-
- require - load some module
- Buffer
- log - Gives out the message into log
- exec - execute some OS command, like "cp file1 file2"
- on - Subscribe on changes or updates of some state
- once
- subscribe - same as on
- unsubscribe
- getSubscriptions
- getFileSubscriptions
- schedule
- scheduleById
- getSchedules
- clearSchedule
- getAttr
- getAstroDate
- isAstroDay
- compareTime
- setState
- setStateAsync
- setBinaryState
- setStateDelayed
- clearStateDelayed
- getStateDelayed
- getState
- getStateAsync
- existsState
- getObject
- setObject
- existsObject
- extendObject
- deleteObject
- getIdByName
- getEnums
- createState
- createStateAsync
- deleteState
- deleteStateAsync
- sendTo
- sendToAsync
- sendToHost
- sendToHostAsync
- setInterval
- clearInterval
- setTimeout
- clearTimeout
- setImmediate
- formatDate
- formatTimeDiff
- getDateObject
- formatValue
- adapterSubscribe
- adapterUnsubscribe
- $ - Selector
- readFile
- writeFile
- delFile
- renameFile
- onFile
- offFile
- onStop
- getHistory
- runScript
- runScriptAsync
- startScript
- startScriptAsync
- stopScript
- stopScriptAsync
- isScriptActive
- name
- instance
- messageTo
- messageToAsync
- onMessage
- onMessageUnregister
- onLog
- onLogUnregister
- wait
- sleep
If in the script, some modules or functions are used with callbacks or cyclic calls, except setTimeout/setInterval, so they will be called again and again even if the new version of a script exists or a script is deleted. For example, the following script:
const http = require('http');
// Read www.google.com page
http.request('www.google.com', function (res) {
res.setEncoding('utf8');
res.on('data', function (chunk) {
log('BODY: ' + chunk);
});
}).on('error', function (e) {
log('problem with request: ' + e.message, 'error');
});
was deleted by user before callback returns. The callback will be executed anyway. To fix this feature restart the javascript adapter.
You can use cb
function to wrap you callback, like this
http.request('www.google.com', cb(function (res) {
res.setEncoding('utf8');
res.on('data', function (chunk) {
log('BODY: ' + chunk);
});
})).on('error', cb(function (e) {
log('problem with request: ' + e.message, 'error');
}));
to be sure, that no callback will be called if a script is deleted or modified.
You can define the global scripts in the global
folder.
All global scripts are available on all instances. If a global script is disabled, it will not be used.
Global script will be just prepended to the normal script and compiled, so you cannot share data between scripts via global scripts. Use states for it.
To use global functions in TypeScript, you have to declare
them first, so the compiler knows about the global functions. Example:
// global script:
// ==============
function globalFn(arg: string): void {
// actual implementation
}
// normal script:
// ==============
declare function globalFn(arg: string): void;
// use as normal:
globalFn('test');
Create two instances of javascript adapter: one "test" and one "production". After the script is tested in the "test" instance, it can be moved to "production". By that you can restart the "test" instance as you want.
const mod = require('module_name');
The following modules are preloaded: fs
, crypto
, wake_on_lan
, request
, suncalc2
, util
, path
, os
, net
, events
, dns
.
To use other modules, enter the name of the module in the instance configuration. ioBroker will install the module. You can require and use it in your scripts afterwards.
Notice - module request is available via variable request. There is no need to write const request = require('request');
.
Buffer - Node.js Buffer, read here http://nodejs.org/api/buffer.html
Usage is the same as in javascript
exec(cmd, [options], callback);
Execute system command and get the outputs.
// reboot linux system :)
exec('reboot');
// Get the list of files and directories in /var/log
exec('ls /var/log', function (error, stdout, stderr) {
log('stdout: ' + stdout);
});
Node.js uses /bin/sh to execute commands. If you want to use another shell, you can use the option object as described in the Node.js documentation for child_process.exec. It is the best practice to always use fill path names to commands to make sure the right command is executed.
Notice: you must enable Enable command "setObject" option to call it.
on(pattern, callbackOrId, value);
The callback function will return the object as parameter with the following content:
{
'id': 'javascript.0.myplayer',
'state': {
'val': 'new state',
'ts': 1416149118,
'ack': true,
'lc': 1416149118,
'from': 'system.adapter.sonos.0'
},
'oldState': {
'val': 'old state',
'ts': 1416148233,
'ack': true,
'lc': 1416145154,
'from': 'system.adapter.sonos.0'
}
}
Note: state
was previously called newState
. That is still working.
