Wiktionary:Grease pit

Wiktionary > Discussion rooms > Grease pit

Click here to start a new Grease pit discussion.

Wiktionary discussion rooms (edit) see also: requests
Information desk start a new discussion | this month | archives Newcomers’ questions, minor problems, specific requests for information or assistance.	Tea room start a new discussion | this month | archives Questions and discussions about specific words.	Etymology scriptorium start a new discussion | this month | archives Questions and discussions about etymology—the historical development of words.	Beer parlour start a new discussion | this month | archives General policy discussions and proposals, requests for permissions and major announcements.	Grease pit start a new discussion | this month | archives Technical questions, requests and discussions.

All Wiktionary: namespace discussions 1 2 3 4 5 – All discussion pages 1 2 3 4 5

A grease pit

Welcome to the Grease pit!

This is an area to complement the Beer parlour and Tea room. Its purpose is specifically for discussing the future development of the English Wiktionary, both as a dictionary and thesaurus and as a website.

The Grease pit is a place to discuss technical issues such as templates, Lua modules, CSS, JavaScript, the MediaWiki software, extensions to it, Toolforge, etc. It is also the second-best place, after the Beer parlor, to think in non-technical ways about how to make the best, free, open online dictionary of “all words in all languages”.

Others have understood this page to explain the “how” of things, while the Beer parlour addresses the “why”.

Permanent notice

• Tips and tricks about customization or personalization of CSS and JS files are listed at WT:CUSTOM.
• Other tips and tricks are at WT:TAT.
• Find information and helpful links about modules, Lua in general, and the Scribunto extension at WT:LUA.
• Everyone is encouraged to expand both pages, or to come up with more such stuff. Other known pages with “tips-n-tricks” are to be listed here as well.
  
  

July 2024

My poiuytrewq page edit.

I don't really know what happened as I tried to edit the poiuytrewq page to add more content, but it was flagged for vandalism. I honestly don't think what I did was vandalism, and I find my changes constructive. I am hoping my hour of work of changes aren't a waste of time. :/

Also, I left the poiuytrewq edit page when heading to this link, and I have no idea where my edits went, so I can't really show them. If they are gone forever, double screw me. :/ :/

Ehh the page will probably disappear like the rest of the qwertyuiop pages.Metanight01 (talk) 03:48, 1 July 2024 (UTC)

@Metanight01 your edit was in good faith (not vandalism), but it is not in scope for this website. We are in the business of documenting words, not keyboard trivia. To help improve the poiuytrewq entry, you could assist us by finding durably archived citations to help verify that this string is actually used as a word with some definable meaning. This, that and the other (talk) 02:01, 2 July 2024 (UTC)

Quotation template showing up as linked to entries in which it is not used

Why are AFAB, AMAB, theyfab, and theymab showing up as “pages that link to ‘Template:RQ:Burney Evelina’”? This template, which is for an 18th-century book, is not used anywhere in these entries. The only connection which I can see is that the template is used in the entry birth, which also appears in the {{etymon}} template. — Sgconlaw (talk) 03:44, 3 July 2024 (UTC)

@Sgconlaw It's an inherent part of parsing entries, unfortunately, because it involves checking each template in turn to see if it's relevant, and part of that involves attempting to resolve it as a redirect (so that {{l}} resolves to {{link}} or whatever). However, this seems to trigger a "transclusion" in the back-end somewhere; I suspect because something further up the chain (i.e. out of our direct control) has to check the template's page for #REDIRECT ... or some such. I will have a think if there's a way to avoid triggering a transclusion, but if you're trying to find a list of pages which contain {{RQ:Burney Evelina}} then it's probably better to use the searchbar rather than Special:WhatLinksHere, since I assume want to exclude any pages that are merely transcluding it via some other template. Theknightwho (talk) 07:57, 3 July 2024 (UTC)

@Theknightwho: hmmm. Have to say that seems unexpected and weird to my (mostly non-technical) mind. — Sgconlaw (talk) 11:10, 3 July 2024 (UTC)

It's annoying, I agree. The page is technically being read, but it's so far removed from what people understand by "transclusion" that I don't think it should count. Theknightwho (talk) 11:16, 3 July 2024 (UTC)

@Theknightwho IMO this is going to be a big problem as {{etymon}} is used more often. I use "references of" (which is underlyingly the same as Special:WhatLinksHere) extensively in my bot work for determining which pages to operate on. The search bar is not a suitable replacement and I'm not even sure it can be automated by bot. We should file a Phabricator ticket to have them try to fix this but in the meantime I think it is probably best to allow the auto-redirecting feature of the template parser to be turned off, and turn it off for {{etymon}}. I would rather sacrifice auto-redirecting (and figure out how to work around it) than have "references of" become unusable. Benwing2 (talk) 21:56, 3 July 2024 (UTC)

@Benwing2 I think a better solution is to cut down on indiscriminate page parsing, by having {{etymon}} only parse the text in etymology sections. That also has the advantage of being less work for the template parser, too. Theknightwho (talk) 18:51, 9 July 2024 (UTC)

It's not great that WhatLinksHere is being polluted in this way. If the only reason that this is occurring is because of a need to resolve template redirects, it seems it would be better to store a bot- or script-updated list of relevant template redirects in a Lua data module. This, that and the other (talk) 11:33, 3 July 2024 (UTC)

I guess we have to get past thinking of transclusion as only a realis phenomenon and allow for irrealis. DCDuring (talk) 19:06, 3 July 2024 (UTC)

@Ioaxxere: you may know something about this. — Sgconlaw (talk) 04:51, 3 July 2024 (UTC)

Template is rendering wrong

As User:Rasmusklump helpfully pointed out, {{R:sh:RJA}} is rendering incorrectly (e.g. see leden#References_5). Currently, there is an error message asking us to see Module:checkparams, but that is wildly obscure and not helpful. @AutoDooz, JeffDoozan: who recently edited it. —Justin (koavf)❤T☮C☺M☯ 10:41, 3 July 2024 (UTC)

This was probably caused by recent edits by @Theknightwho at mod:checkparams. It looks like it's not safe to make preview-only error messages that go over multiple lines of wikitext. This, that and the other (talk) 11:44, 3 July 2024 (UTC)

Hmm - this was working before over multiple lines of text, but I did change the implementation recently to avoid a hack, and it did require reworking some things, so that may be the cause. Theknightwho (talk) 11:47, 3 July 2024 (UTC)

