Bump docusaurus to 3.0.0-alpha.0#1745
Conversation
| "@docusaurus/mdx-loader", | ||
| "@docusaurus/plugin-content-blog", | ||
| "@docusaurus/preset-classic" | ||
| "react-json-view" |
There was a problem hiding this comment.
We still need this peer dependency override until facebook/docusaurus#9116 is released
| "pnpm": { | ||
| "patchedDependencies": { | ||
| "@docusaurus/theme-search-algolia@2.3.1": "patches/@docusaurus__theme-search-algolia@2.3.1.patch" | ||
| "@docusaurus/theme-search-algolia@3.0.0-alpha.0": "patches/@docusaurus__theme-search-algolia@2.3.1.patch" |
There was a problem hiding this comment.
this patch seemed to apply cleanly, and still is having the intended effect in local testing, so that's a relief 😅
96e7596 to
58fbe93
Compare
| extends: toPosixPath( | ||
| path.join(pathFromPackageToRoot, "tsconfig.base.json"), | ||
| ), | ||
| extends: getExtends(pathFromPackageToRoot, input.extends), |
There was a problem hiding this comment.
need this because we are now extending docusaurus tsconfig in our docs site
- Fixes #1749 - Will do a `pnpm remove` on all our typedoc dependencies once #1745 is merged so we don't get massive merge conflicts on our `pnpm.lock` 😅 - Our website deploy has gone from more than 12 minutes to 1 min 20 sec 🙌 ## Checklist - [ ] I have added [tests](https://www.cursorless.org/docs/contributing/test-case-recorder/) - [ ] I have updated the [docs](https://github.com/cursorless-dev/cursorless/tree/main/docs) and [cheatsheet](https://github.com/cursorless-dev/cursorless/tree/main/cursorless-talon/src/cheatsheet) - [ ] I have not broken the cheatsheet
#1745 is almost impossible to review 😅 I'd suggest running `pnpm dedupe` on the PR but it turns out `main` has a whole bunch of duplicate transitive deps too, so running it against `main` first. ## Checklist - [ ] ~I have added [tests](https://www.cursorless.org/docs/contributing/test-case-recorder/)~ - [ ] ~I have updated the [docs](https://github.com/cursorless-dev/cursorless/tree/main/docs) and [cheatsheet](https://github.com/cursorless-dev/cursorless/tree/main/cursorless-talon/src/cheatsheet)~ - [ ] I have not broken the cheatsheet
9ec62c3 to
a7ac988
Compare
|
ok @auscompgeek I've merged your dedupe. Fwiw I just ran a dedupe on this branch and did a soft reset over your dedupe; didn't feel like fighting with merge conflicts in pnpm lockfile 😅 |
|
Yep makes sense, I think the end result should be the same however you slice it 😄 |
There was a problem hiding this comment.
Nothing that leapt out as majorly alarming to me. Left some comments on the more curious new deps coming in (although the comments are a bit of a stream of consciousness).
A few cases where there's -1 +1 versions of a dep, largely different semver major versions (judging by the node engine constraints on them, I'm guessing ES6 conversions or something). A few cases where we have one fewer version of an existing dep though, which is good to see.
|
|
||
| /@babel/core@7.21.4: | ||
| resolution: {integrity: sha512-qt/YV149Jman/6AfmlxJ04LMIu8bMoyl3RB91yTFrxQmgbrSvQMy7cI8Q62FHx1t8wJ8B5fu0UDoLwHAhUo1QA==} | ||
| /@babel/core@7.22.10: |
There was a problem hiding this comment.
Interesting, we had two versions of @babel/core. Looks like the old come was being pulled in by mdx
| /@mdx-js/mdx@1.6.22: | ||
| resolution: {integrity: sha512-AMxuLxPz2j5/6TpF/XSdKpQP1NlG0z11dFOlq+2IP/lSgl11GY8ji6S/rgsViN/L0BDvHvUMruRb7ub+24LUYA==} | ||
| dependencies: | ||
| '@babel/core': 7.12.9 |
There was a problem hiding this comment.
