[01:17:25] [1/2] Might've found a bug in BucketGet [01:17:25] [2/2] When querying for `where=page_name="{{PAGENAME}}"`, if the `{{PAGENAME}}` in question has a quote in it (which is allowed), then no results will be returned. Querying the same bucket through Lua directly via a where clause, comparing against something like `mw.title.getCurrentTitle(frame).text` or `frame:preprocess("{{PAGENAME}}")` will return results correctly [01:40:40] [1/2] Extension:CategoryExplorer [01:40:40] [2/2] Yet another time we tell folks to not enable random extensions. [01:41:34] Ah yes thank you. I thought it was probably something like that from early on [01:53:20] Quotation marks are used to delimit page names and BucketGet doesn't have a mechanism to support escaping. I hate escaping but if you really need it I might be able to figure out how to do that. [02:00:53] yeah it makes sense. Unfortunately I am somewhat in a situation where my alternative is making my own lua functions for these calls, which I can do ultimately if it's too much hassle to add, but it would be more convenient. Up to you on if/how you want to handle it! [02:16:48] I think your easiest workaround is to use `filter` instead of `where` and query using `page_name CONTAINS "something"`. The performance shouldn't be too bad unless you have more than thousands of rows in the bucket. [02:21:49] Ah that works! I'll try it, I had assumed filter would have the same issue [05:25:40] Filter has the same issue except that you can work around it by avoiding quotation marks using `CONTAINS`. [08:02:56] @ayothepizzahere5039 go woke go broke [08:36:35] rather be woke than asleep at this point tbh [08:36:55] we're all getting broke out of the economy anyway unless you're a billlionare [08:44:07] time to sleep [08:44:13] woke or asleep, im most definitely broke either way [08:44:21] same!!!!!!! [08:44:36] I'm definitely more asleep than woke this Monday morning [08:44:47] 5 hours of sleep today [08:44:48] worth it [08:46:34] isnt miraheze rich and they have a massive hotub at the miraheze hq [08:49:10] yeah they do [08:49:29] a sick ass hot tub [08:56:11] oh yeah I remember that tub [09:08:59] when is miraheze opening Miraheze Park, just like apples HQ in california [09:18:31] [1/2] Quick question, trying to upload a file and it says that I'm overriding some random logo from 2010 that was "Transferred from en.wikipedia to Commons by SreeBot". But there is no name collision that I can see. [09:18:31] [2/2] https://cdn.discordapp.com/attachments/407537962553966603/1505862136851599511/image.png?ex=6a0c2ae6&is=6a0ad966&hm=aa8e957a752bf15341ea8cd06b9f101f16dd5a2639fc70bcd294299dd175e7bc& [09:19:34] [1/2] With a slightly different name I hit another collision but at least this time I can see that the name is the same. Why is it a problem though? Shouldn't these random files be in separate namespaces or something? [09:19:34] [2/2] https://cdn.discordapp.com/attachments/407537962553966603/1505862401914835106/image.png?ex=6a0c2b26&is=6a0ad9a6&hm=a9f1b832d0e665da10ffe1eafb762601a2874aedd46d06459499227253643318& [09:20:21] nah, miraheze feels like they're just scraping by [09:27:43] Name collisions with Wikimedia Commons. You can disable it in ManageWiki/settings [10:36:07] Thanks, found it. [10:58:34] https://www.reddit.com/r/wikia/comments/znzo0o/fandom_guidelines_staff_and_judgement/ [11:01:46] A control issue happened on Fandom,I think MH won't follow the same step [12:05:50] Funny thing, I checked recently and I have over 1000 edits?? That feels very odd to me, I did mass upload files one time and I'm curious if that would have caused it? [12:06:44] Yep thats what it was [12:20:23] I uploaded 1200 images of elements of a game so my actual edit count is the more reasonable about 400 [12:45:00] I have no idea what my reasonable edit count is at this point. I’ve done so many mass uploads and mass edits with AWB [12:53:58] Hmm, is recommendation of Miraheze as platform for "personal website" a good recommendation or not? Just making sure I'm not out of line by saying that it's not really a platform for this (espcially considering scoping content policy). [13:02:46] I would say yes [13:02:56] I have my own personal wiki [13:03:26] But also, depends on what you wanna do [13:03:54] Do you wanna use it as a sandbox or make it about your own lore? If so, then yes [13:04:16] hmm [13:05:04] But if it is more like a personal blog, bio website, or portfolio, then