@Theknightwho It will fail specifically after *, as in * {{R:sh:RJA|id=hello}} - is this what you tested it with? This, that and the other (talk) 13:15, 3 July 2024 (UTC)

@This, that and the other No, it isn't - previously I was using a hacky way to get working <pre> tags, but I wanted to avoid it because it was awkward, so decided to do it a different way that didn't involve the weird hack. However, it seems to make it a lot more "exposed" now, so I'll just reinstate the old version. It's not that bad - just a bit convoluted - but I think having fully-functioning <pre> tags is more important, since it avoids any issues like this. Theknightwho (talk) 13:27, 3 July 2024 (UTC)

Deprecating the head= parameter in Serbo-Croatian headword templates

Currently, none of the Serbo-Croatian headword-line templates make use of the 1=, 2=, ... parameters. Since every one of these templates uses the head= parameter, it makes sense to me to just use the numbered parameters for it instead. I made a rewritten version of mod:sh-headword, based on mod:zlw-lch-headword, which sets head= (and tr=) as a default parameter for each template, and uses the numbered parameters as an alias for head=. It can be found here: mod:User:Stujul/sh-headword.

My question is if someone more experienced than me could look at the rewrite to verify that I made no errors in the code, and if we could then implement this module, and run a bot to remove the head= parameter from Serbo-Croatian headwords. Stujul (talk) 13:52, 3 July 2024 (UTC)

Extra horizontal line below {{also}}

In entries which contain {{also}} at the top of the page but have no table of contents, an extra horizontal line is added between the See also line and the language header. Is this intentional? It appeared a few days ago and I don't remember having seen it before. (See e.g. Tolón.) Einstein2 (talk) 17:48, 3 July 2024 (UTC)

@Surjection This commit of yours [1] to MediaWiki:Common.css could be the culprit, but I'm not sure as it doesn't have any changelog message and I'm not sure what its purpose is. Can you take a look (and add a comment explaining what's going on)? Benwing2 (talk) 21:19, 6 July 2024 (UTC)

The comment says it - it's for the new heading HTML generated by Parsoid. — SURJECTION ^{/ T / C / L /} 21:50, 6 July 2024 (UTC)

@Surjection Can you be more specific as to why this is needed (I don't know Parsoid very well or CSS) and whether it's causing the issue that User:Einstein2 reported? Just saying "Parsoid" doesn't explain in specific what all the CSS is doing. Benwing2 (talk) 21:57, 6 July 2024 (UTC)

Parsoid changes the HTML that is generated for headings, and the old rule did not match them anymore. The actual reason the extra line shows up is because, for some reason, {{also}} now uses a different class when there are exactly two links. — SURJECTION ^{/ T / C / L /} 23:40, 6 July 2024 (UTC)

Thank you. That different class used when there are two items dates back to 2013 and doesn't seem to do anything, so I removed it and it seems to have fixed the issue. Benwing2 (talk) 23:56, 6 July 2024 (UTC)

(Are the changes to headers why things like aWa that 'hook' onto headers stopped working?) - -sche (discuss) 00:06, 7 July 2024 (UTC)

This may well be the case, as they seem to have happened around the same time. But someone who knows JavaScript needs to look into the code. Benwing2 (talk) 00:54, 7 July 2024 (UTC)

Template:R:en:DAFN calling for non existent quote tracking templates

noticed this, not sure what's causing it. Akaibu (talk) 22:48, 4 July 2024 (UTC)

What's the problem? It seems fine in the places where it's properly used. It's intended principally for English surname entries. DCDuring (talk) 23:12, 4 July 2024 (UTC)

go to edit mode on a page that uses it, then page options, then "templates used" you should see a bunch of redlinks to Wiktionary:Tracking/quote/param/publisher and what not Akaibu (talk) 23:16, 4 July 2024 (UTC)

I believe we've been told it's a feature. not a bug. DCDuring (talk) 23:35, 4 July 2024 (UTC)

@Akaibu please see Wiktionary:Tracking for more info. In short, you are free to ignore these tracking templates. This, that and the other (talk) 10:39, 5 July 2024 (UTC)

Courtesy link: {{R:en:DAFN}} —Justin (koavf)❤T☮C☺M☯ 00:24, 5 July 2024 (UTC)

Sorting sufflamen correctly in Category:Latin terms prefixed with sub-

As I said in June, adding "af|la|sub-|t1=under-|sort=flamen" seems to do nothing, so I'm not sure how to get it to sort in this category with other prefixed words built on f-initial roots. Urszag (talk) 05:59, 6 July 2024 (UTC)

Should be fixed. Benwing2 (talk) 21:15, 6 July 2024 (UTC)

Unsupported title

Hello, in Rhymes:Swedish/ʉːt one of the words wasn't isn't linked to correctly, namely Q:t/q:t. The problem is that it linked to wikiquote, see Q:t/q:t. I added {{unsupported}} but I think the module also needs updating since now there is a module error. I don't dare to do it myself so I'd appreciate if someone could do it. Jonteemil (talk) 14:48, 6 July 2024 (UTC)

@Jonteemil: I figured out how to do it by adding the entries to Module:unsupported titles/data, so it's working now. The workaround to hard-code links would be [[Unsupported titles/Q:t|Q:t]], which gives: Q:t. Chuck Entz (talk) 16:30, 6 July 2024 (UTC)

@Chuck Entz @Jonteemil The solution is to do this: {{l|sv|Q\:t}}, where \ "escapes" the colon so that it's treated as a literal colon, not a prefix separator. This also has the advantage of generating the correct pagename, which the old template can't do; this should work in any link template, too. {{unsupported}} is only still used for a tiny handful of links where escaping doesn't work (page names which resemble HTML tags), and will eventually be deleted once those issues have been resolved. Theknightwho (talk) 22:20, 6 July 2024 (UTC)

Thanks for the knowledge! Jonteemil (talk) 22:56, 6 July 2024 (UTC)

Template:gem-decl-noun

The dative singular for *bredą is showing i-mutation where there should not be any. Leasnam (talk) 22:23, 8 July 2024 (UTC)

