Wiktionary:Grease pit/2018/June

I've gone on a bit of a regionalisms bender with Bavarian, Alemannic and Cimbrian. And I noticed that the words in the "Regionalisms" categories don't link to their respective languages. Is there a way to fix this? (btw, it's not a recent module edit thing; Category:American English does the same thing). – Julia (talk) ^{• formerly Gormflaith •} 16:42, 1 June 2018 (UTC)

You can add {{catfix|language code}} to the category pages. DTLHS (talk) 17:09, 1 June 2018 (UTC)

Thanks, I'll go around doing that. – Julia (talk) ^{• formerly Gormflaith •} 18:31, 1 June 2018 (UTC)

I added the template to {{dialectboiler}} as well, though a lot of regionalisms pages don't use it. – Julia (talk) ^{• formerly Gormflaith •} 18:44, 1 June 2018 (UTC)

{{der|en|ira-old||*Sanatrūk}} → Old Iranian

What's going on? Benwing2 (talk) 23:57, 3 June 2018 (UTC)

Fixed it, I think. Changed the parent from "ira" to "ira-pro". – Julia (talk) ^{• formerly Gormflaith •} 02:27, 4 June 2018 (UTC)

Old Iranian is not a full language and its parent is not a full language, therefore it can't have entries, therefore the module is removing the text. Is my guess. DTLHS (talk) 04:25, 4 June 2018 (UTC)

@Julia: I undid your "fix". Those etymology-only codes shouldn't be used for lemma -- Old Iranian and Middle Iranian are developmental periods of the language family, not actual languages.

@Benwing2, what you're probably looking for is Proto-Iranian ira-pro, but looking at *Sanatruk, I would say it's pretty outdated and most certainly incorrect. --Victar (talk) 04:35, 4 June 2018 (UTC)

@Victar I don't understand why you undid the fix. Julia set the parent to "ira-pro", which is a real language, not an etymology-only language. The current state where the parent is a family is totally broken. Please put her fix back or fix it properly, thanks! Benwing2 (talk) 05:22, 4 June 2018 (UTC)

@Benwing2:, as I said, ira-old is a developmental period of the language family, and not an actual language, and as such should never be used for lemma, so using ira, the language family, is the correct parent code, not ira-pro. --Victar (talk) 05:49, 4 June 2018 (UTC)

We tend to use ira-old to mean "from some unknown archaic Iranian language". If we have enough phonological information to reconstruct it, it should either be assigned to a particular language or to Proto-Iranian. ira-old exists separately precisely because we don't know enough about the word's phonological shape. —*i̯óh₁n̥C 05:56, 4 June 2018 (UTC)

@Victar Okay, I get that Old Iranian doesn’t have lemmas. Though I do think that the information is useful, given that it’s correct (what did you mean by “it” in “it’s pretty outdated...”? Old Iranian reconstructions in general?) – Julia (talk) ^{• formerly Gormflaith •} 13:56, 4 June 2018 (UTC)

@Julia: I was referring to the form given, which doesn't not conform to Proto-Iranian. In the entry in question, Սանատրուկ (Sanatruk), the {{m|und||*Sanatruk}} should be removed completely. --Victar (talk) 14:49, 4 June 2018 (UTC)

I'm with Victar on this, that is not a good Proto-Iranian reconstruction, and Old Iranian ≠ Proto-Iranian. —Aryaman^{A (मुझसे बात करें • योगदान)} 01:45, 5 June 2018 (UTC)

@Victar, AryamanA Just to be clear: what's the best way to go about Old and Middle Iranian reconstructions – {{der|en|ira-old}} {{m|und|foo}}? – Julia (talk) ^{• formerly Gormflaith •} 02:32, 5 June 2018 (UTC)

@Julia: no, like JohnC5 said, if it can't be properly reconstructed in Proto-Iranian or some PIr descendent, it should just plainly say from Old Iranian and nothing more. --Victar (talk) 02:43, 5 June 2018 (UTC).

@Victar: Okay. There are quite a few etymologies using that method, just wondering. – Julia (talk) ^{• formerly Gormflaith •} 02:48, 5 June 2018 (UTC)

Cheers. --Victar (talk) 03:34, 5 June 2018 (UTC)

Why does the senseid in the etymology of Uniat produce a bullet point? It looks ugly, because it is further left than the rest of the test (including the rest of the bullet points) on the page. - -sche (discuss) 20:38, 4 June 2018 (UTC)

I think that's because it's supposed to be used behind #, otherwise it bugs. --Per utramque cavernam 20:43, 4 June 2018 (UTC)

Collaboration team is announcing plans to graduate the New Filters for Edit Review out of beta on Watchlist by late June or early July. After launch, this suite of improved edit-search tools will be standard on all wikis. Individuals who prefer the existing Watchlist interface will be able to opt out by means of a new preference.

