#docs-website
1 messages · Page 17 of 1
theres docs linked in one of the info things in the action menu
$lib/tab and $lib/components/pane/pane.svelte
wouldn't take an example of that code lol
god, whoever said kotlin sucked to reverse engineer was not joking
you might notice that actually has quite a bit of resugaring lol
it was written with svelte 4 reactivity originally and it also has handling for dumb stuff like destroying and recreating the entire tab's component lazily
$lib/tab not task bruh
yeah I figured lol
resugaring?
any other decompiler would output a java approximation but that's vineflower which has a kotlin plugin, so it outputs kotlin-ish
it's recaf but read only and in the browser 
xD
The process of re-adding all the syntactic sugar that the compiler eats
what are you using virtua for
funny how I looked in your package json and half the libs is stuff I normally use
search results
it started to lag with minecraft loaded in
didn't add anything like the tabs in slicer yet but I got sidebar resizing working pretty good
one thing I wish paneforge had was the ability to make a pane avoid resizing when the container resizes
so that if I shrink or grow the window the sidebar won't change size to keep proportion
it would only shrink if the window got smaller than it
yeah maybe
I actually found kind of a cursed solution
// Maintain sidebar size when resizing window
watch.pre(
() => this.windowInnerWidth,
(newValue, oldValue) => {
if (oldValue !== undefined && newValue !== undefined && this.sidebarPane) {
const oldPx = (this.sidebarPane.getSize() / 100) * oldValue;
const newProportion = this.getProportion(oldPx, newValue);
this.sidebarPane.resize(newProportion);
}
},
);
it still glitches a little bit but by using pre it doesn't lag at least
actually it's pretty much perfect when sidebar is on the left, only glitchy on the right the sidebar fidgets when you resize the window from the opposite side it's docked to
updated with that change, still one thing I'm not happy with mentioned in the TODO
but I think I can live with it for now
Is there any way we can make these messages not appear? I know that on the docs github they don't; maybe apply the same change to the website webhook? Idk how that is managed
@opal flare
does this look like the appropriate channel?
Hey, what can be contributed to the paper docs?
I think this section could use a little more info on database frameworks/libraries like jooq or exposed for a more type-safe approach or help with auto-generated code, instead of handling everything with raw java.sql
Is that something that would fit into the docs?
https://docs.papermc.io/paper/dev/using-databases/
Hello and welcome to Paper! I think the idea is nice but let's wait till the doc team members to see what their thoughts are 
Sure 
Yeah, the docs might contain a mistake
Read the conv
I think mentioning them and explaining them with like a sentence or two is fine, I wouldn't explain how to use them, only link to relevant documentation
might aswell explains ORMs and mention hibernate and throw jdbi into the mix, but can always be expanded
Indeed, I share Mini's opinion. We don't really want to go in-depth into not Paper specific concepts since there are plenty of docs explaining how to use databases with Java already and there is no need to clutter our docs with that. But mentioning it so that people become aware is good
They can do their own research afterwards and find much more specific docs on those subjects if they were to just look it up themselves
Most of the goal of the "using a DB" page was just to provide a simple "getting started" type of deal, documenting 3rd party projects in there is just extra burden especially when those providers already have good docs, would generally lean more towards noting some of the caveats of using a DB in paper, "it works just like a normal java app but you might want to be mindful of were you do your IO" type stuff
pointing towards common solutions is fine, if we ever get an example repo running we might even allow contributions for such (but, such does come with the fear of this stuff falling out of date)
Yh, the idea I had was to basically mention those tools, the advantages over java.sql and provide a snippet comparing how a query looks, enough for people to know the existence of said tools before forwarding them to the actual documentation
It think it would be slightly shorter than the SQL Injection section
I suck at english, I'm generally concerned over the wording of the PR for ^ in that it feels somewhat loaded and maybe kinda vague; I worry it's maybe overly pushy for tools we can't really support?
In wider terms of that page, the nature of being useful for beginners, I think a fair pressing need for expansion exists in ecosystem specifics, such as avoiding your IO on the main thread, but it also feels that every few months I have a discussion of "you should probably avoid caching eco" type discussions
I made it somewhat vague to not make it sound like advertising for those tools
Plus trying to explain the context of those tools, but at the same time not diving too much into it
I'm semi worried that the way in which it's worded kinda makes it feel like it's an imperative for developers to start using those tools
And I tried to avoid sounding like that 
I did ask chatgpt because I suck at language, it generally suggested this which feels a bit more natural;
When working with SQL, developers often need to balance security, type-safety, and code readability. Using a library can help handle these concerns automatically, letting you focus on the business logic rather than the low-level SQL details.
Libraries take different approaches, such as SQL builders or ORMs. Popular options in Java include JDBI and JOOQ.
For example, I avoided terms like "popular" because I truly don't know how popular these tools are amongst plugins devs
Referring to the last line btw
I mean, those are popular options for people wanting to use those types of tools in java
The first paragraph does look better, I'll try to word something like that
I'm not 100% on the chatgpt output, but I think it leans in a nicer direction, I'm generally pondering what the other docs people think as my English made my English teacher upset 😄
I mean, in general yh, those tools are definitely popular, I'm just not so sure how much they are in the context of minecraft plugins
Plugins I know using jooq are relatively few, can't say much about jdbi tho, didn't payed much attention to it
I'll also wait what the rest of the team says. but will be trying to follow a similar structure to the sample you sent 🫡
We also mention Hikari and even have an example for it.
I feel like mentioning some commonly used tools is fine, but I would remove the JOOQ example - that is something people can then look up if they decide on a tool. e.g. I would prefer JDBI there
And maybe remove the " leaving the developer with clean, secure and readable code." part. that sounds too much like ChatGPT
Don't tell me I talk like chatgpt 
I'll have to look at some jdbi examples, I mostly use jooq and exposed
-# ^ Forget it I just understood what you meant about jdbi
I guess providing an example for both would bloat the doc, so that's out of question. I'll see if I can figure out a way to give a peek into how tools reduce the code compared to plain jdbc
Yeah I didn't mean that you should add more examples, I think just directing people to the tools should be good enough. e.g. they all have some examples in their docs or similar
Dw, I understood I should remove them
I took a different approach this time.
I removed the part mentioning sql-builders and orm, plus focused on the advantages of these tools in the context of what is already on the docs (connecting with hikari, building queries, keeping an eye for sql injections) and added a bit about parsing, since what is fetched from the db will most likely be used in an object.
It is extremely short this time (this is the whole section), what do you think? I also tried following a similar wording to the example from ecat
## Database tools
Given the complexity of working with databases (managing connections, building and securing queries, or just parsing the data) several tools
exist in the world of Java to leverage this work.
Some plugin developers use tools like [JDBI](https://jdbi.org/) or [JOOQ](https://www.jooq.org/doc/latest/manual/), which are lightweight
and easy to use.
Since there are no examples this time, maybe I could mentioned exposed as well
Small fix because I felt it ended up too dry
## Database tools
Given the complexity of working with databases (managing connections, building and securing queries, or just parsing the data) several tools
exist in the world of Java to leverage this work.