@Leasnam: I believe this is caused by Module:gem-pronunc#L-308 (called by Module:gem-decl-noun#L-126), where any appearance of "i/ī/į/ǐ/j" in the next syllable causes this syllable to replace "e" by "i". I can fix this by accounting for diphthongs, but for the avoidance of doubt, could you specify which diphthongs should trigger i-mutation? Wikipedia recognises the diphthongs "au, ai, eu, iu, ōu, ōi". (For my curiosity it would be nice if you could also provide a soruce.) --kc_kennylau (talk) 14:00, 11 July 2024 (UTC)

The only diphthong that might potentially trigger i-mutation would be iu (but this is uncommon since iu is itself a triggered diphthong), otherwise none of the others above would. Leasnam (talk) 14:59, 11 July 2024 (UTC)

MediaWiki:Gadget-column-hacks.css

Seems to be unnecessary in 2024 (tested on Firefox 126.0.1). @Erutuon. Ioaxxere (talk) 17:26, 9 July 2024 (UTC)

I removed it. The templates concerned seem to be using a completely different CSS approach now. This, that and the other (talk) 23:37, 9 July 2024 (UTC)

MediaWiki:Gadget-SpecialSearch

Seems to be nonfunctional (tested on Firefox 126.0.1, Chrome 126.0.6478.127, Edge 126.0.2592.87). @This, that and the other, Fytcha (I might add that this doesn't seem like a very useful gadget in the first place...) Ioaxxere (talk) 17:49, 9 July 2024 (UTC)

Yes, it only manages to redirect you to a generic Google/Bing/Yahoo search page. Your actual search query gets lost.

Looking at the JS, it seems to be conflicting with some (new?) code added on the MediaWiki side to the Special:Search form.

I'm not really sure why we have this gadget either. I imagine it dates from an era when the built-in MediaWiki search engine was primitive and inferior to Google etc. (The code almost certainly pre-dates the Gadgets system so tracing this back would involve some wikiarchaeology.) Might be worth posting at BP about getting rid of it. This, that and the other (talk) 23:31, 9 July 2024 (UTC)

Category:en:Neighborhoods... creates by U.S. state but not by U.S. city

Shouldn't we also have "Category:en:Neighborhoods in Los Angeles", "Category:en:Neighborhoods in New York City", etc...? Purplebackpack89 12:38, 11 July 2024 (UTC)

@Purplebackpack89 Yes we should IMO. Something to add to the Module:place code when I have a chance ... Benwing2 (talk) 21:26, 12 July 2024 (UTC)

AjaxEdit gadget makes it hard to locate changes on long pages, like discussion pages

Using AjaxEdit means that Watchlist, Revision history, etc. don't show heading information. That means that specific changes can be hard to locate on long pages, like discussion pages. Can the behavior of the gadget be altered so that using it is just like normal editing of page sections? DCDuring (talk) 14:48, 14 July 2024 (UTC)

Tried to change Proto-Slavic (praxolditi) to (proxolditi), because I made a mistake.

Read title. Ejwarburton11 (talk) 07:56, 15 July 2024 (UTC)

This bot needs some work.

No really, like when I tried to add another grease pit discussion, I got a "this is harmful" message. No I'm not. Ejwarburton11 (talk) 08:47, 15 July 2024 (UTC)

Can you give us any context at all here? —Justin (koavf)❤T☮C☺M☯ 09:00, 15 July 2024 (UTC)

Tried to add ==Russian== on one, for example. Ejwarburton11 (talk) 09:01, 15 July 2024 (UTC)

On one... what? You tried to add a second-level subheading that reads "Russian" on a Grease Pit discussion? Why would you do that? —Justin (koavf)❤T☮C☺M☯ 09:08, 15 July 2024 (UTC)

Context: not a Grease Pit discussion, on prohladit' (been solved, by the way.) Ejwarburton11 (talk) 09:10, 15 July 2024 (UTC)

Then made a Grease Pit discussion discussing that, I got a "this is harmful" message. Ejwarburton11 (talk) 09:14, 15 July 2024 (UTC)

It is so hard to understand what you mean, as there is no entry for prohladit or prohladit'. If you have an issue, please be clear about what it is like "I went to page and tried to do thing but got bad response". —Justin (koavf)❤T☮C☺M☯ 09:42, 15 July 2024 (UTC)

The user has been blocked, but the edits the user tried to make, which were stopped by abuse filters, are visible in the abuse log. (This is not to say the edits were or were not abusive; that's just what the filters and the log are called because that's what they aim to filter out and log.) - -sche (discuss) 15:17, 15 July 2024 (UTC)

Anyone interested in a cognate-tree displayer built atop wiktionary?

I've been working on this: http://radix.ink/word/lang:en/cling. It's very much unfinished and a work in progress. I'd greatly appreciate help, if anyone's interested. T039mwftulnm0l (talk) 22:29, 15 July 2024 (UTC)

@T039mwftulnm0l this is very impressive stuff! I know Ioaxxere is interested in etymology-related technicalities; I suspect you would see eye-to-eye when it comes to the {{from}} template for instance.. This, that and the other (talk) 07:44, 19 July 2024 (UTC)

Category error

I recently created a category and got warning that it is some error in it because it is not added in some module. Searching further, I found a whole bunch of same errors in Category:Categories that are not defined in the category tree. How can I fix all of them? It doesn't look like some glitch, but there are many such lists. Is it bad? Tollef Salemann (talk) 18:59, 17 July 2024 (UTC)

P.S. There you find also some "Nynorsk stem nouns" categories created by @Eiliv which I find very useful to develop ASAP. Tollef Salemann (talk) 19:03, 17 July 2024 (UTC)

@Tollef Salemann These need to be fixed one by one by modifying the appropriate modules. In general it's best to always preview a page using {{auto cat}} before saving, to make sure you don't get this error; otherwise it just makes for more work for the people who clean up this category. Benwing2 (talk) 07:48, 19 July 2024 (UTC)

But how to clean up it? Which exactly module should i search? Tollef Salemann (talk) 19:24, 19 July 2024 (UTC)

@Tollef Salemann Well ... it's a bit tricky, you have to first make sure it would be a good idea to add the category, and then you have to add it to the correct module, which depends on what type of category. There are several topic cat modules for categories of the form 'en:Foo' and lang-specific poscat categories for things like 'Norwegian Nynorsk stem nouns' that are specific to that particular language as well as overall poscat categories for things like 'English foo bars' that apply to several languages. What particular category are you wanting to add? Benwing2 (talk) 20:22, 19 July 2024 (UTC)

