Jump to content

Wikipedia:Village pump (technical)

From Wikipedia, the free encyclopedia

This is an old revision of this page, as edited by Iustinus (talk | contribs) at 01:49, 10 November 2016 (→Insert – — ° ′ ″ ≈ ≠ ≤ ≥ ± − × ÷ ← → · § etc. etc.: fmt). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

 Policy Technical Proposals Idea lab WMF Miscellaneous 
Shortcuts
The technical section of the village pump is used to discuss technical issues about Wikipedia. Bug reports and feature requests should be made in Phabricator (see how to report a bug). Bugs with security implications should be reported differently (see how to report security bugs).

Newcomers to the technical village pump are encouraged to read these guidelines prior to posting here. Questions about MediaWiki in general should be posted at the MediaWiki support desk.


Category sort by number: now grouped in single 0–9

Resolved

Apparently, category sorting now groups numbers 0–9 in one single caption. (Used to be ten captions: 0 ... 9). Does anyone have a link for the background ratio of this? -DePiep (talk) 12:17, 2 November 2016 (UTC)[reply]

The purpose is to sort all numbers in numerical order which is not possible if you group by leading digit. See phab:T8948, meta:2015 Community Wishlist Survey/Categories#Numerical sorting, Wikipedia:Village pump (technical)/Archive 149#Sorting in categories unreliable for a few days. PrimeHunter (talk) 12:33, 2 November 2016 (UTC)[reply]
Answered, thanks. -DePiep (talk) 20:05, 2 November 2016 (UTC)[reply]
one late question, just curious. With this, within the 0–9 category caption, are articles sorted numerically truly? (how are sorted articles: '500 X' '51 X' '050 X'?) — Preceding unsigned comment added by DePiep (talk • contribs) 23:43, 5 November 2016 (UTC)[reply]
@DePiep: See this category from 50; and then same category from 500. --Redrose64 (talk) 00:12, 6 November 2016 (UTC)[reply]
Perfect example demo! Just see how article $2000 ends up there. -DePiep (talk) 00:23, 6 November 2016 (UTC)[reply]
Well, $2000 isn't in that cat; however the page that it redirects to ($2,000) is, but it has {{DEFAULTSORT:2000 dollars}} which is why it doesn't sort with $1 etc. but between 2000 Continental Championships and 2000 Emmy Awards. The thing to do is to look at pages where there isn't a {{DEFAULTSORT:...}} - such as $1000. --Redrose64 (talk) 00:36, 6 November 2016 (UTC)[reply]
Of course. I get it. -DePiep (talk) 20:46, 6 November 2016 (UTC)[reply]

WikiProject quality rating on Talk page not showing

Talk:Yohannes IV has "WikiProject Ethiopia| class=C" but it displays as "has not yet received a rating on the project's quality scale". Any ideas? Nurg (talk) 09:48, 4 November 2016 (UTC)[reply]

@Nurg: In {{WikiProject Ethiopia/class}} the line c=no means that this WikiProject doesn't grade articles as C class. -- John of Reading (talk) 09:59, 4 November 2016 (UTC)[reply]
Thank you very much. Nurg (talk) 19:57, 4 November 2016 (UTC)[reply]

Strange problem with Module:navbar?

at the bottom of ZiL#External_links, the last navbox says "parent" next to the title. but, when I look at Template:ZIS/ZIL timeline 1930-1960, I don't see the same thing. It seems like a LUA problem? @Jackmcbarn and Mr. Stradivarius: thank you. Frietjes (talk) 14:27, 4 November 2016 (UTC)[reply]

ZiL#External links actually transcludes {{ZIS passenger car since 1937 to 1960}} but that template has code which generates a "V" link to {{ZIS/ZIL timeline 1930-1960}} instead of to itself. PrimeHunter (talk) 14:38, 4 November 2016 (UTC)[reply]
This edit should fix it. --Redrose64 (talk) 15:28, 4 November 2016 (UTC)[reply]
thank you. not sure why I was looking at the wrong template (probably the broken v-t-e links). Frietjes (talk) 16:50, 4 November 2016 (UTC)[reply]

I am copying the footnote syntax used in United States v. Progressive, Inc., in which if you click on the link in the Notes section you jump to the relevant highlighted entry in the References section. In the Doris Grumbach bio I am working on in my sandbox, User:HowardMorland/Sandbox-6, the link from Notes to References fails to work. What am I doing wrong? Thanks. HowardMorland (talk) 19:48, 4 November 2016 (UTC)[reply]

There were multiple reasons: (i) a typo Nowak -> Nowack; (ii) you used simple dash instead of ndash in two anchors; (iii) citation template generates anchors, which include only years but not full dates. Ruslik_Zero 20:14, 4 November 2016 (UTC)[reply]
Thanks. The ndash seems to fix it. HowardMorland (talk) 20:38, 4 November 2016 (UTC)[reply]

HELP - ImageMap Problem

QUESTION: Seems some of my imagemap work may have gone well - esp "Template:Features and artificial objects on Mars" and my newly created template "Template:Features and memorials on Mars" (imagemap image for each = "File:Mars Map.JPG"; imagemap coordinates determined via http://imagemap-generator.dariodomi.de/) - However - adapting another image ("File:Pierre-Auguste Renoir - Luncheon of the Boating Party - Google Art Project.jpg") (related to the article => "Luncheon of the Boating Party") does not seem to work at all - ie, coordinates related to people pictured in the painting (via http://imagemap-generator.dariodomi.de/) do not seem to be noticed on a newly created template page similar to the templates mentioned earlier - what might I do to make this imagemap effort work with this Renoir image? - in any regards - Thank you for considering the issue - and - Enjoy! :) Drbogdan (talk) 21:06, 4 November 2016 (UTC)[reply]