Example:
let timer;
// Create state "javascript.0.counter"
createState('counter', 0);
// On change
on('adapter.0.device.channel.sensor', (data) => {
// But not ofter than 30 seconds
if (!timer) {
timer = setTimeout(() => {
timer = null;
}, 30000);
// Set acknowledged value
setState('counter', 1 + getState('counter'), true/*ack*/);
// Or to set unacknowledged command
setState('adapter.0.device.channel.actor', true);
}
});
You can use the following parameters to specify the trigger:
parameter | type/value | description |
---|---|---|
logic | string | "and" or "or" logic to combine the conditions (default: "and") |
id | string | id is equal to given one |
RegExp | id matched to regular expression | |
Array | id matched to a list of allowed IDs | |
name | string | name is equal to given one |
RegExp | name matched to regular expression | |
Array | name matched to a list of allowed names | |
change | string | "eq", "ne", "gt", "ge", "lt", "le", "any" |
"eq" | (equal) New value must be equal to old one (state.val == oldState.val) | |
"ne" | (not equal) New value must be not equal to the old one (state.val != oldState.val) If pattern is id-string this value is used by default | |
"gt" | (greater) New value must be greater than old value (state.val > oldState.val) | |
"ge" | (greater or equal) New value must be greater or equal to old one (state.val >= oldState.val) | |
"lt" | (smaller) New value must be smaller than old one (state.val < oldState.val) | |
"le" | (smaller or equal) New value must be smaller or equal to old value (state.val <= oldState.val) | |
"any" | Trigger will be raised if just the new value comes | |
val | mixed | New value must be equal to given one |
valNe | mixed | New value must be not equal to given one |
valGt | mixed | New value must be greater than given one |
valGe | mixed | New value must be greater or equal to given one |
valLt | mixed | New value must be smaller than given one |
valLe | mixed | New value must be smaller or equal to given one |
ack | boolean | Acknowledged state of new value is equal to given one |
q | number | Quality code state of new value is equal to given one. You can use '*' for matching to any code. If not provided q = 0 is set as pattern! |
oldVal | mixed | Previous value must be equal to given one |
oldValNe | mixed | Previous value must be not equal to given one |
oldValGt | mixed | Previous value must be greater than given one |
oldValGe | mixed | Previous value must be greater or equal to given one |
oldValLt | mixed | Previous value must be smaller than given one |
oldValLe | mixed | Previous value must be smaller or equal to given one |
oldAck | bool | Acknowledged state of previous value is equal to given one |
oldQ | number | Quality code state of previous value is equal to given one. You can use '*' for matching to any code |
ts | string | New value time stamp must be equal to given one (state.ts == ts) |
tsGt | string | New value time stamp must be not equal to the given one (state.ts != ts) |
tsGe | string | New value time stamp must be greater than given value (state.ts > ts) |
tsLt | string | New value time stamp must be greater or equal to given one (state.ts >= ts) |
tsLe | string | New value time stamp must be smaller than given one (state.ts < ts) |
oldTs | string | Previous time stamp must be equal to given one (oldState.ts == ts) |
oldTsGt | string | Previous time stamp must be not equal to the given one (oldState.ts != ts) |
oldTsGe | string | Previous time stamp must be greater than given value (oldState.ts > ts) |
oldTsLt | string | Previous time stamp must be greater or equal to given one (oldState.ts >= ts) |
oldTsLe | string | Previous time stamp must be smaller than given one (oldState.ts < ts) |
lc | string | Last change time stamp must be equal to given one (state.lc == lc) |
lcGt | string | Last change time stamp must be not equal to the given one (state.lc != lc) |
lcGe | string | Last change time stamp must be greater than given value (state.lc > lc) |
lcLt | string | Last change time stamp must be greater or equal to given one (state.lc >= lc) |
lcLe | string | Last change time stamp must be smaller than given one (state.lc < lc) |
oldLc | string | Previous last change time stamp must be equal to given one (oldState.lc == lc) |
oldLcGt | string | Previous last change time stamp must be not equal to the given one (oldState.lc != lc) |
oldLcGe | string | Previous last change time stamp must be greater than given value (oldState.lc > lc) |
oldLcLt | string | Previous last change time stamp must be greater or equal to given one (oldState.lc >= lc) |
oldLcLe | string | Previous last change time stamp must be smaller than given one (oldState.lc < lc) |
channelId | string | Channel ID must be equal to given one |
RegExp | Channel ID matched to regular expression | |
Array | Channel ID matched to a list of allowed channel IDs | |
channelName | string | Channel name must be equal to given one |
RegExp | Channel name matched to regular expression | |
Array | Channel name matched to a list of allowed channel names | |
deviceId | string | Device ID must be equal to given one |
RegExp | Device ID matched to regular expression | |
Array | Device ID matched to a list of allowed device IDs | |
deviceName | string | Device name must be equal to given one |
RegExp | Device name matched to regular expression | |
Array | Device name matched to a list of allowed device names | |
enumId | string | State belongs to given enum |
RegExp | One enum ID of the state satisfies the given regular expression | |
Array | One enum ID of the state is in the given list of enum IDs | |
enumName | string | State belongs to given enum |
RegExp | One enum name of the state satisfies the given regular expression | |
Array | One enum name of the state is in the given list of enum names | |
from | string | New value is from defined adapter |
RegExp | New value is from an adapter that matches the regular expression | |
Array | New value is from an adapter that appears in the given list of allowed adapters | |
fromNe | string | New value is not from defined adapter |
RegExp | New value is not from an adapter that matches the regular expression | |
Array | New value is not from an adapter that appears in the given list of forbidden adapters | |
oldFrom | string | Old value is from defined adapter |
RegExp | Old value is from an adapter that matches the regular expression | |
Array | Old value is from an adapter that appears in the given list of allowed adapters | |
oldFromNe | string | Old value is not from defined adapter |
RegExp | Old value is not from an adapter that matches the regular expression | |
Array | Old value is not from an adapter that appears in the given list of forbidden adapters |
Examples:
Trigger on all states with ID '*.STATE'
if they are acknowledged and have new value true
.
{
"id": /\.STATE$/,
"val": true,
"ack": true,
"logic": "and"
}
Note: you can use RegExp directly:
on(/^system\.adapter\..*\.\d+\.memRss$/, function (obj) {
});
// same as
on({id: /^system\.adapter\..*\.\d+\.memRss$/, change: "ne"}, function (obj) {
});
To simply connect two states with each other, write:
on('stateId1', 'stateId2');
All changes of stateId1 will be written to stateId2.
If the value
parameter is set in combination with state id as the second parameter, on any change the state will filled with the value
.
on('stateId1', 'stateId2', 'triggered');
setState('stateId1', 'new value');
// stateId2 will be set to 'triggered'.
Function on
returns handler back. This handler can be used by unsubscribe.
Notice: By default only states with quality 0x00 will be passed to callback function. If you want to get all events, add {q: '*'}
to pattern structure.
Notice: Please note, that by default "change" is equal to "any", except when only id as string is set (like on('id', () => {});
). In last case change will be set to "ne".
Notice: If you want to also get state deletions/expires as trigger, you need to use change with ne
or any
AND q with *
as filter!
Notice: from 4.3.2 it is possible to write a type of trigger as second parameter: on('my.id.0', 'any', obj => log(obj.state.val));
Registers a one-time subscription which automatically unsubscribes after the first invocation. Same as on, but just executed once.
once(pattern, callback);
subscribe - same as on
unsubscribe(id);
// or
unsubscribe(handler);
Remove all subscriptions for given object ID or for given handler.
// By handler
let mySubscription = on({ id: 'javascript.0.myState', change: 'any' }, (data) => {
// unsubscribe after first trigger
if (unsubscribe(mySubscription)) {
log('Subscription deleted');
}
});
// by Object ID
on({ id: 'javascript.0.myState1', change: 'ne' }, (data) => {
log('Some event');
});
on({ id: 'javascript.0.myState1', change: 'any' }, (data) => {
// unsubscribe
if (unsubscribe('javascript.0.myState1')) {
log('All subscriptions deleted');
}
});
Get the list of subscriptions.
