Added the cleanupUnusedCatalogs configuration. When set to true, pnpm will remove unused catalog entries during installation #9793.
Automatically load pnpmfiles from config dependencies that are named @*/pnpm-plugin-* #9780.
pnpm config get now prints an INI string for an object value #9797.
pnpm config get now accepts property paths (e.g. pnpm config get catalog.react, pnpm config get .catalog.react, pnpm config get 'packageExtensions["@​babel/parser"].peerDependencies["@​babel/types"]'), and pnpm config set now accepts dot-leading or subscripted keys (e.g. pnpm config set .ignoreScripts true).
pnpm config get --json now prints a JSON serialization of config value, and pnpm config set --json now parses the input value as JSON.
Patch Changes
Semi-breaking. When automatically installing missing peer dependencies, prefer versions that are already present in the direct dependencies of the root workspace package #9835.
When executing the pnpm create command, must verify whether the node version is supported even if a cache already exists #9775.
When making requests for the non-abbreviated packument, add */* to the Accept header to avoid getting a 406 error on AWS CodeArtifact #9862.
The standalone exe version of pnpm works with glibc 2.26 again #9734.
Fix a regression in which pnpm dlx pkg --help doesn't pass --help to pkg #9823.
When re-deploying a Worker using wrangler deploy, if the configuration has been modified in the Cloudflare dashboard, the local configuration will overwrite the remote one. This can lead to unexpected results for users. To address this, currently wrangler deploy warns users about potential configuration overrides (without presenting them) and prompts them to confirm whether they want to proceed.
The changes here improve the above flow in the following way:
If the local changes only add new configurations (without modifying or removing existing ones), the deployment proceeds automatically without warnings or prompts, as these changes are non-destructive and safe.
If the local changes modify or remove existing configurations, wrangler deploy now displays a git-like diff showing the differences between the dashboard and local configurations. This allows users to review and understand the impact of their changes before confirming the deployment.
#10334cadf19a Thanks @jonesphillip! - Added queues subscription command to Wrangler including create, update, delete, get, list
A patched version of debug was previously introduced that resolved the package to a custom implementation. However, this caused issues due to CJS/ESM interop problems. We now resolve the debug package to use the Node.js implementation instead.
#1035680e964c Thanks @WillTaylorDev! - fix: Update regex for valid branch name to remove 61 char length requirement, allowing for longer branch names to be specified for preview aliases.
#10289a5a1426 Thanks @emily-shen! - Cleanup container images created during local dev if no changes have been made.
We now untag old images that were created by Wrangler/Vite if we find that the image content and configuration is unchanged, so that we don't keep accumulating image tags.
#103150c04da9 Thanks @emily-shen! - Add rollout_active_grace_period option to containers configuration.
This allows users to configure how long an active container should keep running for during a rollout, before the upgrade is applied.
#10253eb32a3a Thanks @emily-shen! - fix redeploying container apps when previous deploy failed or container (but not image) was deleted.
Previously this failed with No changes detected but no previous image found as we assumed there would be a previous deployment when an image exists in the registry.
#10312bd8223d Thanks @devin-ai-integration! - Added --domain flag to wrangler deploy command for deploying to custom domains. Use --domain example.com to deploy directly to a custom domain without manually configuring routes.
#83188cf47f9 Thanks @gnekich! - Introduce json output flag for wrangler pages deployment list
Patch Changes
#10232e7cae16 Thanks @emily-shen! - fix: validate wrangler containers delete ID to ensure a valid ID has been provided. Previously if you provided the container name (or any non-ID shaped string) you would get an auth error instead of a 404.
#101393b6ab8a Thanks @dom96! - Removes mention of cf-requirements when Python Workers are enabled
#10259c58a05c Thanks @dario-piotrowicz! - Ensure that maybeStartOrUpdateRemoteProxySession considers the potential account_id from the user's wrangler config
Currently if the user has an account_id in their wrangler config file, such id won't be taken into consideration for the remote proxy session, the changes here make sure that it is (note that the auth option of maybeStartOrUpdateRemoteProxySession, if provided, takes precedence over this id value).
The changes here also fix the same issue for wrangler dev and getPlatformProxy (since they use maybeStartOrUpdateRemoteProxySession under the hook).