Maybe it is not a good idea after all. The main problem i see is that modern Nynorsk spelling doesnt reflect plural form endings of strong and weak feminine nouns. Not really a big deal, but one of ways for sorting it out is to create root categories like for Old Norse. Tollef Salemann (talk) 11:59, 21 July 2024 (UTC)

Activating a gadget

I added the auto-glossary gadget to MediaWiki:Gadgets-definition a few hours ago. It appears in the preferences menu but it doesn't seem to be working. Does anyone know what I'm doing wrong? (@Surjection, This, that and the other, Erutuon) Ioaxxere (talk) 19:42, 19 July 2024 (UTC)

@Ioaxxere works for me! It is rather slow to load for some reason though. Is there a reason the code doesn't progressively render rows as the pages are loaded, or in (say) batches of 10? This, that and the other (talk) 22:59, 20 July 2024 (UTC)

@This, that and the other: I was able to fix my original issue by loading the code from userspace rather than copying the code ([2]). The reason why the code doesn't progressively render the table is that the API will randomly take 10 seconds to respond to a query, and I didn't want there to be rows getting inserted between other rows as the table loads. I did add a batch system but it drastically slowed down the loading speed, so I changed it to only render the table at two points: at 500 entries and at the end. By the way, it's much faster when you're logged out. Ioaxxere (talk) 17:01, 21 July 2024 (UTC)

@Ioaxxere I'm not a fan of storing gadget code in userspace. The main problem is, when something needs to be deprecated (e.g. a CSS class or JS function), doing a search of User namespace turns up a lot of old custom user scripts that nobody uses anymore and don't need fixing. Restricting the search to MediaWiki namespace gets you a more relevant set of results... but it omits gadgets that are loaded from userspace. So it would nice to find a solution to the issue you were having (as well as a couple of Yair rand's gadgets which are still loaded from that user's userspace). It may have been a race condition - perhaps check mw:Gadget kitchen#Running code on page load. This, that and the other (talk) 00:34, 22 July 2024 (UTC)

Evident issue with Japanese template

See Wiktionary:Feedback#マッチ. This may be known to editors who work on CJK languages, but in case it's not, I'm raising it here for visibility. —Justin (koavf)❤T☮C☺M☯ 13:09, 20 July 2024 (UTC)

I thought this was fixed. I'll have a look today. Theknightwho (talk) 13:19, 20 July 2024 (UTC)

Domo arigato. —Justin (koavf)❤T☮C☺M☯ 13:20, 20 July 2024 (UTC)

Bot request for Central Nahuatl language name category cleanup

Prior to today there were no regional labels defined for Central Nahuatl (made the module today, MOD:labels/data/lang/nhn) and so all of the regional categories right now are using raw markup. If anyone has a bot and would be willing to clean this up (around 1400 entries) please reply below. -saph668 (user—talk—contribs) 17:18, 20 July 2024 (UTC)

This is not bottable because we need to decide whether to use {{lb}} or {{tlb}}. --kc_kennylau (talk) 17:31, 20 July 2024 (UTC)

Update: I have used my bot to deal with those that are automatable, i.e. those that already have a {{lb}} or {{tlb}} usage, and that totalled to 152 entries, which unfortunately leave 1291 entries to be dealt with manually. --kc_kennylau (talk) 19:43, 20 July 2024 (UTC)

I think T:tlb is sufficiently infrequently used (and T:lb is sufficiently frequently used for both kinds of labels) that it wouldn't really be making anything any more wrong to bot all the manual formatting to T:lb. (But obviously you don't have to do that if you prefer a more conservative approach / aren't comfortable doing so.) - -sche (discuss) 20:30, 20 July 2024 (UTC)

MediaWiki:CustomSearch.js

Does this do anything in 2024? It is apparently only used at Help:Tips_and_tricks#Browser-integrated_search. @-sche (last active editor). Ioaxxere (talk) 17:54, 20 July 2024 (UTC)

I don't know. It doesn't seem to have been maintained in quite some time. I think I only added a documentation link to remove some kind of 'scripts lacking or not linking to their documentation' error, without regard for whether the script worked. Based on the description on the Help page, it seems like its functionality could be duplicated by telling people to include site:en.wiktionary.org or site:wiktionary.org in their search (although this will then fail inasfar as Google seems to include Wiktionary content as Wikipedia content and vice versa). - -sche (discuss) 18:55, 20 July 2024 (UTC)

Bizarre triggering of "Crosswiki abuse and Anti-harassment" protection

Bizarre triggering of page protection at Wiktionary:Wiktionary_for_Wikipedians#What_can_be_included.

Original text [bold added here only]:

There is a practical reason for this approach to redirects. A word can be both a lemma and an inflected form of another lemma. While saw is the past tense of see in English, it is also a noun. Additionally, Wiktionary includes terms in all languages, not just English, so it's possible what is a lemma in one language is an inflected form of another lemma in another language.

Attempted edit [bold added here only]:

There is a practical reason for this approach to redirects. A word can be both a lemma and an inflected form of another lemma. While saw is the past tense of see in English, it is also a noun. Additionally, Wiktionary tries to include terms in all languages, not just English, so it's possible that what is a lemma in one language is an inflected form of another lemma in another language.

Attempted edit summary:

/* What can be included */ logic & grammar emended. —DIV

Warning:

This action has been automatically identified as harmful, and therefore disallowed. If you believe your action was constructive, please start a new Grease pit discussion and describe what you were trying to do. A brief description of the abuse rule which your action matched is: Crosswiki abuse and Anti-harassment

I cannot see anything in my attempted edit that should have triggered that. Furthermore, I had just finished successfully making another minor edit to that very same section. —DIV (1.129.106.197 12:42, 21 July 2024 (UTC))

FWIW, I have just tested to see whether I can make either of the above two changes individually: it turns out that each of the two changes triggers the same 'protection'. —DIV (1.129.106.197 12:46, 21 July 2024 (UTC))

But I have had no problem subsequently making a bunch of other minor edits to that page. Strange! —DIV (1.129.106.197 13:19, 21 July 2024 (UTC))

This is a global abuse filter, which is hidden even from Wiktionary admins. You would need to report this issue somewhere on Meta (m:) - but good luck finding the proper place to report global abuse filter false positives. I couldn't find such a place when I looked.

