Перейти к основному содержанию

· 5 мин. прочитано

Electron 22.0.0 вышел! It includes a new utility process API, updates for Windows 7/8/8.1 support, and upgrades to Chromium 108, V8 10.8, and Node.js 16.17.1. Read below for more details!


Команда Electron рада объявить о выпуске Electron 22.0.0! You can install it with npm via npm install [email protected] or download it from our releases website. Continue reading for details about this release.

If you have any feedback, please share it with us on Twitter, or join our community Discord! Bugs and feature requests can be reported in Electron's issue tracker.

Notable Changes

Stack Changes

Highlighted Features

UtilityProcess API #36089

The new UtilityProcess main process module allows the creation of a lightweight Chromium child process with only Node.js integration while also allowing communication with a sandboxed renderer using MessageChannel. The API was designed based on Node.js child_process.fork to allow for easier transition, with one primary difference being that the entry point modulePath must be from within the packaged application to allow only for trusted scripts to be loaded. Additionally the module prevents establishing communication channels with renderers by default, upholding the contract in which the main process is the only trusted process in the application.

You can read more about the new UtilityProcess API in our docs here.

Windows 7/8/8.1 Support Update

Electron 22 will be the last Electron major version to support Windows 7/8/8.1. Electron follows the planned Chromium deprecation policy, which will deprecate Windows 7/8/8.1 support in Chromium 109 (read more here).

Windows 7/8/8.1 will not be supported in Electron 23 and later major releases.

Additional Highlighted Changes

  • Added support for Web Bluetooth pin pairing on Linux and Windows. #35416
  • Added LoadBrowserProcessSpecificV8Snapshot as a new fuse that will let the main/browser process load its v8 snapshot from a file at browser_v8_context_snapshot.bin. Any other process will use the same path as is used today. #35266
  • Added WebContents.opener to access window opener and webContents.fromFrame(frame) to get the WebContents corresponding to a WebFrameMain instance. #35140
  • Added support for navigator.mediaDevices.getDisplayMedia via a new session handler, ses.setDisplayMediaRequestHandler. #30702

Критические изменения в API

Below are breaking changes introduced in Electron 22. You can read more about these changes and future changes on the Planned Breaking Changes page.

Obsoleto: webContents.incrementCapturerCount(stayHidden, stayAwake)

webContents.incrementCapturerCount(stayHidden, stayAwake) has been deprecated. It is now automatically handled by webContents.capturePage when a page capture completes.

const w = new BrowserWindow({ show: false })

- w.webContents.incrementCapturerCount()
- w.capturePage().then(image => {
- console.log(image.toDataURL())
- w.webContents.decrementCapturerCount()
- })

+ w.capturePage().then(image => {
+ console.log(image.toDataURL())
+ })

Obsoleto: webContents.decrementCapturerCount(stayHidden, stayAwake)

webContents.decrementCapturerCount(stayHidden, stayAwake) has been deprecated. It is now automatically handled by webContents.capturePage when a page capture completes.

const w = new BrowserWindow({ show: false })

- w.webContents.incrementCapturerCount()
- w.capturePage().then(image => {
- console.log(image.toDataURL())
- w.webContents.decrementCapturerCount()
- })

+ w.capturePage().then(image => {
+ console.log(image.toDataURL())
+ })

Removed: WebContents new-window event

The new-window event of WebContents has been removed. Заменяется на webContents.setWindowOpenHandler().

- webContents.on('new-window', (event) => {
- event.preventDefault()
- })

+ webContents.setWindowOpenHandler((details) => {
+ return { action: 'deny' }
+ })

Deprecated: BrowserWindow scroll-touch-* events

The scroll-touch-begin, scroll-touch-end and scroll-touch-edge events on BrowserWindow are deprecated. Instead, use the newly available input-event event on WebContents.

// Deprecated
- win.on('scroll-touch-begin', scrollTouchBegin)
- win.on('scroll-touch-edge', scrollTouchEdge)
- win.on('scroll-touch-end', scrollTouchEnd)

// Replace with
+ win.webContents.on('input-event', (_, event) => {
+ if (event.type === 'gestureScrollBegin') {
+ scrollTouchBegin()
+ } else if (event.type === 'gestureScrollUpdate') {
+ scrollTouchEdge()
+ } else if (event.type === 'gestureScrollEnd') {
+ scrollTouchEnd()
+ }
+ })

End of Support for 19.x.y

Electron 19.x.y has reached end-of-support as per the project's support policy. Developers and applications are encouraged to upgrade to a newer version of Electron.

