Jump to content

Wikipedia:Village pump (technical): Difference between revisions

From Wikipedia, the free encyclopedia
Content deleted Content added
Line 630: Line 630:
<span class="plainlinks">'''[https://en.wikipedia.org/w/index.php?title=Wilt_the_Stilt&redirect=no Wilt the Stilt]'''</span> is a redirect with the following content:<br /><nowiki>#REDIRECT [[Wilt Chamberlain]]</nowiki><br /><nowiki>{{R from alternative name}}</nowiki><br />It categorizes to:<br />[[:Category:Redirects from alternative names]]<br /><br /><nowiki>{{R from nickname}}</nowiki> is a template redirect to:<br /><nowiki>{{R from alternative name}}</nowiki><br /><br />As such, <span class="plainlinks">'''[https://en.wikipedia.org/w/index.php?title=Wilt_the_Stilt&redirect=no Wilt the Stilt]'''</span> with the following content:<br /><nowiki>#REDIRECT [[Wilt Chamberlain]]</nowiki><br /><nowiki>{{R from nickname}}</nowiki><br />Will also categorizes to:<br />[[:Category:Redirects from alternative names]]<br /><br />Is it possible for the template redirect to influence categorization such that the second page could categorize to:<br />[[:Category:Redirects from alternative names]] and<br />[[:Category:Nicknames]] as well?--[[User:John Cline|John Cline]] ([[User talk:John Cline#Top|talk]]) 10:26, 14 May 2018 (UTC)
<span class="plainlinks">'''[https://en.wikipedia.org/w/index.php?title=Wilt_the_Stilt&redirect=no Wilt the Stilt]'''</span> is a redirect with the following content:<br /><nowiki>#REDIRECT [[Wilt Chamberlain]]</nowiki><br /><nowiki>{{R from alternative name}}</nowiki><br />It categorizes to:<br />[[:Category:Redirects from alternative names]]<br /><br /><nowiki>{{R from nickname}}</nowiki> is a template redirect to:<br /><nowiki>{{R from alternative name}}</nowiki><br /><br />As such, <span class="plainlinks">'''[https://en.wikipedia.org/w/index.php?title=Wilt_the_Stilt&redirect=no Wilt the Stilt]'''</span> with the following content:<br /><nowiki>#REDIRECT [[Wilt Chamberlain]]</nowiki><br /><nowiki>{{R from nickname}}</nowiki><br />Will also categorizes to:<br />[[:Category:Redirects from alternative names]]<br /><br />Is it possible for the template redirect to influence categorization such that the second page could categorize to:<br />[[:Category:Redirects from alternative names]] and<br />[[:Category:Nicknames]] as well?--[[User:John Cline|John Cline]] ([[User talk:John Cline#Top|talk]]) 10:26, 14 May 2018 (UTC)
:{{tl|R from nickname}} is currently a redirect to {{tl|R from alternative name}}. As long as that is the case, it is not possible to behave differently. {{tl|R from nickname}} could be changed to get its own code and do anything, e.g. add a new [[:Category:Redirects from nicknames]]. It shouldn't add [[:Category:Nicknames]]. It includes many articles, and redirect templates should not automatically add article categories. Individual redirects can be categorized. See [[Wikipedia:Categorizing redirects]]. [[:Category:Nicknames]] seems to be for articles and disambiguation pages ''about'' nicknames so I wouldn't add the category to a redirect to a specific biography. [[User:PrimeHunter|PrimeHunter]] ([[User talk:PrimeHunter|talk]]) 11:32, 14 May 2018 (UTC)
:{{tl|R from nickname}} is currently a redirect to {{tl|R from alternative name}}. As long as that is the case, it is not possible to behave differently. {{tl|R from nickname}} could be changed to get its own code and do anything, e.g. add a new [[:Category:Redirects from nicknames]]. It shouldn't add [[:Category:Nicknames]]. It includes many articles, and redirect templates should not automatically add article categories. Individual redirects can be categorized. See [[Wikipedia:Categorizing redirects]]. [[:Category:Nicknames]] seems to be for articles and disambiguation pages ''about'' nicknames so I wouldn't add the category to a redirect to a specific biography. [[User:PrimeHunter|PrimeHunter]] ([[User talk:PrimeHunter|talk]]) 11:32, 14 May 2018 (UTC)
::Thank you, I understand.--[[User:John Cline|John Cline]] ([[User talk:John Cline#Top|talk]]) 12:11, 14 May 2018 (UTC)