Needless to say, creating an account for yourself would mean you no longer run into these sorts of issues. This, that and the other (talk) 02:16, 23 July 2024 (UTC)

I looked through the list of Meta admins and recognize one person who also edits here: @Jusjih, do you know where on Meta to report this? (The filter triggered was Meta's global abuse filter 300.) - -sche (discuss) 04:22, 23 July 2024 (UTC)

Orange links broken?

@Dixtosa, Erutuon and anyone else interested: orange links aren't working for me (using Firefox on Windows 10 on a desktop) today. For several months now, they haven't worked when linking to the same page, but as of today they aren't working at all. I just double-checked my Preferences, and they are still switched on. Any ideas? —Mahāgaja · talk 13:16, 21 July 2024 (UTC)

Broken for me too. Chrome, Windows 10 desktop. Vininn126 (talk) 13:19, 21 July 2024 (UTC)

Can you give a page they're broken on? I just tried going to jump upon (just created) and the green links at least work for me (Chrome on Mac OS). @Ioaxxere might also be interested. Benwing2 (talk) 13:57, 21 July 2024 (UTC)

tembr, the alt form. Vininn126 (talk) 13:59, 21 July 2024 (UTC)

OK, broken for me too. I'm guessing either this is MediaWiki breakage or due to a change by Ioaxxere, dunno. Benwing2 (talk) 14:05, 21 July 2024 (UTC)

Or Irish Artúr at Arthur#Translations. (Or right here, for that matter). —Mahāgaja · talk 14:05, 21 July 2024 (UTC)

Works for me for both examples mentioned: Win 10, FF128.0, Vector legacy. DCDuring (talk) 14:07, 21 July 2024 (UTC)

 Fixed Ioaxxere (talk) 16:24, 21 July 2024 (UTC)

Color change? DCDuring (talk) 18:10, 21 July 2024 (UTC)

Automatically transcluding notes and refn from Module:Quotations

@JeffDoozan @Erutuon — Is there a way to get all the information from a Quotations data module to automatically appear on an entry? I'm currently working on Proto-Norse inscriptions and this would be useful.

• Data module: Module:Quotations/gmq-pro/data.
• Example entry: ᚾᚨᚢᛞᛁᚷᚨᛊᛏᛁᛉ.

Currently, the values year, wLink and style are transcluded, while notes and refn are not. Inputting these manually is not a huge hassle, but it would be useful to have it done automatically, so that changes to them can be made at one place (the data module) rather than at every entry quoting from said inscription. ᛙᛆᚱᛐᛁᚿᛌᛆᛌ ᛭ Proto-Norsing ᛭ Ask me anything 17:39, 22 July 2024 (UTC)

Revert creation of the entry

I have accidentally created the wrong form of the noun in Esperanto, can someone delete it please? I tried to revert it myself, but it doesn't allow it for "vandalism", even though it's my edit - sxtaloj FitikWasTaken (talk) 15:02, 23 July 2024 (UTC)

You can't leave pages blank. Use {{d}}. — SURJECTION ^{/ T / C / L /} 15:03, 23 July 2024 (UTC)

Thanks for the information FitikWasTaken (talk) 15:36, 23 July 2024 (UTC)

Link to unsupported titles in MediaWiki:Search-nonefound

Can we link to e.g. Unsupported titles/: on Special:Search/: by placing something like {{#ifexist:Unsupported titles/$1|See entry on [[Unsupported titles/$1|$1]].}}? Nardog (talk) 15:18, 23 July 2024 (UTC)

@Nardog: I put it in there and it works. Nice. — Eru·tuon 20:45, 25 July 2024 (UTC)

Maloi name detected as harmful

Hello, Can someone to fix that “Maloi” detected as harmful word? I don’t think it’s harmful because it refers to a name of girl. Royiswariii (talk) 00:02, 24 July 2024 (UTC)

Hi, so just to be clear, you were trying to edit Maloi to add that entry and you couldn't do it because the site stopped you, correct? —Justin (koavf)❤T☮C☺M☯ 00:06, 24 July 2024 (UTC)

yes, correct. Royiswariii (talk) 00:08, 24 July 2024 (UTC)

Gotcha. So looking at Special:AbuseLog, it looks like there were a few problems. First off, you were editing "maloi" and not "Maloi". These are separate entries at Wiktionary, but would go to the same place at Wikipedia (note, for instance the difference between polish and Polish). Secondly, it looks like many of your edits were too short to be an entry like just creating some IPA text and nothing else. I'm glad you're here and want to welcome you, but you may want to take a look at the message I left on your talk page and familiarize yourself with other existing entries before you try adding new ones. It will just make things easier in the future. I'm not personally familiar with the name Maloi, can you tell me anything about where it comes from? —Justin (koavf)❤T☮C☺M☯ 00:14, 24 July 2024 (UTC)

Maloi comes from a name in a woman in Philippines and member of Filipino girl group BINI. Royiswariii (talk) 00:20, 24 July 2024 (UTC)

bini is a filipino group. Royiswariii (talk) 00:21, 24 July 2024 (UTC)

Thanks. I really think you should focus on trying to add to existing entries for a while before you make a new one. Making entries at Wiktionary isn't that difficult, but it can be confusing at first if you're not familiar with our policies and how things work on a technical level. Maybe if you can add to some entries slowly for a couple of weeks, you can come back to trying to make a new entry. You can even experiment at Wiktionary:Sandbox or make your own personal sandbox at User:Royiswanii/sandbox. —Justin (koavf)❤T☮C☺M☯ 00:29, 24 July 2024 (UTC)

I will figure it out. Thank you very much! Royiswariii (talk) 00:35, 24 July 2024 (UTC)

On a tangential note, it looks like you added a link at Wikipedia already. That was probably a little premature. —Justin (koavf)❤T☮C☺M☯ 00:15, 24 July 2024 (UTC)

How am I supposed to find U+FFFD in a really large hunk of text I'm adding?