Ah, mdx 1 had babel 7.12 pinned. Looks like it doesn't depend on babel any more - pulling in estree instead.
| @@ -3886,7 +3820,11 @@ packages: | |||
| engines: {node: '>=12.22.0'} | |||
| dev: true | |||
|
|
|||
| /@pnpm/config@16.6.4(@pnpm/logger@5.0.0)(@yarnpkg/core@4.0.0-rc.14)(typanion@3.12.1): | |||
| /@pnpm/config.env-replace@1.1.0: | |||
There was a problem hiding this comment.
interesting, pulling in a second version of this, weird
| '@pnpm/network.ca-file': 1.0.2 | ||
| config-chain: 1.1.13 | ||
| dev: true | ||
|
|
||
| /@pnpm/npm-lifecycle@2.0.0(typanion@3.12.1): | ||
| /@pnpm/npm-conf@2.2.2: |
| /@szmarczak/http-timer@4.0.6: | ||
| resolution: {integrity: sha512-4BAffykYOgO+5nzBWYwE3W90sBgLJoUPRWWcL8wlyiM8IB8ipJz3UMJ9KXQd1RKQXpKp8Tutn80HZtWsu2u76w==} | ||
| engines: {node: '>=10'} | ||
| dependencies: | ||
| defer-to-connect: 2.0.1 | ||
| dev: true | ||
|
|
||
| /@szmarczak/http-timer@5.0.1: |
There was a problem hiding this comment.
another one here too, but it's a different major version 🤷
| /cacheable-request@6.1.0: | ||
| resolution: {integrity: sha512-Oj3cAGPCqOZX7Rz64Uny2GYAZNliQSqfbePrgAQ1wKAihYmCUnraBtJtKcGR4xz7wF+LoJC+ssFZvv5BgF9Igg==} | ||
| engines: {node: '>=8'} | ||
| /cacheable-lookup@7.0.0: |
| @@ -7115,6 +7100,11 @@ packages: | |||
| resolution: {integrity: sha512-Gmy6FhYlCY7uOElZUSbxo2UCDH8owEk996gkbrpsgGtrJLM3J7jGxl9Ic7Qwwj4ivOE5AWZWRMecDdF7hqGjFA==} | |||
| engines: {node: '>=10'} | |||
|
|
|||
| /camelcase@7.0.1: | |||
There was a problem hiding this comment.
3 major versions...
is this what it says on the tin? 😂
There was a problem hiding this comment.
If it does what it says on the tin, then the fact that we pull in 3 major versions of a one-line function is pretty much the npm ecosystem in a nutshell 😂
| @@ -7716,6 +7713,14 @@ packages: | |||
| /crypto-random-string@2.0.0: | |||
| resolution: {integrity: sha512-v1plID3y9r/lPhviJ1wrXpLeyUIGAZ2SHNYTEapm7/8A9nLPoyvVp3RK/EPFqn5kEznyWgYZNsRtYYIWbuG8KA==} | |||
| engines: {node: '>=8'} | |||
| dev: true | |||
|
|
|||
| /crypto-random-string@4.0.0: | |||
There was a problem hiding this comment.
a second major version here too.
| resolution: {integrity: sha512-x8dy3RnvYdlUcPOjkEHqozhiwzKNSq7GcPuXFbnyMOCHxX8V3OgIg/pYuabl2sbUPfIJaeAQB7PMOK8DFIdoRA==} | ||
| engines: {node: '>=12'} | ||
| dependencies: | ||
| type-fest: 1.4.0 |
There was a problem hiding this comment.
that's an interesting transitive dependency 😂
There was a problem hiding this comment.
Ha yeah that's...interesting. Should type-fest every be anything other than dev dependency?
There was a problem hiding this comment.
yeah I don't think this should be a non-dev dependency, but I haven't checked how it's actually being used here
| /mem@6.1.1: | ||
| resolution: {integrity: sha512-Ci6bIfq/UgcxPTYa8dQQ5FY3BzKkT894bwXWXxC/zqs0XgMO2cT20CGkOqda7gZNkmK5VP4x89IGZ6K7hfbn3Q==} | ||
| engines: {node: '>=8'} | ||
| dependencies: | ||
| map-age-cleaner: 0.1.3 | ||
| mimic-fn: 3.1.0 | ||
| dev: true | ||
|
|
||
| /mem@8.1.1: |
There was a problem hiding this comment.
wat
... oh, github diff didn't notice these already existed?
We have some security warnings on a couple packages that are dependencies of docusaurus, so I decided to upgrade to the latest to remove those dependencies. Docusaurus uses their nightly version on their own docs, and things seem to work for us on this version with no issue, so I think it's ok. In addition, we were violating their peer dependencies by using react 18; we now no longer need to do that. Also, this version allows us to use mdx v2, which has a lot of niceties that we'll probably want as we do our big docs refactor (#867)
Checklist