Some plugin developers use lightweight tools like [JDBI](https://jdbi.org/), [JOOQ](https://www.jooq.org/doc/latest/manual/)
or [Exposed](https://www.jetbrains.com/help/exposed/get-started-with-exposed.html), which take care of all the heavy lifting,
allowing the developers to focus on their plugins rather than the database.
what is paper mc
not a #docs-website question, ask in #general
How do you even end up here if you have no idea what it is
And this description here is basically everywhere where you can get the invite link from
Bro using vanilla core since his born 💀
Let's keep on topic.
now that adventure and paper docs are merged, perhaps it would be good to merge the GlobalTranslator docs in the Paper docs and the MiniMessage translator ones under an Adventure i18n section
I always end up linking the two when people ask about server-side translations so I feel like they should just be one honestly
what's globaltranslator? am I missing smt?
^
Is it just me or does every doc page switch cause the entire page to reload and the sidebar has to redo its long animation? I tried disabling my extensions, and even switch browsers. I haven't seen other starlight astro docs have this issue
the entire page reloads because it's navigating, it's not a spa
the animation is us scrolling to the active page because the sidebar is too long
Also on slower internet, it seems the font isnt being cached, cause the entire page content is twitching every single time u switch pages
why do no other astro docs have this problem then?
they don't scroll?
?
i just tested it on the starlight official docs
and made my window small
they dont have this problem
the scroll animation is cool for very first page load, but should not be a thing when switching pages
here's the poor accessibility w/ the page shifting due to the font changing on every page swap
yeah the font issue is indeed bad
they don't have this problem because they don't scroll the sidebar programmatically, the shift depends on your browser's handling of scrollIntoView when the element is actually visible, I would accept a PR to manually check for that
what level of throttling is this
i set it to 4g fast, but i have this issue anytime im at my parents place
and they dont have crappy internet, but its not fiber or anything
I'll try to reproduce it
im assuming this is caused by font loading
and my guess is the browser is not caching it
I don't see any shift on my default connection speed in my firefox instance since the font & logo have been preloaded
if you don't have the issue on the main site that also uses the poppins font, just embedding the font into the stylesheet might be our best bet
well the mainsite, u dont click around on pages with the same page structure as u would on the docs site, so its much harder to notice
but the mainsite also has the problem
the issue is on the font assests ur giving a no cache header, there shouldnt be a need to redownload the font on every single page loads, or well any of these assests on every page load
yea the main site doesn't have the same rules as docs do
I thought it embedded the fonts in the stylesheet though, might be pre-astro rewrite behavior
That is a request header, you set that. I cant check myself right now, but by default devtools disable caching if open
[07:17:27 ERROR]: [StarRaid] Uncaught exception in scheduled task
java.lang.IllegalStateException: Asynchronous getNearbyEntities!
at org.spigotmc.AsyncCatcher.catchOp(AsyncCatcher.java:10) ~[paper-1.21.10.jar:1.21.10-113-9fc21bc]
at org.bukkit.craftbukkit.CraftWorld.getNearbyEntities(CraftWorld.java:979) ~[paper-1.21.10.jar:1.21.10-113-9fc21bc]
at org.bukkit.craftbukkit.CraftWorld.getNearbyEntities(CraftWorld.java:974) ~[paper-1.21.10.jar:1.21.10-113-9fc21bc]
at org.bukkit.World.getNearbyEntities(World.java:1681) ~[paper-api-1.21.10-R0.1-SNAPSHOT.jar:?]
at org.bukkit.World.getNearbyEntitiesByType(World.java:1114) ~[paper-api-1.21.10-R0.1-SNAPSHOT.jar:?]
at org.bukkit.World.getNearbyEntitiesByType(World.java:1068) ~[paper-api-1.21.10-R0.1-SNAPSHOT.jar:?]
at org.bukkit.World.getNearbyPlayers(World.java:987) ~[paper-api-1.21.10-R0.1-SNAPSHOT.jar:?]
at com.destroystokyo.paper.ParticleBuilder.receivers(ParticleBuilder.java:247) ~[paper-api-1.21.10-R0.1-SNAPSHOT.jar:?]
at com.destroystokyo.paper.ParticleBuilder.receivers(ParticleBuilder.java:194) ~[paper-api-1.21.10-R0.1-SNAPSHOT.jar:?]
at com.destroystokyo.paper.ParticleBuilder.receivers(ParticleBuilder.java:148) ~[paper-api-1.21.10-R0.1-SNAPSHOT.jar:?]
at StarRaid-1.0.jar/dev.threeadd.feature.emp.EMPManager.lambda$startEMPTicker$1(EMPManager.java:67) ~[StarRaid-1.0.jar:?]
at StarRaid-1.0.jar/dev.threeadd.core.util.ScheduleUtil.lambda$wrap$0(ScheduleUtil.java:88) ~[StarRaid-1.0.jar:?]
at org.bukkit.craftbukkit.scheduler.CraftTask.run(CraftTask.java:78) ~[paper-1.21.10.jar:1.21.10-113-9fc21bc]
at org.bukkit.craftbukkit.scheduler.CraftAsyncTask.run(CraftAsyncTask.java:57) ~[paper-1.21.10.jar:1.21.10-113-9fc21bc]
at com.destroystokyo.paper.ServerSchedulerReportingWrapper.run(ServerSchedulerReportingWrapper.java:22) ~[paper-1.21.10.jar:?]
at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1144) ~[?:?]
at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:642) ~[?:?]
at java.base/java.lang.Thread.run(Thread.java:1583) ~[?:?]```
```Sends the particle to all receiving players (or all). This method is safe to use Asynchronously
Returns:
a reference to this object.```Wrongly documented as thread safe?
When I changed to sync it worked fine (see #paper-dev message)
create an issue on PaperMC/Paper
okay
also what is ScheduleUtil
@wispy python your problem is you are using non async thread safe methods (namely receivers) on the ParticleBuilder in async
only spawn() method is async safe
private static void spawnParticle(final Location someLocation) {
ParticleBuilder explosionParticleBuilder = Particle.HAPPY_VILLAGER.builder()
.location(someLocation)
.offset(0, 10, 0)
.count(0)
.receivers(32, true);
Bukkit.getAsyncScheduler().runNow(ParticleTest.getPlugin(ParticleTest.class), task -> {
explosionParticleBuilder.spawn();
});
}
My method works fine
Spawn is documented wrong aswell tho
it's not?
When running async it lags out my connection, we’re talking bout it in #paper-dev
how is that documented wrong
the docs are factually correct, there is no mention of performance of what you're talking about
If you see something wrong with our docs, submit a pull request to resolve it. This channel is about docs.papermc.io, the docs website.
oh mb
https://papermc-feat-line-selection.diffs.pages.dev/?github_url=https%3A%2F%2Fgithub.com%2FPaperMC%2Fdiff-viewer%2Fpull%2F49#web%2Fsrc%2Flib%2Fcomponents%2Fdiff%2Fconcise-diff-view.svelte.ts:R1255:R1268 svelte attachment api is cool (totally not just showing off the new line linking feature)
if someone wants to figure out how to make an action that comments a link to look at a diff-viewer PR diff in the PR's preview deployment that would be cool 
I cannot really test whether it works
, but this would be my proposal: https://github.com/PaperMC/diff-viewer/pull/50
I think the wording is a bit weird
But nice idea
Makes perfect sense to have this imo
Although using a random paper commit could also be funny
You've got a better suggestion?