I've had this kind of problem before: something strange in text I'd like to add to a user page triggers a filter which keeps me from saving the text. The message just directs me to GP. "This action has been prevented as it is most probably in error." (What if it isn't? Why is there a presumption that my time to format the text in the preview window is worth so much less than the possible consequence of what is not even certainly an error?) DCDuring (talk) 02:00, 24 July 2024 (UTC)

One quick kludge is to copy and paste half of the text into the window and leave out the other half and then keep on iterating on this until you find the errant character. Now, if you have multiple characters, this may not help. A few online sources that may help:

• https://www.soscisurvey.de/tools/view-chars.php
• https://invisible-characters.com/view.html
• https://stackoverflow.com/questions/39258979/how-can-i-view-hidden-characters-in-notepad

I hope that's useful. :/ —Justin (koavf)❤T☮C☺M☯ 02:15, 24 July 2024 (UTC)

@Koavf: Thanks. I actually had saved a link to the sosscisurvey.de's tool, which I hadn't put in a good bookmarks folder, but eventually found. If I was more familiar with Notepad++, it looks to have all kinds of relevant capability. I ended up manually scanning to find about 5 strange-looking "vernacular names". I'll have to do some kind of special dump-processing to get at the underlying problem. DCDuring (talk) 03:11, 24 July 2024 (UTC)

� (U+FFFD) is (almost) always going to be an error, and is generally a sign that the text being added is corrupted. That abuse filter seems fair to me (and has been in place for longer than I've been editing, I should add). Theknightwho (talk) 16:23, 28 July 2024 (UTC)

@Theknightwho: "How am I supposed to find U+FFFD in a really large hunk of text I'm adding?" DCDuring (talk) 16:27, 28 July 2024 (UTC)

@DCDuring The main question had already been answered, and I was specifically addressing the part where you asked “What if it isn't? Why is there a presumption that my time to format the text in the preview window is worth so much less than the possible consequence of what is not even certainly an error?” Theknightwho (talk) 19:18, 28 July 2024 (UTC)

@Theknightwho: The "filter" message seemed to imply that U+FFFD was not necessarily the offending character. The file that I was uploading was 856K, about 30,000 lines. The tools mentioned above offer windows that are less than 10 lines. Manually locating the characters by paging down 3,000 times is not a plausible means of locating offending characters. I suppose that I was thinking that if we have "filters" (actually blocking tools) that can detect funny characters, then we could also have useful tools to help users locate and eliminate or quarantine such characters. I addressed the problem by deleting about 90% of the file and manually inspecting the remainder. I still don't know the underlying cause, so I will face the problem again. DCDuring (talk) 01:37, 29 July 2024 (UTC)

There are a couple of rather crude ways I might suggest: First of all, since you're an admin, you can view the abuse logs. The simplest would be to click the "abuse log" link at the top of your contributions page. On the line for the edit in question, click "Examine". Since this filter looks at the Added_Lines variable, you can scroll down to that variable in the "Variables generated for this change" section. This lists all the lines in the file. You would then scroll down the list until you see one that's garbled. I was able to spot it scrolling really fast- basically holding down the down-arrow or even the page-down key. Since your eyesight isn't that good, you would probably have to go slower. The offending line is # 6380, between "# leaping mullet - 2" and "# Appalachian gentian - 2"

The other way is a divide and conquer strategy. Normally I would do this using HTML comments: <!-- before and --> after, or nowikis: <nowiki></nowiki> before and </nowiki> after. In this case, the abuse filter no doubt ignores those, so the alternative is creating multiple pages.

The idea is to split your text into two parts, and seeing which one triggers the abuse filter. If only one does, you've reduced the amount you need to search by half. Next, split the bad page in two and repeat. The bad page you need to search is now a quarter of the whole. Keep on repeating until you've found the bad part. You can keep new pages to a minimum by adding the text from each good page to the first good page and reusing the page you just emptied. It may be simpler to have a top holding page and a bottom holding page to make it easier to keep the text in order. If both pages are bad, you'll have to leave the second page for the moment and go through the whole process starting with the first bad page.

This may seem complicated, but you're only dealing with two pages at a time, and the size reduces by powers of 2. 65,536 lines take a maximum of 16 splits to reduce to a single line, 32,768 15 splits, 16,384 14 splits, 8,192 13 splits, 4,096 12 splits, etc.- and that's the worst case scenario.

Let's look at how it would work with the page that led to this discussion:

Pages: Holding I, Holding II, A, B. Total lines: 8,490, bad line: 6,380

1. A good, B bad. Holding I <-A, Move half of B to A. Remaining lines 4,245, Holding I 4,245 lines
2. A good, B bad. Holding I <-A, Move half of B to A. Remaining lines 2,123, Holding I 6,367 lines
3. A bad, B good. B-> Holding II, Move half of A to B. Remaining lines 1,062, Holding II 1,061 lines
4. A bad, B good. B-> Holding II, Move half of A to B. Remaining lines 531, Holding II 1,592 lines
5. A bad, B good. B-> Holding II, Move half of A to B. Remaining lines 266, Holding II 1,857 lines
6. A bad, B good. B-> Holding II, Move half of A to B. Remaining lines 133, Holding II 1,990 lines
7. A bad, B good. B-> Holding II, Move half of A to B. Remaining lines 67, Holding II 2,056 lines
8. A bad, B good. B-> Holding II, Move half of A to B. Remaining lines 34, Holding II 2,089 lines
9. A bad, B good. B-> Holding II, Move half of A to B. Remaining lines 17, Holding II 2,106 lines
10. A good, B bad. Holding I <-A, Move half of B to A. Remaining lines 9, Holding I 6,375 lines
11. A good, B bad. Holding I <-A, Move half of B to A. Remaining lines 5, Holding I 6,379 lines
12. A bad, B good. B-> Holding II, Move half of A to B. Remaining lines 3, Holding II 2,108 lines
13. A bad, B good. B-> Holding II, Move half of A to B. Remaining lines 2, Holding II 2,109 lines
14. A bad, B good. B-> Holding II, Move half of A to B. Remaining lines 1, Holding II 2,110 lines

Of course, you would probably be able to spot the bad line after the remaining lines got down to the size of the edit window, so you would only need perhaps up to line 8. After deleting the bad line, you would add the remainder to the end of Holding I, then the contents of Holding II.

In this notation, <- means the contents of the page on right are added to the bottom of the page on the left and -> means the contents of the page on the left are added to the top of the page on the right. The names are just for the notation: in reality Holding I would be your original page, and you would start by moving the first half to A and the other half to B.