Where is the template? I couldn't find it in your contributions or uses of the image. PrimeHunter (talk) 21:38, 4 November 2016 (UTC)[reply]
@PrimeHunter: Thank you for your comments - my earlier template efforts were "Preview" only - and not saved since the effort did not seem ok - a very recent return try seems to have worked ok and, if interested, the newly created template is at => "Template:Renoir-BoatingParty-ImageMap" - seems the coordinates worked with the original image file size (at 3973x2933) - but not with a smaller image file size (ie, 800x591) - I should now be ok with this - Thanks again for your comments - and for considering the issue - it's *greatly* appreciated - Enjoy! :) Drbogdan (talk) 22:38, 4 November 2016 (UTC)[reply]

Ref "Wrapping"

Tracked in Phabricator
Task T20231

How plausible and helpful would it be to set up ref's in such a way that upon putting a cursor over the number of the ref, it would highlight anything in the section that it was verifying. This would likely be achieved by putting certain words, perhaps <startref"something"> and then <endref"something"> before and after (Respectively) the sentences the ref pertained to, it would pull from the ref (one of the main weaknesses I can see in my proposal is that it would likely require that all refs use a refname= format.) What do you guys think? Iazyges Consermonor Opus meum 23:44, 4 November 2016 (UTC)[reply]

It's an interesting thought. I'd find it helpful; it's not satisfactory that there's a lack of clarity as to which sentences or sentence fragments are covered by a ref. Equally, I have some slight UI concerns, and a more general pessimism about uptake ... but then, references and categories once were new on wikipedia. --Tagishsimon (talk) 23:48, 4 November 2016 (UTC)[reply]
@Tagishsimon: I have found that very few people even use refnames, I personally love them, mostly because I work with classical history, where most of the article is from a few books, but I haven't come across much usage of them. I don't know how many people will use it as their first choice, but it may be useful when clarity is requested. Iazyges Consermonor Opus meum 23:53, 4 November 2016 (UTC)[reply]
Refnames are ace, but {{sfn}} is more ace ;) --Tagishsimon (talk) 23:59, 4 November 2016 (UTC)[reply]
@Tagishsimon: I'm gonna have to disagree with you their. Iazyges Consermonor Opus meum 00:03, 5 November 2016 (UTC)[reply]
Unfeasible. Was discussed very recently. Forget where. --Redrose64 (talk) 23:49, 4 November 2016 (UTC)[reply]
@Redrose64: what exactly about it is unfeasible? The coding or the usage by users? Iazyges Consermonor Opus meum 23:53, 4 November 2016 (UTC)[reply]
Ironically, we have a template to mark the inverse case: the lack of citation for a span of text (Template:Citation needed span). – Finnusertop (talk ⋅ contribs) 23:56, 4 November 2016 (UTC)[reply]
@Finnusertop: so (Arguably) it would be very helpful for those cases? Iazyges Consermonor Opus meum 00:03, 5 November 2016 (UTC)[reply]
Arguably, but I'm not arguing for that. Personally, whenever I find that, say only half of a paragraph is sourced to the ref that sits at the end of it, I split the paragraph. It's a fairly standard convention that references apply to the preceding paragraph, from start to finish. – Finnusertop (talk ⋅ contribs) 00:07, 5 November 2016 (UTC)[reply]
Agreed, but in some cases, the only logical place to add content with the ref is in the middle of a paragraph, and sometimes you have a lot of these, which makes it somewhat confusing. Iazyges Consermonor Opus meum 00:09, 5 November 2016 (UTC)[reply]
It is not a secret that sloppy referencing by previous editors causes confusion and hard work for others. – Finnusertop (talk ⋅ contribs) 00:12, 5 November 2016 (UTC)[reply]
@Finnusertop: Which is why i'm making a proposal for something that can help minimize future confusions and hard work for others. Iazyges Consermonor Opus meum 00:16, 5 November 2016 (UTC)[reply]
This is tracked as a Phabricator task. I'm aware of a handful of discussions onwiki as well. --Izno (talk) 02:03, 5 November 2016 (UTC)[reply]
@Izno: so will it likely be added? I admit my lack of knowledge with phabricator. Iazyges Consermonor Opus meum 02:15, 5 November 2016 (UTC)[reply]

CXT

Inside Special:SpecialPages, the link to the Special:ContentTranslation seems to have something wrong. Maybe it should be fixed. --Stang 01:47, 5 November 2016 (UTC)[reply]

It appears MediaWiki:Cx by Xaosflux cannot currently parse wikilinks when it's used. PrimeHunter (talk) 02:20, 5 November 2016 (UTC)[reply]
 Done I removed the first pipe, the second one is left for copy and paste purposes from the translation screen. — xaosflux Talk 02:25, 5 November 2016 (UTC)[reply]
Can these text appear inside Special:CX? --Stang 02:27, 5 November 2016 (UTC)[reply]
I'm up for any suggestions where to move this - the developers for that special page didn't seem to make it very customizable. — xaosflux Talk 02:33, 5 November 2016 (UTC)[reply]
see also WT:CXT where this was discussed. — xaosflux Talk 02:33, 5 November 2016 (UTC)[reply]
Ok, thanks for your help:) --Stang 02:34, 5 November 2016 (UTC)[reply]

Need template adjustment at Template talk:Infobox television

Resolved

Hello! Per the discussion at Template talk:Infobox television#Proposal: Add "native_name" parameter could someone please lend a hand to add |native_name= to Template:Infobox television underneath |show_name_2=? I'm not sure what it all entails to do this, but the idea was to duplicate Template:Infobox film's |film name= parameter. So for instance at the film article for Drishyam:

| film name = {{Infobox name module|ml|ദൃശ്യം}}

produces:

Malayalam       ദൃശ്യം

in the infobox. We want that same thing to be possible for TV articles, but we're changing the parameter to |native_name=. Thanks in advance for the help, Cyphoidbomb (talk) 02:41, 5 November 2016 (UTC)[reply]

URL encoding colon and slash

Given this URL:

http://www.webcitation.org/66gupqQDM?url=http%3A%2F%2Fviewer.nationalmap.gov%2Fviewer%2F

The "http%3A%2F%2F" has been percent-encoded because it's in the query portion of the URL (the part following the "?"). This is in accordance with RFC 3986, which states in section 2.2 that : and / are reserved characters, meaning they should be encoded in the path and query strings.