E19 (May'22)E20 (Aug'22)E21 (Sep'22)E22 (Nov'22)E23 (Jan'23)
19.x.y20.x.y21.x.y22.x.y23.x.y
18.x.y19.x.y20.x.y21.x.y22.x.y
17.x.y18.x.y19.x.y20.x.y21.x.y

Что дальше

The Electron project will pause for the the month of December 2022, and return in January 2023. More information can be found in the December shutdown blog post.

In the short term, you can expect the team to continue to focus on keeping up with the development of the major components that make up Electron, including Chromium, Node, and V8.

You can find Electron's public timeline here.

More information about future changes can be found on the Planned Breaking Changes page.

· 2 мин. прочитано

Electron will end support of Windows 7, Windows 8 and Windows 8.1 beginning in Electron 23.


In line with Chromium’s deprecation policy, Electron will end support of Windows 7, Windows 8 and Windows 8.1 beginning in Electron 23. This matches Microsoft's end of support for Windows 7 ESU and Windows 8.1 extended on January 10th, 2023.

Electron 22 will be the last Electron major version to support Windows versions older than 10. Windows 7/8/8.1 will not be supported in Electron 23 and later major releases. Older versions of Electron will continue to function on Windows 7, and we will continue to release patches for Electron the 22.x series until May 30 2023, when Electron will end support for 22.x (according to our support timeline).

Why deprecate?

Electron follows the planned Chromium deprecation policy, which will deprecate support in Chromium 109 (read more about Chromium's timeline here). Electron 23 will contain Chromium 110, which won’t support older versions of Windows.

Electron 22, which contains Chromium 108, will thus be the last supported version.

Deprecation timeline

The following is our planned deprecation timeline:

  • December 2022: The Electron team is entering a quiet period for the holidays
  • January 2023: Windows 7 & 8 related issues are accepted for all supported release branches.
  • February 7 2023: Electron 23 is released.
  • February 8 2023 - May 29 2023: Electron will continue to accept fixes for supported lines older than Electron 23.
  • May 30 2023: Electron 22 reaches the end of its support cycle.

What this means for developers:

  • The Electron team will accept issues and fixes related to Windows 7/8/8.1 for stable supported lines, until each line reaches the end of its support cycle.
    • This specifically applies to Electron 22, Electron 21 and Electron 20.
  • New issues related to Windows 7/8/8.1 will be accepted for Electron versions older than Electron 23.
    • New issues will not be accepted for any newer release lines.
  • Once Electron 22 has reached the end of its support cycle, all existing issues related to Windows 7/8/8.1 will be closed.

What's next

Please feel free to write to us at [email protected] if you have any questions or concerns. You can also find community support in our official Electron Discord.

· Одна мин. чтения

The Electron project will pause for the month of December 2022, then return to full speed in January 2023.

via GIPHY


What will be the same in December

  1. Zero-day and other major security-related releases will be published as necessary. Security incidents should be reported via SECURITY.md.
  2. Code of Conduct reports and moderation will continue.

What will be different in December

  1. No new Stable releases in December. No Nightly and Alpha releases for the last two weeks of December.
  2. With few exceptions, no pull request reviews or merges.
  3. No issue tracker updates on any repositories.
  4. No Discord debugging help from maintainers.
  5. No social media content updates.

Why is this happening?

With the success of December Quiet Month 2021, we wanted to bring it back for 2022. December continues to be a quiet month for most companies, so we want to give our maintainers a chance to recharge. Everyone is looking forward to 2023, and we expect good things to come! We encourage other projects to consider similar measures.

· 6 мин. прочитано

Мы рады сообщить, что вышел Electron Forge v6.0.0! This release marks the first major release of Forge since 2018 and moves the project from electron-userland into the main electron organization on Github.

Продолжайте чтение, чтобы узнать о новинках и как ваше приложение может внедрить Electron Forge!