The New Filters introduce an easier yet more powerful user interface to Watchlist as well as a whole list of filters and other tools that make reviewing edits more efficient, including live page updating, user-defined highlighting,the ability to create special-purpose filter sets and save them for re-use and (on wikis with ORES enabled) predictive filters powered by machine learning. If you’re not familiar with the New Filters, please give them a try on Watchlist by activating the New Filters beta feature. In particular, it would be very helpful if you can test the new functionality with your local gadgets and configurations. The documentation pages provide guidance on how to use the many new tools you’ll discover.

Over 70,000 people have activated the New Filters beta, which has been in testing on Watchlist for more than eight months. We feel confident that the features are stable and effective, but if you have thoughts about these tools or the beta graduation, please let us know on the project talk page. In particular, tell us if you know of a special incompatibility or other issue that makes the New Filters problematic on your wiki. We’ll examine the blocker and may delay release on your wiki until the issue can be addressed. - -Kaartic (talk) 17:03, 5 June 2018 (UTC)

I posted edit requests at

• Wiktionary talk:Tea room/2006
• Wiktionary talk:Word of the day/Nominations/Archive 2011

They have been ignored so far. Is this the right place to draw attention to these requests? —Anomalocaris (talk) 00:35, 6 June 2018 (UTC)

I really doubt if anyone is watching such old pages. Perhaps you should have posted the edit requests at the current talk pages of those pages, "Wiktionary talk:Tea room" and "Wiktionary talk:Word of the day". — SGconlaw (talk) 02:48, 6 June 2018 (UTC)

Also, why not just correct the errors yourself? — SGconlaw (talk) 02:51, 6 June 2018 (UTC)

The pages are (/were) protected, although I'm not entirely sure why. Probably you and I and a few other users could put them on our watchlists (to be able to spot any vandalism of them) and then unprotect them. (I've unprotected them now.) And @ Anomalocaris, yes, people don't necessarily watch the talk pages of such old pages, but this is a fine place to draw attention to such requests. - -sche (discuss) 02:57, 6 June 2018 (UTC)

When I checked, "Wiktionary:Word of the day/Nominations/Archive 2011" wasn't protected. Maybe you got to it before I looked. — SGconlaw (talk) 03:17, 6 June 2018 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ -sche: Would you please unprotect Wiktionary:Beer parlour/2006/September and Wiktionary:Beer parlour/2006/April so I can remove the lint (while carefully preserving appearance, of course) from those as well? —Anomalocaris (talk) 15:03, 6 June 2018 (UTC)

 Done. — SGconlaw (talk) 15:11, 6 June 2018 (UTC)

SGconlaw: Thanks! Edits on these two items are complete. There are two remaining items with Multiple unclosed formatting tags. One is a protected user talk page and I've already asked its owner to fix it or unprotect it so I can fix it. The other is Module:User:Suzukaze-c/zh-pron/documentation, and I know how to edit Wikitext, but not this. So I leave that final one to greater minds. —Anomalocaris (talk) 15:55, 6 June 2018 (UTC)

@Suzukaze-c, can you assist? — SGconlaw (talk) 15:59, 6 June 2018 (UTC)

@Anomalocaris Connel is not around anymore, I unprotected his talk page. Let me know when you are done and I'll re-protect it. - TheDaveRoss 16:13, 6 June 2018 (UTC)

TheDaveRoss:  Done. Feel free to re-protect User talk:Connel MacKenzie/archive-2006-08 and the two Beer parlours as well. —Anomalocaris (talk) 16:22, 6 June 2018 (UTC)

I'm having trouble getting {{IPA}} to correctly indicate the number of syllables in the word scour. As far as I'm aware, the pronunciations are as follows:

• (Received Pronunciation) IPA^(key): /ˈskaʊə/ (two syllables)
• (General American) IPA^(key): /ˈskouɚ/, /skouɹ/ (two syllables and one syllable respectively).

However, I am getting a three-syllable count somewhere. It may be that I am not using the correct IPA symbols to indicate diphthongs. (Pinging @Erutuon in case you can help.) — SGconlaw (talk) 17:19, 6 June 2018 (UTC)

@Sgconlaw: Yes, it was the wrong symbols. /ou/ is not a symbol in Appendix:English pronunciation, and the GA pronunciation is similar to the RP one. — Eru·tuon 18:29, 6 June 2018 (UTC)

Ah, I see. Thanks! — SGconlaw (talk) 20:19, 6 June 2018 (UTC)

@Erutuon, Sgconlaw FWIW, the US pronunciations of flour, sour, scour and dour are each indicated differently but the four should rhyme exactly, whether they are reckoned as one or two syllables. For GA I don't really think there exist two distinct one-syllable vs. two-syllable pronunciations, for me at least there's only one pronunciation I've ever heard (which is more like two syllables, same as with flower and power); the one-syllable pronunciation AFAIK is only found dialectally, e.g. in Southern American dialects where you might hear more like and certain East-Coast dialects where you might hear more like . Benwing2 (talk) 14:37, 29 June 2018 (UTC)

