electron/docs/api/session.md
Samuel Attard 0ea1985ec4
docs: improve documentation on spellchecker download URL (#22383)
* docs: improve documentation on spellchecker download URL

* Update session.md
2020-02-26 16:04:27 -08:00

631 lines
22 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

# session
> Manage browser sessions, cookies, cache, proxy settings, etc.
Process: [Main](../glossary.md#main-process)
The `session` module can be used to create new `Session` objects.
You can also access the `session` of existing pages by using the `session`
property of [`WebContents`](web-contents.md), or from the `session` module.
```javascript
const { BrowserWindow } = require('electron')
let win = new BrowserWindow({ width: 800, height: 600 })
win.loadURL('http://github.com')
const ses = win.webContents.session
console.log(ses.getUserAgent())
```
## Methods
The `session` module has the following methods:
### `session.fromPartition(partition[, options])`
* `partition` String
* `options` Object (optional)
* `cache` Boolean - Whether to enable cache.
Returns `Session` - A session instance from `partition` string. When there is an existing
`Session` with the same `partition`, it will be returned; otherwise a new
`Session` instance will be created with `options`.
If `partition` starts with `persist:`, the page will use a persistent session
available to all pages in the app with the same `partition`. if there is no
`persist:` prefix, the page will use an in-memory session. If the `partition` is
empty then default session of the app will be returned.
To create a `Session` with `options`, you have to ensure the `Session` with the
`partition` has never been used before. There is no way to change the `options`
of an existing `Session` object.
## Properties
The `session` module has the following properties:
### `session.defaultSession`
A `Session` object, the default session object of the app.
## Class: Session
> Get and set properties of a session.
Process: [Main](../glossary.md#main-process)
You can create a `Session` object in the `session` module:
```javascript
const { session } = require('electron')
const ses = session.fromPartition('persist:name')
console.log(ses.getUserAgent())
```
### Instance Events
The following events are available on instances of `Session`:
#### Event: 'will-download'
Returns:
* `event` Event
* `item` [DownloadItem](download-item.md)
* `webContents` [WebContents](web-contents.md)
Emitted when Electron is about to download `item` in `webContents`.
Calling `event.preventDefault()` will cancel the download and `item` will not be
available from next tick of the process.
```javascript
const { session } = require('electron')
session.defaultSession.on('will-download', (event, item, webContents) => {
event.preventDefault()
require('request')(item.getURL(), (data) => {
require('fs').writeFileSync('/somewhere', data)
})
})
```
#### Event: 'preconnect'
Returns:
* `event` Event
* `preconnectUrl` String - The URL being requested for preconnection by the
renderer.
* `allowCredentials` Boolean - True if the renderer is requesting that the
connection include credentials (see the
[spec](https://w3c.github.io/resource-hints/#preconnect) for more details.)
Emitted when a render process requests preconnection to a URL, generally due to
a [resource hint](https://w3c.github.io/resource-hints/).
### Instance Methods
The following methods are available on instances of `Session`:
#### `ses.getCacheSize()`
Returns `Promise<Integer>` - the session's current cache size, in bytes.
#### `ses.clearCache()`
Returns `Promise<void>` - resolves when the cache clear operation is complete.
Clears the sessions HTTP cache.
#### `ses.clearStorageData([options])`
* `options` Object (optional)
* `origin` String (optional) - Should follow `window.location.origin`s representation
`scheme://host:port`.
* `storages` String[] (optional) - The types of storages to clear, can contain:
`appcache`, `cookies`, `filesystem`, `indexdb`, `localstorage`,
`shadercache`, `websql`, `serviceworkers`, `cachestorage`. If not
specified, clear all storage types.
* `quotas` String[] (optional) - The types of quotas to clear, can contain:
`temporary`, `persistent`, `syncable`. If not specified, clear all quotas.
Returns `Promise<void>` - resolves when the storage data has been cleared.
#### `ses.flushStorageData()`
Writes any unwritten DOMStorage data to disk.
#### `ses.setProxy(config)`
* `config` Object
* `pacScript` String (optional) - The URL associated with the PAC file.
* `proxyRules` String (optional) - Rules indicating which proxies to use.
* `proxyBypassRules` String (optional) - Rules indicating which URLs should
bypass the proxy settings.