Что такое Electron Forge?

Electron Forge is a tool for packaging and distributing Electron applications. It unifies Electron's build tooling ecosystem into a single extensible interface so that anyone can jump right into making Electron apps.

Highlight features include:

  • 📦 Application packaging and code signing
  • 🚚 Customizable installers on Windows, macOS, and Linux (DMG, deb, MSI, PKG, AppX, etc.)
  • ☁️ Automated publishing flow for cloud providers (GitHub, S3, Bitbucket, etc.)
  • ⚡️ Easy-to-use boilerplate templates for webpack and TypeScript
  • ⚙️ Native Node.js module support
  • 🔌 Extensible JavaScript plugin API
Further reading

Visit the Why Electron Forge explainer document to learn more about Forge's philosophy and architecture.

What's new in v6?

Completely rewritten

From v1 to v5, Electron Forge was based on the now-discontinued electron-compile project. Forge 6 is a complete rewrite of the project with a new modular architecture that can be extended to meet any Electron application's needs.

In the past few years, Forge v6.0.0-beta has achieved feature parity with v5 and code churn has slowed down dramatically, making the tool ready for general adoption.

Don't install the wrong package

For versions 5 and below, Electron Forge was published to the electron-forge package on npm. Starting with the v6 rewrite, Forge is instead structured as a monorepo project with many smaller projects.

Officially supported

Historically, Electron maintainers have been unopinionated about build tooling, leaving the task to various community packages. However, with Electron maturing as a project, it has become harder for new Electron developers to understand which tools they need to build and distribute their apps.

To help guide Electron developers in the distribution process, we have have decided to make Forge the official batteries-included build pipeline for Electron.

Over the past year, we have been slowly integrating Forge into the official Electron documentation, and we have recently moved Forge over from its old home in electron-userland/electron-forge to the electron/forge repo. Now, we are finally ready to release Electron Forge to a general audience!

Начало работы

Initializing a new Forge project

Scaffolding a new Electron Forge project can be done using the create-electron-app CLI script.

yarn create electron-app my-app --template=webpack
cd my-app
yarn start

The script will create an Electron project in the my-app folder with completely JavaScript bundling and a preconfigured build pipeline.

For more info, see the Getting Started guide in the Forge docs.

First-class webpack support

The above snippet uses Forge's Webpack Template, which we recommend as a starting point for new Electron projects. This template is built around the @electron-forge/plugin-webpack plugin, which integrates webpack with Electron Forge in a few ways, including:

  • enhancing local dev flow with webpack-dev-server, including support for HMR in the renderer;
  • handling build logic for webpack bundles before application packaging; and
  • adding support for Native Node modules in the webpack bundling process.

If you need TypeScript support, consider using the Webpack + TypeScript Template instead.

Importing an existing project

The Electron Forge CLI also contains an import command for existing Electron projects.

cd my-app
yarn add --dev @electron-forge/cli
yarn electron-forge import

When you use the import command, Electron Forge will add a few core dependencies and create a new forge.config.js configuration. If you have any existing build tooling (e.g. Electron Packager, Electron Builder, or Forge 5), it will try to migrate as many settings as possible. Some of your existing configuration may need to be migrated manually.

Manual migration details can be found in the Forge import documentation. If you need help, please stop by our Discord server!

Why switch to Forge?

If you already have tooling for packaging and publishing your Electron app, the benefits associated with adopting Electron Forge can still outweigh the initial switching cost.

We believe there are two main benefits to using Forge:

  1. Forge receives new features for application building as soon as they are supported in Electron. In this case, you won't need to wire in new tooling support yourself, or wait for that support to be eventually implemented by other packages before upgrading. For recent examples, see macOS universal binaries and ASAR integrity checking.

  2. Forge's multi-package architecture makes it easy to understand and extend. Since Forge is made up of many smaller packages with clear responsibilities, it is easier to follow code flow. In addition, Forge's extensible API design means that you can write your own additional build logic separate from the provided configuration options for advanced use cases. For more details on writing custom Forge plugins, makers, and publishers, see the Extending Electron Forge section of the docs.