I very much struggled with how I wanted to word it
i'm surprised that refined cloudflare pages action doesn't give you control over the content of the comment
personally i would incorporate it within the cloudflare deployment message and include it as another column, e.g. "PR Diff Preview"?
The PR itself looks good, maybe "Click here to preview this PR's diff using it's deployment URL"?
I mean, we could do a fork of the cloudflare pages action, that's not a big problem
And thinking about it, I could probably also just straight up modify the comment
That actually sounds like a fun idea
Be nicer if this was an action on the repo u wanted it to do
But generally a nice idea
wouldn't make sense with the deployment url part since that's specifically for diff-viewer PRs
but in general using the diff-viewer website yeah i'd agree
maybe just "Click here to preview this PR's diff"
but yeah it would be nice to customize the message
I'm planning to make an extension to the troubleshooting guide with something along the lines of "I/someone can't join the server"
Is it fair game to copy a factoid into there if I point this out?
#paper-help message
@random pollen https://github.com/PaperMC/packages-maven-mirror/blob/master/src/index.ts#L40
Is this HEAD on the wrong line?
that is the wrong line for sure
Head: not-present (France)
https://docs.papermc.io/paper/migration/#:~:text=Should you ever want to go back to Vanilla%2C follow the Vanilla Migration Guide closely%2C as manual changes will be required.
is that correct to say in English?
Yeah, that is perfectly acceptable English grammar
what do you think is wrong with the sentence?
English is not my native language.
Never saw this usage for "should"
https://docs.papermc.io/misc/downloads-service/
You should always serve & use the stable builds. Experimental builds are prone to error and do not receive support.
- Ampersand instead of and?
- Experimental builds don't get support? But... we do help fix issues with them, that's how we fix many bugs!
- Should this be modified from always serve to always offer by default (or similar phrasing) to encourage stable but acknowledge beta and alpha exist and some users might want to try beta at the very least?
Yea, the wording there looks kinda weird; Experimental builds come with a higher expectation on the user to know what they're doing,is I guess what is trying to be conveyed there
the sentence should probably just be scrapped in favor of an entire section on the channel nomenclature
@sharp pivot is that property already inside of Paper, meaning it could be merged right now?
I cannot be bothered to validate it atm
it is
Great, all I needed to hear
might it be worth not having the word "stable" in the tagline for experimental paper builds?
experimental […]. proceed with caution
I don’t think paper provides support or guarantees it’s 100% stable so people can’t sue papermc for it.
can i ask questions in here if i have trouble wit paperMC plugins? >My core i use on my server is paperMC, where can i get help?
im new to this server
this server is generally not the place for seeking help with plugin administration, you can try helpchat from #community-guilds
okay thank you sir!
<@&1190722804849582152> https://docs.papermc.io/misc/downloads-service/ needs updating, did not get updated after the pagination change to GraphQL endpoint. Also, the "lastest stable build" ones don't work if the most recent version has no STABLE channel downloads - those need adjusting
somebody borked docs deployment?
https://docs.papermc.io/paper/vanilla/#paper_global_spam_limiter_incoming_packet_threshold - is also incorrect, it's sorta like, how frequently you can send stuff, setting it that high just means that once you bust the limit you never recover
outlines on this type of icon button or no
(the color is changed as part of a broader change where I'm reducing the blue-500 spam
)
what did blue-500 do?
Looks weird at first glance without the blue
the blue is too much imo
Looks like diff viewer can't read initial commit from github without parent
@random pollen fix it 
known, there’s a comment on the URL parsing function
might also be an issue idr
get out of my account
get out of my head
add file extension filters sir
And a filter for moved files without content changes?
@violet bough sooo, regarding PaperMC/docs#697, which talks about the removal of the project setup using the MCDev plugin, I'd like to talk to you as a representative of MCDev.
To be frank with you, I fully agree. The default plugin template is really not that great. There are some unnecessary parts in the build script and generally also some straight up wrong stuff (I don't know if that is relevant anymore, but I remember that the default setup did not work the config cache; maybe that has changed since the last time I've used it, which has been ages frankly).
But I think this being the case is sad, as the plugin has other great uses as well, which are lesser known (like the event method generator, which I have seen nobody use). Instead, I wanted to directly ask you, whether you would accept a direct contribution to the template itself to modernise and fix it in order to keep that plugin as part of our Paper docs.
I realise I could've simply opened an issue on your repo, but I think having this in discussion openly like this could be nicer and more reachable.
only real thing is that the templates aren't kotlin scripts, which kinda sucks a bunch
rage bait 😆
Not really rage bait, the community is moving away from groovy
poor groovy.
we encourage people to PR improvements to the templates https://github.com/minecraft-dev/templates/
Alright very well, I will shoot up a PR and you can tell me what you think of it
I think that's the nicer option to just completely nuke it off the docs site
Planning on getting rid of all the groovy build scripts too
Great, we should encourage people to use kts Gradle anyways
only use groovy since those templates were originally put together prior to when kotlin build scripts even existed
Fair enough
Never done any IJ templates, how do I test these?
Oh well, it is explained in the README, fair enough
unfortunately if you go to test this the local repo is currently broken because jetbrains did what jetbrains does
So, just give us some time to fix that
Unrelated but I am a bit taken aback by how you worded your original message, if I would "accept a direct contribution to the template" - mcdev has been an open source project since day 1, over 10 years ago. I've never been antagonistic towards accepting contributions, the way you presented the message feels as if you expected some kind of push back from me. I assume that isn't what you meant, but it was jarring for me to read that, confused by the tone I perceived
Oh god, excuse me a lot for that; that was obviously not my intention 😅, please forgive me for the wording
it's all good, just a classic case of tone coming across wrong via text. It happens, and sorry I read it that way
I just wrote that on a whim, so maybe re-reading that could've been a good idea
I don’t understand why you want to do that like it is the worst part of the documentation.