There is also a special rule in section 3.4 that in the query string for characters / and ?, because they are frequently used in URIs "it is sometimes better for usability to avoid percent-encoding those characters". Thus this would also be acceptable:

http://www.webcitation.org/66gupqQDM?url=http%3A//viewer.nationalmap.gov/viewer/

However the question is, why don't we do this?

http://www.webcitation.org/66gupqQDM?url=http://viewer.nationalmap.gov/viewer/

Drop the percent encoding of the colon %3A. According to section 2.2, percent encoding is needed only if there is a possibility of conflict with interpreting its meaning.. and the only purpose of the colon is in the protocol and port strings. Once we get to the path and query strings there is no possibility of conflict and so it's safe not to encode it. Certainly we do stuff like this all the time:

https://web.archive.org/web/*/http://example.com

Any thoughts on not percent-encoding the : and / in the path and query strings? -- GreenC 04:50, 5 November 2016 (UTC)[reply]

You can do this, but it is harder when you have to use implementations of query encoders. Adapting a generic implementation to follow your bidding can be hard, invites human error. Also server implementations cannot be relied upon 100% to stay consistent on an edge case like that. (though I think you'll be hard pressed these days to find a server implementation anymore that implements this incorrectly, those were actually still quite common up to like 5 years ago). Note that archive.org is following it's own server specific interpretation of that url. It requires work in a server to make sure that you always get the right value if you have your urls specified like that, and it makes it harder to parse with generic solutions and for the software of reusers of the links. —TheDJ (talk • contribs) 11:18, 5 November 2016 (UTC)[reply]
Ok thanks. I guess the question is specific to webcitation.org and their servers have proven to work using : instead of %3A .. the query encoder I'm using Python 2.7's query() which I believe has the option to exclude characters from encoding so shouldn't be a problem. -- GreenC 15:33, 5 November 2016 (UTC)[reply]
I should probably point out that the long-form URL that is supplied in this example, the url value being passed is completely ignored by their server because the snapshot is actually retrieved from the supplied base-62 encoded timestamp.—cyberpowerChat:Limited Access 20:40, 5 November 2016 (UTC)[reply]
@Cyberpower678: You're right it doesn't seem to matter what's in the url query. This presents a problem for the security of blocking blacklisted websites, per the long-form RfC. We probably should have be doing this:
http://webcitation.org/query?url=http://viewer.nationalmap.gov/viewer/&date=20131023000000 .. -- GreenC 14:31, 6 November 2016 (UTC)[reply]
I don't think it's that big of a problem. Since webcite doesn't crawl, the chances of inappropriate cites being archived don't seem as high.—cyberpowerChat:Online 14:35, 6 November 2016 (UTC)[reply]
Well true but someone who wants to link to a blacklisted website can .. manually initiate the archive of the blacklisted page, then link to it on Wikipedia using a false url parameter. But you're right it's probably not going to be a big problem, though it does essentially violate the spirit of the RfC which was to prevent web shortening from hiding blacklisted sites. I don't think this information was known during the RfC, that the url parameter is ignored. Right now WP:Using_WebCite#Use_within_Wikipedia says either format is appropriate. -- GreenC 14:49, 6 November 2016 (UTC)[reply]
I think I'll keep using the base62 + url query because a future bot can always do the work of verifying they match up, searching for hidden blacklisted URLs. If that bot discovers a big problem then we can deal with it then. -- GreenC 15:01, 6 November 2016 (UTC)[reply]

DYKN broken

Is it just be, or does WP:DYKN not display properly after November 1 (the nomination templates no longer transclude)? Satellizer el Bridget (Talk) 05:01, 5 November 2016 (UTC)[reply]

Yes. It's exceeding the post-expand include size. Already discussed in WT:DYK as I see. Jo-Jo Eumerus (talk, contributions) 09:45, 5 November 2016 (UTC)[reply]
This is Wikipedia:Village pump (technical)#DYK template transclusions on nomination page and Wikipedia talk:Did you know#Something wrong with the DYK nominations special occasion holding area?. Quick précis: WP:TLIMIT. --Redrose64 (talk) 13:32, 5 November 2016 (UTC)[reply]

Automatic alteration of section order in all articles (Basque WP)

Hi! Just by chance if anyone has a clue and answers to this it would be really appreciated. Is there any way to alter the section order in the general layout of articles for a specific WP project (Basque WP), I mean automatize it? Regards Iñaki LL (talk) 12:01, 5 November 2016 (UTC)[reply]

You can create a bot to do this. Ruslik_Zero 20:06, 5 November 2016 (UTC)[reply]
Thank you Ruslik, will check this with the bot managers in the Basque WP. Iñaki LL (talk) 22:25, 7 November 2016 (UTC)[reply]

Preview page with this module:Page title

It used to be that I could save a version of my sandbox as a simple test for work to be done in a Lua module. I could then make an edit in the Lua module, put the name of my sandbox in the Page title box and click the Show preview button. Very handy, that. If the module change that I had made was not quite right (not at all uncommon), I could tweak the module and again click Show preview. Repeat ad nauseum.

That worked well because my browser remembered the value in the Page title box so I only had to enter it once (the browser also remembered it session-to-session so on the first module edit preview of a new session, a couple of letters was enough to get the browser to show the full name of my sandbox, thereafter retaining it until I was done.

Now, every time I want to preview a module change, I have to type user:trappi at a minimum to get to my user pages and then scroll down the list to my sandbox. Every time. Who thought that this would be a good idea?