Важные изменения

Forge 6 has spent a long time in the beta phase, and its release cadence has gradually slowed down. However, we have accelerated development in the second half of 2022 and used the last few releases to push some final breaking changes before the v6.0.0 stable release.

If you are an Electron Forge 6 beta user, see the v6.0.0 GitHub release notes for a list of breaking changes made in recent betas (>=6.0.0-beta.65).

A complete list of changes and commits can be found in the repo's CHANGELOG.md.

Submit your feedback!

Tell us what you need! The Electron Forge team is always looking to build the project to better suit its users.

You can help us improve Electron Forge by submitting feature requests, posting issues, or just letting us know your feedback! You can also join us in the official Electron Discord server, where there is a dedicated channel for Electron Forge discussion.

If you want to give any feedback on the Forge docs at https://electronforge.io, we have a GitBook instance synced to the electron-forge/electron-forge-docs repo.

· 5 мин. прочитано

Last month, Electron’s maintainer group met up in Vancouver, Canada to discuss the direction of the project for 2023 and beyond. Over four days in a conference room, core maintainers and invited collaborators discussed new initiatives, maintenance pain points, and general project health.

Group Photo! Taken by @groundwater.

Going forward, the team will still be fully dedicated to releasing regular and rapid Chromium upgrades, fixing bugs, and making Electron more secure and performant for everyone. We also have a few exciting projects in the works we would love to share with the community!

Transformative new APIs

Major API proposals in the Electron project that require consensus go through a Request for Comments (RFC) process, which gets reviewed by members of our API Working Group.

This year, we have driven forward two major proposals that have the potential to unlock a new dimension of capabilities for Electron apps. These proposals are highly experimental, but here’s a sneak peek of what to expect!

New native addon enhancements (C APIs)

This proposal outlines a new layer of Electron C APIs that will allow app developers to write their own Native Node Addons that interface with Electron’s internal resources, similar to Node’s own Node-API. More information about the proposed new API can be found here.

Example: Supercharging apps with Chromium resources

Many Electron apps maintain their own forks to interact directly with Chromium internals that would otherwise be inaccessible with vanilla (unmodified) Electron. By exposing these resources in the C API layer, this code can instead live as a native module alongside Electron, potentially reducing app developer maintenance burden.

Exposing Chromium’s UI layer (Views API)

Under the hood, the non-website parts of Chrome’s user interface (UI), such as toolbars, tabs, or buttons, are built with a framework called Views. The Views API proposal introduces parts of this framework as JavaScript classes in Electron, with the eventual goal of allowing developers to create non-web UI elements to their Electron applications. This will prevent apps from having to hack together web contents.

The groundwork to make this new set of APIs possible is currently in progress. Here are a few of the first things you can expect in the near future.

Example: Refactoring the window model with WebContentsView

Our first planned change is to expose Chrome’s WebContentsView to Electron’s API surface, which will be the successor to our existing BrowserView API (which, despite the name, is Electron-specific code unrelated to Chromium Views). With WebContentsView exposed, we will have a reusable View object that can display web contents, opening the door to making the BrowserWindow class pure JavaScript and eliminating even more code complexity.

Although this change doesn’t provide a lot of new functionality to app developers, it is a large refactor that eliminates a lot of code under the hood, simplifying Chromium upgrades and reducing the risk of new bugs appearing between major versions.

If you’re an Electron developer using BrowserViews in your app: don’t worry, we haven’t forgotten about you! We plan on making the existing BrowserView class a shim for WebContentsView to provide a buffer as you transition to the newer APIs.

See: electron/electron#35658

Example: Scrollable web contents with ScrollView

Our friends at Stack have been driving an initiative to expose the Chromium ScrollView component to Electron’s API. With this new API, any child View component can be made scrollable horizontally or vertically.

Although this new API fulfills a single smaller functionality, the team’s eventual goal is to build a set of utility View components that can be used as a toolkit to build more complex non-HTML interfaces.

Getting involved

Are you an Electron app developer interested in either of these API proposals? Although we’re not quite ready to receive additional RFCs, please stay tuned for more details in the future!

Electron Forge v6 stable release

Since the framework’s inception, Electron’s build tooling ecosystem has been largely community-driven and has consisted of many small single-purpose packages (e.g. electron-winstaller, electron-packager, electron-notarize, electron-osx-sign). Although these tools work well, it’s intimidating for users to piece together a working build pipeline.