Sorry if this is a bit dense. I would spend more time making it easier to read, but I have to be out the door in 8 hours for work and I still have stuff to do tonight. Chuck Entz (talk) 05:51, 29 July 2024 (UTC)

Thanks. I wasn't aware of the abuse filter log. But it didn't help find the source of the problem. I was able to use Notepad++ to find that the problem characters were some kind of misreading of Cyrillic синни́нгия краси́вая (sinníngija krasívaja) in the entry синнингия (sinningija). The Perl script I use wasn't ready for Cyrillic, I guess, so I'll filter such items, probably any instance of {{vern}} with a lang parameter. DCDuring (talk) 16:21, 29 July 2024 (UTC)

@DCDuring - I had a look at the filter logs to see what you were trying to do, and I've disabled the filter in the User and User talk namespaces, which is something we already do for various other similar filters. Theknightwho (talk) 21:22, 29 July 2024 (UTC)

@Theknightwho: Thanks. That would have been a help. I don't plan any large move from User space to, say, Appendix space, but would a filter catch the same kind of problem for such a move? DCDuring (talk) 12:42, 30 July 2024 (UTC)

@DCDuring Not at the moment, as it just checks the lines added by an edit, but it probably should. The reason U+FFFD exists is to be a placeholder if text contains an invalid byte sequence, which usually happens if a multi-byte character has been accidentally chopped in half, so it's a good sign that there might be something wrong with more than just that part of the text. In the (very rare) instances where it's really needed, it's possible to use � to display it on purpose. Theknightwho (talk) 13:43, 30 July 2024 (UTC)

Chinese terms spelled with CHAR

Categories like Category:Chinese terms spelled with 嵊 suggest that

Note that categories of the form LANG terms spelled with CHAR are intended for characters not part of the standard repertoire of a language (e.g. Cyrillic characters in English or Latin characters in Russian).

I am not sure whether the categories are created as intended or from incorrect template edits. Categories for recurring characters might be necessary, but in this case the note should be removed. — 物灵 (talk) 18:14, 25 July 2024 (UTC)

@物灵 Done. Benwing2 (talk) 04:46, 27 July 2024 (UTC)

Hide the Anagrams section

I've been using this common.js (lines 20-33) to hide the Anagrams section, but about a month ago it stopped working and I don't know why. I use the Vector (2022) skin. Any help to get it working again would be appreciated. Vuccala (talk) 21:56, 25 July 2024 (UTC)

@Vuccala try

// Hide TOC links to #Anagrams
document.querySelectorAll(`#toc [href^="#Anagrams"], #vector-toc [href^="#Anagrams"]`).forEach(e => e.parentElement.style.display = "none");
// Hide #Anagrams section(s)
document.querySelectorAll(`.mw-heading > [id^="Anagrams"]`).forEach(e => {
    e.parentElement.style.display = "none";
    e = e.parentElement.nextElementSibling;
    while(e && !e.classList.contains("mw-heading")) {
        e.style.display = "none";
        e = e.nextElementSibling;
    }
});

Your script broke last month due to mw:Heading HTML changes which caused the selector h3 > span#Anagrams to stop working. Ioaxxere (talk) 06:03, 26 July 2024 (UTC)

@Ioaxxere thank you very much! The problem is now fixed! Vuccala (talk) 00:08, 27 July 2024 (UTC)
 Done

Broken PMID links

PMID links (|pmid=) inside quote templates are broken since June. I suspect this change introduced the error. Einstein2 (talk) 15:01, 27 July 2024 (UTC)

@Einstein2 Can you give me an example? Benwing2 (talk) 03:36, 28 July 2024 (UTC)

@Benwing2 buccoapical, nucleosome, neurostabilizer. Einstein2 (talk) 10:37, 28 July 2024 (UTC)

Module errors for reconstructed headwords without *

Since @Theknightwho added a requirement for headwords to have an asterisk when they're in the Reconstruction namespace, there have been a number of entries periodically popping up in CAT:E as the change has propagated out to the various headword templates, the templates called by those templates, the modules invoked by the various templates, etc. So far, most of those have been fixed by adding an * to the |head= parameter. That hasn't worked with Reconstruction:Okinawan/あー: it has a head parameter, but throws a module error regardless of whether there's an * or not. I'm guessing that's because one of the language-specific modules gets the headword independently in order to use it for things like categories.

At any rate, some questions occurred to me:

1. If we already know that the headword must begin with an * in Reconstruction space, why not just tag it as missing the * and add it to the headword? After all, the idea isn't to punish those who don't use the right syntax, but to display the headword and add the right categories. As I've said before, CAT:E is for emergencies, and overuse of module errors is like setting off a fire alarm and evacuating the entire building because someone is smoking in one of the restrooms. It's hard enough getting new editors to even use headword templates without making it more complicated to use them.
2. Do the headword modules allow for requesting a headword without displaying it? Or is that something that shouldn't be done?
3. How do we handle languages with a mix of attested and reconstructed forms, more specifically where attestation is different between the lemma and one or more of the forms linked to in the headword line? What if the lemma is in mainspace and a principle part is reconstructed, or vice versa? Most of the world's attested languages have gaps in their attestation. The sources/literature for the better known ones have ways to get around that, but there are lots of exceptions.
4. We have a lot of language-specific headword modules, many written by editors who are no longer active. What happens when our changes to other modules break them?

Please note that this is an attempt to start a discussion, not an interrogation. The idea is to make things easier by considering the issues outside of emergency problem-solving mode. I spend a lot of my time fixing things, and I respect the people I work with. Chuck Entz (talk) 22:17, 27 July 2024 (UTC)

I agree that this should have been implemented through tracking rather than throwing an error, until all the headwords were fixed. There are still a lot more that are throwing errors but aren't showing up yet in CAT:E. I am in the process of fixing them by downloading the whole contents of the Reconstruction namespace and looking for cases where an explicit |head= is given without a * in it. Benwing2 (talk) 22:37, 28 July 2024 (UTC)

@Benwing2 @Chuck Entz I tracked this for a day, and changed it to an error when I saw nothing had shown up. Theknightwho (talk) 22:45, 28 July 2024 (UTC)