Please see the discussion at Template talk:Q#Misnested tags and discuss it there, not here. Reply here if you think this is the wrong place to even mention that discussion. —Anomalocaris (talk) 23:00, 7 June 2018 (UTC)

 Done The problem is fixed now! —Anomalocaris (talk) 00:21, 8 June 2018 (UTC)

Does Wiktionary have a template comparable to Wikipedia:Template:div col? Please compare Wikipedia:User:Anomalocaris/sandbox/Lint Test to Appendix:Colours/Alphabetical listing#A and you see a clean solution to three problems:

1. fostered content lint errors
2. Bombastic code that is difficult to maintain if items are added, because the editor rather than the computer has to determine column breaks
3. Inflexible display that may not display well on small screens or for users who require unexpectedly high magnifications

This will not work here because we don't have Template:div col, but do we have something else that works the same way? If not, consider this a request for it. —Anomalocaris (talk) 06:40, 8 June 2018 (UTC)

Anybody home? —Anomalocaris (talk) 05:56, 14 June 2018 (UTC)

Module:columns, used for example in Template:der3. DTLHS (talk) 06:00, 14 June 2018 (UTC)

DTLHS: Module:columns says, "This module ... should not be used directly in entries, but in templates such as {{der2}} or {{der3}}..." So, I'm not supposed to use it directly. As for {{der4}}, it automatically hides the body, which is not desired here. —Anomalocaris (talk) 08:32, 15 June 2018 (UTC)

What about {{top3}} et al? Chuck Entz (talk) 08:38, 15 June 2018 (UTC)

Could the template/module be made to automatically put these into a category similar to Category:German terms with red links in their inflection tables, without that using too much memory, so they can be found and created, potentially en masse by bot? Examples: Floridize (because Floridized, Floridizing etc have not been created as of this writing), Gallicize, Hispanize, Japanize. - -sche (discuss) 06:01, 9 June 2018 (UTC)

@DTLHS I thought that NadandoBot generated inflected forms from header templates. - TheDaveRoss 01:07, 11 June 2018 (UTC)

Definitely not, I really only generate Spanish and Catalan forms. DTLHS (talk) 01:14, 11 June 2018 (UTC)

I wrote a little script to fix up cases like this:

From {{etyl|de|sh}} {{m|de|Spage}}, from {{etyl|it|sh}} {{m|it|spago}}. Compare Albanian {{m|sq|spango}}.

to instead read:

From {{etyl|de|sh}} {{m|de|Spage}}, from {{etyl|it|sh}} {{m|it|spago}}. Compare {{cog|sq|spango}}.

I want to make sure people are OK with this change. Basically, it looks for a punctuation mark followed by "ognate with" or "ompare (to)" followed by either "LANGNAME {{m|CODE|...}}" (where LANGNAME is the canonical name of the language whose code is CODE) or "{{etyl|CODE|-}} {{m|CODE|...}}" (where the two codes match). It correctly handles sequences of cognates, e.g.:

From {{inh|ru|orv|рудъ}}, from {{inh|ru|sla-pro|*rudъ}}, ultimately from {{etyl|ine-pro|ru}} {{m|ine-pro|*h₁rewdʰ-}}. Compare to Ukrainian {{m|uk|руди́й}}, Czech {{m|cs|rudý}}, Polish {{m|pl|rudy}}.

It only does these changes in Etymology sections. Eventually I'd like to run it for all pages that contain {{m}}, although that will take awhile. Comments? Benwing2 (talk) 23:58, 10 June 2018 (UTC)

Sounds good; I'd like to not have to do this manually. Could you add "ognates include" and "ognate of" as well? – Julia ☺ ☆ ^{• formerly Gormflaith •} 01:10, 11 June 2018 (UTC)

Also: "ompare with", and "the" before LANGNAME (e.g. Cognate with the Latin verbum). – Julia ☺ ☆ ^{• formerly Gormflaith •} 01:40, 11 June 2018 (UTC)

@Julia Implemented, thanks. Benwing2 (talk) 01:57, 11 June 2018 (UTC)

On a related note, perhaps markup like "{{etyl|de|sh}} {{m|de|Spage}}" could be replaced with "{{der|sh|de|Spage}}"? However, there should not be any replacement if there is text between {{etyl}} and {{m}}. — SGconlaw (talk) 03:04, 11 June 2018 (UTC)

This is not correct; any of {{der}}, {{inh}} or {{bor}} might be correct depending on the situation. However, it may be possible to do bot replacement of strings of "From X, from Y, from Z, ...", like this:

From {{etyl|enm|en}} {{m|enm|adweschen}}, from {{etyl|ang|en}} {{m|ang|ādwǣscan||to extinguish, quench, put out, put an end to, put down, suppress, staunch, blot out, appease, destroy}}, from {{etyl|gem-pro|en}} {{m|gem-pro|*uz-||out}} + {{etyl|gem-pro|en}} {{m|gem-pro|*dwaiskijaną||to suffocate}}, from {{etyl|ine-pro|en}} {{m|ine-pro|*dʰew-||to dwindle, die}}. Compare {{l|en|adush}}, {{l|en|dush}}.