To help build a friendlier experience for Electron developers, we built Electron Forge, an all-in-one solution that combines all this existing tooling into a single interface. Although Forge has been in development since 2017, the project has lain dormant for the last few years. However, given community feedback on the state of build tooling in Electron, we have been hard at work on releasing the next-gen stable major version of Forge.

Electron Forge 6 comes with first-class TypeScript and Webpack support, as well as an extensible API that allows developers to create their own plugins and installers.

Stay tuned: announcement coming soon

If you’re interested in building a project with Forge or building templates or plugins with Forge’s extensible third-party APIs, stay tuned for our official announcement on the Forge v6 stable release sometime this month!

What’s next?

Aside from the above, the team is always thinking of a bunch of exploratory projects to make the Electron experience better for app developers and end users. Updater tooling, API review processes, and enhanced documentation are other things we are experimenting with. We hope to have more news to share in the near future!

· 3 мин. прочитано

Electron 21.0.0 вышел! Он включает обновления Chromium 106, V8 10.6 и Node.js 16.16.0. Read below for more details!


Команда Electron рада объявить о выпуске Electron 21.0.0! You can install it with npm via npm install [email protected] or download it from our releases website. Continue reading for details about this release.

If you have any feedback, please share it with us on Twitter, or join our community Discord! Bugs and feature requests can be reported in Electron's issue tracker.

Notable Changes

Stack Changes

New Features

  • Added webFrameMain.origin. #35534
  • Added new WebContents.ipc and WebFrameMain.ipc APIs. #35231
  • Added support for panel-like behavior. Window can float over full-screened apps. #34388
  • Added support for push notifications from APNs for macOS apps. #33574

Breaking & API Changes

Below are breaking changes introduced in Electron 21.

V8 Memory Cage Enabled

Electron 21 enables V8 sandboxed pointers, following Chrome's decision to do the same in Chrome 103. This has some implications for native modules. This feature has performance and security benefits, but also places some new restrictions on native modules, e.g. use of ArrayBuffers that point to external ("off-heap") memory. Please see this blog post for more information. #34724

Refactored webContents.printToPDF

Refactored webContents.printToPDF to align with Chromium's headless implementation. See #33654 for more information.

More information about these and future changes can be found on the Planned Breaking Changes page.

End of Support for 18.x.y

Electron 18.x.y has reached end-of-support as per the project's support policy. Developers and applications are encouraged to upgrade to a newer version of Electron.

E18 (Mar'22)E19 (May'22)E20 (Aug'22)E21 (Sep'22)E22 (Dec'22)
18.x.y19.x.y20.x.y21.x.y22.x.y
17.x.y18.x.y19.x.y20.x.y21.x.y
16.x.y17.x.y18.x.y19.x.y20.x.y

Что дальше

In the short term, you can expect the team to continue to focus on keeping up with the development of the major components that make up Electron, including Chromium, Node, and V8.

You can find Electron's public timeline here.

More information about future changes can be found on the Planned Breaking Changes page.

· 4 мин. прочитано

Electron 20.0.0 вышел! Он включает обновления Chromium 104, V8 10.4 и Node.js 16.15.0. Read below for more details!


Команда Electron рада объявить о выпуске Electron 20.0.0! You can install it with npm via npm install [email protected] or download it from our releases website. Continue reading for details about this release and please share any feedback you have!

Notable Changes

New Features

  • Added immersive dark mode on Windows. #34549
  • Added support for panel-like behavior. Window can float over full-screened apps. #34665
  • Updated Windows Control Overlay buttons to look and feel more native on Windows 11. #34888
  • Renderers are now sandboxed by default unless nodeIntegration: true or sandbox: false is specified. #35125
  • Added safeguards when building native modules with nan. #35160

Stack Changes

Breaking & API Changes

Below are breaking changes introduced in Electron 20. More information about these and future changes can be found on the Planned Breaking Changes page.

