Conversation
|
Claude finished @cngonzalez's task —— View job Code Review
|
Coverage Report
File Coverage
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
69da6d3 to
3bcb396
Compare
This stack of pull requests is managed by Graphite. Learn more about stacking. |
4b456f7 to
a9021c0
Compare
|
I humbly acknowledge the Claude PR feedback but all of these are issues that existed in the existing utility, so I'm not inclined to fix all of them right now (I fixed a few) |
laurenashpole
left a comment
There was a problem hiding this comment.
LGTM! I left some suggestions to get rid of anys in the mockBrowserEnvironment.ts file but otherwise I was able to plug this into my branch and everything worked fine.
bd60edc to
f22fcc6
Compare
f22fcc6 to
2950118
Compare
laurenashpole
left a comment
There was a problem hiding this comment.
LGTM! Dropped it into my branch and called mockBrowserEnvironment and everything was working as expected.


Description
This PR exposes some of the logic that was previously invoked in getting browser globals and environment variables set up in
studioLoader.worker.tsin a new function calledmockBrowserEnvironment, meant to simulate themockBrowserEnvironmentfunction inoriginal-cli. WhilestudioWorkerTaskis probably a good fit for most commands that invokedmockBrowserEnvironment, moving user scripts to be executed by a worker probably constitutes a breaking change, so there has to be some way to set up the environment in a child process forsanity exec.Some of the functionality in the original
mockBrowserEnvironment-- notablyregisterESBuildno longer seem necessary, but I may be wrong about this.What to review
Does the code re-organization make sense? Any missing logic?
Testing
I don't see tests for utils, so they're omitted here in the interest of time.
execwill hopefully test all of this.