[SE-3401] Upgrades Webpack to v4.44.2 to support content hashing - #265
[SE-3401] Upgrades Webpack to v4.44.2 to support content hashing#265nizarmah wants to merge 5 commits into
Conversation
gabor-boros
left a comment
There was a problem hiding this comment.
👍 🎉
- I tested this: Tested asset loading from a normal window
- I tested this: Tested asset loading from a private window
- I tested this: Tested asset loading from a tor window
- I tested this: Tested asset checksum match on the instances
- I read through the code
- [ ] NA I checked for accessibility issues
- [ ] NA Includes documentation
- [ ] NA I made sure any change in configuration variables is reflected in the corresponding client'sconfiguration-securerepository.
I totally missed that part. Sorry about that! Thanks for paying attention to that! |
f2d6e7e to
d2dba5c
Compare
I compared the file to other instances; I also compared the loading time of the file with other instances. And I can't relate to what you are mentioning. The asset seems to be as "compressed" as it already were. In addition, the loading time is closely similar between different instances. The cherry-on-top is that this file is the same between the LMS and Studio. So by content-hashing it, it gets cached the first time it loads. Meanwhile, on other instances, the file has to be downloaded on LMS and Studio so that it is cached for each (since it is chunk hashed). Can you confirm what I said, and compare the file on different instances, so that I can mark this as "Ready for Review" please? Side note: the |
gabor-boros
left a comment
There was a problem hiding this comment.
👍 🎉
- I tested this: Tested asset loading from a normal window
- I tested this: Tested asset loading from a private window
- I tested this: Tested asset loading from a tor window
- I tested this: Tested asset checksum match on the instances
- I read through the code
- [ ] NA I checked for accessibility issues
- [ ] NA Includes documentation
- [ ] NA I made sure any change in configuration variables is reflected in the corresponding client'sconfiguration-securerepository.
Ok, I have to apologize here. It turned out that the output I've seen seems compressed, but a "cool new feature" (which is really good but should be promoted), tricked me. The screenshot below is one line, but displayed as a "formatted" string...
In reality that's just one line. Sorry for the previous.
|
Hahaha no worries 😄 Thanks for the review 👍 and for making sure what I mentioned was correct 🙂 I'll try to search for a core member to review this. |
359109a to
13135b5
Compare
8a082d8 to
dd4c89a
Compare
ab82f44 to
d51c620
Compare
13f6b0b to
f12a0ed
Compare
e9f5733 to
7473c3e
Compare
|
I'm closing this PR since it's no longer necessary. And I'll work on this contribution during my own personal time 👍 https://github.com/edx/edx-platform/pull/25635 |

Upgrades Webpack from v2.7.0 to v4.44.2 so that content hashing is possible.
JIRA tickets: SE-3401
Sandbox URL:
Testing instructions:
commons.js.ls /edx/var/edxapp/staticfiles/studio/bundles/ | grep -i commonsis the same among the different instances./edx/var/edxapp/staticfiles/studio/bundles/and/edx/var/edxapp/staticfiles/bundles/between both App Servers and make sure they have the same filename (including hashes)./edx/var/edxapp/staticfiles/studio/bundles/and/edx/var/edxapp/staticfiles/bundles/between app servers with old instances (without the fix) and new ones (with the fix) and make sure everything is similar/acting normal.Reviewers