Example of a result:
{
"megad.0.dataPointName": [
{
"name" : "script.js.NameOfScript",
"pattern" : {
"id" : "megad.0.dataPointName",
"change" : "ne"
}
}
]
}
Get the list of file subscriptions.
Example of a result:
{
"vis.0$%$main/*": [
{
"name" : "script.js.NameOfScript",
"id" : "vis.0",
"fileNamePattern": "main/*"
}
]
}
schedule(pattern, callback);
Time scheduler with astro-function.
Pattern can be a string with Cron-Syntax, which consists of 5 (without seconds) or 6 (with seconds) digits:
* * * * * *
│ │ │ │ │ │
│ │ │ │ │ │
│ │ │ │ │ └───── day of week (0 - 6) (0 to 6 are Sunday to Saturday, or use names; 7 is Sunday, the same as 0)
│ │ │ │ └────────── month (1 - 12)
│ │ │ └─────────────── day of month (1 - 31)
│ │ └──────────────────── hour (0 - 23)
│ └───────────────────────── min (0 - 59)
└───────────────────────────── [optional] sec (0 - 59)
// Example with 5 digits:
schedule('*/2 * * * *', () => {
log('Will be triggered every 2 minutes!');
});
// Example with 6 digits:
schedule('*/3 * * * * *', () => {
log('Will be triggered every 3 seconds!');
});
The pattern can also be an object, it is used especially if seconds are required:
the object could have the following properties:
second
minute
hour
date
month
year
dayOfWeek
schedule({ second: [20, 25] }, () => {
log('Will be triggered at xx:xx:20 and xx:xx:25 of every minute!');
});
schedule({ hour: 12, minute: 30 }, () => {
log('Will be triggered at 12:30!');
});
Pattern can be a Javascript Date object (some specific time point) - in this case only it will be triggered only one time.
If start or end times for a schedule are needed, this could also be implemented with usage of an object. In this scenario the object has the properties:
start
end
rule
start and end defines a Date object a DateString or a number of milliseconds since 01 January 1970 00:00:00 UTC. Rule is a schedule string with Cron-Syntax or an object:
let startTime = new Date(Date.now() + 5000);
let endTime = new Date(startTime.getTime() + 5000);
schedule({ start: startTime, end: endTime, rule: '*/1 * * * * *' }, () => {
log('It will run after 5 seconds and stop after 10 seconds');
});
The rule itself could be also an object:
let today = new Date();
let startTime = new Date(today.getFullYear(), today.getMonth(), today.getDate() + 1);
let endTime = new Date(today.getFullYear(), today.getMonth(), today.getDate() + 7);
let ruleData = { hour: 12, minute: 30 };
schedule({ start: startTime, end: endTime, rule: ruleData }, () => {
log('Will be triggered at 12:30, starting tomorow, ending in 7 days');
});
Astro-function can be used via "astro" attribute:
schedule({ astro: 'sunrise' }, () => {
log("Sunrise!");
});
schedule({ astro: 'sunset', shift: 10 }, () => {
log("10 minutes after sunset!");
});
The attribute "shift" is the offset in minutes. It can be negative, too, to define time before astro event.
The following values can be used as attribute in astro-function:
"sunrise"
: sunrise (top edge of the sun appears on the horizon)"sunriseEnd"
: sunrise ends (bottom edge of the sun touches the horizon)"goldenHourEnd"
: morning golden hour (soft light, the best time for photography) ends"solarNoon"
: solar noon (sun is in the highest position)"goldenHour"
: evening golden hour starts"sunsetStart"
: sunset starts (bottom edge of the sun touches the horizon)"sunset"
: sunset (sun disappears below the horizon, evening civil twilight starts)"dusk"
: dusk (evening nautical twilight starts)"nauticalDusk"
: nautical dusk (evening astronomical twilight starts)"night"
: night starts (dark enough for astronomical observations)"nightEnd"
: night ends (morning astronomical twilight starts)"nauticalDawn"
: nautical dawn (morning nautical twilight starts)"dawn"
: dawn (morning nautical twilight ends, morning civil twilight starts)"nadir"
: nadir (darkest moment of the night, sun is in the lowest position)
Note: to use "astro"-function the "latitude" and "longitude" must be defined in javascript adapter settings.
Note: in some places sometimes it could be so, that no night/nightEnd exists. Please read here about it.
Note: you can use "on" function for schedule with small modification:
on({ time: '*/2 * * * *' }, () => {
log((new Date()).toString() + " - Will be triggered every 2 minutes!");
});
on({ time: { hour: 12, minute: 30 }}, () => {
log((new Date()).toString() + " - Will be triggered at 12:30!");
});
on({ astro: 'sunset', shift: 10 }, () => {
log((new Date()).toString() + " - 10 minutes after sunset!");
});
scheduleById(id, callback);
scheduleById(id, ack, callback);
Allows to create a schedule based on a state value. If the state value changes, the old schedule will be deleted and a new schedule is created automatically.
Supported formats:
[h]h:[m]m:ss
(e.g.12:42:15
,15:3:12
,3:10:25
)[h]h:[m]m
(e.g.13:37
,9:40
)
scheduleById('0_userdata.0.configurableTimeFormat', () => {
log('Executed!');
});
Example: Create state and register schedule on changes:
createState(
'0_userdata.0.myTime',
'00:00:00', // default value
{
type: 'string',
read: true,
write: true
},
() => {
scheduleById('0_userdata.0.myTime', () => {
log('Executed!');
});
}
);
const list = getSchedules(true);
Returns the list of all CRON jobs and schedules (except astro).