definitely not [13:06:17] I have a semi-personal wiki I use for game design stuff and it works for that. [13:09:27] Persoal sites: Neocities, google sites, wordpress, tumblr [13:09:43] For context, the sentence in question goes "If you are looking to create a personal website, there are many other choices" (miraheze is one of the links in this sentence) so I proposed changing Miraheze to something like Neocities since that's much better fitted for a personal *website* than a wiki platform. Other suggestions there are https://carrd.co/ and Wordpress [13:09:46] My semi-personal wiki contains info about my original lore, code snippets that were made on Fandom at the time, and some other meta wiki-related pages [13:13:03] Yeah in your case I think neocities is a better fit [13:19:33] thanks for y'alls input :3 Have a bat as a thanks: 🦇 [14:03:20] yo how can I rename the link of the main site? i have to rebrand my community because of copyright issues [14:07:24] oh wait a sec [14:07:43] do i change via Repository Database ($wmgWikibaseRepoDatabase)? [14:13:15] [1/2] If it wouldn't be a problem could someone give me a hand with setting up my Miraheze wiki and help me fork over my Fandom Wiki? [14:13:15] [2/2] I'm a fish out of water and have little idea on what to do, it would be greatly appreciated 🙂 [14:14:25] if you want to change the url from a.miraheze.org -> b.miraheze.org, submit a task on Phorge with a reason why [14:15:15] where do I access that? [14:15:20] https://issue-tracker.miraheze.org/ [14:15:39] https://cdn.discordapp.com/attachments/407537962553966603/1505936912102461581/image.png?ex=6a0c708a&is=6a0b1f0a&hm=7011b2052e2225850b3af5622cf18511c77c37740fb6fe467920d464fac385d9& [14:20:01] k im here in the form, which extension do I mention? [14:21:39] Use Form 22 [14:42:55] What if I don't have stewards approval? 🤔 [14:48:50] uh which option do I choose? a or b? [14:49:01] idk what any of this steward thing means [14:50:00] what do you see right now [14:50:03] take a screenshot [14:50:45] https://cdn.discordapp.com/attachments/407537962553966603/1505945748657930400/Captura_de_ecra_2026-05-11_165512.png?ex=6a0c78c5&is=6a0b2745&hm=8ab2885d77ccd89ac8e9037386e4d4fd38390091aede63146488a6f56e889c71& [14:51:15] what do they mean by scope? is it like...the objective or purpose of the wiki? [14:52:34] kinda yeah [14:52:41] so...do I check option A? [14:52:43] the topic/purpose [14:52:58] yes. if its the same, change from `[ ]` to `[X]` [14:53:05] alr [14:53:50] alr done, created the "task" [14:56:00] you need to fill out the reason for change too [14:56:08] you can edit the existing task or send a follow-up repl [14:56:09] https://meta.miraheze.org/wiki/Stewards [15:01:42] Do you mean the two urls u wanna change? [15:01:52] Yeah [15:05:09] Well..... then url can't be changed 👀 [15:05:51] But you have given your reasons for the changes, and it seems reasonable. Still, wait for stewards approval [15:17:15] then... get approval? [15:19:43] Force your way in [15:49:52] woah woah woah, dont force your win (dont listen to the bad voice) [16:55:43] [1/2] If the topic and scope aren't meaningfully changing from what was approved, steward approval isn't needed. [16:55:44] [2/2] If they are, a quick heads up on the new scope via [[SR/M]] is the way to go. The tech team will likely ask you to file a request if there's a substantial unapproved change [16:55:54]

[16:56:58] already done [16:57:17] Great, I am about to go through and clear out the queue [17:09:41] Alright, approved, tech is clear to proceed [17:15:22] I have heard a rumor that on wikis that use Cargo, pages that insert hundreds of rows would time out the server upon being saved. Our wiki is implementing Bucket which seems very similar to Cargo and we have a page that inserts over 7000 rows into one bucket. If we were using Cargo to do the same thing how badly would it break everything [17:15:51] Hypothetically [17:20:00] ("why are you doing that" the revolutionary addition of species-biome relations as a searchable thing) [17:21:09] Gonna be a narrow slice of our technology team ( @posix_memalign and @thewwrnerdguy ) who are deeply versed in bucket, so might need a #support thread for an answer if they're not available [17:21:34] This would be a very funny use of a support thread [17:32:45] this is more of a cargo question than a bucket question, Petrus found issues at around 400 insertions I think? [17:54:25] Oh, d'oh, I read it the