#1028842aafa3 Thanks @tgarg-cf! - Do not attempt to update queue producer settings when deploying a Worker with a queue binding
Previously, each deployed Worker would update a subset of the queue producer's settings for each queue binding, which could result in broken queue producers or at least conflicts where different Workers tried to set different producer settings on a shared queue.
The async_hooks module is now provided natively by workerd, making the polyfill unnecessary. This improves performance and ensures better compatibility with Node.js async_hooks APIs.
#100609aad334 Thanks @edmundhung! - refactor: switch getPlatformProxy() to use Miniflare's dev registry implementation
Updated getPlatformProxy() to use Miniflare's dev registry instead of Wrangler's implementation. Previously, you had to start a wrangler or vite dev session before accessing the proxy bindings to connect to those workers. Now the order doesn't matter.
#1021928494f4 Thanks @dario-piotrowicz! - fix NonRetryableError thrown with an empty error message not stopping workflow retries locally
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR contains the following updates:
Release Notes
nuxt/eslint (@nuxt/eslint-config)v1.9.0
Compare Source
🚀 Features
🐞 Bug Fixes
View changes on GitHub
antfu-collective/bumpp (bumpp)v10.2.3
Compare Source
🐞 Bug Fixes
View changes on GitHub
antfu-collective/local-pkg (local-pkg)v1.1.2
Compare Source
No significant changes
View changes on GitHub
microsoft/playwright (playwright-core)v1.55.0
Compare Source
pnpm/pnpm (pnpm)v10.15.0
Compare Source
Minor Changes
Patch Changes
- Semi-breaking. When automatically installing missing peer dependencies, prefer versions that are already present in the direct dependencies of the root workspace package #9835.
- When executing the pnpm create command, must verify whether the node version is supported even if a cache already exists #9775.
- When making requests for the non-abbreviated packument, add */* to the Accept header to avoid getting a 406 error on AWS CodeArtifact #9862.
- The standalone exe version of pnpm works with glibc 2.26 again #9734.
- Fix a regression in which pnpm dlx pkg --help doesn't pass --help to pkg #9823.
vitejs/vite (vite)v7.1.3
Compare Source
Features
Bug Fixes
Performance Improvements
Miscellaneous Chores
Code Refactoring
Tests
v7.1.2
Compare Source
Bug Fixes
Miscellaneous Chores
- deps: migrate to @jridgewell/remapping from @ampproject/remapping (#20577) (0a6048a)
- deps: update rolldown-related dependencies (#20586) (77632c5)
vuejs/language-tools (vue-tsc)v3.0.6
Compare Source
cloudflare/workers-sdk (wrangler)v4.31.0
Compare Source
Minor Changes
#10314 9b09751 Thanks @dario-piotrowicz! - Show possible local vs. dashboard diff information on deploys
When re-deploying a Worker using wrangler deploy, if the configuration has been modified in the Cloudflare dashboard, the local configuration will overwrite the remote one. This can lead to unexpected results for users. To address this, currently wrangler deploy warns users about potential configuration overrides (without presenting them) and prompts them to confirm whether they want to proceed.
The changes here improve the above flow in the following way:
#10334 cadf19a Thanks @jonesphillip! - Added queues subscription command to Wrangler including create, update, delete, get, list
Patch Changes
#10374 20520fa Thanks @edmundhung! - Simplify debug package resolution with nodejs_compat
A patched version of debug was previously introduced that resolved the package to a custom implementation. However, this caused issues due to CJS/ESM interop problems. We now resolve the debug package to use the Node.js implementation instead.
#10249 875197a Thanks @penalosa! - Support JSRPC for remote bindings. This unlocks:
Updated dependencies [565c3a3, ddadb93, 20520fa, 875197a]:
v4.30.0
Compare Source
Minor Changes
Patch Changes
#10217 979984b Thanks @veggiedefender! - Increase the maxBuffer size for capnp uploads
#10356 80e964c Thanks @WillTaylorDev! - fix: Update regex for valid branch name to remove 61 char length requirement, allowing for longer branch names to be specified for preview aliases.
#10289 a5a1426 Thanks @emily-shen! - Cleanup container images created during local dev if no changes have been made.