Argument must be true
if you want to get the list for every running script. Otherwise only schedules in the current script will be returned.
const list = getSchedules(true);
list.forEach(schedule => log(JSON.stringify(schedule)));
// clear all schedules in all scripts!
list.forEach(schedule => clearSchedule(schedule));
Example output:
2020-11-01 20:15:19.929 - {"type":"cron","pattern":"0 * * * *","scriptName":"script.js.Heizung","id":"cron_1604258108384_74924"}
2020-11-01 20:15:19.931 - {"type":"schedule","schedule":"{"period":{}}","scriptName":"script.js.Heizung","id":"schedule_19576"}
If no "astro" function is used, you can cancel the schedule later. To allow this, the schedule object must be saved:
let sch = schedule('*/2 * * * *', () => { /* ... */ });
// later:
clearSchedule(sch);
getAttr({ attr1: { attr2: 5 } }, 'attr1.attr2');
Returns an attribute of the object. Path to attribute can be nested, like in the example.
If the first attribute is string, the function will try to parse the string as JSON string.
getAstroDate(pattern, date, offsetMinutes);
Returns a javascript Date object for the specified astro-name (e.g. "sunrise"
or "sunriseEnd"
). For valid values see the list of allowed values in the Astro section in the schedule function.
The returned Date object is calculated for the passed date. If no date is provided, the current day is used.
let sunriseEnd = getAstroDate('sunriseEnd');
log(`Sunrise ends today at ${sunriseEnd.toLocaleTimeString()}`);
let today = new Date();
let tomorrow = new Date(today.getFullYear(), today.getMonth(), today.getDate() + 1);
let tomorrowNight = getAstroDate('night', tomorrow);
Note: Depending on your geographical location, there can be cases where e.g. 'night'/'nightEnd' do not exist on certain time points (e.g. locations north in May/June each year!
You can use webpages like suncalc.net to check if the time points are correct.
isAstroDay();
Returns true
if the current time is between the astro sunrise and sunset.
compareTime(startTime, endTime, operation, timeToCompare);
Compare given time with limits.
If timeToCompare
is not given, so the actual time will be used.
The following operations are possible:
">"
- if given time is greater thanstartTime
">="
- if given time is greater or equal tostartTime
"<"
- if given time is less thanstartTime
"<="
- if given time is less or equal tostartTime
"=="
- if given time is equal tostartTime
"<>"
- if given time is not equal tostartTime
"between"
- if given time is betweenstartTime
andendTime
"not between"
- if given time is not betweenstartTime
andendTime
Time can be Date object or Date with time or just time.
You can use astro-names for the time definition. All 3 parameters can be set as astro time.
Following values are possible: sunrise
, sunset
, sunriseEnd
, sunsetStart
, dawn
, dusk
, nauticalDawn
, nauticalDusk
, nightEnd
, night
, goldenHourEnd
, goldenHour
.
See Astro for detail.
log(compareTime('sunsetStart', 'sunsetEnd', 'between') ? 'Now is sunrise' : 'Now is no sunrise');
It is possible to define the time with offset too:
log(compareTime({ astro: 'sunsetStart', offset: 30 }, { astro: 'sunrise', offset: -30 }, '>') ? 'Now is at least 30 minutes after sunset' : 'No idea');
Structure of an astro object.
{
astro: 'sunsetStart',// mandatory, can be written as string and not as object if offset and date are default
offset: 30, // optional
date: new Date() // optional
}
setState(id, state, ack, callback);
Note: The following commands are identical
setState('myState', 1, false);
setState('myState', { val: 1, ack: false });
setState('myState', 1);
Please refer to https://github.com/ioBroker/ioBroker/wiki/Adapter-Development-Documentation#commands-and-statuses for usage of ack
.
Short:
ack
= false : Script wants to send a command to be executed by the target device/adapterack
= true : Command was successfully executed, and state is updated as a positive result
await setStateAsync(id, state, ack);
Same as setState, but with promise
.
Attention: This method is deprecated!
setBinaryState(id, state, callback);
Same as setState, but for the binary states, like files, images, buffers.
The difference is that such a state has no ack, ts, lc, quality and so on flags und should be used only for binary things.
The object's common.type
must be equal to file
.
setStateDelayed(id, state, isAck, delay, clearRunning, callback);
Same as setState but with delay in milliseconds. You can clear all running delays for this ID (by default). E.g.
// Switch ON the light in the kitchen in one second
setStateDelayed('Kitchen.Light.Lamp', true, 1000);
// Switch OFF the light in the kitchen in 5 seconds and let first timeout run.
setStateDelayed('Kitchen.Light.Lamp', false, 5000, false, () => {
log('Lamp is OFF');
});
This function returns the handler of the timer, and this timer can be individually stopped by clearStateDelayed
clearStateDelayed(id);
Clears all delayed tasks for specified state ID or some specific delayed task.
setStateDelayed('Kitchen.Light.Lamp', false, 10000); // Switch OFF the light in the kitchen in ten second
let timer = setStateDelayed('Kitchen.Light.Lamp', true, 5000, false); // Switch ON the light in the kitchen in five second
clearStateDelayed('Kitchen.Light.Lamp', timer); // Nothing will be switched on
clearStateDelayed('Kitchen.Light.Lamp'); // Clear all running delayed tasks for this ID
getStateDelayed(id);
This is a synchronous call, and you will get the list of all running timers (setStateDelayed) for this id. You can call this function without id and get timers for all IDs. In case you call this function for some specific object ID, you will get the following answer:
getStateDelayed('hm-rpc.0.LQE91119.1.STATE');
// returns an array like
[
{ timerId: 1, left: 1123, delay: 5000, val: true, ack: false },
{ timerId: 2, left: 12555, delay: 15000, val: false, ack: false },
]
If you ask for all IDs the answer will look like:
getStateDelayed();
// returns an object like
{
'hm-rpc.0.LQE91119.1.STATE': [
{ timerId: 1, left: 1123, delay: 5000, val: true, ack: false },
{ timerId: 2, left: 12555, delay: 15000, val: false, ack: false },
],
'hm-rpc.0.LQE91119.2.LEVEL': [
{ timerId: 3, left: 5679, delay: 10000, val: 100, ack: false }
]
}
left
is the time left in millisecondsdelay
is the initial delay value in milliseconds
You can ask by timerId directly. In this case, the answer will be:
getStateDelayed(3);
// returns an object like
{ id: 'hm-rpc.0.LQE91119.2.LEVEL', left: 5679, delay: 10000, val: 100, ack: false }
getState(id);
Returns state with the given id in the following form:
{
val: value,
ack: true/false,
ts: timestamp,
lc: lastchanged,
from: origin
}
If state does not exist, a warning will be printed in the logs and the object { val: null, notExist: true }
will be returned.
