Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Don't process CSS imports by default #5874

Merged
merged 2 commits into from
Jan 5, 2017
Merged

Don't process CSS imports by default #5874

merged 2 commits into from
Jan 5, 2017

Conversation

wagenet
Copy link
Contributor

@wagenet wagenet commented May 2, 2016

This is a potentially dangerous default in clean-css. Ideally they'd change it there, but it doesn't seem that it's likely. The benefits of importing files by URL is also greatly reduced in ember-cli apps. See clean-css/clean-css#767

@rwjblue
Copy link
Member

rwjblue commented May 2, 2016

Seems good to me.

@stefanpenner
Copy link
Contributor

seems like a breaking change are there somespecial steps we need to take?

@Turbo87
Copy link
Member

Turbo87 commented May 11, 2016

@wagenet could you fix the broken test?

@wagenet
Copy link
Contributor Author

wagenet commented May 23, 2016

Its not a breaking change because old versions of the minifier didn't do stuff with imports. I'll look into the tests.

On May 2, 2016, at 1:10 PM, Stefan Penner [email protected] wrote:

seems like a breaking change are there somespecial steps we need to take?


You are receiving this because you authored the thread.
Reply to this email directly or view it on GitHub

@stefanpenner
Copy link
Contributor

@wagenet but now there is an interim time were they are process? How big is that window (i would like to write this off as a non-breaking change, just want to make sure we have all the facts).

@homu
Copy link
Contributor

homu commented May 24, 2016

☔ The latest upstream changes (presumably #5928) made this pull request unmergeable. Please resolve the merge conflicts.

This is a potentially dangerous default in clean-css. Ideally they'd change it there, but it doesn't seem that it's likely. The benefits of importing files by URL is also greatly reduced in ember-cli apps. See clean-css/clean-css#767
@wagenet
Copy link
Contributor Author

wagenet commented Jul 13, 2016

@homu I've fixed the merge conflict.
@stefanpenner Failing to pre-import http imports shouldn't break anything, it just could make things a bit slower at build time. Given that doing the import can break things this seems like a worthwhile trade-off, especially given that this was the previous behavior.

@nathanhammond
Copy link
Contributor

@wagenet How long was the timeframe where where https://github.com/jakubpawlowicz/clean-css/ inlined imports that we had this as a side effect? (AKA, when did they ship it?)

@wagenet
Copy link
Contributor Author

wagenet commented Jul 13, 2016

@nathanhammond The chain is long and confusing.

It looks like ember-cli-preprocess-registry (the first package in the long chain to clean-css) was updated to ^2.0.0 from ^1.0.3 for ember-cli 2.3. Going back a step to ember-cli-preprocessor-registry shows us that 1.0.3 used broccoli-clean-css 0.2.0 which in turn uses clean-css ^2.2.1. ember-cli-preprocessor-registry 1.2.0 uses broccoli-clean-css ^1.1.0, which uses clean-css-promise which uses clean-css ^3.4.5.

So all that to say that prior to ember-cli 2.3 you could end up with clean-css ^2.2.1 or ^3.4.5 depending on what version of ember-cli-preprocessor-registry you were on. Once we hit 2.3 you were guaranteed to be on clean-css ^3.4.5.

Presumably the issues I'm seeing have something to do with that clean-css version change. This goes back long enough and is inconsistent enough that maybe we shouldn't do anything about this, though I really would like to make this less confusing to users, since I think that clean-css has the wrong defaults (especially in regards to importing // urls).

Another option that I realized is that we could do { processImportFrom: ['local'] } to still allow local imports but not allow remote. However, this might be considered a breaking change. Maybe this is something we could call out in the docs somewhere instead?

@nathanhammond
Copy link
Contributor

@wagenet First, thank you for doing all of that research. That looks like it was a pain in the ass. In human time that means that we arbitrarily switched behavior on February 16, 2016 with no complaints (except yours), and could arbitrarily switch back with likely the same story. I consider this undefined behavior and am in favor of shipping a default configuration option of local.

Since it's easily configurable in consuming apps this seems like a solid win. I'm in favor. @rwjblue as caretaker of upgrade pain, you're the final arbiter on this.

@homu
Copy link
Contributor

homu commented Jan 4, 2017

☔ The latest upstream changes (presumably #6629) made this pull request unmergeable. Please resolve the merge conflicts.

Copy link
Member

@rwjblue rwjblue left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Lets do it!

@nathanhammond
Copy link
Contributor

@homu r+

@homu
Copy link
Contributor

homu commented Jan 5, 2017

📌 Commit bd3cc13 has been approved by nathanhammond

homu added a commit that referenced this pull request Jan 5, 2017
Don't process CSS imports by default

This is a potentially dangerous default in clean-css. Ideally they'd change it there, but it doesn't seem that it's likely. The benefits of importing files by URL is also greatly reduced in ember-cli apps. See clean-css/clean-css#767
@homu
Copy link
Contributor

homu commented Jan 5, 2017

⌛ Testing commit bd3cc13 with merge 4ea756a...

@homu
Copy link
Contributor

homu commented Jan 5, 2017

💔 Test failed - status

@homu
Copy link
Contributor

homu commented Jan 5, 2017

☀️ Test successful - status

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

6 participants