Skip to content

Optimize /datum/datacore/proc/manifest#12567

Draft
noelle-lavenza wants to merge 1 commit into
cmss13-devs:masterfrom
noelle-lavenza:optimize/manifest
Draft

Optimize /datum/datacore/proc/manifest#12567
noelle-lavenza wants to merge 1 commit into
cmss13-devs:masterfrom
noelle-lavenza:optimize/manifest

Conversation

@noelle-lavenza

@noelle-lavenza noelle-lavenza commented Jul 2, 2026

Copy link
Copy Markdown
Contributor

About the pull request

/datum/datacore/proc/manifest now uses an assoc lookup for valid job titles to make the looping step much faster.

Removes the sleep from manifest injection. It's entirely unnecessary, see here: tgstation/tgstation#20038

Explain why it's good for the game

No longer will there be an arbitrary 4 second pause before the manifest loads.

Testing Photographs and Procedure

Will take screenshots once I'm done PRing all the commits I've split out of #11826. I believe it was testmerged and no one reported issues with the manifest, though.

Screenshots & Videos

Put screenshots and videos here with an empty line between the screenshots and the <details> tags.

Changelog

🆑
code: removed an unnecessary pause before crew manifest generation
/:cl:

@github-project-automation github-project-automation Bot moved this to Awaiting Review in Review Backlog Jul 2, 2026
@cmss13-ci cmss13-ci Bot added Missing Changelog Maintainers always document their changes. size/S Denotes a PR that changes 19-49 lines, ignoring generated files. Code Improvement Make the code longer and removed Missing Changelog Maintainers always document their changes. labels Jul 2, 2026
@fira

fira commented Jul 3, 2026

Copy link
Copy Markdown
Member

I'm pretty sure last time i tried to shoot down a sleep involving datacore on roundstart, it resulted in a race condition. Job/Spawn code is an absolute mess and can't be relied upon.
I'd advocate for live testing this

@fira fira added the Needs Testing Need to test it on the guinea pigs (production server) label Jul 3, 2026
@noelle-lavenza

Copy link
Copy Markdown
Contributor Author

I'm pretty sure last time i tried to shoot down a sleep involving datacore on roundstart, it resulted in a race condition. Job/Spawn code is an absolute mess and can't be relied upon. I'd advocate for live testing this

theres a very real possibility that the whole thing needs a once-over, i think the biggest risk is that it works without the sleep on other codebases because they removed even more ones, or made something faster, or dont have as much pop, etc

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Code Improvement Make the code longer Needs Testing Need to test it on the guinea pigs (production server) size/S Denotes a PR that changes 19-49 lines, ignoring generated files.

Projects

Status: Awaiting Review

Development

Successfully merging this pull request may close these issues.

2 participants