To suppress the warning check if the state exists before calling getState (see existsState).
const stateObject = await getStateAsync(id);
Same as getState, but with promise
.
Attention: This method is deprecated!
getBinaryState(id, (err, data) => {});
Same as getState, but for the binary states, like files, images, buffers.
The difference is that such a state has no ack, ts, lc, quality and so on flags und should be used only for binary "things".
The object's common.type
must be equal to file
.
This function must always be used with callback. "data" is a buffer.
existsState(id, (err, isExists) => {});
If option "Do not subscribe all states on start" is deactivated, you can use simpler call:
existsState(id)
the function returns in this case true or false.
Check if a state exists.
getObject(id, enumName);
Get description of object id as stored in a system. You can specify the enumeration name. If this is defined, two additional attributes will be added to result: enumIds and enumNames. These arrays have all enumerations, where ID is a member of. E.g:
getObject('adapter.N.objectName', 'rooms');
gives back in enumIds all rooms, where the requested object is a member. You can define "true" as enumName to get back all enumerations.
setObject(id, obj, callback);
Write an object into DB. This command can be disabled in adapter's settings. Use this function carefully, while the global settings can be damaged.
You should use it to modify an existing object you read beforehand, e.g.:
const obj = getObject('adapter.N.objectName');
obj.native.settings = 1;
setObject('adapter.N.objectName', obj, (err) => {
if (err) log('Cannot write object: ' + err);
});
existsObject(id, function (err, isExists) {});
If the option "Do not subscribe all states on start" is deactivated, you can use simpler call:
existsObject(id)
the function returns in this case true or false.
Check if an object exists.
extendObject(id, obj, callback);
It is almost the same as setObject
, but first it reads the object and tries to merge all settings together.
Use it like this:
// Stop instance
extendObject('system.adapter.sayit.0', {common: {enabled: false}});
deleteObject(id, isRecursive, callback);
Delete an object from DB by ID. If the object has type state
, the state value will be deleted too.
Additional parameter isRecursive
could be provided, so all children of given ID will be deleted. Very dangerous!
Use it like this:
// Delete state
deleteObject('javascript.0.createdState');
Notice: isRecursive
option is available only with js-controller >= 2.2.x
getIdByName(name, alwaysArray);
Returns id of the object with given name.
If there is more than one object with this name, the result will be an array.
If alwaysArray
flag is set, the result will always be an array if some ID found.
getEnums(enumName);
Get the list of existing enumerations with members, like:
getEnums('rooms');
// returns:
[
{
"id":"enum.rooms.LivingRoom",
"members":["hm-rpc.0.JEQ0024123.1","hm-rpc.0.BidCoS-RF.4"],
"name": "Living room"
},
{
"id":"enum.rooms.Bath",
"members":["hm-rpc.0.JEQ0024124.1","hm-rpc.0.BidCoS-RF.5"],
"name": "Bath"
}
]
createState(name, initialValue, forceCreation, common, native, callback);
Create state and object in javascript space if it does not exist, e.g. javascript.0.mystate
.
!! Prefer to create own data points with the full ID 0_userdata.0.mystate
!!!
name
: name of the state without namespace, e.g.mystate
initialValue
: variable can be initialized after created. Value "undefined" means do not initialize value.forceCreation
: create/overwrite state independent of if state yet exists or not.common
: common description of object see description herenative
: native description of an object. Any specific information.callback
: called after state is created and initialized.
If you set in common
the flag alias
to true
, then alias will be created with the same name (but in alias.0
namespace) as the state.
Alias is created only if it does not exist yet.
The following settings for aliases are valid too:
common => {
alias: {
id: 'alias.0.myOtherState', // will be created automatically if not already exists
write: 'val * 1000', // convert function for write to created state
read: 'val / 1000' // convert function to read from created state
}
}
or
common => {
alias: {
id: 'alias.0.myOtherState', // will be created automatically if not already exists
}
}
It is possible short type of createState:
createState('myDatapoint')
- simply create datapoint if it does not existcreateState('myDatapoint', 1)
- create datapoint if it does not exist and initialize it with value 1createState('myDatapoint', { name: 'My own datapoint', unit: '°C' }, () => { log('created'); });
createState('myDatapoint', 1, { name: 'My own datapoint', unit: '°C' })
- create datapoint if it does not exist with specific name and units
await createStateAsync(name, initialValue, forceCreation, common, native);
Same as createState
, but the promise will be returned.
deleteState(name, callback);
Delete state and object in javascript space, e.g. javascript.0.mystate
. States from other adapters cannot be deleted.
deleteState('myDatapoint')
simply delete datapoint if exists.
await deleteStateAsync(name);
Same as deleteState
, but the promise will be returned.
createAlias(name, alias, forceCreation, common, native, callback);
Create alias in alias.0
space if it does not exist, e.g. javascript.0.myalias
and reference to a state or read/write states.
The common definition is taken from the read alias id object, but a provided common takes precedence.
name
: id of the new alias state with (possible without alias namespace), e.g.test.mystate
(namespacealias.0.
will be added =alias.0.test.mystate
)alias
: can be either an existing state id as string or an object with full alias definition including read/write ids and read/write functions. Note: Alias definitions can not be set as part of the common parameter!forceCreation
: create/overwrite alias independent of if state yet exists or not.common
: common description of alias object see description here. Values provided here will take precedence over the common definition of the read alias id object. Not: Alias definitions can not be set as part of this common parameter, see alias parameter!native
: native description of an object. Any specific information.callback
: called after state is created and initialized.
It is possible a short type of createAlias:
createAlias('myAlias', 'myDatapoint')
- simply create alias.0.myAlias that refernces to javascript.X.myDatapoint if it does not existcreateAlias('myAlias', { id: { read: 'myReadDatapoint', write: 'myWriteDatapoint' }})
- creates alias and reference to different read/write states
For other details, see createState, it is similar.
await createAliasAsync(name, alias, forceCreation, common, native);
Same as createAlias
, but the promise will be returned.
sendTo(adapter, command, message, callback);
sendTo(adapter, command, message, options, callback);
Send a message to a specific or all adapter instances. When using the adapter name, the message is sent to all instances.