wrong way around. Thanks WWR [18:03:09] [1/2] how do i shrink the header so its not covering the side logo? [18:03:09] [2/2] https://cdn.discordapp.com/attachments/407537962553966603/1505994166562455865/Screenshot_2026-05-18_120129.png?ex=6a0ca5dd&is=6a0b545d&hm=99238624f81dea22950e5f743a6da797488567edb1c18ded40ec8990cf975e38& [18:12:59] I'll say it again I'm glad I'm not trying to do this with cargo xD [18:13:26] But I'm absolutely curious just how much it would break if I did [18:41:24] uh, bucket = modern cargo? [18:48:17] New extension being developed by weird gloop, definitely some similarities to cargo conceptually but comparable or better performance in our implementation tests [18:48:22] [1/2] More info: [18:48:22] [2/2] https://bucket.miraheze.org/wiki/Main_Page [19:01:26] Why so many data management add-ons [19:03:02] People have a need that no extension perfectly meets (or have meaningfully different needs), so multiple folks have made attempts targeting different problems over the years. [19:03:25] Excited to see where bucket gets to in a few years, I think it's got the most promise of the current crop to meet several use cases. [19:12:27] Several attempts to replace Semantic MediaWiki that failed to do so either because they had performance just as bad or got rid of the stuff that made Semantic MediaWiki good [19:13:06] What do you think of each one? [19:17:48] [1/5] - Semantic MediaWiki: great and intuitive but so so slow [19:17:49] [2/5] - Cargo: super slow and why did they think inline SQL was a good idea it's both unintuitive to non-programmers and a security nightmare [19:17:49] [3/5] - WikiDB: not tested [19:17:49] [4/5] - Most of the rest: all of them have a data table that you can't write to with inline code. storing data in two different places is unintuitive to wiki editors who expect the wiki page to be the source of truth, and if you have a pre-existing very very large wiki, good luck. [19:17:50] [5/5] - Bucket: It's just cargo without the lag or the inline SQL, it looks amazing so far [19:21:02] I think (at least one version of) DPL has semantic data features too but documentation on them is very hard to find. But its performance is worse than SMW anyway [19:23:17] Bucket is exceptional. It's very simple. Feels like a nice "blank slate" to work with [19:26:21] [1/2] as an actual question, what does bucket accomplish that lua modules can't/do worse [19:26:21] [2/2] might check it out sometime, but I dunno if it's better than just using lua [19:26:45] You can't exactly store and retrieve semantic data with pure lua [19:26:51] Lua can't - yeah [19:27:23] All these structured data extensions are used to solve the same problem: Page A wants to use data from Page B, without needing Page C [19:28:10] Fair enough, though not sure what the use cases for doing that would be as I can't think of any off the top of my head, at least not for the type of stuff I typically use modules for [19:28:34] [1/3] for example: [19:28:34] [2/3] https://outlanders.miraheze.org/wiki/Bread_(Outlanders_2) Bread has a recipe using sticks and flour and stores the recipe in a bucket. [19:28:35] [3/3] https://outlanders.miraheze.org/wiki/Stick_(Outlanders_2) stick queries the recipe bucket for recipes that use sticks as an ingredient. This means I did not manually write the recipe on two separate pages [19:28:51] wrap ur links in < > to avoid embed [19:29:06] oh thanks didn't know that I'm always manually removing them lol [19:29:47] Bucket, Cargo, SMW, WikiDB--they all let you basically write on a page code that defines, "these are the stats for this monster" and then pull that information to be displayed on any page and update instantly when the monster is edited [19:30:16] You can accomplish this in lua with a data module as far as I'm concerned, but I do see the use case there, defo [19:30:30] + simpler to use than lua, so I guess there's that too [19:30:45] And then you can search for all monsters with an attack stat of 38 [19:31:41] I guess you mean that you could just have a module for the data itself, but that means your editors are all required to know Lua syntax and debug Lua syntax to write your wiki [19:31:42] It's probably most useful for things like very large games that get regular updates [19:32:10] true, I guess I'm just thinking of smaller scope that can be maintained by 1-2 people. For large scope usage, that makes sense [19:32:59] Sagan 4's use case is making it where our