Returns `Promise<void>` - Resolves when the proxy setting process is complete.
Sets the proxy settings.
When `pacScript` and `proxyRules` are provided together, the `proxyRules`
option is ignored and `pacScript` configuration is applied.
The `proxyRules` has to follow the rules below:
```sh
proxyRules = schemeProxies[";"<schemeProxies>]
schemeProxies = [<urlScheme>"="]<proxyURIList>
urlScheme = "http" | "https" | "ftp" | "socks"
proxyURIList = <proxyURL>[","<proxyURIList>]
proxyURL = [<proxyScheme>"://"]<proxyHost>[":"<proxyPort>]
```
For example:
* `http=foopy:80;ftp=foopy2` - Use HTTP proxy `foopy:80` for `http://` URLs, and
HTTP proxy `foopy2:80` for `ftp://` URLs.
* `foopy:80` - Use HTTP proxy `foopy:80` for all URLs.
* `foopy:80,bar,direct://` - Use HTTP proxy `foopy:80` for all URLs, failing
over to `bar` if `foopy:80` is unavailable, and after that using no proxy.
* `socks4://foopy` - Use SOCKS v4 proxy `foopy:1080` for all URLs.
* `http=foopy,socks5://bar.com` - Use HTTP proxy `foopy` for http URLs, and fail
over to the SOCKS5 proxy `bar.com` if `foopy` is unavailable.
* `http=foopy,direct://` - Use HTTP proxy `foopy` for http URLs, and use no
proxy if `foopy` is unavailable.
* `http=foopy;socks=foopy2` - Use HTTP proxy `foopy` for http URLs, and use
`socks4://foopy2` for all other URLs.
The `proxyBypassRules` is a comma separated list of rules described below:
* `[ URL_SCHEME "://" ] HOSTNAME_PATTERN [ ":" <port> ]`
Match all hostnames that match the pattern HOSTNAME_PATTERN.
Examples:
"foobar.com", "*foobar.com", "*.foobar.com", "*foobar.com:99",
"https://x.*.y.com:99"
* `"." HOSTNAME_SUFFIX_PATTERN [ ":" PORT ]`
Match a particular domain suffix.
Examples:
".google.com", ".com", "http://.google.com"
* `[ SCHEME "://" ] IP_LITERAL [ ":" PORT ]`
Match URLs which are IP address literals.
Examples:
"127.0.1", "[0:0::1]", "[::1]", "http://[::1]:99"
* `IP_LITERAL "/" PREFIX_LENGTH_IN_BITS`
Match any URL that is to an IP literal that falls between the
given range. IP range is specified using CIDR notation.
Examples:
"192.168.1.1/16", "fefe:13::abc/33".
* `<local>`
Match local addresses. The meaning of `<local>` is whether the
host matches one of: "127.0.0.1", "::1", "localhost".
#### `ses.resolveProxy(url)`
* `url` URL
Returns `Promise<String>` - Resolves with the proxy information for `url`.
#### `ses.setDownloadPath(path)`
* `path` String - The download location.
Sets download saving directory. By default, the download directory will be the
`Downloads` under the respective app folder.
#### `ses.enableNetworkEmulation(options)`
* `options` Object
* `offline` Boolean (optional) - Whether to emulate network outage. Defaults
to false.
* `latency` Double (optional) - RTT in ms. Defaults to 0 which will disable
latency throttling.
* `downloadThroughput` Double (optional) - Download rate in Bps. Defaults to 0
which will disable download throttling.
* `uploadThroughput` Double (optional) - Upload rate in Bps. Defaults to 0
which will disable upload throttling.
Emulates network with the given configuration for the `session`.
```javascript
// To emulate a GPRS connection with 50kbps throughput and 500 ms latency.
window.webContents.session.enableNetworkEmulation({
latency: 500,
downloadThroughput: 6400,
uploadThroughput: 6400
})
// To emulate a network outage.
window.webContents.session.enableNetworkEmulation({ offline: true })
```
#### `ses.preconnect(options)`
* `options` Object
* `url` String - URL for preconnect. Only the origin is relevant for opening the socket.
* `numSockets` Number (optional) - number of sockets to preconnect. Must be between 1 and 6. Defaults to 1.