Revision as of 12:11, 14 May 2018

 Policy Technical Proposals Idea lab WMF Miscellaneous 
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. If you want to report a JavaScript error, please, follow this guideline. Questions about MediaWiki in general should be posted at the MediaWiki support desk.


Wikipedia bug?

black squares

Is it just my computer or is there a bug on Wikipedia. I've been noticing black squares appearing now & then on articles during both editing & viewing. GoodDay (talk) 00:54, 27 April 2018 (UTC)[reply]

GoodDay, next time you see one like that, please link the article here, along with where the black squares are appearing. By any chance, have you been reading any articles that deal with foreign people or places that use a non-European language? Mathglot (talk) 01:09, 27 April 2018 (UTC)[reply]
They appear randomly & disappear randomly. Sometimes flickering into different places on the same article, each time I press a key. GoodDay (talk) 01:41, 27 April 2018 (UTC)[reply]
GoodDay, please try to get a screenshot of that, if possible. In the meantime, what's your browser and OS? Maybe people with a similar setup could try watching out for it. Whatamidoing (WMF) (talk) 02:13, 27 April 2018 (UTC)[reply]
I don't understand those terminologies. If anyone else has noticed these things, then I'll know it's Wikipedia related, not just my computer. GoodDay (talk) 02:15, 27 April 2018 (UTC)[reply]
See Wikipedia:Screenshots of Wikipedia. The most popular web browsers are Chrome, Firefox, Safari, Internet Explorer, and Edge. OS means operating system, e.g. Windows 10. PrimeHunter (talk) 02:31, 27 April 2018 (UTC)[reply]
This is also happening to me on bn.wiki (see right). Browser google chrome, latest version, windows 8. This appear randomly. --আফতাব (talk) 20:33, 27 April 2018 (UTC)[reply]
It appears to be a Chrome problem also affecting other websites. https://bugs.chromium.org/p/chromium/issues/list?q=black+boxes shows several reports this week. PrimeHunter (talk) 21:40, 27 April 2018 (UTC)[reply]
I do see such boxes on my Firefox browser sometimes, also on other websites. Usually if I have a very large amount of tabs open. Jo-Jo Eumerus (talk, contributions) 09:19, 28 April 2018 (UTC)[reply]

Those black cubes are still floating around. GoodDay (talk) 21:55, 5 May 2018 (UTC)[reply]

I've had something like this intermittently for years. See Wikipedia:Village pump (technical)/Archive 144#Interminable flashing from 2016, for example. Nyttend (talk) 22:12, 13 May 2018 (UTC)[reply]
PS, GoodDay, I meant to ping you at first. Nyttend (talk) 22:12, 13 May 2018 (UTC)[reply]
Ah hah. GoodDay (talk) 22:38, 13 May 2018 (UTC)[reply]

"a failed attempt to log in to your account" alerts

two-factor authorization

As I sit here reading a notification that there has been a failed attempt to log in to my account, I find myself wondering how it is the site I spend the most time on is the only site I spent much time on that doesn't have multi-factor authorization. I'm quite sure this has come up before, but apparently I don't know a good search term for the VP archives. Now, I have a pretty strong password, but it's unsettling nonetheless, as it most certainly was not me who tried to log in. — Rhododendrites talk \\ 13:34, 3 May 2018 (UTC)[reply]