That's what the issue requested, not what I want to do

Right, that's fun
@drowsy zinc omg they finally closed it https://github.com/expressive-code/expressive-code/issues/342
Oh sorry
I read it as we can do this and that
And not yeah this is a better alternative
Sorry
Yeah, I've been kinda wanting to improve the mcdev templates for a while now, I just kinda never got the push I needed
even if the template is improved, I honestly think that part of the documentation is redundant to say the least
The clicking buttons part, perhaps
a recommendation to use mcdev on the top as an alternative to manual setup would've been enough
But the mcdev template project setup step is fine I believe
Remember, this is for people who have never made a plugin before and might have very little Gradle knowledge
I don't think that's productive given we will in fact just send people to learn java if they don't know how to follow that guide
Knowing Java is a different topic from knowing Gradle
but well, it is already part of the documentation so there's no harm in just leaving it there at this point, I am not that against it
You can know Java without having tooling knowledge
I would disagree, you need to be able to understand the tooling at least on a basic level in order to understand java on a basic level
Or you might come from a different background, like C#, meaning you are not entirely lost when doing Java dev, but might still not have any Gradle knowledge
like, I don't need to understand Gradle to just add compileOnly("...") to my build script
That's very wrong. IntelliJ default project creation is their own system which does not concern itself with neither Gradle nor Maven
The template does other stuff as well, like set the toolchain java version or add a processResources task for plugin yml templating
that is not necessary to get your plugin going, it is just a nice to have
I am also planning to include the run-paper plugin as a clickable button as that's kinda nice
So what? Are we not allowed to include nice to haves for people wanting to start plugin dev?
I don't see how having a QoL step will hurt anyone
the point is that they have to understand that, and they don't
???, I said they need to understand it on a basic level, at least to the point they can understand where to place the compileOnly
if they can't understand that, then they'll be severly lost when doing anything else
.
Again, Gradle != Java
and again, it doesn't matter
you dont
they don't need to know how to make gradle plugins to understand how to add a dependency to the dependencies block
99% of gradle features you dont need for basic plugins
Why exactly does it matter then? Again, is there anything actively bad about us recommending to use the mcdev wizard for project setup?
Alright, I'll be abundantly clear so as to not cause more confusion. As mentioned above, I don't think it is bad to recommend the mcdev plugin, the plugin itself is fine regardless of the templates being "outdated", but I think it is redundant if not counter-productive to the people reading the project setup documentation to tell them how to setup the project manually, then go on to tell them how to do the same thing but with an IntelliJ addon. If they understand how to install the plugin, I don't believe they need help to know how to use the wizard, and if they do that'd be up to MCDev's documentation and not Paper's. The only reason I see it as counter productive is because people often follow guides on a step-by-step basis, and I've helped people who have followed that guide setup a project twice just because the project setup guide explains how to do the same thing twice
the recommendation to the plugin could just be a note on the top of that guide, but the way it is now feels redundant if not hard to understand
i dont see it
well, you don't have to agree with me, that's just the way I see it
if it says
Alternatively, you can...
how is that confusing or hard to understand?
And if people cant read docs that not papers fault
I am semi certain most positions in this have been stated 😅
I am not going to argue this point any more than I have, so yeah ^
I just wanted to complain about fill-ui looking awful but that was just dark reader, wtf
I think it looks a bit weird. But I can’t tell why.
would be surprised if it sucked since it's just shadcn if I recall
By the way, could you ping me for when you've fixed it? No rush on that, I'd just like to know
https://github.com/PaperMC/docs/issues/117 can probably be closed with data driven enchant now
I think that's fair; it is kinda documented in the registries page as well after all
https://github.com/PaperMC/diff-viewer/pull/58 filter: Modified kts files not in test-plugin
~~not sure why it's showing the AT diff though
~~fixed
hi docs team here https://github.com/PaperMC/docs/pull/698 thank you :)
Related, the graphql playground's starting example doesn't work.
CC: @pliant canopy as riley is currently busy. 
ya
Great, all I needed to hear
it's just the new pagination stuff and a more robust shell script
pagination is just the standard graphql shiz
is the sort needed
you can thank gemini 3 for that
oops accidentally pushed a local branch
lol
smh scorp
Hello
Uh hi? If you're looking for a general chat/conversation channel it's here: #general
I need paper file 1.20.4
is there reason downloading the .bat n running it in a file w server.jar it says 'java is not a internal or external command'
(on windows 11)
it worked on windows 10
This is the channel for discussing our documentation and website. Follow these directions and if you need more help, visit #paper-help
oh am I an idiot and forgot to install java on new pc 🤦♂️
if somebody is bored, I think we need better error handling in javaducks, like making sure errors dont get cached and maybe check that all filesystem things are handled correctly https://github.com/seiama/javaducks/issues/30
the cache is not a cloudflare thing?
thats ontop
that restart command was more a spigot command (might need a new section) but looking at it there's several missing command permissions here such as:
/spigot reload (spigot)
/mspt (paper)
/tps (spigot)
also not sure why https://docs.papermc.io/paper/reference/permissions/#paper-permissions include command permissions that are not in the first section
Docs channel got demoted below build-tooling-help :(
-# (I noticed earlier, but only now thought of something funny to write)
@ cat stop fat fingering
It certainly messed up my workflow
This caused my plugin to not compile
Because it depends on the position of the docs-website channel in my channel list for sha generation
I mean build-tooling-help is probably more active / needs more prominence for help topics.
But yeah I'm guessing it was an accident.
another famous cat drag and drop
It's always the brits
or mbax and his “occasional hand tremor thing” 
build tooling ended up entirely out of this section earlier so I put it back where I thought it went.
It's real and sucks, thanks. ❤️
plugin messaging could use a section on how those custom payloads work outside of bungee stuff
that a modded client or whatever puts what it wants to receive into a minecraft:register custom payload, and only then can you send on that channel via player.sendPluginMessage
any idea what this is? https://github.com/PaperMC/docs/actions/runs/20937495228/job/60163537735?pr=714#step:6:64 didn't get the error locally when i built the docs
it's been happening on and off on the ci for a while and I haven't gotten to looking into it
I re-ran the job and it should be fine I think
I blame @scenic gull
he infected github actions with kittens
and they are eating the pages
what do you have to say for yourself @scenic gull!
._.
can you infect my house with kittens instead?
:(
I blame scorp
No other possibility
He's the astro guy
The frontend dev of us
The typescript enjoyer
5 ignored messages
I diagnose it with failed build
@sharp pivot https://github.com/PaperMC/docs/pull/709
my bad, didn't think anyone else ran into it yet
I could merge it if you think it's good
I don't know maven so I left it for comments
already approved but, maybe having separate tabs for maven 3 and 4 would be good
i could see someone not reading and copying the whole thing into their pom.xml
will do
dont be stupid like this guy
8
Do you have a comment about our docs/website?
It was an accident i think
hello
https://github.com/PaperMC/docs/pull/720/ should we also mention velocity 3.5.0 requiring 21?
Yes
that can be another PR if wanna rephrase that
How do you change the version of the Minecraft server? I wanted to set it to 1.21.1
thanks friend
i just tried to access adventure JD and as usual went to jd.papermc.io but couldnt find it there. Did you already discuss/think about adding it? Is there any reason to not add it, now that is is part of papermc?
it's probably on the bottom of someone's todo list, given the various random changes that would be necessary post-migration and the fact that it works fine
adventures JDs are a lot more split out vs the primary platforms we have, I think it's sorta on the todo list but not a high priority and we'd need to figure out how to deal with 20 dozen artifacts
we could just add a button that links to jd.advntr.dev
that's what i meant ^^
you can suggest in the website repo for consider that
i decided to ask here first to find out if this has already been discussed. I didn't want to open an unnecessary issue
dude Strokkur24 was just acting like an ass wipe in my pr rn to me
bro didn't even bother to check
closed it and thats it
strokk is sometimes stupid
and cannot read two lines above
where LiteCommands is also in that table
yep
bro should at least bother to ask
if too lazy to check
am i supposed to open it again?