We now untag old images that were created by Wrangler/Vite if we find that the image content and configuration is unchanged, so that we don't keep accumulating image tags.
#10315 0c04da9 Thanks @emily-shen! - Add rollout_active_grace_period option to containers configuration.
This allows users to configure how long an active container should keep running for during a rollout, before the upgrade is applied.
#10321 b524a6f Thanks @emily-shen! - print prettier errors during container deployment
#10253 eb32a3a Thanks @emily-shen! - fix redeploying container apps when previous deploy failed or container (but not image) was deleted.
Previously this failed with No changes detected but no previous image found as we assumed there would be a previous deployment when an image exists in the registry.
#9990 4288a61 Thanks @penalosa! - Fix startup profiling when sourcemaps are enabled
Updated dependencies [d54d8b7, ae0c806]:
v4.29.1
Compare Source
Patch Changes
v4.29.0
Compare Source
Minor Changes
#10283 80960b9 Thanks @WillTaylorDev! - Support long branch names in generation of branch aliases in WCI.
#10312 bd8223d Thanks @devin-ai-integration! - Added --domain flag to wrangler deploy command for deploying to custom domains. Use --domain example.com to deploy directly to a custom domain without manually configuring routes.
#8318 8cf47f9 Thanks @gnekich! - Introduce json output flag for wrangler pages deployment list
Patch Changes
#10232 e7cae16 Thanks @emily-shen! - fix: validate wrangler containers delete ID to ensure a valid ID has been provided. Previously if you provided the container name (or any non-ID shaped string) you would get an auth error instead of a 404.
#10139 3b6ab8a Thanks @dom96! - Removes mention of cf-requirements when Python Workers are enabled
#10259 c58a05c Thanks @dario-piotrowicz! - Ensure that maybeStartOrUpdateRemoteProxySession considers the potential account_id from the user's wrangler config
Currently if the user has an account_id in their wrangler config file, such id won't be taken into consideration for the remote proxy session, the changes here make sure that it is (note that the auth option of maybeStartOrUpdateRemoteProxySession, if provided, takes precedence over this id value).
The changes here also fix the same issue for wrangler dev and getPlatformProxy (since they use maybeStartOrUpdateRemoteProxySession under the hook).
#10288 42aafa3 Thanks @tgarg-cf! - Do not attempt to update queue producer settings when deploying a Worker with a queue binding
Previously, each deployed Worker would update a subset of the queue producer's settings for each queue binding, which could result in broken queue producers or at least conflicts where different Workers tried to set different producer settings on a shared queue.
#10242 70bd966 Thanks @devin-ai-integration! - Add experimental API to expose Wrangler command tree structure for documentation generation
#10258 d391076 Thanks @nikitassharma! - Add the option to allow all tiers when creating a container
#10248 422ae22 Thanks @emily-shen! - fix: re-push container images on deploy even if the only change was to the Dockerfile
#10179 5d5ecd5 Thanks @pombosilva! - Prevent defining multiple workflows with the same "name" property in the same wrangler file
#10232 e7cae16 Thanks @emily-shen! - include containers API calls in output of WRANGLER_LOG=debug
#10243 d481901 Thanks @devin-ai-integration! - Remove async_hooks polyfill - now uses native workerd implementation
The async_hooks module is now provided natively by workerd, making the polyfill unnecessary. This improves performance and ensures better compatibility with Node.js async_hooks APIs.
#10060 9aad334 Thanks @edmundhung! - refactor: switch getPlatformProxy() to use Miniflare's dev registry implementation
Updated getPlatformProxy() to use Miniflare's dev registry instead of Wrangler's implementation. Previously, you had to start a wrangler or vite dev session before accessing the proxy bindings to connect to those workers. Now the order doesn't matter.
#10219 28494f4 Thanks @dario-piotrowicz! - fix NonRetryableError thrown with an empty error message not stopping workflow retries locally
Updated dependencies [1479fd0, 05c5b28, e3d9703, d481901]:
Configuration
📅 Schedule: Branch creation - "on Monday" (UTC), Automerge - At any time (no schedule defined).
🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.
♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.
👻 Immortal: This PR will be recreated if closed unmerged. Get config help if that's undesired.
This PR was generated by Mend Renovate. View the repository job log.