@Rhododendrites: 2FA is in pilot right now, see meta:Help:Two-factor authentication for information on the implementation. There are many issues preventing large-scale rollout, see the phabricator workboard for an overview of the work to be done. — xaosflux Talk 13:40, 3 May 2018 (UTC)[reply]
@Rhododendrites: See also Wikipedia:Simple 2FA. --Izno (talk) 13:41, 3 May 2018 (UTC)[reply]
Thanks. As I'm not an admin, it doesn't look like there's anything I can do at this point. Reading over the meta page, however, I wonder what reason there is for this to be seemingly more involved than other systems. Why not just use the texted code like everyone else? I'm sure that would require a little investment from WMF, but this sure seems like a good use of its cash... — Rhododendrites talk \\ 13:50, 3 May 2018 (UTC)[reply]
Texted codes are not all that secure, so the privacy implications of storing people's phone numbers and the expense of an SMS gateway don't seem to be worth it. Others, such as Google and NIST, are moving away from SMS-based 2FA. Anomie 12:37, 4 May 2018 (UTC)[reply]
Also, though it's not something typically offered with this sort of mechanism, it might also be useful to determine the IP that tried to log in. In this case, the notification came shortly after my first edit of the day, this revert. With an unregistered user, there's not a whole lot of point except as an additional reason to block, but it makes me think that registered users might do it, too, in which case that information would be important if confirmed via checkuser. Getting into a whole can of worms there, I know. — Rhododendrites talk \\ 13:56, 3 May 2018 (UTC)[reply]
@Rhododendrites: if you really really really want it, you can request tester access at meta SRGP - there is no good way to manage these and you could lose access to your account forever. Some recovery options are being discussed on some of the tasks I linked to in the phab workboard. As far as the logon notifications, much discussion is in place regarding improving these and including more information that may be useful (see phab:T174388). — xaosflux Talk 14:24, 3 May 2018 (UTC)[reply]
Rhododendrites, I just also got one of those (for the first time in 11 years). Does WP record the IP of these attempts? If so, is this information available for discussion at a venue like this? Why should anyone try to do this when it's pretty obvious I'm logged in and editing? Perhaps it's some kind of stupid bot?| Thanks. Martinevans123 (talk) 14:51, 3 May 2018 (UTC)[reply]
This type of notification has only been available for about a year. Right now it is informational only, bots try to hack accounts all the time. — xaosflux Talk 15:09, 3 May 2018 (UTC)[reply]
Do they try to hack my account all the time? Or have they tried all the time in the last year? Or are you saying that bots won't trigger this warning? Thanks. Martinevans123 (talk) 15:23, 3 May 2018 (UTC)[reply]
Note, the bots I refer to are not "wikipedia bots" - these are malicious parties on the internet; and they try to hack lots of accounts all the time. I got a failed logon notification myself today. For the most part, you don't need to do anything on these - if you find the notifications annoying, you can turn them off in preferences. — xaosflux Talk 15:34, 3 May 2018 (UTC)[reply]
Understood. From your last reply it seems like you are assuming the notifications that all three of us got were triggered my malicious bots. I'm quite happy that I don't need to do anything (although if I start to get one every day, I might do something anyway). I'm just wondering how often other folks get these notifications, whether IP's are recorded and if anyone does anything as a result. Sorry is this has careered off topic. Thanks. Martinevans123 (talk) 15:45, 3 May 2018 (UTC)[reply]
A lot of other people have received this notification, it doesn't seem to stop, even on frwiki we receive a notification from enwiki. See [1]. Lofhi (talk) 16:09, 3 May 2018 (UTC)[reply]
That's a very interesting graph! That looks quite cataclysmic. Can the WMF issue any information on where these attempts are being made? Martinevans123 (talk) 19:11, 3 May 2018 (UTC)[reply]
Unless any of the attempts are succeeding, I'm guessing someone's re-using the list that got Jimbo and some other admins hacked last year. If you've changed your password in the last few months then you should be safe. Ian.thomson (talk) 19:19, 3 May 2018 (UTC)[reply]
Wait, I see that there was a new data dump in February that seems to have built off of the one from the Anti Public Combo List and other breaches. I haven't grabbed it yet, and hopefully I'm right that if no accounts have been compromised so far, then it's probably older passwords. It'd be a good idea to check haveibeenpwned.com. Ian.thomson (talk) 20:02, 3 May 2018 (UTC)[reply]
Yep, I got one (my first) about a half-hour ago, too. Does the Foundation take note of bursts of failed login attempts? TenOfAllTrades(talk) 16:31, 3 May 2018 (UTC)[reply]
Similarly to TenOfAllTrades, I've just got my first such notification ever where it hasn't been me on a different device. Is this sort of rate common or is something going on? Red Fiona (talk) 17:37, 3 May 2018 (UTC)[reply]
Same here, got my first today. DuncanHill (talk) 18:13, 3 May 2018 (UTC)[reply]