like 3 maintainers don't have to edit thousands of pages at once on a regular basis [19:33:25] And so that contributors can find info to make their own species very quickly [19:34:12] storing what are semantically similar to typed objects in an untyped language where objects are untyped tables seems silly on any significant scale beyond something that one person is maintaining, especially on an interface with no LSP and no way to utilize things like LuaCATS [19:36:25] To be fair, you can enforce LuaCATS if you use `libraryUtil` [19:36:51] :o didn't know that. I'll have to look into it. I do write a lot of LuaCATS for my modules. [19:37:17] I have my own type checking functions haha [19:37:19] Enforce might be a loose term, but you can do typechecks on runtime and also have unit tests if you really want [19:37:44] Ah yes that's basically what I did on my own implementation. Wish I'd known about libraryUtil before doing it [19:38:55] The species-biome relations bucket is revolutionary because it was humanly impossible to keep a species' current habitat up to date and even smw still has limitations there [19:45:34] Yay! I'm glad my suggestion was helpful. 😄 [19:53:05] Hello I have a small question: does anyone know if CC BY-SA 4.0 and CC BY-NC-SA 3.0 CN are compatible? I've found mixed replies online :xsob: [19:55:15] (Inclusion of unique unidentified biomes is also a big deal bc it makes it easier to find biomes that we missed...) [20:04:44] Generally non-commercial and commercial-permitted licenses won't be mutually compatible, but it highly depends on the context and sequence [20:05:45] (Also, we are not end-to-end experts on all things licensing, though we do try our best) [20:05:56] What if information is taken from the first one and put on the second one? [20:06:12] Dw I know that I just thought someone might know since my issue tbh is with Moegirl [20:10:59] [1/2] If you have a moment can you provide some more info on what this should look like? For example, if I have a page name which contains quotation marks, and use this to filter against with `{{PAGENAME}}` it still must be surrounded in quotation marks for filter, right? so `|select=page_name|filter=page_name CONTAINS {{PAGENAME}}|` still does not seem able to ma [20:10:59] [2/2] tch anything. How does one avoid the quotation marks in this situation? [20:12:16] Hey, does anyone know how to change the favicon on a new wiki page? [20:12:51] Special:ManageWiki/settings#mw-section-styling has the setting you're looking for. [20:12:58] That will set the favicon wiki-wide [20:13:31] [1/2] MGP has a fair bit of CC BY-SA templates from Wikipedia. For those you can simply make a note about the license on top of the page stating that the template is licensed differently from the rest of the wiki. For content pages it's the same thing: simply put the whole page under a different license. [20:13:31] [2/2] If you mix content on the same page it is a lot more tricky. Generally this is not permitted, but you can work around it by making them explicitly separate (just like how images can be licensed separately from text). I intend to provide an example in when I get to finishing it. [20:15:49] Is there something I can do if I don't want a moegirl user to use our info and put it there? Tbh I don't want anything from them especially because of all of the ads [20:16:17] I think I explained it wrong but my isssue is that there's an user that just edits here to move stuff to moegirl [20:16:59] Hmm... actually, hang on. Tokenize tokenizes via either ' or " (whichever is found first). So in theory, something like `page_name CONTAINS '{{PAGENAME}}'` would be okay... right? [20:17:56] So if the page title is `PartA "PartB" PartC` then the query could be `filter=page_name CONTAINS "PartA" AND page_name CONTAINS "PartB" AND page_name CONTAINS "PartC"`. The query matches both `PartA PartB PartC`, `PartA "PartB" PartC`, and `PartCPartBPartA`, so it's not perfect and very odd. [20:20:21] Ahh, gotcha. Hm.. [20:22:34] [1/3] Depends on what the user is writing. If they write original content and only port their own contributions to MGP, then it is perfectly fine because they retain the copyright to their own work and can submit it in as many places as they want. [20:22:34] [2/3] If they move the work of other editors (which are licensed under CC BY-SA) to MGP then it likely constitutes a license violation unless they are extremely careful about it. MGP used