I would use the following logic:

1. The first in the series gets replaced with {{inh}} (if the source language code is a parent), and {{bor}} if not.
2. Further entries get replaced with {{inh}} (if and only if the source language code is a parent and all preceding entries are {{inh}}), else {{der}}.
3. As an exception, if the source word ends in a hyphen and the lemma doesn't, always use {{der}}. This should catch cases like the above where a PIE root is used as a source.
4. As another exception, if an entry in the series is directly followed by a +, stop and don't proceed any further (or use {{der}} for the entries on both sides of the +). This should catch cases like the above where a compound is given in the etymology.

Benwing2 (talk) 03:15, 11 June 2018 (UTC)

Just pointing out: this would be problematic for languages that borrow from parents, particularly those descended from Latin or Sanskrit. I think for this sequence only 100% certainties would be replacing everything after the first in the series with {{der}} iff the first is not a parent. – Julia ☺ ☆ ^{• formerly Gormflaith •} 03:33, 11 June 2018 (UTC)

@Julia That is a good point. I was thinking in terms of Russian and English where this situation rarely (if ever?) occurs. Benwing2 (talk) 03:43, 11 June 2018 (UTC)

It's rare, but it does happen in English (in Russian, I only spot one word). - -sche (discuss) 08:26, 11 June 2018 (UTC)

@Julia, Benwing2: User:Erutuon/scripts/cleanup.js (lines 752 et sqq.) already does both of these (adding {{cog}}, and convert {{etyl}} to {{der}}). It doesn't go as far as deciding between {{inh}} and {{bor}} though. Per utramque cavernam 07:57, 11 June 2018 (UTC)

@Per utramque cavernam This script has various issues, e.g. it doesn't properly check that the language name and code match (and it won't handle many common intro words like "Cognate with" or lowercase "cognate with" after a paren, etc.). And the other script just blindly converts {{etyl}} to {{der}}, which is wrong, and again doesn't check that the language codes match in the two parts of the {{etyl}}/{{m}} sequence. These problems may be acceptable in an interactive script but definitely not in a batch script. Benwing2 (talk) 10:26, 11 June 2018 (UTC)

@Julia, Benwing2, Per utramque cavernam, Erutuon Could someone point me to the documentation for using a JavaScript "cleanup" program such as this? I think it would make my life a LOT easier. — This unsigned comment was added by Jawitkien (talk • contribs) at 18:05, 12 July 2018 (UTC).

@Jawitkien: I haven't written much documentation. The only documentation is in User:Erutuon/scripts/cleanup.js. If you want to do something specific with the script and it isn't explained, I can try to help. — Eru·tuon 20:20, 12 July 2018 (UTC)

I have been suggesting via attempted "corrections", a host of past and present participles that are legitimately useful and used as adjectives and nouns, but have received no reaction thereto. I have also suggested specific words, such as the noun "verga" which is a word commonly used in USA weather reports, meaning rain that evaporates before reaching the earth, and the noun "peek", meaning a quick often surreptitious look at someone or something. Except for "verga", which was apparently redacted or rejected, without explanation, my other suggestions seem to be ignored completely. Can anyone suggest a way I can qualify as a contributor and have any of my suggestions considered favorably? I was advised to add a message to the bottom of the list of redacted and rejected words, but could find no place to do so! I would appreciate any help and helper I can find.

Scott MacStravic (*Scottmacstra")

@Scottmacstra: The word is virga, not **verga. Per utramque cavernam 18:01, 11 June 2018 (UTC)

@Scottmacstra thanks for your interest in contributing to Wiktionary. I would suggest that you review the links that were left on your talk page by Koavf, there are some helpful tips on how we format entries here and other practices and policies. If you take a look at a this change which you made, you accidentally replaced the content instead of adding to it, which is why that one got undone. If you would like some help in getting used to how editing here works please let me know, you can reach out to me on my talk page and I would be happy to help with further links or by checking out edits you make and suggesting what you might want to do differently going forward. Again, thanks for your interest in helping out around here. - TheDaveRoss 18:45, 11 June 2018 (UTC)

CKoerner (WMF) (talk) 20:58, 12 June 2018 (UTC)

I currently have an addition for it, and until it is added, accelerator is downright broken on all Finnish entries. I am asking because the page has not been updated in a while and there seems to be multiple requests on the talk page. SURJECTION _{·talk·contr·log·} 00:05, 13 June 2018 (UTC)

Added your request. DTLHS (talk) 00:20, 13 June 2018 (UTC)

My common.js script literally doesn't do anything anymore, since at least about a week ago. Does anyone happen to know why this might be happening? PseudoSkull (talk) 14:52, 13 June 2018 (UTC)

Are there any error messages in your browser console? DTLHS (talk) 16:01, 13 June 2018 (UTC)

@DTLHS No, and this is happening on multiple devices. PseudoSkull (talk) 01:18, 29 June 2018 (UTC)