Is there any way to get around this 'improvement'?

—Trappist the monk (talk) 14:11, 5 November 2016 (UTC)[reply]

Template changes, too. I first noticed it at approximately 15:00, 3 November 2016. --Redrose64 (talk) 15:12, 5 November 2016 (UTC)[reply]
Works for me in Firefox 49.0.2. PrimeHunter (talk) 15:36, 5 November 2016 (UTC)[reply]
Not a browser issue, then: I have Firefox 49.0.2 --Redrose64 (talk) 16:21, 5 November 2016 (UTC)[reply]
Works correctly for me too, after submitting a preview the box continues to be pre-filled with the same title. And I note the HTML served includes the title as expected. You should check to see if you have a user script or gadget that is interfering with it somehow. Anomie⚔ 17:34, 5 November 2016 (UTC)[reply]
For me, chrome on both winxp and win7. Does not work on either regardless of whether I'm logged in or not logged in.
—Trappist the monk (talk) 17:47, 5 November 2016 (UTC)[reply]

How to widen clickable space for linked "I"?

In the periodic table of chemical elements, we have character/symbol "I" for iodine. How can I widen the clickable space? So I better →  I , but of course we won't use spaces-to-format (and it copy/pastes badly). Any ideas, css maybe? -DePiep (talk) 00:16, 6 November 2016 (UTC)[reply]

Certainly not non-breaking spaces, but a <span>...</span> inside the link (with suitable styling) does work: I. --Redrose64 (talk) 00:21, 6 November 2016 (UTC)[reply]
An easy solution would be to wrap the symbol in a <div> inside the [[ ... ]] in Template:Element cell-named. That would expand the clickable area to the width of the table cell. It would need a bit of testing, considering how many pages use it. --Unready (talk) 01:18, 6 November 2016 (UTC)[reply]
Hmm, maybe not so easy. The parser doesn't want to let a <div> through the [[ ... ]]. Forcing it with 
$('a[title="Iodine"]').contents().wrapAll('<div></div>');
works nicely, but it's not practical. --Unready (talk) 01:43, 6 November 2016 (UTC)[reply]
Redrose64 provided the solution above. Here it is, with "abc" before and " xyz" after:
abc[[iodine|<span style="margin-left: 1ex; margin-right: 1ex">I</span>]] xyz → abcI xyz
I had to put a space before "xyz" to avoid it being part of the link. Johnuniq (talk) 01:57, 6 November 2016 (UTC)[reply]
I would suggest instead <span style="display: block;">I</span>. It makes the span into a div and gets it through the parser. It doesn't inflate the element artificially, but lets its width expand to 100% of its containing block. --Unready (talk) 02:07, 6 November 2016 (UTC)[reply]
Using Unready's display: block gives:
abc[[iodine|<span style="display: block;">I</span>]] xyz → abcI xyz
... so, with newlines. Or do I mistake? Anyway, I see the underline does not expand in any situation, even though the sensitive area is wider. Methinks this could be by w3c design. -DePiep (talk) 20:43, 6 November 2016 (UTC)[reply]
You're missing that free text (some text, the block, some more text) creates a new block in the middle of some text. The periodic table has only a centered line of text (a new block inside another block). Therefore there are no new newlines in the application. (Feel free to try it there yourself.) The underline text decoration is not longer, but the clickable area is wider, which was the original request. To have a longer decoration, you need to have longer text. --Unready (talk) 22:26, 6 November 2016 (UTC)[reply]
I get it, the span acts as a div-block. Now nicely covers the whole table cell, not just some text area any more :-). Deployed for the "I", planning to do this for all cells (had to add |3= parameter naming, obviously). -DePiep (talk) 14:18, 7 November 2016 (UTC)[reply]
Without using 3=, you could also use {{=}} for = in style=. Even better IMO would be to put the span in Template:Element cell-named for all the symbols. I'm not certain if that'll break any other pages, though, so it would need testing. --Unready (talk) 07:09, 8 November 2016 (UTC)[reply]
Yes, in the subtemplate. Now deployed for Template:element cell-compact for all cells indeed. See this article (not navbox, so mobile check possible too!). Template:Element cell-named will follow, though people now start asking to have the whole cell sensitive (over 3 text lines) ;-) . What do you think of the Template:Element cell-named/sandbox2 construct? (testing is in the first, hydrogen cell in Template:Periodic table/sandbox). Too much of a hack, or acceptable w3c design? -DePiep (talk) 12:05, 8 November 2016 (UTC)[reply]

Mathematical equations' font too tiny in mobile view.

Tracked in Phabricator
Task T150052

I use Opera v.30 on a Samsung Android 5.1.1 device to read most articles on here. But, whenever I get to the math equations, they virtually atomise. I can only see them after zooming so far in or switching to desktop. The mobile view is a lot more convenient on a phone for me. Please, what can be done? — Preceding unsigned comment added by Daimler Ben (talk • contribs) 02:17, 6 November 2016 (UTC)[reply]

Corresponding ticket in the issue tracker is phab:T150052. --AKlapper (WMF) (talk) 19:15, 6 November 2016 (UTC)[reply]

Accessing "Cite" in the wikitext edit window

In Wikipedia:Village pump (technical)/Archive 144#Can't always access "Cite" in the wikitext edit window, I mentioned a problem I was having in able able to use the Cite function in the wikitext edit window, so as to use the pop-up forms for the "cite web", "cite book", and other cite templates. Well, the problem has recurred. Not only do I only get the "Cite" function to appear above the edit window only occasionally, but when it does appear, clicking on "cite web", "cite news", "cite book", or "cite journal" has no effect at all; no form pops up to enable me to supply the author, title, etc. This problem appears to apply to all articles I have been trying to edit today, but in case anyone is wondering, the first one I noticed it on was Leopold Frankenberger, Jr. --Metropolitan90 (talk) 02:43, 6 November 2016 (UTC)[reply]