Preconnects the given number of sockets to an origin.
#### `ses.disableNetworkEmulation()`
Disables any network emulation already active for the `session`. Resets to
the original network configuration.
#### `ses.setCertificateVerifyProc(proc)`
* `proc` Function | null
* `request` Object
* `hostname` String
* `certificate` [Certificate](structures/certificate.md)
* `validatedCertificate` [Certificate](structures/certificate.md)
* `verificationResult` String - Verification result from chromium.
* `errorCode` Integer - Error code.
* `callback` Function
* `verificationResult` Integer - Value can be one of certificate error codes
from [here](https://code.google.com/p/chromium/codesearch#chromium/src/net/base/net_error_list.h).
Apart from the certificate error codes, the following special codes can be used.
* `0` - Indicates success and disables Certificate Transparency verification.
* `-2` - Indicates failure.
* `-3` - Uses the verification result from chromium.
Sets the certificate verify proc for `session`, the `proc` will be called with
`proc(request, callback)` whenever a server certificate
verification is requested. Calling `callback(0)` accepts the certificate,
calling `callback(-2)` rejects it.
Calling `setCertificateVerifyProc(null)` will revert back to default certificate
verify proc.
```javascript
const { BrowserWindow } = require('electron')
let win = new BrowserWindow()
win.webContents.session.setCertificateVerifyProc((request, callback) => {
const { hostname } = request
if (hostname === 'github.com') {
callback(0)
} else {
callback(-2)
}
})
```
#### `ses.setPermissionRequestHandler(handler)`
* `handler` Function | null
* `webContents` [WebContents](web-contents.md) - WebContents requesting the permission. Please note that if the request comes from a subframe you should use `requestingUrl` to check the request origin.
* `permission` String - Enum of 'media', 'geolocation', 'notifications', 'midiSysex',
'pointerLock', 'fullscreen', 'openExternal'.
* `callback` Function
* `permissionGranted` Boolean - Allow or deny the permission.
* `details` Object - Some properties are only available on certain permission types.
* `externalURL` String (optional) - The url of the `openExternal` request.
* `mediaTypes` String[] (optional) - The types of media access being requested, elements can be `video`
or `audio`
* `requestingUrl` String - The last URL the requesting frame loaded
* `isMainFrame` Boolean - Whether the frame making the request is the main frame
Sets the handler which can be used to respond to permission requests for the `session`.
Calling `callback(true)` will allow the permission and `callback(false)` will reject it.
To clear the handler, call `setPermissionRequestHandler(null)`.
```javascript
const { session } = require('electron')
session.fromPartition('some-partition').setPermissionRequestHandler((webContents, permission, callback) => {
if (webContents.getURL() === 'some-host' && permission === 'notifications') {
return callback(false) // denied.
}
callback(true)
})
```
#### `ses.setPermissionCheckHandler(handler)`
* `handler` Function<Boolean> | null
* `webContents` [WebContents](web-contents.md) - WebContents checking the permission. Please note that if the request comes from a subframe you should use `requestingUrl` to check the request origin.
* `permission` String - Enum of 'media'.
* `requestingOrigin` String - The origin URL of the permission check
* `details` Object - Some properties are only available on certain permission types.
* `securityOrigin` String - The security orign of the `media` check.
* `mediaType` String - The type of media access being requested, can be `video`,
`audio` or `unknown`
* `requestingUrl` String - The last URL the requesting frame loaded
* `isMainFrame` Boolean - Whether the frame making the request is the main frame
Sets the handler which can be used to respond to permission checks for the `session`.
Returning `true` will allow the permission and `false` will reject it.
To clear the handler, call `setPermissionCheckHandler(null)`.
```javascript
const { session } = require('electron')
session.fromPartition('some-partition').setPermissionCheckHandler((webContents, permission) => {
if (webContents.getURL() === 'some-host' && permission === 'notifications') {
return false // denied
}
return true
})
```
#### `ses.clearHostResolverCache()`
Returns `Promise<void>` - Resolves when the operation is complete.
Clears the host resolver cache.
#### `ses.allowNTLMCredentialsForDomains(domains)`
* `domains` String - A comma-separated list of servers for which
integrated authentication is enabled.
Dynamically sets whether to always send credentials for HTTP NTLM or Negotiate
authentication.