To get specific information about messages, you must read the documentation for a particular adapter.
Example (with custom timeout):
sendTo('telegram', { user: 'UserName', text: 'Test message' }, { timeout: 2000 });
Some adapters also support responses to the sent messages. (e.g. history, sql, telegram) The response is only returned to the callback if the message is sent to a specific instance!
Example (with callback):
sendTo('telegram.0', { user: 'UserName', text: 'Test message' }, (res) => {
log(`Sent to ${res} users`);
});
Default timeout is 20000 milliseconds (if a callback function has been defined)
sendTo('telegram.0', { user: 'UserName', text: 'Test message' }, { timeout: 2000 }, (res) => {
log(`Sent to ${res} users`);
});
await sendToAsync(adapter, command, message);
await sendToAsync(adapter, command, message, options);
Same as sendTo, but with promise
.
Example:
const res = await sendToAsync('sql.0', 'getEnabledDPs', {});
log(JSON.stringify(res));
sendToHost(hostName, command, message, callback);
Send a message to controller instance.
The following commands are supported:
"cmdExec"
"getRepository"
"getInstalled"
"getVersion"
"getDiagData"
"getLocationOnDisk"
"getDevList"
"getLogs"
"getHostInfo"
It is rather specific commands and are not required often.
Example:
sendToHost('myComputer', 'cmdExec', { data: 'ls /' }, (res) => {
log('List of files: ' + res.data);
});
Notice: you must enable Enable command "setObject" option to call it.
await sendToHostAsync(hostName, command, message);
Same as sendToHost, but with promise
.
setInterval(callback, ms, arg1, arg2, arg3, arg4);
Same as javascript setInterval
.
clearInterval(id);
Same as javascript clearInterval
.
setTimeout(callback, ms, arg1, arg2, arg3, arg4);
Same as javascript setTimeout
.
clearTimeout(id);
Same as javascript clearTimeout
.
setImmediate(callback, arg1, arg2, arg3, arg4);
Same as javascript setImmediate
and almost the same as setTimeout(callback, 0, arg1, arg2, arg3, arg4)
but with higher priority.
formatDate(millisecondsOrDate, format);
millisecondsOrDate
: number of milliseconds from state.ts or state.lc (Number milliseconds from 1970.01.01 00:00:00) or javascript new Date() object or number of milliseconds from (new Date().getTime())format
: Can benull
, so the system time format will be used, otherwise
- YYYY, JJJJ, ГГГГ - full year, e.g 2015
- YY, JJ, ГГ - short year, e.g 15
- MM, ММ(cyrillic) - full month, e.g. 01
- M, М(cyrillic) - short month, e.g. 1
- DD, TT, ДД - full day, e.g. 02
- D, T, Д - short day, e.g. 2
- hh, SS, чч - full hours, e.g. 03
- h, S, ч - short hours, e.g. 3
- mm, мм(cyrillic) - full minutes, e.g. 04
- m, м(cyrillic) - short minutes, e.g. 4
- ss, сс(cyrillic) - full seconds, e.g. 05
- s, с(cyrillic) - short seconds, e.g. 5
- sss, ссс(cyrillic) - milliseconds
- WW, НН(cyrillic) - full week day as text
- W, Н(cyrillic) - short week day as text
- OO, ОО(cyrillic) - full month as text
- OOO, ООО(cyrillic) - full month as text as genitiv
- O, О(cyrillic) - short month as text
formatDate(new Date(), "YYYY-MM-DD") // => Date "2015-02-24"
formatDate(new Date(), "hh:mm") // => Hours and minutes "17:41"
formatDate(state.ts) // => "24.02.2015"
formatDate(state.ts, "JJJJ.MM.TT SS:mm:ss.sss") // => "2015.02.15 17:41:98.123"
formatDate(new Date(), "WW") // => Day of week "Tuesday"
formatDate(new Date(), "W") // => Day of week "Tu"
formatTimeDiff(milliseconds, format);
milliseconds
: difference in milliseconds*format
: Can benull
, so thehh:mm:ss
format will be used, otherwise
- DD, TT, ДД - full day, e.g. 02
- D, T, Д - short day, e.g. 2
- hh, SS, чч - full hours, e.g. 03
- h, S, ч - short hours, e.g. 3
- mm, мм(cyrillic) - full minutes, e.g. 04
- m, м(cyrillic) - short minutes, e.g. 4
- ss, сс(cyrillic) - full seconds, e.g. 05
- s, с(cyrillic) - short seconds, e.g. 5
formatTimeDiff(60000, "mm:ss") // => "01:00"
const diff = 172800000 + 10800000 + 540000 + 15000; // 2 days, 3 hours, 9 minutes + 15 secoonds
formatTimeDiff(diff); // "51:09:15"
formatTimeDiff(diff, "DD hh:mm"); // "02 03:09"
formatTimeDiff(diff, "D hh:mm"); // "2 03:09"
formatTimeDiff(diff, "hh:mm:ss"); // "51:09:15"
formatTimeDiff(diff, "h:m:s"); // "51:9:15"
formatTimeDiff(diff, "hh:mm"); // "51:09"
formatTimeDiff(diff, "mm:ss"); // "3069:15"
formatTimeDiff(diff, "hh"); // "51"
formatTimeDiff(diff, "mm"); // "3069"
getDateObject (stringOrNumber);
Converts string or number to a Date object. If only hours are given, it will add current date to it and will try to convert.
getDateObject("20:00") // => "Tue Aug 09 2016 20:00:00 GMT+0200"
formatValue(value, decimals, format);
Formats any value (strings too) to number. Replaces point with comma if configured in system. Decimals specify digits after comma. The default value is 2. Format is optional:
- '.,': 1234.567 => 1.234,56
- ',.': 1234.567 => 1,234.56
- ' .': 1234.567 => 1 234.56
adapterSubscribe(id);
Send to an adapter message "subscribe" to inform adapter. If adapter has the common flag "subscribable" in case of function "subscribe" this function will be called automatically.
adapterUnsubscribe(id);
Sends to an adapter the message unsubscribe
to inform adapter to not poll the values.