Yea please do 👍
pressed the button for you
Look, I made a mistake, no need to be personal about it
thats not it
I didn't notice the exact section you PRed it to
i sent this on 2 lines
he said you cant read 2 lines
ok jokes are done now
🤝
im sorry for calling you an asswipe tho, you seem cool
today was not my best day is all
mb
Yeah, it apparently wasn't my best day either
Glad we got it figured out, and sorry for closing your PR
@_@
For transparency you should really provide a reason why you don’t accept their PR
Since there seems to be another command framework present there as well
(This user is no longer in this discord.)
Transphobia/homophobia, bigotry, and ban evasion. The user was originally banned over at Kyori back in 2024, but during the organization move they became unbanned.
Okay I see thanks for clarification
The lifecycle example seems a bit confusing to me because it doesn't show that you need to register the handler in the lifecycle manager or not?
-# https://docs.papermc.io/paper/dev/lifecycle/#lifecycleevents
It's in the https://docs.papermc.io/paper/dev/lifecycle/#registering thing afterwards, but I agree with you, it's a bit confusing
I think that page could use some general love with an example or two anyways
ah yea thanks, that's what I get for reading the full page :)
that's kinda what the registries and datapack discovery page are for but not sure if they are linked?
They aren't really linked directly
I think it needs a cleanup anyways
Also need to add in the tag flattening things
Just popping in to note --add-plugin-dir seems missing from live docs 🙂
I don't think we have any docs on the jar args actually?
That's system properties
my docs where is code
?
Hangar allows you to find and download the best Minecraft plugins for your Minecraft server
??
You want #hangar-help instead, you should write your plugin description in Markdown
Maybe note in the Velocity docs that the annotation processor only works for java and for kotlin you have to manually use the velocity-plugin.json or kapt
falls into that "and what about scala?" territory; Annoyingly, for as niche as kotlin is in this community, it seems to pop up a lot, especially with fairly common traps and kotlins documentation last I rember wasn't the most amazing here
kotlin is not that niche considering how often i see people ask about kapt
not even in the same ocean as scala for plugins haha
niche was maybe a bit harsh; it's used but I rarely see it here; we have 152 results for kapt
I still don't understand why anyone would use the AP, it's more complicated and annoying to setup than just generating the file especially if you want to substitute any values from your build script
I'm partial to https://github.com/jpenilla/resource-factory but even just good old processResources and a json file seems more convenient than the AP to me
yea I think we should prefer the ap in the docs if not deprecate it
once someone does get the AP working, their next thought will be "ok now how do I insert the version number from my project" and with the AP the answer is you need to use source code templating or preprocessing
which is super over the top
I mean, the AP does feel kinda cludgy, like it's often come up especially with stuff like maven no longer scanning for them anymore
(also kapt sucks, if people actually want to care about the AP I had a PR adding KSP support a while back)
honestly, it's more in a area where the json file is basically undocumented and the AP is the only documented way to do it
probably better to document the json file at this point tbh
Yea
yeah I had to open the velocity src to figure it out when I implemented resource factory
i agree with documenting the json file
That sounds sane
Perhaps we should also update... uh... the mcdev templates
@violet bough god damnit

I had a stroke writing the description for it, but here we go https://github.com/PaperMC/docs/issues/731
Either someone gets to it relatively soon-ish, or I will fix it whenever I have my next docs writing spree and go through some of the issues
fill me in on context I ain't readin up
The Velocity @Plugin annotation, which combined with the Velocity annotation processor used to generate the plugin sjon file (basically a bukkit plugin.yml, but it's a json file) should no longer be the main recommended way in the docs, and for that reason the mcdev template needs updating to instead generate a json file which should be edited manually. Basically this issue: minecraft-dev/templates#46
@drowsy zinc opinion on removing the status: accepted label, as I don't have a feeling it does anything?
isn't that for issues (feature requests)?
don't know if my opinion matters, but if you want to make a PR for an issue it's surely better to see maintainers agree with it
if a maintainer disagrees with whatever the issue poses then it'd just be closed, feels redundant
Perhaps, but then again what if an issue has been open for a while and hasnt received any comments from maintainers? Best you can do is assume the maintainers are okay with it, whereas a label makes it clear
Currently, all issues which are open would be accepted, at least to some extend. And if an issue is labelled in general, you can already treat it as accepted, as that is a sort of maintainer comment itself
I guess the one issue not accepted how it was posted was that mcdev plugin removal issue, but that was dealt with already (including a discussion in the comments)
And as abby said, if you are unsure, it takes about 10 seconds to open this channel, link the issue, and ask if you may open a PR for it
fair enough
Could it be nice to have a feedback post in #1120783097802276906 for the Pre-Player-Obj-Perms API?
-# And close the old/completed ones
<@&748618676189528155>
Maybe can be consider a redirect or something for old javadoc urls like https://papermc.io/javadocs/paper/1.18/org/bukkit/entity/ThrowableProjectile.html ?
Currently just go to the default 404 page (still in queue a PR for modify that 404 page)
part of the reason for moving stuff to the new URLs is because adding redirects in the new infra is a bit of a headache, especially if you're wanting them to be smart
I guess maybe with our cloudflare plan we can get away with it now
i believe this section needs to be corrected as its no longer true after the hardfork
https://docs.papermc.io/paper/migration/#from-craftbukkit-or-spigot
Paper retains full compatibility with all Spigot plugins, allowing a seamless transition.
Yes, probably a good idea. Open an issue!
Does this work in terms of rephrase (cc @scenic gull)
Up until 1.20.5, Paper retained full compatibility with all Spigot plugins. However, Spigot plugins made after that version may not fully work. Such plugins may often provide Paper-compatible builds. If they do not do that, you may need to look for an alternative.
I would be more specific, as in, spigot plugins that use spigot api released after 1.20.5 will not work on paper
but additionally something like most plugins are written for paper? idk
Up until 1.20.5, Paper retained full compatibility with all Spigot plugins. However, plugins using Spigot API added after 1.20.5 version will not work. Such plugins may often provide Paper-compatible builds. If they do not do that, you will need to look for alternatives.
Look for*
Probably a final remark that most plugins are built and tested against paper these days?
Up until 1.20.5, Paper retained full compatibility with all Spigot plugins. However, plugins using Spigot API added after 1.20.5 version will not work. Such plugins may often provide Paper-compatible builds. If they do not do that, you will need to look for alternatives. However, nowadays most plugins are built and tested against Paper, so you typically will not encounter incompatible plugins.