Default Changed: renderers without nodeIntegration: true are sandboxed by default

Previously, renderers that specified a preload script defaulted to being unsandboxed. This meant that by default, preload scripts had access to Node.js. In Electron 20, this default has changed. Beginning in Electron 20, renderers will be sandboxed by default, unless nodeIntegration: true or sandbox: false is specified.

If your preload scripts do not depend on Node, no action is needed. If your preload scripts do depend on Node, either refactor them to remove Node usage from the renderer, or explicitly specify sandbox: false for the relevant renderers.

Fixed: spontaneous crashing in nan native modules

In Electron 20, we changed two items related to native modules:

  1. V8 headers now use c++17 by default. This flag was added to electron-rebuild.
  2. We fixed an issue where a missing include would cause spontaneous crashing in native modules that depended on nan.

For the most stability, we recommend using node-gyp >=8.4.0 and electron-rebuild >=3.2.9 when rebuilding native modules, particularly modules that depend on nan. See electron #35160 and node-gyp #2497 for more information.

Removed: .skipTaskbar on Linux

On X11, skipTaskbar sends a _NET_WM_STATE_SKIP_TASKBAR message to the X11 window manager. There is not a direct equivalent for Wayland, and the known workarounds have unacceptable tradeoffs (e.g. Window.is_skip_taskbar in GNOME requires unsafe mode), so Electron is unable to support this feature on Linux.

End of Support for 17.x.y

Electron 17.x.y has reached end-of-support as per the project's support policy. Developers and applications are encouraged to upgrade to a newer version of Electron.

E18 (Mar'22)E19 (May'22)E20 (Aug'22)E21 (Sep'22)E22 (Dec'22)
18.x.y19.x.y20.x.y21.x.y22.x.y
17.x.y18.x.y19.x.y20.x.y21.x.y
16.x.y17.x.y18.x.y19.x.y20.x.y
15.x.y--------

Что дальше

In the short term, you can expect the team to continue to focus on keeping up with the development of the major components that make up Electron, including Chromium, Node, and V8. Although we are careful not to make promises about release dates, our plan is to release new major versions of Electron with new versions of those components approximately every 2 months.

You can find Electron's public timeline here.

More information about future changes can be found on the Planned Breaking Changes page.

· 7 мин. прочитано

Electron 21 and later will have the V8 Memory Cage enabled, with implications for some native modules.


Update (2022/11/01)

To track ongoing discussion about native module usage in Electron 21+, see electron/electron#35801.

In Electron 21, we will be enabling V8 sandboxed pointers in Electron, following Chrome's decision to do the same in Chrome 103. This has some implications for native modules. Also, we previously enabled a related technology, pointer compression, in Electron 14. We didn't talk about it much then, but pointer compression has implications for the maximum V8 heap size.

These two technologies, when enabled, are significantly beneficial for security, performance and memory usage. However, there are some downsides to enabling them, too.

The main downside of enabling sandboxed pointers is that ArrayBuffers which point to external ("off-heap") memory are no longer allowed. This means that native modules which rely on this functionality in V8 will need to be refactored to continue working in Electron 20 and later.

The main downside of enabling pointer compression is that the V8 heap is limited to a maximum size of 4GB. The exact details of this are a little complicated—for example, ArrayBuffers are counted separately from the rest of the V8 heap, but have their own limits.

The Electron Upgrades Working Group believes that the benefits of pointer compression and the V8 memory cage outweigh the downsides. There are three main reasons for doing so:

  1. It keeps Electron closer to Chromium. The less Electron diverges from Chromium in complex internal details such as V8 configuration, the less likely we are to accidentally introduce bugs or security vulnerabilities. Chromium's security team is formidable, and we want to make sure we are taking advantage of their work. Further, if a bug only affects configurations that aren't used in Chromium, fixing it is not likely to be a priority for the Chromium team.
  2. It performs better. Pointer compression reduces V8 heap size by up to 40% and improves CPU and GC performance by 5%–10%. For the vast majority of Electron applications which won't bump into the 4GB heap size limit and don't use native modules that require external buffers, these are significant performance wins.
  3. It's more secure. Some Electron apps run untrusted JavaScript (hopefully following our security recommendations!), and for those apps, having the V8 memory cage enabled protects them from a large class of nasty V8 vulnerabilities.