$(selector).on(function(obj) {});
$(selector).each(function(id, i) {});
$(selector).setState(value, ack);
$(selector).getState();
Format of selector:
"name[commonAttr=something1](enumName=something2){nativeName=something3}[id=idfilter][state.id=idfilter]"
name can be: state, channel, device or schedule
idfilter
can have wildcards '*'
Prefixes (not implemented - should be discussed) :
- # - take by name and not by id
- . - filter by role
- § - filter by room
Example:
$('state[id=*.STATE]')
or$('state[state.id=*.STATE]')
or$('*.STATE')
- select all states where id ends with ".STATE".$('state[id='hm-rpc.0.*]')
or$('hm-rpc.0.*')
- returns all states of adapter instance hm-rpc.0$('channel(rooms=Living room)')
- all states in room "Living room"$('channel{TYPE=BLIND}[state.id=*.LEVEL]')
- Get all shutters of Homematic$('channel[role=switch](rooms=Living room)[state.id=*.STATE]').setState(false)
- Switch all states with .STATE of channels with role "switch" in "Living room" to false$('channel[state.id=*.STATE](functions=Windows)').each(function (id, i) {log(id);});
- print all states of enum "windows" in log$('schedule[id=*65]').each(function (id, i) {log(id);});
- print all schedules with 65 at the end$('.switch §"Living room")
- Take states with all switches in 'Living room' (not implemented - should be discussed)$('channel .switch §"Living room")
- Take states with all switches in 'Living room' (not implemented - should be discussed)
Explanation Lets take a look at:
$('channel[role=switch][state.id=*.STATE](rooms=Wohnzimmer)').on(obj => {
log('New state ' + obj.id + ' = ' + obj.state.val);
});
This code searches in channels.
Find all channels with common.role="switch"
and belongs to enum.rooms.Wohnzimmer
.
Take all their states, where id ends with ".STATE"
and make subscription on all these states.
If some of these states change, the callback will be called like for "on" function.
Following functions are possible, setState, getState (only from first), on, each
// Switch on all switches in "Wohnzimmer"
$('channel[role=switch][state.id=*.STATE](rooms=Wohnzimmer)').setState(true);
You can interrupt the "each" loop by returning the false value, like:
// print two first IDs of on all switches in "Wohnzimmer"
$('channel[role=switch][state.id=*.STATE](rooms=Wohnzimmer)').each((id, i) => {
log(id);
if (i == 1) {
return false;
}
});
readFile(adapter, fileName, (error, bytes) => {});
The result will be given in callback.
Read file from DB from folder javascript.0
.
Argument adapter can be omitted.
// read vis views
readFile('vis.0', '/main/vis-views.json', (error, data) => {
log(data.substring(0, 50));
});
// The same as
//readFile('/../vis.0/main/vis-views.json', (error, data) => {
// log(data.substring(0, 50));
//});
By default, working directory/adapter is javascript.0
.
writeFile(adapter, fileName, bytes, (error) => {});
The optional error code will be given in callback. Argument adapter can be omitted. fileName is the name of file in DB. All files are stored in the folder "javascript". if you want to write to other folders, e.g. to "/vis.0/" use setFile for that.
The file that looks like '/subfolder/file.txt'
will be stored under "/javascript/subfolder/file.txt"
and can be accessed over web server with "http://ip:8082/javascript/subfolder/file.txt"
// store screenshot in DB
const fs = require('fs');
let data = fs.readFileSync('/tmp/screenshot.png');
writeFile(null, '/screenshots/1.png', data, (error) => {
log('file written');
});
// The same as
//writeFile('/screenshots/1.png', data, function (error) {
// log('file written');
//});
// store file in '/vis.0' in DB
const fs = require('fs');
let data = fs.readFileSync('/tmp/screenshot.png');
writeFile('vis.0', '/screenshots/1.png', data, (error) => {
log('file written');
});
delFile(adapter, fileName, (error) => {});
Delete file or directory. fileName is the name of file or directory in DB.
The alternative name of this method is unlink
renameFile(adapter, oldName, newName, (error) => {});
Rename file or directory. oldName is the name of file or directory in DB and is renamed to newName.
The alternative name of this method is rename
onFile(id, fileName, withFile, (id, fileName, size, fileData, mimeType) => {});
// or
onFile(id, fileName, (id, fileName, size) => {});
Subscribe to file changes:
id
is ID of an object of typemeta
, likevis.0
fileName
is file name or pattern, likemain/*
ormain/vis-view.json
withFile
if the content of file should be delivered in callback or not. the delivery of file content costs memory and time, so if you want to be just informed about changes, setwithFile
to false.
Arguments in callback:
id
- ID ofmeta
object;fileName
- file name (not pattern);size
- new file size;fileData
- file content of typeBuffer
if file is binary (detected by extension) orstring
. Delivered only ifwithFile
;mimeType
- mime type of file, likeimage/jpeg
. Delivered only ifwithFile
;
Important: this functionality is only available with [email protected] or newer.
offFile(id, fileName);
// or
onFile(id, fileName);
Unsubscribe from file changes:
id
is ID of an object of typemeta
, likevis.0
fileName
is file name or pattern, likemain/*
ormain/vis-view.json
Important: this functionality is only available with [email protected] or newer.
onStop (() => { /* do something when script is stopped */ }, timeout);
Install callback, that will be called if a script stopped. Used, e.g., to stop communication or to close connections.
// establish connection
const conn = require('net');
// ...
// close connection if script stopped
onStop((callback) => {
if (conn) {
// close connection
conn.destroy();
}
callback();
}, 2000 /*ms*/);
timeout
is 1000ms by default.
getHistory(instance, options, (error, result, options, instance) => {});
Read history from specified instance. If no instance is specified, the system default history instance will be taken.
// Read history of 'system.adapter.admin.0.memRss' from sql driver
const end = new Date().getTime();
getHistory(
'sql.0',
{
id: 'system.adapter.admin.0.memRss',
start: end - 3600000,
end: end,
aggregate: 'm4',
timeout: 2000
},
(err, result) => {
if (err) console.error(err);
if (result) {
for (let i = 0; i < result.length; i++) {
log(result[i].id + ' ' + new Date(result[i].ts).toISOString());
}
}
}
);
Possible options you can find here.