The last time something like this came up, it was related to user language settings; are you using an English variant as your use language (e.g. Canadian English, British English etc) rather than just plain English? That might be the cause of this issue. — foxj 04:08, 6 November 2016 (UTC)[reply]
In my preferences under Internationalization, it's just en - English, not Canadian or British English. --Metropolitan90 (talk) 18:16, 6 November 2016 (UTC)[reply]
I've had this quirk, but not often. That is, when it happens to me, I click the dropdown for Templates, and select one of the cites. And absolutely nothing happens. Repeatedly. How I clear that, is to "Show preview", and after it goes through the motions of that, it seems to clear the issue. Then I can do a cite. It's been happening a few times the last week or so. It's happened in the more distance past, also. And then it doesn't happen again for long periods. — Maile (talk) 18:20, 6 November 2016 (UTC)[reply]
This problem is JavaScript that is either failing to load completely, or failing to run to completion. For the first, it might be slow servers, a poor connection, or a memory issue on your machine. For the latter, it might be a bug in the JavaScript, or a browser problem at your end. There are other possibilities for both. In Firefox, if the page tab doesn't display the "W" favicon but instead remains at a spinny thing, that's a giveaway for a page that hasn't finished loading/executing. Usually a hard reload (Ctrl+F5 in Firefox) will fix it, but will refresh the edit window so you will lose your changes. Going for the Preview button also loads a fresh set of JavaScript files. --Redrose64 (talk) 19:24, 6 November 2016 (UTC)[reply]
@Metropolitan90 and Maile66: You need to have the "Enable wizards for links, formatting, tables, citations, and the search and replace function" option checked in Special:Preferences#mw-prefsection-editing for the Reftoolbar gadget to work. Please see if that's the case. The preferences and documentation for the various toolbars are very confusing. - NQ (talk) 19:51, 6 November 2016 (UTC)[reply]
Mine was already checked. — Maile (talk) 20:59, 6 November 2016 (UTC)[reply]
@Maile66: I'm unable to reproduce it using your personal javascript. Do you see any errors in the console (WP:JSERROR) when you load the editing window? Might be a conflict with one of the gadgets causing the js to fail. See Wikipedia:Village pump (technical)/Archive 148#What happened on or about 6 August?. - NQ (talk) 21:13, 6 November 2016 (UTC)[reply]
From my console, are these error messsages?— Maile (talk) 21:25, 6 November 2016 (UTC)[reply]
Extended content
Use of "wgPageName" is deprecated. Use mw.config instead.(unknown)
Use of "wgPageName" is deprecated. Use mw.config instead.(unknown)
Use of "escapeRE" is deprecated. Use mediawiki.RegExp instead.(unknown)
Use of "wgNamespaceNumber" is deprecated. Use mw.config instead.(unknown)
Use of "addPortletLink" is deprecated. Use mw.util.addPortletLink instead(unknown)
Use of "wgPageName" is deprecated. Use mw.config instead.(unknown)
Use of "wgContentLanguage" is deprecated. Use mw.config instead.
@Maile66: Everything seems fine. Could you try checking while logged out? If you still encounter the issue then it could be due to any number of things Redrose64 said above. - NQ (talk) 21:44, 6 November 2016 (UTC)[reply]
It's not that important to me. Why don't we just drop this? It take so little effort for me to clear it on my own. Thanks for your input. — Maile (talk) 21:47, 6 November 2016 (UTC)[reply]
According to the documentation at Wikipedia:RefToolbar/2.0#Troubleshooting and Help:Edit toolbar, if I have the "Enable enhanced editing toolbar" option checked and the "Enable wizards" options unchecked, I should get RefToolbar 2.0a with {{}} icon in middle. Instead, I get "RefToolbar 2.0b" with Cite button on right but no working popup. - NQ (talk) 20:08, 6 November 2016 (UTC)[reply]

A navbox

Is it possible to use custom bullets in an hlist? Jc86035 (talk) Use {{re|Jc86035}}
to reply to me 05:37, 6 November 2016 (UTC)[reply]

As I understand, no. Ruslik_Zero 19:58, 6 November 2016 (UTC)[reply]

How come Yahoo isn't showing up in the Finance links at Goldman Sachs#External links? The template doc says you can put in the Finance ID, but not to include it for public companies. Kendall-K1 (talk) 18:52, 6 November 2016 (UTC)[reply]

{{Finance links}} has been edited without changing the documentation. The current behaviour is to only make a Yahoo link if the yahoo parameter is set to a non-empty value. The value is ignored. The link will always be decided by the symbol paramter: http://finance.yahoo.com/q?s={{{symbol}}}. PrimeHunter (talk) 21:44, 6 November 2016 (UTC)[reply]
That seems like very unhelpful behavior, even if it were documented. Kendall-K1 (talk) 21:55, 6 November 2016 (UTC)[reply]

language=bn in cite templates changed to "in Bangla" from "in Bengali"

Please see this discussion at Help Talk:CS1. It appears that somewhere, the language code "bn" has been changed to return "Bangla" instead of "Bengali". These are two names for the same language, according to Wikipedia, but most other parts of en.WP use "Bengali" as the name for this language.

