Fixes #37775 - include hashes in production builds of assets #10125
Conversation
| @@ -153,6 +153,12 @@ const coreConfig = function() { | |||
| 'webpack/assets/javascripts/bundle.js' | |||
| ); | |||
| config.context = path.resolve(__dirname, '..'); | |||
| if (config.mode == 'production') { | |||
There was a problem hiding this comment.
for some reason, this is always true, and I don't understand it.
There was a problem hiding this comment.
Looking at
foreman/config/webpack.config.js
Lines 56 to 59 in a837946
NODE_ENV to production.
There was a problem hiding this comment.
foreman/lib/tasks/webpack_compile.rake
Line 10 in fd21a88
that could be it, indeed
|
Okay, maybe I'm not as smart as I thought… |
|
I feel like I need |
evgeni
force-pushed
the
webpack-hashes
branch
3 times, most recently
from
478727b
to
a837946
Compare
| } | ||
| config.output = { | ||
| path: outputPath, | ||
| publicPath: '/webpack/' + pluginName + '/', | ||
| filename: chunkFilename, |
There was a problem hiding this comment.
I am almost sure this will break plugins, but I've not tested it yet
There was a problem hiding this comment.
amazingly, it does not!
old foreman and old foreman_puppet:
new foreman (with hashes) and old foreman_puppet:
new foreman (with hashes) and new foreman_puppet (with hashes):
|
Thank you for your contribution! This PR has been inactive for 3 months, closing for now. Feel free to reopen when you return to it. This is an automated process. |
|
This is still relevant |
test/integration/host_js_test.rb
Outdated
| @@ -96,7 +96,7 @@ class HostJSTest < IntegrationTestWithJavascript | |||
|
|
|||
| test "edit page" do | |||
| visit host_details_page_path(@host) | |||
| click_button 'Edit' | |||
| find('.host-details-header-section').find_button('Edit').click # the Edit button for the host details, not the host comment | |||
There was a problem hiding this comment.
I think this is a regression introduced in 37093a5, but it's weird if we've been ignoring this test failure since May.
There was a problem hiding this comment.
something, something, doesn't always load quickly enough or whatever.
this test HAS been super flaky, which could be the regression here.
There was a problem hiding this comment.
now to find out why the global params one is failing
ekohl
changed the title
|
3.11 or 3.12? |
|
I think it was introduced as a regression in 89c6104 so: $ git tag --contains 89c610451c9ea22eb426826e405518e04d68972f | grep -v rc
3.10.0
3.11.0
3.11.1That means also 3.10. |
|
Should the commit message include a |
Uh, thanks GitHub for not refreshing the comments, I guess… It could, but now it does not :) |
draft, as this only does it for core, not plugins