@Theknightwho: It may have seemed like a day, but it looks to me like it was a little over 7 hours. At any rate, the job queue wouldn't know what namespaces the change was relevant to, so it would include all 8 million transclusions for the module in no particular order, of which only 36 thousand are in the Reconstruction namespace- less than half a percent. Of those, only a small percentage would be expected to have what you were testing for. On top of that, unrelated edits that would bypass the job queue are far less likely to happen in the Reconstruction namespace, and the job queue serves all transclusions in all the projects, so there were no doubt other edits ahead of yours. The odds of the job queue or a page edit not hitting a page that would get tagged in even a day or two would be actually pretty high. Needless to say, such things should be allowed a week before making a decision. Chuck Entz (talk) 00:07, 29 July 2024 (UTC)

Alright, fair enough. Theknightwho (talk) 20:44, 29 July 2024 (UTC)

אויססדרן

The transliterations for the infinitive, present participle, conjugated present-tense forms and imperative forms are wrong, but it seems to only be when I have to specify the tr parameter for the yi-conj template PLUS a converb. Can someone with more technical knowledge than myself go and see if they can fix this issue with the yi-conj template? Alternatively, please tell me what I did wrong when inputting the parameters into yi-conj. Thanks. Insaneguy1083 (talk) 07:14, 29 July 2024 (UTC)

fungusses

How do I make fungusses appear as a "nonstandard, rare" plural on fungus, alongside "funguses"? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:44, 29 July 2024 (UTC)

You could do a usage note, like at octopus and that could explain why and how one might use this particular version. —Justin (koavf)❤T☮C☺M☯ 17:34, 29 July 2024 (UTC)

Thank you, but my point more specifically was how to amened the current {{en-noun|~|fungi|pl3qual=nonstandard, rare|+}}; by trial and error I found that {{en-noun|~|fungi|pl3qual=nonstandard, rare|+|fungusses}} works. Is there a better way? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:09, 29 July 2024 (UTC)

https://en.wiktionary.org/w/index.php?title=fungus&diff=80914399&oldid=80914384, mayhaps? —Justin (koavf)❤T☮C☺M☯ 18:10, 29 July 2024 (UTC)

Ah, you edited your comment as I was responding. I think the way that you did it was totally intelligible and the way that I amended it maybe makes it a little more clear at the risk of being a little more clunky. —Justin (koavf)❤T☮C☺M☯ 18:11, 29 July 2024 (UTC)

gmw-infl-noun-i template

Please see the inflection table at Proto-West Germanic *anagripi, as it's yielding an error. Thanks Leasnam (talk) 01:41, 30 July 2024 (UTC)

Courtesy link: Template:gmw-infl-noun-i —Justin (koavf)❤T☮C☺M☯ 01:49, 30 July 2024 (UTC)

@Victar: —Justin (koavf)❤T☮C☺M☯ 01:50, 30 July 2024 (UTC)

Multisyllable i-stems sometimes need to specify |j=. Fixed. --{{victar|talk}} 02:39, 30 July 2024 (UTC)

*teng-. —Justin (koavf)❤T☮C☺M☯ 02:56, 30 July 2024 (UTC)

Thank you All ! I usually remember the j param, but I didn't think of it this time Leasnam (talk) 02:58, 30 July 2024 (UTC)

Me too. I feel like it could be fixed so we wouldn't have to, but I haven't bothered. --{{victar|talk}} 03:19, 30 July 2024 (UTC)

Norwegian expression det årnær sæ

Hey ! I am a novice but I wanted to create a wiktionnary page about the norwegian expression "det årnær sæ". I have tried to publish but I can't. Can you read it and say me if it's possible to publish it ? It it can't be published can you let me know why and I would try to correct the page. I can't find the way to make a draft on wiktionary but here is what I've done. Thank you for reading me :D ! (I'm a french speaking user so sorry for my mistakes in English.

Norwegian

Det årnær sæ

Grease pit

1. It’ll work out, Expression of a happy, optimistic state of mind.

2020, Hagle (lyrics and music), “Det årnær sæ”‎^[3]:

Droppa ut av barneskolen, men det årnær sæ

I left elementary school early, but things will work out.

2017 December 13, Bengt Børresen, “Foreslår «Kjepper i jula» hvis «Det årnær sæ» refuseres”, in Fredriksstad Blad:

[...] «det årnær sæ» faktisk er et gammelt dialektuttrykk, som på en treffende måte formidler det joviale, gemyttlige, optimistiske og inkluderende lynnet hos Fredrikstads befolkning.

‘det årnær sæ’ is actually an old dialect expression that aptly conveys the jovial, cheerful, optimistic and inclusive nature of the people of Fredrikstad.

Synonyms

• (It'll work out): Det ordner seg

Further reading

Visit Norway
Fredriksstad Blad Ledébutantinexpérimenté (talk) 19:14, 30 July 2024 (UTC)

@Ledébutantinexpérimenté welcome to Wiktionary! The edit was picked up as spam owing to the YouTube link. I've created the page for you, but without the external links at the bottom, as these are not especially relevant to the dictionary entry. Sorry about this; I hope you still feel encouraged to continue contributing to Wiktionary. This, that and the other (talk) 01:25, 31 July 2024 (UTC)

Thank you very much!! :) Ledébutantinexpérimenté (talk) 01:47, 31 July 2024 (UTC)

imbalanced square brackets causing template(s) to fail to render as templates

I noticed the issue I outline here today. I would understand the template failing to render because of an imbalance of curly braces, but an imbalance of square brackets making it fail to render (rather than e.g. throw an error message) intrigued me. I'm not sure if it's something that needs to be fixed so that the template itself displays as a template, rather than outputting {{template..., and the only broken output is the one broken square-bracketed [[link], or whether people are happy with it as-is. (PS don't worry, I'm only putting that many links inside {{quote-book}} in my sandbox, where I'm looking for words we're missing, not in mainspace.) - -sche (discuss) 00:18, 31 July 2024 (UTC)

@-sche This isn't something we can fix, as it relates to how templates are parsed by MediaWiki. I can give you a detailed explanation if you want, but the short version is that the parser also keeps track of pairs of [[ and ]] while parsing templates as well, so you need to close that "block" before you close the template. The specifics aren't documented anywhere official, but I wrote about them in the template parser comments, since it needs to deal with all the various parser weirdnesses like this (see Module:template parser#L-759). I essentially had to work them all out on my own by going over the native parser's code. Theknightwho (talk) 01:00, 31 July 2024 (UTC)

Ah! Interesting (and unfortunate but unsurprising that it was not documented anywhere official). Thanks for that explanation. - -sche (discuss) 03:10, 31 July 2024 (UTC)