Additionally, to these parameters you must specify "id" and you may specify timeout (default: 20000ms).
One more example:
// Get last 50 entries from default history instance with no aggregation:
getHistory({
id: 'system.adapter.admin.0.alive',
aggregate: 'none',
count: 50
}, (err, result) => {
if (err) console.error(err);
if (result) {
for (let i = 0; i < result.length; i++) {
log(result[i].id + ' ' + new Date(result[i].ts).toISOString());
}
}
});
**Note: ** of course, history must be first enabled for selected ID in admin.
runScript('scriptName', () => {
// Callback is optional
log('Srcipt started, but not yet executed');
});
Starts or restarts other scripts (and itself too) by name.
// restart script
runScript('groupName.scriptName1');
Same as runScript, but with promise
.
runScriptAsync('scriptName')
.then(() => log('Script started, but not yet executed'));
// or
await runScriptAsync('scriptName');
log(`Script was restarted`);
startScript('scriptName', ignoreIfStarted, callback);
Starts the script. If ignoreIfStarted set to true, nothing will be done if a script yet running, otherwise the script will be restarted.
startScript('scriptName', true); // start script if not started
Same as runScript, but with promise
.
startScriptAsync('scriptName', ignoreIfStarted)
.then(started => log(`Script was ${started ? 'started' : 'already started'}`));
// or
const started = await startScriptAsync('scriptName', ignoreIfStarted);
log(`Script was ${started ? 'started' : 'already started'}`);
Starts the script. If ignoreIfStarted set to true, nothing will be done if a script yet running, otherwise the script will be restarted.
startScript('scriptName', true); // start script if not started
stopScript('scriptName', callback);
If stopScript is called without arguments, it will stop itself:
stopScript();
Same as stopScript, but with promise
:
stopScriptAsync('scriptName')
.then(stopped => log(`Script was ${stopped ? 'stopped' : 'already stopped'}`));
//or
const stopped = await stopScriptAsync('scriptName');
log(`Script was ${stopped ? 'stopped' : 'already stopped'}`);
If stopScript is called without arguments, it will stop itself:
stopScript();
isScriptActive('scriptName');
Returns if a script enabled or disabled. Please note that that does not give back if the script is now running or not. The script can be finished, but still activated.
It is not a function. It is a variable with javascript instance, that is visible in script's scope.
Just pause the execution of the script.
Warning this function is promise
and must be called as follows:
await wait(1000);
Same as wait
messageTo({ instance: 'instance', script: 'script.js.common.scriptName', message: 'messageName' }, data, {timeout: 1000}, result =>
log(JSON.stringify(result)));
Send via the "message bus" the message to some other script. Or even to some handler in the same script.
Timeout for callback is 5 seconds by default.
The target could be shorted to:
messageTo('messageName', data, result => {
log(JSON.stringify(result));
});
Callback and options are optional and timeout is by default 5000 milliseconds (if callback provided).
messageTo('messageName', dataWithNoResponse);
onMessage('myTopic', async (data, callback) => {
log(data);
if (!data.myPayload) {
// return error (promise reject)
callback({ error: 'something went wrong!!' });
} else {
// return result (promise resolve)
callback({ result: 'ok' });
}
});
(async () => {
try {
const msg = await messageToAsync({ instance: 0, script: 'script.js.test2', message: 'myTopic' }, { myPayload: true }, { timeout: 1000 });
log(`Done with: ${JSON.stringify(msg)}`);
} catch (error) {
// contents of result.error
console.error(error);
}
})();
onMessage('messageName', (data, callback) => {
log(`Received data: ${data}`); callback({ result: Date.now() });
});
Subscribes on javascript adapter message bus and delivers response via callback. The response from script which sends response as first will be accepted as answer, all other answers will be ignored.
To send a message to a JavaScript script which is then received by this handler, use messageTo.
To send a message from any other adapter use
adapter.sendTo('javascript.0', 'toScript', {
script: 'script.js.messagetest',
message: 'messageName',
data: {
flag: true
}
});
to send a message from CLI use
iob message javascript.0 toScript '{"script": "script.js.messagetest", "message": "messageName", "data": { "flag": true }}'
const id = onMessage('messageName', (data, callback) => {
log(data);
callback(Date.now());
});
// unsubscribe specific handler
onMessageUnregister(id);
// or unsubscribe by name
onMessageUnregister('messageName');
Unsubscribes from this message.
onLog('error', data => {
sendTo('telegram.0', { user: 'UserName', text: data.message });
log('Following was sent to telegram: ' + data.message);
});
Subscribe on logs with specified severity.
Important: you cannot output logs in handler with the same severity to avoid infinite loops.
E.g., this will produce no logs:
onLog('error', data => {
console.error('Error: ' + data.message);
});
To receive all logs the *
could be used. In this case, the log output in handler will be disabled completely.
onLog('*', data => {
console.error('Error: ' + data.message); // will produce no logs
});
function logHandler(data) {
console.error('Error: ' + data.message);
}
const id = onLog('warn', logHandler);
// unsubscribe by ID
onLogUnregister(id);
// or unsubscribe by function handler
onLogUnregister(logHandler);
// or unsubscribe all handlers with specific severity
onLogUnregister('warn');
Unsubscribes from these logs.
scriptName
- The name of the script.
log('Script ' + scriptName + ' started!');
It is not a function. It is a variable with script name, that is visible in script's scope.
The javascript instance where script is executed.
log('Script ' + name + ' started by ' + instance + '!');
There are two modes of subscribe to states:
- Adapter subscribes to all changes at start and receives all changes of all states (it is easy to use getStates(id), but requires more CPU and RAM):
log(getState('someID').val);
- Adapter subscribes every time on specified ID if "on/subscribe" called. In this mode, the adapter receives only updates for desired states. It is very performed and RAM efficiency, but you cannot access states directly in getState. You must use callback to get the result of state:
getState('someID', (error, state) => {
log(state.val);
});
It is because the adapter does not have the value of state in RAM and must ask central DB for the value.
There is a possibility to enable and disable scripts via states. For every script, the state will be created with the name javascript.INSTANCE.scriptEnabled.SCRIPT_NAME
.
Scripts can be activated and deactivated by controlling this state with ack=false
.