This works:

But this doesn't:

The only difference is quotation marks in the Arabic text. @Benwing2, Wikitiki89: Can something be done to make it work with double-quotes? --Anatoli T. ^{(обсудить}/^вклад) 12:39, 14 June 2018 (UTC)

@Atitarev OK, I understand how to fix this but don't have time right now, I'll do it later. Benwing2 (talk) 15:32, 14 June 2018 (UTC)

@Atitarev Should be fixed now. Benwing2 (talk) 02:03, 15 June 2018 (UTC)

@Benwing2: Thank you! The example is used here: بَحَثَ (baḥaṯa). It's the title of this old article here. Someone kindly provided the audio on YouTube (v=pJblBiAojtc).--Anatoli T. ^{(обсудить}/^вклад) 02:09, 15 June 2018 (UTC)

I noted two requests for examples of the poetic use of the word "hover" as a noun, though I recall another example in the poem by Archibald McLeish called "The End of the World", in which he mentions: "...and there in the starless dark, the poise the hover, of vast wings across the cancelled skies...."

There is a discussion at Template talk:categoryTOC-Armenian-TAO#misnested tags about Template:categoryTOC-Armenian-TAO generating 76 misnested tags of <span>. Please contribute to that discussion there, not here. —Anomalocaris (talk) 01:42, 18 June 2018 (UTC)

 Done It was fixed the same day! —Anomalocaris (talk) 07:14, 27 June 2018 (UTC)

One of the Arabic translations at Yugoslavian is يُوجُوسْلَافِيَّة, which includes (near the end) a sequence consisting of the letter for /y/ (yeh), a gemination mark (shadda), a Unicode right-to-left mark (U+200F), and the diacritic for /a/ (fatha); so, something like /yy←a/, instead of /yya/. Does anyone know why the right-to-left mark is there? Should it be removed? —Ruakh_TALK 03:58, 18 June 2018 (UTC)

@Ruakh: I think so, because a right-to-left mark can only be useful at the beginning or end of a run of right-to-left text, as one end of a right-to-left sandwich that contains characters that would otherwise render as left-to-right. — Eru·tuon 04:21, 18 June 2018 (UTC)

@Erutuon: User:Ruakh refers to the POSITION of the symbol (U+200F) (left-to-right). It's in the middle of the word between fatḥa and šadda, not at the end of the word. (BTW, Arabic ي is called yāʾ and ye is the Persian name for letter ی.) --Anatoli T. ^{(обсудить}/^вклад) 05:51, 18 June 2018 (UTC)

@Atitarev: Perhaps I should have been clearer – I meant it should be removed. — Eru·tuon 06:05, 18 June 2018 (UTC)

@Ruakh, Erutuon: Removed and saved as يُوجُوسْلَافِيَّة f (yūgoslāviyya). --Anatoli T. ^{(обсудить}/^вклад) 06:50, 18 June 2018 (UTC)

OK, thanks! —Ruakh_TALK 07:04, 18 June 2018 (UTC)

Can anyone tell me how I can pull out all Wiktionary entries which include Hittite cognates in their etymologies, please? — This unsigned comment was added by Rdurkan (talk • contribs).

@Rdurkan: This search should help. It might not find cognates that are not "standardly" formatted though. —Suzukaze-c◇◇ 19:39, 20 June 2018 (UTC)

@Suzukaze-c, I'm curious, why doesn't this work? --Victar (talk) 20:06, 20 June 2018 (UTC)

That is a curious bug (unless there are no {{l}}-links to ira-pro terms?), considering that the same search with "m" instead of "l" works. - -sche (discuss) 00:49, 21 June 2018 (UTC)

There are none. Try it with other codes. I confirmed that it's working with a test edit. DTLHS (talk) 00:52, 21 June 2018 (UTC)

Hello. I install MediaWiki 1.31.0 on local. After that I create MediaWiki:Common.css and MediaWiki:Common.js, and Azerbaijani declension-table templates. My problem with hide/show function. When I edit Declension section I can see show/hide button, but after save or show preview when edit all page I can't see buttons like this. What can I do?

Note: When I delete MediaWiki:Common.css and MediaWiki:Common.js, table is working but not collapsible. --Drabdullayev17 (talk) 14:02, 20 June 2018 (UTC)

This seems to be happening a lot. Anyone have any idea why? It's always just "Yes" precisely thus, with capital Y. Maybe we should make an abuse filter. Equinox ◑ 04:56, 25 June 2018 (UTC)

Examples? DTLHS (talk) 05:00, 25 June 2018 (UTC)

Unfortunately/obviously I have deleted all the ones I have seen. There have been about three today, and others recently. (How do you get to the deletion log? I AM 38 YEARS OLD AND WHAT IS THIS SITE) Equinox ◑ 05:04, 25 June 2018 (UTC)

At the top of your contributions page, click on "logs". That page will include deletions. As for the abuse filter: I created one that disallows new pages less than 20 bytes. Feel free to tweak it. Chuck Entz (talk) 05:59, 25 June 2018 (UTC)

