1090599 - Update in-tree jsmin to release-2.0.11 or later

Status: RESOLVED FIXED

(Firefox Build System :: Android Studio and Gradle Integration, defect)

Description

[Nick Alexander :nalexander [he/him] (PTO Feb 24-March 3)]

The in-tree jsmin at python/jsmin has lagged behind upstream jsmin at https://bitbucket.org/dcs/jsmin/.  We should update to at least release-2.0.11.

I'd actually like to upgrade past that, to include my fix for backticks in JS at https://bitbucket.org/dcs/jsmin/pull-request/7/fix-for-issue-14-handle-ecmascript-harmony, but it hasn't landed yet.

Comment 1

[Nick Alexander :nalexander [he/him] (PTO Feb 24-March 3)]

With an updated jsmin, and my backticks fix, I get identical ASTs and a green try build:

https://treeherder.mozilla.org/ui/#/jobs?repo=try&revision=67bde4af32bd

Comment 2

[Nick Alexander :nalexander [he/him] (PTO Feb 24-March 3)]

Attachment: Upgrade in-tree jsmin to release-2.0.11. r=gps

This updates to release-2.0.11 (sha1: 04374472ed18) from
https://bitbucket.org/dcs/jsmin/.

Comment 3

[Nick Alexander :nalexander [he/him] (PTO Feb 24-March 3)]

Attachment: Treat `backticks` like 'quotes' in jsmin. r=gps

This is my patch to handle backticks from
https://bitbucket.org/dcs/jsmin/pull-request/7/fix-for-issue-14-handle-ecmascript-harmony
It hasn't landed upstream yet; it's not clear that jsmin will support
any Harmony features, so we should land locally and patch up conflicts
as they arise.

gps: It looks like the jsmin tests do not run in automation, and I
don't care to enable them just now.

Comment 4

[Gregory Szorc [:gps]]

Comment on attachment 8513659 [details] [diff] [review]
Upgrade in-tree jsmin to release-2.0.11. r=gps

Review of attachment 8513659 [details] [diff] [review]:
-----------------------------------------------------------------

rs=me

Comment 5

[Gregory Szorc [:gps]]

Comment on attachment 8513662 [details] [diff] [review]
Treat `backticks` like 'quotes' in jsmin. r=gps

This is probably OK. But I don't think it buys us much (I think you said making 5 or 10 files out of 1000 work) and isn't worth forking the upstream project over. Let's focus on getting the minifying "framework" in place in automation and then chase the long tail later, if needed.

To explicitly state it: I'm glad you put in this work. Let's get the changes upstreamed then import them into m-c so we can keep running an unmodified package.

Comment 6

[Nick Alexander :nalexander [he/him] (PTO Feb 24-March 3)]

https://hg.mozilla.org/integration/fx-team/rev/25a77efd89bf

Comment 7

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/mozilla-central/rev/25a77efd89bf