Now it's lengthy
people dont need to read it, its like a side note
Docs are for the second pass at learning 😂
Not docs in general, just that disclaimer, lol
I see it as a, hey, we told you to double check! When somebody complains about some plugin by some of the backwards thinking people that go out of their way to break paper compat
But since the fast majority of servers run paper derivstes, not spigot, any self respecting plugin dev tests on paper
I mean yeah, if it took this long to even notice, this probably isn't too often looked at anyways
Well, or it does it's job of not discouraging paper usage ^^
I think the paragraph should start by saying most modern and up-to-date plugins all already depend on Paper. I agree with mini that we should probsbly not even go into too much details about Spig*t
https://docs.papermc.io/paper/dev/command-api/arguments/registry/#registry-key-previews
this is outdated strokkur, zombie nautilus variant, gamerule etc are not included
take your time but tbh this page feels clunky I think people would get the gist with a single example.
also I want to expose the tag argument at some point so that would duplicate everything
Yeahh, it was fine when I initially wrote the docs, but I forgot that it'd need pretty constant maintanance
I will probably indeed simplify it to just one single example and simply list them in a table
Will do that on my 26.1 update branch, where I will update them to the registry entries exposed in 26.1 as well (idk if 26.1 adds anything, but might as well)
bunch of new sound variant registries iirc
Does that add new registries though?
well like wolf currently
i also think that the page is a bit too verbose
What's the stance on having some very basic NMS description page, which gives a rough overview of, for example, bukkit <-> nms object conversion (cast+handle), maybe some help on how packets generally work, perhaps with reference to the multiversion paperweight example plugin, and just generally a guide what to watch out for? I know the whole NMS trickery is super unsupported and docs don't exist as concrete implementations are prone to breaking quickly, but some of the more stable internal things I believe could be useful; also in view of, for example, data components not representable in API or reusing internal logic not exposed
Not much of a fan on my end
i never touch the docs for anything apart the configs anyway let alone do i contribute there so my opinion is probably not worth much
i think people who would profit from such docs the most are those who dont even know the api itself well enough to even consider diving down the internals rabbit hole – which means either they dont read much code and are in some kind of tutorial hell or they just started working with paper/minecraft and shouldnt touch nms until they actually know the code/api a bit
for the people who have a valid use case for using nms that probably would add much value since they most certainly already know how to work with it and how to accomplish certain things with nms
holy macaroni
sorry for the wall of text, didnt realize my two cents are that long
You should join the docs team, you seem to be quite capable of writing
/s
i just have a lot of opinions on a lot of stuff
i cant even keep up documenting my own sht 😄
also you already covered everything anyone could ever dream of
so no idea what i would even document
yea, thank you Strokkur
There are so many things undocumented, I find it impressive
Just check the issues tab, surely you would find something

I think anyone capable of working with NMS can, trivially, identify the things you're proposing documenting. If they cannot, they are not ready to poke the bear.
I generally like this idea, but I don't think it should be part of the Paper docs or anything making it seem official/supported like that
It's time for Strokkur.dev blog
add the kotlin version of the examples
Paper does not use Kotlin anywhere in the codebase, and the API is also written purely for Java. Kotlin is not a directly supported language, so we probably won't add Kotlin examples anytime soon
at least i tried
also not necessary bc you can just paste java code into IDEA and it'll auto convert it to kotlin code
<@&748618676189528155>
going to pull the trigger on the dependency updates, if anything breaks I blame it on the lack of reviews 😛
You are a better front-end dev than me, so just view my lack of review as the trust I place in your hands

when will they update for 26.1
as I expected it fails to build if the version header is set to 26.1 since it expects it to start with 1
What fails?
omg I think I know what is going on, it parses it as a float
26.1
what in the YAML like bullshit is this
[InvalidContentEntryDataError] docs → paper/dev/particles data does not match collection schema.
**: [
{
"expected": "string",
"code": "invalid_type",
"path": [
"version"
],
"message": "version: Expected type `\"string\"`, received `\"number\"`"
}
]
ye error message seems to confirm that
@pliant canopy what did i break https://github.com/PaperMC/docs/actions/runs/23604296141/job/68742642527
it all looks fine to me
is the branch just out of date or smth?
yes it might be
Funnily enough, no, it just likes to do that sometimes
Re-running the job (which happened with your update) typically fixes it
@timid cedar wouldn't you perhaps also want to include the MiniMessage presets in there?
Could be a great chance to do so
yeah good shout
done!
@pliant canopy was just thinking the docs update to 5.0.0 but that isn't released yet
do you wanna maybe hold off for now or should we update them to 5.0.0-SNAPSHOT or smth
I'd just merge it once the first Adventure 5.0 release goes online
fab
that was indeed because the branch was out of date fwiw
Okay cool, fuck me then
Everything I say is wrong
The universe is an egg

the d2 plugin uses d2 in wasm pulled in by a normal npm dependency now so the d2 binaries are no longer installed in the workflow
Ah, did you do that with your dependency up PR?
yes
Well, that's great
Theoretically I'm a Contributor cause I PRed and got merged
it's that easy 
Yup
But it’s on a case by case basis to get contributor role. I also contributed a few PRs and didn’t get the role 🤷🏼♂️
I also never got the role, even after multiple Paper PRs
isn't that one just added for people who aren't in the team to speak in #paper-contrib
Contributor role is for someone who has continuously contributed for a while now
No, that's the paper-contrib role. Contributor also gives access to a few internal channels
I see
I never got contributor role either >:(
This guy is playing two truths and a lie
And he is not that great at it
God our build times are abysmal
Almost all of that is just image preprocessing
Anyways, @mighty dirge, if you have time, I'd greatly appreciate if you'd check over #735 again. I did some changes based on your comment
If you sign off on it, I'd prolly merge it soon
-# np, I'll do it tomorrow though
I think it's fine as-is now. The duplicate entries across the MC dev plugin and the velocity plugin json reference is not really a problem imo. Also that MCdev creator thing will be outdated anyways once my Velocity template PR goes through, so that's something to fix for then I guess
@drowsy zinc if you'd like to give #735 one last look over, I'd appreciate it. Will merge it afterwards
scorp, do you think this looks fine or is that too much text?
Have you considered using https://expressive-code.com/key-features/text-markers/#marking-individual-text-inside-lines ?
How would that help me though? The main thing I am adding here are inline comments, it's less about marking stuff
There's also ofc the alternative of just saying "the marked fields are required", idk which is better
perhaps make both and see which looks better?
not really a fan of the current setup, but if it looks better than the second, why not
For whoever wants to update docs about world storage, the new procedure for going back to vanilla is to move these data files from <level-name (world default)>/dimensions/minecraft/overworld/data/minecraft to <level name>/data/minecraft
GameRuleMap.TYPE);
WeatherData.TYPE);
TimerQueue.TYPE);
WanderingTraderData.TYPE);```
and feel free to ping me with questions about it generally
I will include it in https://github.com/PaperMC/docs/pull/741
We can drop documentation about the pre-26.1 world migration, right?
What about migrating to Spigot? Is that something we should remove from the docs page as well?
Worth keeping up the older vanilla bit at least until we have 26 stable imo.