Does the filter exclude redirect pages, which, in principle, could be as short as 13 characters (eg. #REDIRECTxx)? DCDuring (talk) 10:30, 25 June 2018 (UTC)

Yes. Chuck Entz (talk) 12:31, 25 June 2018 (UTC)

The "No-L2-L3" filter flags these at least. I am guessing that they are mostly mobile edits and that is why the first word is consistently capitalized (thanks auto-correct!). The logs page for everything is at Special:Log. - TheDaveRoss 11:57, 25 June 2018 (UTC)

An example from today: User:Mikepipo's deleted contribs. Equinox ◑ 00:06, 27 June 2018 (UTC)

Please see the discussion at MediaWiki talk:ExtractFirst.xsl#Protected edit request on 18 June 2018 regarding misnested tags of <i> in MediaWiki:ExtractFirst.xsl, and contribute to the discussion there, not here. —Anomalocaris (talk) 07:19, 27 June 2018 (UTC)

There was a recent announcement about the plans to graduate the New Filters for Edit Review out of beta for this Wiki. It stated that the deployment would happen by late June or early July. Since that announcement, we received feedback about a performance issue related to the change which is being actively worked upon. As a consequence, the deployment is postponed until further notice. Sorry for the inconvenience caused, if any.

Please let us know of any other issues or special incompatibility that you may face so that we could make sure they are solved before the feature gets deployed. Thanks, Kaartic (talk) 15:21, 27 June 2018 (UTC)

I added a reference template to an entry and it generated a "triple braces" flag when hit "publish changes". I deleted the template and formatted the reference manually, and I added a warning and an explanation on the template's page. It's fine now. The template has another problem besides the triple-brace issue, and formatting it manually fixes both of them. This is the entry I was working on when I ran into this, and this is the template I was trying to use.

The template creates a link to Trask's Etymological Dictionary of Basque published on line by Sussex University in pdf format. The second problem with the template is that it creates a link to the pdf file, but not to the specific entry you're referencing. If this was fixed, the user would need to copy the template and set parameters for the page number and headword where the relevant entry is found. Currently, without using the template, you need to copy the URL (in brackets) and append the page number and the headword under which the etymology is given.

So my question is: does this template need to exist? It can be fixed, obviously, but I don't see how a even a corrected version would save any time or effort for someone using it versus doing the same thing manually. — This unsigned comment was added by Namnagar (talk • contribs) at 17:42, 27 June 2018 (UTC).

@Namnagar: The reason for the triple-brace warning was that you used the template incorrectly. You're not meant to copy the contents of the template page. This is how to use it. — Eru·tuon 17:56, 27 June 2018 (UTC)

But the person adding the reference should be able to save the user scrolling or searching for the right page for the lemma. Parameter 1 would be the page and optional parameter 2 would be a something other than the pagename. Am I missing something? DCDuring (talk) 18:12, 27 June 2018 (UTC)

@DCDuring: I agree, the page number should be the first parameter and the entry name the second (because it is optional). I can implement that, as it looks like the template is only used once. — Eru·tuon 18:21, 27 June 2018 (UTC)

Maybe it should be used on Basque entries by default. DCDuring (talk) 18:28, 27 June 2018 (UTC)

I'm the one user who used it (and then deleted it). I'm not going to use it for any Basque entries until I know it's fixed. References to Trask's Etymological Dictionary are easy enough to format without worrying about that template. (And like I said earlier, I don't think this template does anything valuable enough to be worth fixing or supplying documentation for.Namnagar (talk) 19:08, 27 June 2018 (UTC)

@Namnagar: As I explained, the template works and it wasn't really broken. Look at zurgin. I don't understand; it seems quite valuable to me to only have to type {{R:EDB|page number|entry name if different from page name}} rather than to type or copy the whole title, author, URL, and whatever else into every entry. — Eru·tuon 19:31, 27 June 2018 (UTC)

OK. So it wasn't broken, it just didn't have documentation, and you added that. All good! Thanks for your help!Namnagar (talk) 23:55, 27 June 2018 (UTC)

Please see the discussion at Template talk:TickerEntry#Misnested tags with diff parameter regarding Misnested tag with different rendering in HTML5 and HTML4 of <span> in Template:TickerEntry, and contribute to the discussion there, not here. —Anomalocaris (talk) 19:51, 27 June 2018 (UTC)

 Done —Anomalocaris (talk) 21:24, 28 June 2018 (UTC)

Template:rfv-etymology has an edit request at Template talk:rfv-etymology#Edit request. —Anomalocaris (talk) 17:35, 28 June 2018 (UTC)

 Done. - -sche (discuss) 17:43, 28 June 2018 (UTC)

Is there any way I can draw out all Wiktionary entries for Sanskrit words of Semitic origin?