But hey, it's World's password day, so... --Edgars2007 (talk/contribs) 18:46, 3 May 2018 (UTC)[reply]

I got one of these. I just ignored it as I have 2FA. Doug Weller talk 19:12, 3 May 2018 (UTC)[reply]
Ah thank goodness. So it's not Vladimir after all. It's just Big Brother reminding us to choose a better one. Martinevans123 (talk) 19:17, 3 May 2018 (UTC)[reply]
Add me to the list of people that got their first "failed login" notification today. --Ahecht (TALK
PAGE
) 21:13, 3 May 2018 (UTC)[reply]
Add me to the "failed login attempts" as well. Multiple attempts 8 hours ago.  Aloha27  talk  14:05, 10 May 2018 (UTC)[reply]

Question: if I simply mistype my password and fail to log-in, will I get another one of these messages? Or does it look at the device used as well? Thanks. Martinevans123 (talk) 10:54, 4 May 2018 (UTC)[reply]

@Martinevans123: if you are using a "known device" there has to be many failed attempts (5) before the notification is generated. — xaosflux Talk 12:54, 4 May 2018 (UTC)[reply]
Thank you for that clarification. I suppose some users may now be tempted to try this on their own accounts just to test the system. Is there any reason they should not? Martinevans123 (talk) 13:11, 4 May 2018 (UTC)[reply]

password resets failing

This seems to be some sort of mass password reset troll. At threads at both WP:AN and WP:TEAHOUSE users are reporting that after receiving the notification they are unable to actually reset their password. One user report trying 17 times and failing. Software glitch? Beeblebrox (talk) 18:39, 3 May 2018 (UTC)[reply]

@Beeblebrox: They may be getting the message "A password reset email has already been sent, within the last 24 hours. To prevent abuse, only one password reset email will be sent per 24 hours." If so, this is expected as it has a throttle on it, there is some discussion of changing this at phab:T181070. — xaosflux Talk 18:53, 3 May 2018 (UTC)[reply]
I can verify that I have not received that message as I don't have email. The only messages I get when I try are the standard "Incorrect username or password entered. Please try again." except once when it told me to wait for five minutes. -- Millionsandbillions (talk) 19:08, 3 May 2018 (UTC)[reply]
I really hope everyone changed their email password at least once (in addition to their WP password) after that incident where Jimbo got hacked. Ian.thomson (talk) 19:21, 3 May 2018 (UTC)[reply]
Just changing a password makes no difference, however many times you do it. You need to change it to something stronger? 86.187.160.63 (talk) 07:50, 4 May 2018 (UTC)[reply]
If whoever is doing this is using a list of passwords from some other sites (which is generally how accounts get compromised here), then yes, changing the password is going to help. If they're going after as many accounts as possible, and most of us are only getting one failed login attempt, then it's save to assume they're not taking the time to brute force every possible combination (which is where password strength matters). Ian.thomson (talk) 14:48, 4 May 2018 (UTC)[reply]

Illicit signin

I just got a message saying someone has tried to sign in to my acc from an unrecognized device. So I changed my password. Thing is - NOBODY can access my stuff - except maybe someone who is on my network. I share my hub with some neighbours; it would be immensely helpful to me if sysops could tell me if the interloper was "coming from inside the house". Sysops know my address - was the interloper from the same address? MarkDask 19:27, 3 May 2018 (UTC)[reply]