```javascript
const { session } = require('electron')
// consider any url ending with `example.com`, `foobar.com`, `baz`
// for integrated authentication.
session.defaultSession.allowNTLMCredentialsForDomains('*example.com, *foobar.com, *baz')
// consider all urls for integrated authentication.
session.defaultSession.allowNTLMCredentialsForDomains('*')
```
#### `ses.setUserAgent(userAgent[, acceptLanguages])`
* `userAgent` String
* `acceptLanguages` String (optional)
Overrides the `userAgent` and `acceptLanguages` for this session.
The `acceptLanguages` must a comma separated ordered list of language codes, for
example `"en-US,fr,de,ko,zh-CN,ja"`.
This doesn't affect existing `WebContents`, and each `WebContents` can use
`webContents.setUserAgent` to override the session-wide user agent.
#### `ses.getUserAgent()`
Returns `String` - The user agent for this session.
#### `ses.getBlobData(identifier)`
* `identifier` String - Valid UUID.
Returns `Promise<Buffer>` - resolves with blob data.
#### `ses.downloadURL(url)`
* `url` String
Initiates a download of the resource at `url`.
The API will generate a [DownloadItem](download-item.md) that can be accessed
with the [will-download](#event-will-download) event.
**Note:** This does not perform any security checks that relate to a page's origin,
unlike [`webContents.downloadURL`](web-contents.md#contentsdownloadurlurl).
#### `ses.createInterruptedDownload(options)`
* `options` Object
* `path` String - Absolute path of the download.
* `urlChain` String[] - Complete URL chain for the download.
* `mimeType` String (optional)
* `offset` Integer - Start range for the download.
* `length` Integer - Total length of the download.
* `lastModified` String (optional) - Last-Modified header value.
* `eTag` String (optional) - ETag header value.
* `startTime` Double (optional) - Time when download was started in
number of seconds since UNIX epoch.
Allows resuming `cancelled` or `interrupted` downloads from previous `Session`.
The API will generate a [DownloadItem](download-item.md) that can be accessed with the [will-download](#event-will-download)
event. The [DownloadItem](download-item.md) will not have any `WebContents` associated with it and
the initial state will be `interrupted`. The download will start only when the
`resume` API is called on the [DownloadItem](download-item.md).
#### `ses.clearAuthCache()`
Returns `Promise<void>` - resolves when the sessions HTTP authentication cache has been cleared.
#### `ses.setPreloads(preloads)`
* `preloads` String[] - An array of absolute path to preload scripts
Adds scripts that will be executed on ALL web contents that are associated with
this session just before normal `preload` scripts run.
#### `ses.getPreloads()`
Returns `String[]` an array of paths to preload scripts that have been
registered.
#### `ses.setSpellCheckerLanguages(languages)`
* `languages` String[] - An array of language codes to enable the spellchecker for.
The built in spellchecker does not automatically detect what language a user is typing in. In order for the
spell checker to correctly check their words you must call this API with an array of language codes. You can
get the list of supported language codes with the `ses.availableSpellCheckerLanguages` property.
**Note:** On macOS the OS spellchecker is used and will detect your language automatically. This API is a no-op on macOS.
#### `ses.getSpellCheckerLanguages()`
Returns `String[]` - An array of language codes the spellchecker is enabled for. If this list is empty the spellchecker
will fallback to using `en-US`. By default on launch if this setting is an empty list Electron will try to populate this
setting with the current OS locale. This setting is persisted across restarts.
**Note:** On macOS the OS spellchecker is used and has it's own list of languages. This API is a no-op on macOS.
#### `ses.setSpellCheckerDictionaryDownloadURL(url)`
* `url` String - A base URL for Electron to download hunspell dictionaries from.
By default Electron will download hunspell dictionaries from the Chromium CDN. If you want to override this
behavior you can use this API to point the dictionary downloader at your own hosted version of the hunspell
dictionaries. We publish a `hunspell_dictionaries.zip` file with each release which contains the files you need
to host here.
If the files present in `hunspell_dictionaries.zip` are available at `https://example.com/dictionaries/language-code.bdic`
then you should call this api with `ses.setSpellCheckerDictionaryDownloadURL('https://example.com/dictionaries/')`. Please
note the trailing slash. The URL to the dictionaries is formed as `${url}${filename}`.
