diff --git a/st-embroider/2024/2026-02-03.md b/st-embroider/2024/2026-02-03.md new file mode 100644 index 0000000..c8fd143 --- /dev/null +++ b/st-embroider/2024/2026-02-03.md @@ -0,0 +1,135 @@ +# 2026-02-03 + +Note Taker: Chris Manson +Time Keeper: Chris Manson + +## Attendees + +Add yourself to the list if you attend and check the box! + +- [x] Ed Faulkner (EF) +- [x] Chris Manson (CM) +- [x] Alex (A) +- [x] Preston Sego (PS) +- [ ] Simon Ihmig (SI) +- [x] Katie Gengler (KG) +- [ ] Krystan HuffMenne (KH) +- [ ] Peter Wagenet (PW) +- [ ] James Davis (JD) +- [x] Liam Potter (LP) +- [x] tommyjr +- [ ] + +## Topics + +### Releasing @glimmer/component update for this DX improvement https://github.com/emberjs/ember.js/pull/21014 <@evoactivity> + +- EF: since the merge we might not have the automation to release this automatically +- KG: I will need to check that it isn't being automatically caught by our new CI +- PS: maybe release plan just for some packages? +- CM: maybe 🙈 +- KG: I have a few ideas + +### who owns https://github.com/emberjs/ember-cli-babel ? shouldn’t that be the Tooling team? <@mansona> + +- CM: all been solved +- just trying to get rid of npm deprecations +- more on this later + +### PRs from nvp <@NullVoxPopuli> + +- https://github.com/embroider-build/embroider/pull/2664 +- merged + +- https://github.com/ember-learn/guides-source/pull/2190 +- merged + +- https://github.com/embroider-build/embroider/pull/2662 +- PS: I can probably add docs before merging but I wanted to get a sense of the correct implementation +- EF: looks a lot better +- why are we adding the types to the public API? we probably don't want to have this added to the public api +- (discussion about the actual implemntation and the timing of when things happen) +- (discussion about it being always runtime) +- A: I would be in favour of always runtime +- EF: that would prevent you from branch elimination for production +- we wouldn't need it to be a macro then +- (discussion of the specifics of the implementation) +- comment about updating the error to be an assertion that the global and the argument to the setTesting() match (you can't toggle it at runtime) + +### can I get a review and merge on https://github.com/broccolijs/broccoli-concat/pull/171 ? is it a decent way to revive a legacy codebase like this? <@mansona> + +- (discussion of the CI approach) +- all agreed that this is ok +- KG: should we just do a major quickly + +### update on the install new app with no npm deprecations situation <@mansona> + +- sumamry of the strategy +- some overrides that have been used so far to make it better: + +```json +"@embroider/compat>broccoli": "^4.0.0", +"@embroider/macros>ember-cli-babel": "^8.0.0", +"async-disk-cache>rimraf": "^6.0.0", +"broccoli-concat": "github:broccolijs/broccoli-concat#lodash", +"broccoli-config-replace": "github:ember-cli/broccoli-config-replace#kitchen-sink", +"broccoli-funnel>rimraf" :"^6.0.0", +"broccoli-persistent-filter>rimraf": "^6.0.0", +"broccoli-plugin>rimraf": "^6.0.0", +"ember-resolver>ember-cli-babel": "^8.0.0", +"fs-updater>rimraf": "^6.0.0", +"hash-for-dep>broccoli-kitchen-sink-helpers": "-", +"hash-for-dep>path-root": "-", +"sync-disk-cache>rimraf": "^6.0.0", +"testem": "github:mansona/testem#remove-npmlog", +"testem>consolidate": "^1.0.0", +"testem>glob": "^13.0.0", +"testem>rimraf": "^6.0.0", +"watch-detector>tmp": "^0.2.0", +"broccoli-caching-writer>rimraf": "^6.0.0", +"temp>rimraf":"^6.0.0", +"broccoli-config-replace>fs-extra": "^11.0.0", +"@ember/optional-features>glob": "^13.0.0", +"body>raw-body": "^3.0.0", +"jsdom": "^28.0.0", +"fast-sourcemap-concat": "github:mansona/fast-sourcemap-concat#inline-source-map-url", +"@embroider/vite>source-map-url": "-" +``` + +- (discussion of the problem with source-map-url) +- EF: happy to inline +- + +### ember-table v1 addon migration - anyone have connections to reach maintainers? <@void-mAlex> + +- https://github.com/Addepar/ember-table/pull/1203 +- https://github.com/Addepar/ember-table/pull/1173 +- A: summary of the situation +- I'm happy to do the work if we can get access +- CM: I am pushing on the v2 addon by default RFC so that we can have a codemod that pushes people in the right direction +- EF: remember that we can ship default compat adapters to prevent people being blocked +- A: better to upstream changes than monkeypatching in compat adapters +- CM: why not both! +- + +### @embroider/try scare with no readme or link to repository https://www.npmjs.com/package/@embroider/try <@void-mAlex> + +- need to have a README and a back-link from the package.json +- (discussion of backlash against npx) +- + +### any one have ideas for reducing the number of packages brought in by tiny tools? <@NullVoxPopuli> + +- https://node-modules.dev/grid/depth#install=@embroider/macros +- (large discussion about the benefit) +- CM: we could use e18e to optimise +- EF: we should focus on common things like `@embroider/addon-shim` that would be good +- for decorator-transforms we could minimise the graph by having a separate runtime package +- CM: we should consider upgrading things like broccoli-plugin to be esm only and then following that upgrade up the graph (broccoli-funnel, addon-shim) +- (all agreed) + + + + + +