to be fine with content pages being under CC BY-SA but that is no longer the case and all content pages are under CC BY-NC-SA now. [20:22:35] [3/3] If you have specific diffs I can raise this issue in MGP group chats. [20:24:46] And as an off topic side note the ads and incompetency of MGP sysadmins (there are several full-time employees btw) were the reasons I left MGP. Fandom is exploitative but as least somewhat competent, the company behind MGP is both exploitative and incompetent. [20:25:02] [1/3] The easiest example I can give of why that might not work is https://battlecats.miraheze.org/wiki/List_of_Enemy_Variants [20:25:02] [2/3] You'd either have to have every single enemy variant perfectly formatted in wikitext inside some lua module, or you'd have to run a bot that looks through template usages and updates the central data store ( [like this](https://battlecats.miraheze.org/wiki/Module:Variant/descs.csv) ). [20:25:02] [3/3] But by making [the template](https://battlecats.miraheze.org/wiki/Template:Variant) use Bucket, each page gets to have its own variant description, maintained by normal wiki editors, and then the list of enemy variants page simply works with no additional effort. [20:26:24] gotcha [20:26:31] they don't really have content edits (they sometimes just fix typos and that's it), I just found out they do this because they added their moegirl edit to the wiki by mistake. They do seem to think that adding a reference section at moegirl is enough but it's not right? [20:31:38] If you can provide the link to the MGP page and the page on your wiki I can take a look. From your descriptions attribution alone is not sufficient because they added the NC clause when copying from your wiki. [20:47:03] [1/5] One example would be something like this: vs in this case they took basically the description and lyrics (I know these aren't ours but we got them from the blu-ray ourselves). The same issue is seen here and [2/5] miraheze.org/wiki/Killer_Stage> [20:47:04] [3/5] Here they took this part of the page / which tbh I could have just made it up [20:47:04] [4/5] It tbh might not be a big deal but the edits are mine and I'm not happy with this. I could give more examples but they're basically all the same: they take descriptions and info like that and they just tl them to chinese ( vs or vs [5/5] ttps://argobdp.miraheze.org/wiki/Egoistic_Sai-%CE%A6>) [20:47:35] I'm sorry this might seem nothingburger but I just feel bad that after researching and getting the info myself it just gets up on a site infested with ads the moment after I save the edit ;; [21:04:02] [1/3] Hmm. Lyrics are weird because the copyright belongs to the original author, so presenting them on-wiki technically constitutes a copyright violation that is rarely enforced. If you took the information off of the blue-ray without modification and then they copied the lyrics over, I'm not sure if there's anything you can do unless you've made changes to it [21:04:03] [2/3] (e.g. by translating). [21:04:03] [3/3] From a cursory glance the description text is different enough from the English version and doesn't seem like a 1-to-1 translation. The images are identical, which raises some eyebrows but again I'm not sure if that's enforceable in terms of copyright. [21:10:02] oh I see, idk chinese but they seemed closed enough to me (◞‸◟). I know lyrics and images are not enforceable but I sent them as an example that they take information from the pages as a whole [21:10:24] even if nothing can be done from moegirl's side, I can at least do it here on miraheze, right? [21:11:01] thanks for the help ദ്ദി ( ᵔ ᗜ ᵔ ) [21:15:41] i have a question [21:16:19] my apologies for budding in a conversation [21:36:59] [1/8] I found a fix for my issue. This is something weird about how `{{PAGENAME}}` encodes characters, I think - it would visually display as tokenizing as the same thing as a working version (where I manually placed `where='Name of the "page"'`, and that would work fine) Replacing my calls to `{{PAGENAME}}` with a hacky function: [21:37:00] [2/8] ``` [21:37:00] [3/8] function p.get_titletext(frame) [21:37:00] [4/8] local title = mw.title.getCurrentTitle(frame) [21:37:01] [5/8] return title.text [21:37:01] [6/8] end [21:37:01] [7/8] ``` [21:37:01] [8/8] works! [21:40:02] ask away [21:41:03] im having problems setting up my icons, background and favicon despite doing what the support wiki instructed me to do [21:42:12]