Category:Sanskrit terms derived from Semitic languages. —Μετάknowledge^{discuss/deeds} 20:31, 28 June 2018 (UTC)

As of a few hours ago or so, when an admin makes a rollback, that edit has to be manually patrolled. Even if the admin is added to the autopatroller group (which is strictly speaking unnecessary), the problem persists). Is this something on our end that just broke, or should we file a bug report? @Chuck Entz —Μετάknowledge^{discuss/deeds} 04:51, 29 June 2018 (UTC)

Whether it's on our end or not, it's probably not something we can fix, so we need a bug report. I don't think I've ever done a bug report before, so I'm probably not the best one to do it.Chuck Entz (talk) 04:58, 29 June 2018 (UTC)

FWIW I just rolled back some vandalism and can confirm that this is happening as described. Incidentally, using "rollback" does (correctly) still mark the edit which is rolled back as patrolled. - -sche (discuss) 05:49, 29 June 2018 (UTC)

Might be this bug T198449. - TheDaveRoss 11:41, 29 June 2018 (UTC)

That's the bug, but it looks like it may not get fixed as soon as we'd like. A solution for the time being that was mentioned there was periodically running a bot to patrol all the edits with the rollback tag. —Μετάknowledge^{discuss/deeds} 23:14, 29 June 2018 (UTC)

@Mx. Granger, Atitarev, Guldrelokk Granger reports that in обняться and убраться, the footnote symbol appears in the conjugation table (a superscripted number 1) but the footnote itself doesn't appear at the bottom (it should say "Becoming dated." for обняться and "Dated." for убраться). I see both footnotes fine. Guldrelokk doesn't see them either, though. Any ideas why some people would see something that others don't? The code that creates the conjugation table and adds the footnotes is at the bottom of Module:ru-verb (look for 'internal_notes'). Benwing2 (talk) 13:32, 29 June 2018 (UTC)

@Benwing2: Have you tried clearing your cache? Guldrelokk (talk) 13:34, 29 June 2018 (UTC)

@Guldrelokk I tried looking at the pages using Safari (which I never use) on Mac OS X, after clearing the cache on that browser, and not logged in, and I see the footnotes fine. Benwing2 (talk) 13:52, 29 June 2018 (UTC)

@Benwing2: Can you get the export.show output through the debug console? You can alter it temporarily without saving so that it use predefined ‘arguments’. Guldrelokk (talk) 13:59, 29 June 2018 (UTC)

@Benwing2 Fine, no need, it’s there:
<caption>Note: For declension of participles, see their entries. Adverbial participles are indeclinable.

</caption>
<caption>¹Dated.
</caption>
Part of the HTML. I see it in the page source, it just doesn’t show up for some reason. I am using Firefox. Guldrelokk (talk) 14:09, 29 June 2018 (UTC)

@Guldrelokk Hmmm. I normally use Chrome on Mac OS X, where it shows up, and Safari displays it too. I'll try Firefox. Benwing2 (talk) 14:13, 29 June 2018 (UTC)

@Guldrelokk I tried using Firefox 52.8.1 (the latest pre-Quantum version) on Mac OS X and it doesn't show. Must be a Firefox bug of some sort. Benwing2 (talk) 14:18, 29 June 2018 (UTC)

(edit conflict) @Benwing2: If I remove the first <caption>…</caption>, with the note, through Web Inspector, the second one with ‘1. Dated’ shows up. From the documentation it appears that there should only be one <caption> element per table, and while other browsers tolerate multiple captions, Firefox seemingly doesn’t. Something must have changed about the output, there was probably only one capture before, because I didn’t update my browser. Guldrelokk (talk) 14:20, 29 June 2018 (UTC)

@Guldrelokk I see ... Let me try to change the code to see if I can fix this issue. Benwing2 (talk) 14:22, 29 June 2018 (UTC)

@Guldrelokk Try it now. Benwing2 (talk) 14:25, 29 June 2018 (UTC)

@Benwing2: Same issue, same HTML. But before you try again: the thing is, you haven’t changed this part of code for some time, right? I fear that the problem is not in the module. Was MediaWiki recently updated or something (I don’t really know how it all works)? Guldrelokk (talk) 14:33, 29 June 2018 (UTC)

@Guldrelokk, Benwing2: I think I've fixed it. — Eru·tuon 00:05, 30 June 2018 (UTC)

Yes, I can see the footnotes on Firefox now. —Granger (talk · contribs) 01:23, 30 June 2018 (UTC)

Works now, thank you. Guldrelokk (talk) 03:57, 30 June 2018 (UTC)

What other entries should have several footnotes? Guldrelokk (talk) 14:41, 29 June 2018 (UTC)

@Mx. Granger, Benwing2, Guldrelokk: My late reply - I can see the footnotes fine. --Anatoli T. ^{(обсудить}/^вклад) 23:10, 29 June 2018 (UTC)

Like Benwing2, I can see them on Safari but not Firefox. —Granger (talk · contribs) 23:30, 29 June 2018 (UTC)