Fair, will move them under a new header, similar to how userdev has the pre-mappings one
there's a new system property too
Hell yeah
Okay so I need a simple yes or no: remove the migration docs for Spigot, both from and to?
Well, I guess from Spigot might still be useful, but I have no idea how their world structure stuff works, it might be so widly different that migrating might just be impossible?
We don't support migration from spigot
Like, and the versions we do support there are generally going to be older versions where we aligned so that's just akin to a paper upgrade as usual
It may be worthwhile to make a note that specifically mentions that we don't support migrations from Spigot, especially after 26.1 due to the difference in world structure.
Yeah, will do that
Could mention users would need to migrate from spigot to vanilla for it. So nobody misses out on being able to enjoy paper?
That's a pretty good point, but does Spigot even offer any tutorials on that?
Or do we just say "figure it out on your own"?
I dunno. I hope they have info on it somewhere.
I don't have anything resembling WanderingTraderData when I started a 26.1.1 Paper test server. What would be the file name for that one?
Maybe it doesn't happen until you get one?
Sure, but I am not gonna wait around until I get one
often the case with world data, that stuff is lazily populated for many parts iirc
did you stop the server yet
I did
did you login
If people could look over https://github.com/PaperMC/docs/commit/b6e60a23bf5d3a1177aadc84bae63866936125b0 for misinformation and style, I'd appreciate that
I would make it say something long ". . .making a direct migration as of 26.1 . . . "
@drowsy zinc we really need to do something against these abysmal build times
I have no idea what to do, so I'll let you look into it 
Which sys property was added?
In case you just happen to know, if not, I'll look into it shortly
Alright @random pollen, I've updated the project setup page: https://papermc-update-26-1.papermc-docs.pages.dev/paper/dev/project-setup/. Done a bit more than just the versioning, also got rid of the groovy dsl because no one should use that + removed the mention of mappings, since it is not needed there. Anyways, mind checking over the dependency declaration stuff if that's all correct?
I mean, are they that bad?
It’s like 4 minutes from cold
I will try and remember to take a look tomorrow cause I think it’s just image processing
Yeah it is. Thank you
I think you can cache it
We theoretically do
But I feel like scorp's D2 update might've caused it to be un-cached again
As a note, the commits in the experimental tab lead to the Paper-Archive repo instead the Paper repo.
What experimental tab?
*At the Downloads page
if it is an old enough build then that would make sense
Oh yeah I see
@hard quartz do something about it, thx
I need to look through Paper source code in search of a newly added system propery sigh
Get AI to do it
https://github.com/PaperMC/website/blob/695a9849a21934ad7b77f3a94b797a0eb00f35b1/src/utils/github.ts#L62 code needs to be updated to support new versioning
I genuinely don't even know where to start looking
If someone just knows, please tell me
Ah? 
the logic is hardcoded for 1.x; needs to be updated to support 26.X
Okay from my side I'd be done with updating the docs for 26.1. If people would be so kind as to look over that (dw I will rephrase that one thing, ocelot), I would greatly appreciate it
I might merge it somewhen tomorrow unless something else comes up
your last change to the workflow essentially made it so the cache would never update even if was invalid
ah thanks for looking at the cache scorp looks good, there are a fair few warnings with cached remote assets but they are just warnings and it looks to be handled gracefully so should be fine
fixed
somebody else will need to continue this https://github.com/PaperMC/website/pull/221
Any issue if use mojira.dev over the attlasian portal in the docs?
the bugs-legacy throw issues by expired SSL and link to comments works bad a few times.
They do seo farm bun tho 
https://docs.astro.build/en/recipes/bun/
2023 issue https://github.com/oven-sh/bun/issues/5951
well better move from bun? xd
Well, it breaks under node with pnpm too, see my PR
i see i try check things like this breaking change... but still cannot find what other thing is broken for even load dev in local...
also try set in adapter the new config for the old behaviour (v5) and just make the error change xd
@spice temple
sorry the ping.. i make "run" with errors but now the site has a preview xd
Does it build? Lol
yep
i ask now the IA in astro Discord for fix the last error because terminal and player count use the env removed... the IDE warns but now "works"
You can push to my branch, that seems like an improvement, lol
wait i can?
i was to open annother PR to that branch xd
remote: Permission to PaperMC/website.git denied to Doc94.
looks like not.. only in the docs repo because has my name or not sure
<
green and also happy to remember i "patch" in the work a prettier for avoid mess with break lines xd
(because most of the people use WSL for touch code.. i use windows and every time touch that project the prettier change all)
You just need to disable core.autocrlf in git
Fuck that setting
Always input, always lf
sure?
i remember try the same inn the work with the same result
try modifying the gitattributes like https://github.com/Incendo/cloud/blob/master/.gitattributes and then running git add --renormalize .
If you change that setting you need to reclone
Or renormalize I guess, not sure if that's enough
working 😄 (i take that for reference in the work xd)
thanks
now only need wait for merge and known if cloudflare not fails in deploy a preview
Oooo Astro 6 update. Nice work.
why are we switching package manager though
That's a good question.
for this this.
Astro also mention bun but its like "if you found issues go to bun repo and report" also a similar report in astro end with.
bun runtime is different from bun package manager
in prod it uses the workers runtime anyways
so idk why we would need to switch package manager, just don't pass --bun or set bun runtime in the bunfig.toml
not sure why in the start the --bun was added.. how all start with the "bun and astro v6 mess with the web" i try first change from bun and later try to read the whole changelog for cloudflare workers
and how docs already use that
already known

Ohh..
pending in merge 
I mean, idk who on docs team usually merges stuff, can hit the button if nobody else minds
oh the old way we did that check is cursed 
the lasts was from Lead team xd
or can apply to docs teams too and self-merge 
Docs? Isn't that website
This is website yea
Last time I merged something on the website, lynx hit me with the
emoji
fwiw doc's changes are trivial and really can't break anything lol

I was already hovering over the green button
Yea, I thought it was on the docs thing 😄
bleh, can't force push it out either
Yeah, that's stuck there now
thanks for the merge
imagine not setting up the repo's rules smh
xd
apply for web maintainer
I mean, I have the same perms on website as on docs to my knowledge
I've just learned to not merge on website
i only in docs but i avoid touch that only for make PR
very good
Is lynx a squasher?
We're all squashers
very based
but in website can be good for merge this and check if deploy a preview xd
that thing about cloudflare where PR created in forks are ignored for previews...
oh yeah forks can't deploy to preview 
in website? nope
and in general when have i always use a fork for branchs xd unless need the repo CI
Huh, the more you know
I know you have commit perms on there though
Did you ever try it out?
i try push to mini branch and get an error for permission
Well rip
mfw I gotta taint my system with bun to do anything with website
Well, you know what
Doc
If mini will be mad
It's your fault
The build failed

@pliant canopy they allow that so...
for that merge was better the not squash xd
riley will show up beside your bed tonight and look at you angrily.
well maybe fails for the change of the package manager... need wait mini or any with access to cloudflare for the logs xd
There's a very good chance that's true
It's the build that fails. Vite struggles to rollup some astro related stuff.