Please respond at the original discussion. Thanks. – Jonesey95 (talk) 04:24, 7 November 2016 (UTC)[reply]

Deletion and page history

To the outsider, one of the harsher aspects of MediaWiki is that, ignoring cases appropriate for Oversight, page deletion hides the page history. One can ask an admin for temporary undeletion to gain access to the history, but that seems odd and inconvenient. Has there ever been a discussion about separating the two functions? Perhaps an option could be added to the delete function to the effect of "page history visible". If it takes many years to schedule such software modifications, perhaps Wikipedia could change its ways. For instance, in closing an AfD, admins could be encouraged to move to draft namespace. Perhaps better than that, a new namespace pair called "Deleted" and "Deleted talk" could be created to move pages to. Such a change might encourage a few more experts and professionals to participate as editors.--172.56.33.76 (talk) 11:17, 7 November 2016 (UTC)[reply]

See Wikipedia:Perennial proposals#Deleted pages should be visible. PrimeHunter (talk) 11:30, 7 November 2016 (UTC)[reply]
Such a change might encourage a few more experts and professionals to participate as editors. Because they're the sort of people most likely to be enthused by writing articles that'll be semi-deleted? What arrant nonesense. --Tagishsimon (talk) 12:31, 7 November 2016 (UTC)[reply]
This is already technically possible in the software, see Wikipedia:User_access_levels#Researcher. The permission to view deleted titles & history is currently only granted by the WMF to official researchers. Mamyles (talk) 15:18, 7 November 2016 (UTC)[reply]

reftool

would appreciate any help w/ Wikipedia_talk:RefToolbar#Autofill_2, thank you --Ozzie10aaaa (talk) 15:13, 7 November 2016 (UTC)[reply]

Login and password reset

I cannot log in or reset my password. I have been a registered Wikipedian since 3 September 2003. I am a system administrator. https://en.wikipedia.org/wiki/User:Hawstom The system will not email me. — Preceding unsigned comment added by 72.201.76.62 (talk) 17:47, 7 November 2016 (UTC)[reply]

Yes, I did indeed make an unsigned comment. 72.201.76.62 (talk) 17:49, 7 November 2016 (UTC)[reply]
I noticed this problem about a week ago when I wanted to make a minor edit to an article. I tried again today with the same result. 72.201.76.62 (talk) 17:52, 7 November 2016 (UTC)[reply]
Do you see an option below the password that asks "Forgot your password?" That takes you to a page that asks for your user name and email address. The only other thing I can think of is Help:Reset password. I believe they used to take shorter passwords, but are now requesting Password strength requirements. — Maile (talk) 17:57, 7 November 2016 (UTC)[reply]
Thanks, but I will probably need to talk to a developer or high level bureaucrat. "The system will not email me." 72.201.76.62 (talk) 19:50, 7 November 2016 (UTC)[reply]
This is a strange question for a system administrator. However you are not actually one of them. You are just a local sysop. Ruslik_Zero 19:53, 7 November 2016 (UTC)[reply]
No signature and a claim to be a system administrator are odd mistakes for an administrator but User:Hawstom has claimed to be a system administrator since 2005.[1] What exactly happens when you try to log in and when you use Special:PasswordReset? Does "The system will not email me" mean you get a message saying something like that, or do you just not receive a mail at the address you think is stored in your account? Try to look in spam folders in your mail software and the website of your mail provider. Try other mail addresses you have access to if an old address may be stored. PrimeHunter (talk) 20:53, 7 November 2016 (UTC)[reply]
Well, one can be a system administrator and not an administrator of this system.. . . That is, the claim of being "a system administrator" and of being "a Wikipedia administrator" are two different things. - Nunh-huh 20:56, 7 November 2016 (UTC)[reply]
The user group used to be called sysop, which is short for system operator. Given the other things on that user page which indicate its age, that it should refer to the permission as "system administrator" isn't particularly bothersome. --Izno (talk) 05:12, 8 November 2016 (UTC)[reply]

Search for articles containing N'

Hi all. I'd like to know how to search for articles whose titles contain N' - e.g. Charles N'Zogbia.

The trouble is, if I search with the apostrophe, the results contain N-, N_ & all manner of not quite N'!

So, is there a way to restrict results to articles containing strictly N'?

Many thanks, Trafford09 (talk) 20:08, 7 November 2016 (UTC)[reply]

Are you talking about titles containing N', or beginning with N'? If the latter, Special:PrefixIndex is your friend; if the former, you're probably better off searching the database dump (follow the instructions here—the file you want is all-titles-in-ns0.gz) of page titles in an external program. ‑ Iridescent 20:30, 7 November 2016 (UTC)[reply]
Boo. Show us the regex. It's defeated me. --Tagishsimon (talk) 20:35, 7 November 2016 (UTC)[reply]
Not to oversimplify this, but would /.*n'.*/gi suffice? Could whip up a Quarry query? -- samtar talk or stalk 20:42, 7 November 2016 (UTC)[reply]
What's the g for? And no, doesn't seem to work from search. Neither does escaping the apostrophe [2]. --Tagishsimon (talk) 20:47, 7 November 2016 (UTC)[reply]
(edit conflict) /g is the 'global flag' - find all matches in string (linky) - wouldn't be needed here though, so apologies for including it. Special:Search doesn't support regex by the looks of it (though CirrusSearch does) so I'm running a quarry query -- samtar talk or stalk 20:58, 7 November 2016 (UTC)[reply]
'g' means global, if one does not include the g, the regex only acts on the first match. -- The Voidwalker Whispers 20:51, 7 November 2016 (UTC)[reply]
Search N' at https://tools.wmflabs.org/grep. I get 543 mainspace results. PrimeHunter (talk) 20:55, 7 November 2016 (UTC)[reply]
@PrimeHunter: Did not know about /grep, thank you! -- samtar talk or stalk 20:59, 7 November 2016 (UTC)[reply]

Thanks, Iridescent and all. I was referring to articles whose titles contain N', so I'll follow up your very helpful links. Trafford09 (talk) 00:10, 8 November 2016 (UTC)[reply]

Subset

Is there a way to find the files that are Unused and Fair use. Xaris333 (talk) 20:39, 7 November 2016 (UTC)[reply]

From the looks of things, none? I did a search for files in the category Fair use images, that have been tagged with {{Orphan image}}. The search produced nothing. I don't know how accurate this is, but it is all I've got. -- The Voidwalker Whispers 21:17, 7 November 2016 (UTC)[reply]
Wikipedia:Database reports/Unused non-free files is updated daily by a bot. Does that satisfy your needs? PrimeHunter (talk) 21:57, 7 November 2016 (UTC)[reply]
Also tried a slightly different search but got the same result. You can start to see some results if you dive into the various subcategories for non-free images here, but we can't do recursive searches in categories onwiki. The DB report above is probably more useful here :) ^demon[omg plz] 01:59, 8 November 2016 (UTC)[reply]

Photo of famous physicists

I would like to upload a photograph of a group of physicists taken at the Dublin Institute for Advanced Studies 1942. I have no idea what the copyright status is. However, the photo (and parts of it) is found in quite a few places on the internet including the 5 different URLs below. Is this an indication that the photo is open source and may be used in WP, say in the article about the Dublin IAS? OTH, If someone is sure that it is hopeless to use this photo, just say so and I will move on to other things.

