧚
sys(3erl) — the system messages one, lowkey essential 🦋

And that's the whole page, bestie, we ate that 🍓🎀. This little masterpiece comes to you courtesy of Ericsson AB, running the stdlib 4.3.1.3 era of the gen_event(3erl) manual — go forth and build your event handlers, main character energy only 👑💜

📦 where this page comes from

ubuntu noble: erlang-manpages 1:25.3.2.8+dfsg-1ubuntu4 · Apache-2.0
ubuntu noble: erlang-manpages 1:25.3.2.8+dfsg-1ubuntu4.8 · Apache-2.0
💖 girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on 2026-11-06: a language model's rewrite of the original, under its terms

gen_event(3erl)

💖 NAME 💖

✨ DESCRIPTION ✨

🧚 DATA TYPES 🧚

request_id()

request_id_collection()

🚀 EXPORTS 🚀

add_handler(EventMgrRef, Handler, Args) -> Result

add_sup_handler(EventMgrRef, Handler, Args) -> Result

call(EventMgrRef, Handler, Request) -> Result

call(EventMgrRef, Handler, Request, Timeout) -> Result

💅 check_response(Msg, ReqId) -> Result

🎀 check_response(Msg, ReqIdCollection, Delete) -> Result

🦋 delete_handler(EventMgrRef, Handler, Args) -> Result

🌈 notify(EventMgrRef, Event) -> ok / sync_notify(EventMgrRef, Event) -> ok

💖 receive_response(ReqId, Timeout) -> Result

🍓 receive_response(ReqIdCollection, Timeout, Delete) -> Result

💌 reqids_add/3

📨 send_request/3

📨📨 send_request/5

🌱 start/0,1,2

🌱🔗 start_link/0,1,2

🌱👀 start_monitor/0,1,2

🛑 stop/1,2

🌸 CALLBACK FUNCTIONS 🌸

✨ EXPORTS ✨

🎀 Module:format_status(Opt, [PDict, State]) -> Status

🦋 Module:handle_call(Request, State) -> Result

🌈 Module:handle_event(Event, State) -> Result

🧚 Module:handle_info(Info, State) -> Result

💫 Module:init(InitArgs) -> {ok,State} | {ok,State,hibernate} | {error,Reason}

🍓 Module:terminate(Arg, State) -> term()

✨ SEE ALSO ✨

`request_id()`

`request_id_collection()`

`add_handler(EventMgrRef, Handler, Args) -> Result`

`add_sup_handler(EventMgrRef, Handler, Args) -> Result`

`call(EventMgrRef, Handler, Request) -> Result`

`call(EventMgrRef, Handler, Request, Timeout) -> Result`

💅 `check_response(Msg, ReqId) -> Result`

🎀 `check_response(Msg, ReqIdCollection, Delete) -> Result`

🦋 `delete_handler(EventMgrRef, Handler, Args) -> Result`

🌈 `notify(EventMgrRef, Event) -> ok` / `sync_notify(EventMgrRef, Event) -> ok`

💖 `receive_response(ReqId, Timeout) -> Result`

🍓 `receive_response(ReqIdCollection, Timeout, Delete) -> Result`

💌 `reqids_add/3`

📨 `send_request/3`

📨📨 `send_request/5`

🌱 `start/0,1,2`

🌱🔗 `start_link/0,1,2`

🌱👀 `start_monitor/0,1,2`

🛑 `stop/1,2`

🎀 `Module:format_status(Opt, [PDict, State]) -> Status`

🦋 `Module:handle_call(Request, State) -> Result`

🌈 `Module:handle_event(Event, State) -> Result`

🧚 `Module:handle_info(Info, State) -> Result`

💫 `Module:init(InitArgs) -> {ok,State} | {ok,State,hibernate} | {error,Reason}`

🍓 `Module:terminate(Arg, State) -> term()`