I use Firefox (version: Quantum 62), and also can't see the footnote. It's placed in a table caption element (a caption tag generated with the wikitext |+), below another table caption element. This seems unusual. Usually a table has just one caption. Maybe Firefox has been programmed to not show the second of two table captions in the same table. Putting the footnote in a different HTML element would probably fix the problem. — Eru·tuon 23:55, 29 June 2018 (UTC)

I don't know whether this is possible, but it would be useful in maintaining and extending the Lojban pages. I know there is a way to look for "needed" pages for the wiki as a whole using a "Special" link. I would like to find when a link in Appendix:Lojban is bad (probably because it was marked up with square brackets instead of the "l" Template) because the Lojban words are no longer in mainspace but now in an Appendix. Could you help me find how to ask Wiktionary for a list of those links in pages that need to be changed? Thank you Jawitkien (talk) 15:36, 29 June 2018 (UTC)

@Jawitkien: I may be misunderstanding, but for entries in Index:Lojban the easiest way to find bad links is probably just to click "edit" and scroll through all of the markup. Looks like they are uniformly using {{l}}. If you are looking for links to Lojban words in other entries, e.g. in translation sections, that is probably difficult or impossible to do with the search, but should be reasonably accomplished by analyzing the database dump. I can probably help you with that if that is what you are trying to do. - TheDaveRoss 20:39, 29 June 2018 (UTC)

@TheDaveRoss, I wish you were right. I sometimes think I am the only one who is actively changing things to make them better. What you see as {{l}} uniformity on recent pages, is largely my work. The situation is twofold:

1) old pages use . Since Lojban is no longer in the mainspace, these need to be converted into {{l|jbo|l-template links}} and with my preference into >{{l|jbo|l-template links|gloss={{l|en|standard gloss for this lojban word}}}} This is brain-dead enough that if I understood bots, I'd probably just setup a table and say do this replacement if you find this pattern.

2) places where the "smart" conversion of text in a template that thinks it is always okay to replace a lojban link but now makes broken links because there is already a {{l|jbo|l-template links}}. This is most evident on rafsi (abbreviation) pages, and on calls to the "ux" template.

The sad thing is that I was replacing examples in Lojban words to use the ux template, and now it creates double Appendix:Lojban links which end up being broken links.

If anyone can advise, I would appreciate the help Jawitkien (talk) 15:06, 12 July 2018 (UTC)

@Jawitkien: Can you provide an example of the square bracket link you are hoping to find all instances of? - TheDaveRoss 15:13, 12 July 2018 (UTC)

Of course, (GRIN) if I knew where they were, I wouldn't be asking for a tool, GRIN. But seriously, they are just simple lojban words in . For example, you see some of them on ka which points out the difficulty of this. When looking at the page, you can't see that the links to is NOT linking to the Lojban word, but to the mainspace word ka.

@Jawitkien: Ah, I see what you are looking for now. Since there is no distinction between words which are Lojban and those which are English (or other languages) it will be a bit tricky. I can easily give you a list of all Lojban appendix pages which contain square-bracket links, which would be a place to start. I can possibly also limit the list of such pages to those which contain square-bracket links to terms which have Lojban appendix pages, but then you would miss any redlinks. There may be alternative ways of getting at this information but none spring readily to mind. - TheDaveRoss 17:49, 12 July 2018 (UTC)

@TheDaveRoss: Is it possible to find the list of square-bracket links to terms that are red-linked, regardless of if they are Lojban words? It might be useful to find those that point to mainspace words that don't exist. This might be an indication of a misspelling, for example.

Now if I can find the reason the {{ux}} template and other templates are over compensating and creating false redlinks that would be nice too. Jawitkien (talk) 17:57, 12 July 2018 (UTC)

You can use the square bracket notation in {{ux}} and the links will correctly point to the appendix. DTLHS (talk) 17:59, 12 July 2018 (UTC)

@DTLHS: thanks for the suggestion. I am concerned that using the square brackets in the {{ux}} and not using the {{l}} form would confuse anyone else editing the Lojban entries, as it is another rule that isn't explicitly stated but would have to be remembered or learned by everyone who used it.

@DTLHS: I know you and I haven’t talked about this in a year or so. Could you please give me your current thoughts about this and ideas about how to do some of this with a dump or not ? especially in the idea of porting some of the content here to the lojban wiktionary ?Thanks Jawitkien (talk) 18:48, 27 July 2019 (UTC)

@Jawitkien User:DTLHS/lojban DTLHS (talk) 05:52, 28 July 2019 (UTC)

How can I pull out all Wiktionary English entries which have a Provencal origin? (I would also be interested in words of Provencal origin in other languages) — This unsigned comment was added by Rdurkan (talk • contribs) at 23:13, 29 June 2018 (UTC).

You've asked many of these questions without providing any indication that you're listening to the responses. DTLHS (talk) 23:15, 29 June 2018 (UTC)

Category:Terms_derived_from_Provençal (surprisingly they exist). Guldrelokk (talk) 05:24, 30 June 2018 (UTC)