Thanks for taking a look.--Toploftical (talk) 20:58, 7 November 2016 (UTC)[reply]

Mere use isn't enough to indicate it's public domain or open source. You need to know when it was first published (presumably 1942), where, and whether it was published with a copyright notice (and if so, if that notice was renewed.) The easiest course would be to e-mail DIAS and ask them. I suspect they hold the copyright, and probably won't release the photo under any licence we use. But asking is free.- Nunh-huh 21:08, 7 November 2016 (UTC)[reply]

Thanks for the great advice. I will try that.--Toploftical (talk) 22:27, 7 November 2016 (UTC)[reply]

Tech News: 2016-45

Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.

Recent changes

  • Victoria Coleman is the new Chief technology officer for the Wikimedia Foundation. [3]
  • First release candidate for MediaWiki 1.28 is now available. [4]
  • .gitreview for MediaWiki branches and extensions switched from targeting a specific branch to using track=1. [5]
  • Section numbers in Table of Contents boxes will use grey to improve readability. [6]

Changes this week

Meetings

  • You can join the next meeting with the VisualEditor team. During the meeting, you can tell developers which bugs you think are the most important. The meeting will be on 8 November 2016 at 20:00 (UTC). See how to join.
  • You can join the next meeting with the Architecture committee. The topics this week are Image Thumbnail API and allow SVG files uploaded on MediaWiki to have XHTML namespaces. The meeting will be on 9 November at 21:00 (UTC). See how to join.

Future changes

Tech news prepared by Tech News writers and posted by bot • Contribute • Translate • Get help • Give feedback • Subscribe or unsubscribe.

23:01, 7 November 2016 (UTC)

Notifications awry.

I have often had problems with notifications through email. When I had it set to send by email they often did not come. For ages I got none. So, I started just using the Wikipedia website and I unchecked all email notifications, which is how it is now. That was quite some time ago, but a few months back some started to trickle through in the email, about one in seven? Jed Stuart (talk) 02:56, 8 November 2016 (UTC)[reply]

Weird logging out/in

I have just got logged out several times, "automatically", after just changing to other browser window. Once I couldn't log in even with correct password (after that, in second try, login was successful).

After third logout and reloading page notification came that I am logged in centrally and it logged me in automatically. Very weird... /WB: latest Chrome, OS: Win7/ --Obsuser (talk) 13:41, 8 November 2016 (UTC)[reply]

It's been logging me out after every single edit today; it's driving me insane, especially when using Huggle. Something's definitely happening. FoCuS contribs; talk to me! 14:40, 8 November 2016 (UTC)[reply]

Looking for small tasks and mentors for Google Code-in - Got something in mind?

Hi everybody! Google Code-in (GCI) will soon take place again - a seven week long contest for 13-17 year old students to contribute to free software projects. Tasks should take an experienced contributed about two-three hours and can be of the categories Code, Documentation/Training, Outreach/Research, Quality Assurance, and User Interface/Design. Do you have an idea for a task and could you imagine mentoring that task? For example, do you have something on mind that needs documentation, research, some gadget or template issues on your "To do" list but you never had the time, and can imagine enjoying mentoring such a task to help a new contributor? If yes, please check out mw:Google Code-in 2016 and become a mentor! Thanks in advance! --AKlapper (WMF) (talk) 14:44, 8 November 2016 (UTC)[reply]

Log in problems

Hi, User:Ms Sarah Welch emailed to inform me that she is having logging in problems. "No active login attempt is in progress for your session." is coming up. Can somebody look into it?♦ Dr. Blofeld 15:44, 8 November 2016 (UTC)[reply]

Fixed. Thanks. Ms Sarah Welch (talk) 17:16, 8 November 2016 (UTC)[reply]
@Dr. Blofeld and Ms Sarah Welch: Can either of you help at Wikipedia:Help desk#Can not login? What was the fix? -- John of Reading (talk) 18:24, 8 November 2016 (UTC)[reply]
Hello Dr. Blofeld or Ms Sarah Welch, can you help me with my login problems (the same as you have today). What was the fix? 84.104.223.208 (talk) 19:29, 8 November 2016 (UTC)[reply]
Have you tried clearing your cache, per suggestions on WP:BYPASS? Then reboot? Let us know if that does or does not work. Sorry, frustrating this must be, Ms Sarah Welch (talk) 20:47, 8 November 2016 (UTC)[reply]
Thanks for your 'support' but it does not work. Can you tell me what they did to fix your login problem? 84.104.223.208 (talk) 21:00, 8 November 2016 (UTC)[reply]
Lets discuss it at one place, where you started the discussion. Hopefully, those who know better than me will join you there, to help you. Ms Sarah Welch (talk) 21:07, 8 November 2016 (UTC)[reply]

Need assist with table formatting

Hey all, if you're good with tables, can you donate a few minutes to List of highest-grossing Indian films#Highest grossing Indian franchise and film series, please? I'm having trouble making head-or-tails of the formatting...it's very confusing. It looks like some attempt was made to jury rig the column widths so that the table works on smaller devices, but the columns do not line up on my large monitor. The double-border of the table is inconsistent with the other tables. We need a Source column which should contain the references for consistency with the other tables, etc. Many thanks in advance, Cyphoidbomb (talk) 19:21, 8 November 2016 (UTC)[reply]

Just throw out all those templates and make a normal table. It is impossible to understand anything in this template mess. Ruslik_Zero 20:41, 8 November 2016 (UTC)[reply]
Yup, same thought here. Not sure who in the film project thought that Template:Highest-grossing films franchise template was a good idea, but... they were wrong :) —TheDJ (talk • contribs) 21:05, 8 November 2016 (UTC)[reply]
Betty Logan Pppery 21:18, 8 November 2016 (UTC)[reply]

How to replace text?