Good news is with a little tweaks it builds fine and works fine in wrangler 
I need to make sure they're actually sound and not dumb
He can
me as much as he likes
You should be
Aren't you like on vacation or something
Technically
the astro 6 branch?
Yeah
It might just not work anyways
you change the packagemanager or just run dev?
well i feel fine if keep the same packagemanager than docs site...
I've been wanting to replace docs with bun for a while now too, but if astro breaks anyways, what's the point
I'm pretty sure it's just broken from weird astro 5 -> 6 things
well i dont try what jmp says about change configs.. but if astro add warnings for bun...
bunx wrangler dev also worked fine
i dont like changes in web things every version has many breaking changes..
You'll have to build first or wrangler will probably bitch
well if wanna try move again to bun and make a PR can test if the mini PR finally deploy xd
Go ahead
I'll make a quick PR to mini's branch
Well, let's see
Okay that's great
Lemme quickly fix the lint issues
So are we using bun or pnpm now?
I'll probably go with pnpm
so probably follow that
so context has a env... i was sure was removed
The type errors are probably a bad import from workers or something
So, the thing does not build for me

Oh wait
I didn't pnpm i
Oh yeah interesting, some runtime missing
I think it's a bad import on my end
okay weird the way mini did KV access originally is correct
Just gotta declare it in env.d.ts
why did it not build then
because u didnt do it correctly
@junior galleon let's stick to helpful remarks, thanks
According to https://docs.astro.build/en/guides/integrations-guide/cloudflare/#environment-variables-and-bindings the whole astro:env/server thing originally there should've worked
Unsure why it didn't
I mean, the thing I did there should hopefully be fine now, we'll see what cf says about it though
It'll be fine, it was solely a type error failing
That whole check workflow is kinda flawed
It just runs a lint and type check, doesn't even build 
Only run on PRs if the source branch is on someone else's repo
What does that mean
So we just don't get linting if it's the same repo?
Lol
yes for some reason
Let's maybe not do that, that smells like bad design
Oh apparently we do cache, just kinda not really
Well, we don't cache sharp
Which feels kinda important I feel like
It would be if the action ran a build
But since it doesn't it doesn't really do anything 
It should probably also pnpm run build in a 2nd job after the lint and checks pass.
Then it'd be a full workflow for testing
Hm, I would almost put that in a separate workflow
linting and building are two things
Can you cross workflow depends 
I mostly meant it so you could not build unless the tests pass since there's no point in building what is bad to begin with
I am not sure, that'd probably be a CF configuration then
I don't have access to that
I can PR it 
Yes but you can still build
Yes I can
And I intend to do that
But I cannot make it so that the preview doesn't happen if the build fails
this is why you put the deploy in as a workflow too 
I believe we do that in the docs
Eh, this is something that needn't be part of the astro 6 branch
still don't get why we're changing package managers
it does, it was wrongly switched due to confusion of bun runtime vs bun package manager
I remember that convo. However, do local dev previews work with bun?
Because I kinda feel like that was one problem with astro 6 and bun
when take the mini branch no
Hm, lemme test this
Oh yeah
Nice
Okay lemme undo the bun->pnpm move
Oh woops, needs node anyways
Since it's not using the bun runtime
Okay there we go, that should do it
if Ci use node 24 then the package.json can declare min 24 too?
xd
Okay fixed itself after a restart
i hope can have write acces...
but you can change also the packageManager in package.json and the runtime in devEngines?
Right
devEngine I already did
(Locally)
need packagemanager
Bun is not technically accepted into the packageManager field
So I guess just remove that field
and now why Cloudflare is angry again ._.
I am guessing node version
Yep
Well whatever
@spice temple we did some changes to your PR. It's back to bun as the package manager, just using node as the runtime, and everything compiles now correctly
the only change can be if wanna touch the worker for make run in node 24
not sure if worker allow define in a file like pages
Honestly, doesn't look like it
looks like you can
https://developers.cloudflare.com/workers/ci-cd/builds/build-image/
in General
what happens when you desync from your wrangler file 
Nice
Is it all working now?
That's not enough certainty for me to merge it on my phone 
Clicking through the preview site looks fine
it looks and works fine from my brief testing
Am still stuck at my parents but I wanna boot my PC once I am at home anyways, gotta do some accounting, woo
Take your time, not like merging this is critical
is, give perms and i merge and also close the renovate PR xd
I mean I can merge it too, no problem
We are not exactly short of mergers
But I think mini should at least look over it once
Are the 26.1.1 javadocs broken for anyone else? My browser doesn't want to connect
yes its not working yet
Oh alright. The link to the javadocs was already updated to 26.1.1, so I expected it to work
thats done automatically
thanks guys
Great work all!
https://docs.papermc.io/paper/dev/project-setup/#the-main-class
The main class
The way this section is worded ever so slightly implies that you should have a class named main, even though the example shows a different entry class name.
Since naming a class "main" is usually considered bad practice in bukkit-derivative development, is this something which should be edited? If so, I'm willing to make a PR.
the entry in plugin.yml is called main:, so I think that naming is generally fine. But maybe sentence about not naming it literally Main could be added there
Yeah I was more talking about editing the section, not changing the title
Is this just because it’s not actually the entrypoint for the program & also could be class path clashes if every plugin has a Main.java?
I mean, if others add your plugin as a dependency, CryptoScamPlugin is much more descriptive than Main
I see it more as a "Class names should be somewhat descriptive"
that was the "reading code" part 😛
Design is my passion xd
i try the guide for tags in content... not like the design but the logic for take tags from news and make lists or show posts with that works xd
not sure if this is the correct channel, but https://jd.papermc.io/paper/26.1.1/org/bukkit/event/block/BlockExplodeEvent.html seems to be broken, returning no content
cloudflare caches a bad response sometimes
ok... with a little of help i think looks better now.
Pulled the latest updates locally and I'm realizing that the image optimization isn't actually working. While it creates multiple copies of the image for different device sizes, they are all the same size as the original. Our dist folder is over 250MB because of this 😛

Been trying to debug it, but I can't really figure out why.
Looks like a known issue:
Opening a PR with a temp fix. I'll make sure to note it.
Oh jk. It doesn't actually apply the fix in the CI for workers, only locally.
that change works?
i add because in local just not start working...
I confirmed what I was going to PR does apply the image optimizations locally, but when done in the worker CI environment, it doesn't apply it. I think we might just have to wait for fixes.
Looks like we're waiting on this PR: https://github.com/withastro/astro/pull/16194
you can open an issue in website? for mark blocked with that issue and keep follow
I can if you'd like. It'll probably be merged in with @astrojs/cloudflare@13.1.9 / 13.2.0 and it'll auto pick up when we update dependencies and fix itself 🙂
nice
Alright, opened an issue for tracking.
added reference, thanks Github for allow that 🙂
Didn't know that was a thing. Cool.
yeah very useful for that things
At work I would take this chance to write unit tests for bundle size and image size, do we care about that here too?
Maybe just one script that checks the size of the dist folder as a safeguard is enough
Why not 