Lastly, there are workarounds for apps that really need a larger heap size. For example, it is possible to include a copy of Node.js with your app, which is built with pointer compression disabled, and move the memory-intensive work to a child process. Though somewhat complicated, it is also possible to build a custom version of Electron with pointer compression disabled, if you decide you want a different trade-off for your particular use case. And lastly, in the not-too-distant future, wasm64 will allow apps built with WebAssembly both on the Web and in Electron to use significantly more than 4GB of memory.


FAQ

How will I know if my app is impacted by this change?

Attempting to wrap external memory with an ArrayBuffer will crash at runtime in Electron 20+.

If you don't use any native Node modules in your app, you're safe—there's no way to trigger this crash from pure JS. This change only affects native Node modules which allocate memory outside of the V8 heap (e.g. using malloc or new) and then wrap the external memory with an ArrayBuffer. This is a fairly rare use case, but some modules do use this technique, and such modules will need to be refactored in order to be compatible with Electron 20+.

How can I measure how much V8 heap memory my app is using to know if I'm close to the 4GB limit?

In the renderer process, you can use performance.memory.usedJSHeapSize, which will return the V8 heap usage in bytes. In the main process, you can use process.memoryUsage().heapUsed, which is comparable.

What is the V8 memory cage?

Some documents refer to it as the "V8 sandbox", but that term is easily confusable with other kinds of sandboxing that happen in Chromium, so I'll stick to the term "memory cage".

There's a fairly common kind of V8 exploit that goes something like this:

  1. Find a bug in V8's JIT engine. JIT engines analyze code in order to be able to omit slow runtime type checks and produce fast machine code. Sometimes logic errors mean it gets this analysis wrong, and omits a type check that it actually needed—say, it thinks x is a string, but in fact it's an object.
  2. Abuse this confusion to overwrite some bit of memory inside the V8 heap, for instance, the pointer to the beginning of an ArrayBuffer.
  3. Now you have an ArrayBuffer that points wherever you like, so you can read and write any memory in the process, even memory that V8 normally doesn't have access to.

The V8 memory cage is a technique designed to categorically prevent this kind of attack. The way this is accomplished is by not storing any pointers in the V8 heap. Instead, all references to other memory inside the V8 heap are stored as offsets from the beginning of some reserved region. Then, even if an attacker manages to corrupt the base address of an ArrayBuffer, for instance by exploiting a type confusion error in V8, the worst they can do is read and write memory inside the cage, which they could likely already do anyway. There's a lot more available to read on how the V8 memory cage works, so I won't go into further detail here—the best place to start reading is probably the high-level design doc from the Chromium team.

Мне нужен рефакторинг нативного модуля Node для поддержки Electron 21+. Как это сделать?

There are two ways to go about refactoring a native module to be compatible with the V8 memory cage. The first is to copy externally-created buffers into the V8 memory cage before passing them to JavaScript. This is generally a simple refactor, but it can be slow when the buffers are large. The other approach is to use V8's memory allocator to allocate memory which you intend to eventually pass to JavaScript. This is a bit more involved, but will allow you to avoid the copy, meaning better performance for large buffers.

To make this more concrete, here's an example N-API module that uses external array buffers:

// Create some externally-allocated buffer.
// |create_external_resource| allocates memory via malloc().
size_t length = 0;
void* data = create_external_resource(&length);
// Wrap it in a Buffer--will fail if the memory cage is enabled!
napi_value result;
napi_create_external_buffer(
env, length, data,
finalize_external_resource, NULL, &result);

This will crash when the memory cage is enabled, because data is allocated outside the cage. Refactoring to instead copy the data into the cage, we get:

size_t length = 0;
void* data = create_external_resource(&length);
// Create a new Buffer by copying the data into V8-allocated memory
napi_value result;
void* copied_data = NULL;
napi_create_buffer_copy(env, length, data, &copied_data, &result);
// If you need to access the new copy, |copied_data| is a pointer
// to it!

This will copy the data into a newly-allocated memory region that is inside the V8 memory cage. Optionally, N-API can also provide a pointer to the newly-copied data, in case you need to modify or reference it after the fact.