What I'd like to do is create a script that hides annotations in the list items in lists, without having to change the source text of articles in any way. The annotations I'm interested in start with an en dash surrounded by spaces. Like in the list items below:

  • Dendrology – involves the study and identification of economically useful tree species
  • Energy forestry – includes specifically managing for the production of energy from biomass or biofuel derived from a fast-growing species of tree or woody shrub
  • Forest ecology – studies the patterns and processes of a forest ecosystem
  • Forest economics – studies the impact of economics on forest management decisions

I found some sample code that homes in on elements with one's desired text in them, and altered it to do a test to find all list items with an en dash surrounded by spaces, and color them orange: 

$(document).ready(function(){
    $("li:contains( – )").css("background-color", "orange");
});

(I then looked around with the script active to see how well it found annotated list entries, and it does so very well. The list above, for example, turns orange.)

Now I'd like to do a test of text replacement. How can the above script
be changed with a regex to replace the " – " and all text that follows it, with nothing?

I look forward to your reply. The Transhumanist 04:18, 9 November 2016 (UTC)[reply]

@The Transhumanist: Use .css("display", "none") to hide the cells from view.
$(document).ready(function(){
    $("li:contains( – )").css("display", "none");
});
Or you can use .remove() to remove the cells from the DOM entirely.
$(document).ready(function(){
    $("li:contains( – )").remove();
});
Hope this helps. — Mr. Stradivarius ♪ talk ♪ 07:44, 9 November 2016 (UTC)[reply]
@Mr. Stradivarius: Thank you for the suggestions.
Both of those appear to make the items disappear, and will help in the future when I need to make entire items go bye bye. What I need currently is for just a part of the items to disappear. The part starting with the en dash surrounded by 2 spaces, and all text that follows that to the end of the item.
In other words, the script should make the above list look like this:
Is there a function that applies regex to elements?
I look forward to your reply. The Transhumanist 08:42, 9 November 2016 (UTC)[reply]

Are you trying to make it possible to transclude a list (that contains descriptions of items) into an outline (without those descriptions)? Whatamidoing (WMF) (talk) 15:35, 9 November 2016 (UTC)[reply]

Math coding (Hyperbolic function)

Can someone finish the job and code \csch \sech \arsinh \arcosh \artanh \arcsch \arsech \arcoth ? יהודה שמחה ולדמן (talk) 14:38, 9 November 2016 (UTC)[reply]

Infobox embed function

Why does the embed function at John Bingham, 7th Earl of Lucan create a box within a box for the "person" embed and form properly for the previous "military" embed? Is it that I am not nesting them properly? Please ping me so I see the answer. --Richard Arthur Norton (1958- ) (talk) 21:53, 9 November 2016 (UTC)[reply]

@Richard Arthur Norton (1958- ): In preview it displayed: Warning: Page using Template:Infobox person with unknown parameter "embed". I fixed it by changing embed=yes to child=yes. PrimeHunter (talk) 22:24, 9 November 2016 (UTC)[reply]
Thanks! Is there a primer on how to use embedded infoboxes? We really need a master biographical infobox that contains all biographical fields and people can just copy and paste the subset that they need for say, tennis people or baseball people. --Richard Arthur Norton (1958- ) (talk) 22:40, 9 November 2016 (UTC)[reply]
Each infobox needs to be set up individually to allow such embedded usage. Not all of them have been done; and those that have been are inconsistent. But see Template:Infobox#Embedding. --Redrose64 (talk) 23:48, 9 November 2016 (UTC)[reply]
Can you add color to the header for an embed? See: Meriwether Lewis Walker where his military info does not have a colored background for the header "Military career" the way "Personal details" does. It makes it hard for the eye to jump to the starting point of the military data. --Richard Arthur Norton (1958- ) (talk) 00:12, 10 November 2016 (UTC)[reply]
{{Infobox officeholder}} sets the |headerstyle= parameter of {{Infobox}}, essentially to the value background:lavender - this is non-standard; whereas {{Infobox military person}} doesn't set the |titlestyle= parameter of {{Infobox}}, which is standard behaviour. So for the sake of consistency, it's the coloured background that should be removed. --Redrose64 (talk) 00:37, 10 November 2016 (UTC)[reply]

Insert – — ° ′ ″ ≈ ≠ ≤ ≥ ± − × ÷ ← → · § etc. etc.

Can anyone tell me where the system message (or whatever it is) is stored, that gives shortcuts to commonly needed unicode glyphs, wiki markup, and so on while you're editing a page? I know I edited it on once a long, long time ago, but I can't find it now, and for all I know it's kept somewhere else now anyway. In case I'm no being clear, I mean the thing that currently says

– — ° ′ ″ ≈ ≠ ≤ ≥ ± − × ÷ ← → · § Sign your posts on talk pages: [[User:Iustinus|Iustinus]] ([[User talk:Iustinus|talk]]) 01:48, 10 November 2016 (UTC) Cite your sources: <ref></ref>

directly beneath where I'm writing. --Iustinus (talk) 01:48, 10 November 2016 (UTC)[reply]

Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia:Village_pump_(technical)&oldid=748736306"