create vendor bundle with hand-picked libraries #30374
Merged
Conversation
This file contains 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
joshlory
reviewed
Aug 22, 2019
| }, | ||
| vendors: { | ||
| name: 'vendors', | ||
| priority: 30, |
There was a problem hiding this comment.
we discussed offline briefly. more details are here: https://webpack.js.org/plugins/split-chunks-plugin/#splitchunkscachegroupscachegrouppriority
|
This is cool! Can you share a before and after snapshot of the |
joshlory
reviewed
Aug 22, 2019
apps/Gruntfile.js
Outdated
| _.keys(appsEntries), | ||
| _.keys(pegasusEntries), | ||
| _.keys(professionalDevelopmentEntries), | ||
| _.keys(internalEntries) |
There was a problem hiding this comment.
Nit: why not Object.keys?
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
joshlory
reviewed
Aug 26, 2019
| %script{src: minifiable_asset_path('js/code-studio-common.js')} | ||
| %script{src: minifiable_asset_path('js/common.js')} | ||
| %script{src: asset_path("js/#{js_locale}/common_locale.js")} | ||
| %script{src: minifiable_asset_path('js/styleguide.js')} |
There was a problem hiding this comment.
Removed because it already inherits from the application template?
There was a problem hiding this comment.
Removed because it unused and would require updating otherwise.
joshlory
approved these changes
Aug 26, 2019
|
Caveat: I haven't tested this locally yet. Lmk if you'd like me to! |
|
I don't need you to test this locally unless you want to but thanks for the careful review! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Background
One problem with our current webpack config is that shared modules are not pulled out of plc entry points. Because using count-based approaches is tricky, I am hand-picking the list of packages included here based on inspection of which packages are used by most entry points as well as which packages are aligned with our current best practices (e.g. react).
According to many sources on the internet, the strategy of bundling npm packages separately from application code should improve our caching performance because vendors.js is likely to remain cached (by Cloudfront, firewalls, browsers, etc.) for days or weeks, compared with our application code which updates daily.
Description
Extract a hand-picked list of npm packages from most dashboard and pegasus entry points. This does not include special entry points like applab-api.js and embedVideo.js which need to be entirely self-contained.
results
before
after
future work
there is some code still shared across many PLC entry points (e.g. react-bootstrap) which could benefit from being extracted into some kind of
plc-common.jsbundle. I didn't include this in vendors.js because react-bootstrap is not used widely outside of PLC code.Ideally, we'd be able to do something like this:
currently we use this strategy to avoid duplication between common.js and code-studio-common.js, however it is not clear whether the trick mentioned earlier will scale as the number of shared bundles increases.
I have a future work item to explore a more dynamic approach to bundle splitting which does not require our configuration to have specific knowledge of our application code, which is part of why I'm choosing this as a stopping point for now.