Refactoring to use V8's memory allocator is a little more complicated, because it requires modifying the create_external_resource function to use memory allocated by V8, instead of using malloc. This may be more or less feasible, depending on whether or not you control the definition of create_external_resource. The idea is to first create the buffer using V8, e.g. with napi_create_buffer, and then initialize the resource into the memory that has been allocated by V8. It is important to retain a napi_ref to the Buffer object for the lifetime of the resource, otherwise V8 may garbage-collect the Buffer and potentially result in use-after-free errors.

· 3 мин. прочитано

Electron 19.0.0 вышел! Он включает обновления Chromium 102, V8 10.2 и Node.js 16.14.2. Read below for more details!


Команда Electron рада объявить о выпуске Electron 19.0.0! You can install it with npm via npm install [email protected] or download it from our releases website. Continue reading for details about this release and please share any feedback you have!

Notable Changes

Electron Release Cadence Change

The project is returning to its earlier policy of supporting the latest three major versions. See our versioning document for more detailed information about Electron versioning and support. This had temporarily been four major versions to help users adjust to the new release cadence that began in Electron 15. You can read the full details here.

Stack Changes

Breaking & API Changes

Below are breaking changes introduced in Electron 19. More information about these and future changes can be found on the Planned Breaking Changes page.

Unsupported on Linux: .skipTaskbar

The BrowserWindow constructor option skipTaskbar is no longer supported on Linux. Changed in #33226

Removed WebPreferences.preloadURL

The semi-documented preloadURL property has been removed from WebPreferences. #33228. WebPreferences.preload should be used instead.

End of Support for 15.x.y and 16.x.y

Electron 14.x.y and 15.x.y have both reached end-of-support. This returns Electron to its existing policy of supporting the latest three major versions. Developers and applications are encouraged to upgrade to a newer version of Electron.

E15 (Sep'21)E16 (Nov'21)E17 (Feb'22)E18 (Mar'22)E19 (May'22)
15.x.y16.x.y17.x.y18.x.y19.x.y
14.x.y15.x.y16.x.y17.x.y18.x.y
13.x.y14.x.y15.x.y16.x.y17.x.y
12.x.y13.x.y14.x.y15.x.y--

Что дальше

In the short term, you can expect the team to continue to focus on keeping up with the development of the major components that make up Electron, including Chromium, Node, and V8. Although we are careful not to make promises about release dates, our plan is to release new major versions of Electron with new versions of those components approximately every 2 months.

You can find Electron's public timeline here.

More information about future changes can be found on the Planned Breaking Changes page.

· 2 мин. прочитано

Electron is changing its primary S3 bucket, you may need to update your build scripts


What is happening?

A significant amount of Electron's build artifacts are uploaded to an S3 bucket called gh-contractor-zcbenz. As part of ongoing infrastructure/ownership migrations that started way back in 2020, we will be changing everything that used gh-contractor-zcbenz from its old home in S3 to a new storage system hosted at https://artifacts.electronjs.org. The path prefix that most of our assets use is changing slightly as well. Examples are included below:

Before: https://gh-contractor-zcbenz.s3.amazonaws.com/atom-shell/dist/v17.0.0/node.lib
After: https://artifacts.electronjs.org/headers/dist/v17.0.0/node.lib

The important things here are the Hostname changed and the /atom-shell prefix changed. Another example, this time for debug symbols:

Before: https://gh-contractor-zcbenz.s3.amazonaws.com/atom-shell/symbols/path/to/symbol.pdb
After: https://artifacts.electronjs.org/symbols/path/to/symbol.pdb

Again, the hostname changed and the /atom-shell prefix was changed.

How might this impact you?

Anyone using standard build tooling such as electron-rebuild, electron-packager or @electron/get won't have to do anything. This should be the majority of people.

For anyone directly referencing the S3 bucket, you must update your reference to point at the hostname and update the path as well.

What about existing data?

Most data that existed on the gh-contractor-zcbenz bucket has been cloned into the new storage system. This means all debug symbols and all headers have been copied. If you relied on some data in that bucket that hasn't been copied over please raise an issue in electron/electron and let us know.

The current gh-contractor-zcbenz S3 bucket will not be actively deleted. However, we can't guarantee how long that bucket will be left alive. We strongly recommend updating to target the new bucket as soon as possible.