**Note:** On macOS the OS spellchecker is used and therefore we do not download any dictionary files. This API is a no-op on macOS.
#### `ses.listWordsInSpellCheckerDictionary()`
Returns `Promise<String[]>` - An array of all words in app's custom dictionary.
Resolves when the full dictionary is loaded from disk.
#### `ses.addWordToSpellCheckerDictionary(word)`
* `word` String - The word you want to add to the dictionary
Returns `Boolean` - Whether the word was successfully written to the custom dictionary.
**Note:** On macOS and Windows 10 this word will be written to the OS custom dictionary as well
#### `ses.removeWordFromSpellCheckerDictionary(word)`
* `word` String - The word you want to remove from the dictionary
Returns `Boolean` - Whether the word was successfully removed from the custom dictionary.
**Note:** On macOS and Windows 10 this word will be removed from the OS custom dictionary as well
#### `ses.loadExtension(path)`
* `path` String - Path to a directory containing an unpacked Chrome extension
Returns `Promise<Extension>` - resolves when the extension is loaded.
This method will raise an exception if the extension could not be loaded. If
there are warnings when installing the extension (e.g. if the extension
requests an API that Electron does not support) then they will be logged to the
console.
Note that Electron does not support the full range of Chrome extensions APIs.
See [Supported Extensions APIs](extensions.md#supported-extensions-apis) for
more details on what is supported.
Note that in previous versions of Electron, extensions that were loaded would
be remembered for future runs of the application. This is no longer the case:
`loadExtension` must be called on every boot of your app if you want the
extension to be loaded.
```js
const { app, session } = require('electron')
const path = require('path')
app.on('ready', async () => {
await session.defaultSession.loadExtension(path.join(__dirname, 'react-devtools'))
// Note that in order to use the React DevTools extension, you'll need to
// download and unzip a copy of the extension.
})
```
This API does not support loading packed (.crx) extensions.
**Note:** This API cannot be called before the `ready` event of the `app` module
is emitted.
**Note:** Loading extensions into in-memory (non-persistent) sessions is not
supported and will throw an error.
#### `ses.removeExtension(extensionId)`
* `extensionId` String - ID of extension to remove
Unloads an extension.
**Note:** This API cannot be called before the `ready` event of the `app` module
is emitted.
#### `ses.getExtension(extensionId)`
* `extensionId` String - ID of extension to query
Returns `Extension` | `null` - The loaded extension with the given ID.
**Note:** This API cannot be called before the `ready` event of the `app` module
is emitted.
#### `ses.getAllExtensions()`
Returns `Extension[]` - A list of all loaded extensions.
**Note:** This API cannot be called before the `ready` event of the `app` module
is emitted.
### Instance Properties
The following properties are available on instances of `Session`:
#### `ses.availableSpellCheckerLanguages` _Readonly_
A `String[]` array which consists of all the known available spell checker languages. Providing a language
code to the `setSpellCheckerLanaguages` API that isn't in this array will result in an error.
#### `ses.cookies` _Readonly_
A [`Cookies`](cookies.md) object for this session.
#### `ses.serviceWorkers` _Readonly_
A [`ServiceWorkers`](service-workers.md) object for this session.
#### `ses.webRequest` _Readonly_
A [`WebRequest`](web-request.md) object for this session.
#### `ses.protocol` _Readonly_
A [`Protocol`](protocol.md) object for this session.
```javascript
const { app, session } = require('electron')
const path = require('path')
app.whenReady().then(() => {
const protocol = session.fromPartition('some-partition').protocol
protocol.registerFileProtocol('atom', (request, callback) => {
let url = request.url.substr(7)
callback({ path: path.normalize(`${__dirname}/${url}`) })
}, (error) => {
if (error) console.error('Failed to register protocol')
})
})
```
#### `ses.netLog` _Readonly_
A [`NetLog`](net-log.md) object for this session.
```javascript
const { app, session } = require('electron')
app.whenReady().then(async () => {
const netLog = session.fromPartition('some-partition').netLog
netLog.startLogging('/path/to/net-log')
// After some network events
const path = await netLog.stopLogging()
console.log('Net-logs written to', path)
})
```