I got the same notification today, which was first time ever. So either someone is trying all usernames with "123456", "qwerty", "password", and other common passwords ow someone is testing new notification features. --Jarekt (talk) 19:39, 3 May 2018 (UTC)[reply]
See Wikipedia:Administrators' noticeboard#Please help- who tried to break into my account? and #two-factor authorization just above. Amaury (talk | contribs) 19:44, 3 May 2018 (UTC)[reply]
Have you checked haveibeenpwned.com? Ian.thomson (talk) 20:02, 3 May 2018 (UTC)[reply]
Received the same message today, also for the first time. Is this perhaps a large-scale attack? GregorB (talk) 21:00, 3 May 2018 (UTC)[reply]
@GregorB: I guess they don't like me as I haven't received any such notifications so far today, but yes, it appears to be a widespread and planned attack, not just one of those every now and then things. Amaury (talk | contribs) 21:02, 3 May 2018 (UTC)[reply]

Thanks all. I think two factor authorization on here would be a step too far. That said - I dont like being told someone might be trying to hack my space so I've dissed all my hangers-on. (Sigh) - I've taken back control of my hub - (screaming in the background). MarkDask

I also woke up to three failed log-in attempts. Odd. Anarchyte (work | talk) 22:46, 3 May 2018 (UTC)[reply]
There's a thread on this happening at WP:AN I got a notice of a failed attempt to hack my account. Others are getting it also. I already have the 2-factor enabled, but I also changed my password after the failed attempt.— Maile (talk) 00:01, 4 May 2018 (UTC)[reply]
  • From what I’ve been told, there were 60,000-70,000 failed logins per hour for several hours. So, tens of thousands of users got these messages, but they do not represent an actual security breach, just some troll with a little bit of programming knowledge that set up a trollbot.
That being said, it’s worth it for everyone to review WP:STRONGPASS and WP:SECURITY Beeblebrox (talk) 00:15, 4 May 2018 (UTC)[reply]
According to this graph, the attack is still underway. Anyone who has a weak password needs to change it immediately please. — Diannaa 🍁 (talk) 00:51, 4 May 2018 (UTC)[reply]

Attempted login breach on my account

Good morning. I wonder if you could help me out please? Approximately 9 hours ago (9pm UTC, May 3rd) an attempt on this Wikipedia was made to login to my account. The wrong password was used, thankfully, and I got flagged. Does Wikipedia record logins and where they originate from please, and if so, is there anything you can do? Obviously you can't tell me who it was, but I'm curious as to why someone would attempt to access my account. Thanks. Dane|Geld 05:49, 4 May 2018 (UTC)[reply]

I received the same message regarding a login attempt within the same time frame.--John Cline (talk) 05:54, 4 May 2018 (UTC)[reply]
Yes, see above (especially the phab ticket). Killiondude (talk) 06:01, 4 May 2018 (UTC)[reply]
I and my fellow mates on Marathi wikipedia too have got this notification from English Wikipedia. I think we must also have a two factor authentication like gmail on wikipedia. --✝iѵɛɳ२२४०†ลℓк †๏ мэ 13:37, 4 May 2018 (UTC)[reply]

Detecting suspicious failed logins

Is there anything in the software that would notice if one IP address were making a lot of failed login attempts to different accounts? (I got an email that here had been an attempt to log into my account from a new computer and to make sure I had a strong password. A lot of people have commented on getting this message. If it's part of a lot of attempts has noticing it been automated?) RJFJR (talk) 14:12, 4 May 2018 (UTC)[reply]

RJFJR Moving this thread up to appropriate section. See all above, been happenig to everybody.Two sections up, Diannaa has said the attack is still happening, and included this link: this graph — Maile (talk) 14:33, 4 May 2018 (UTC)[reply]
Are they all coming from the same IP? Would blocking the IP help? RJFJR (talk) 19:34, 4 May 2018 (UTC)[reply]
My understanding is that technically this isn't possible. It's not possible to see the burglar until he gets through the front door with his counterfeit key? Am surprised though. Martinevans123 (talk) 08:28, 5 May 2018 (UTC)[reply]

I changed my password, and now I can't log in

Resolved

Hey, i received the same message two days ago and changed my password. The next day i was logged out and my new and old password don't work. I requested a new password to my actual mail adress but that was not the one, which was listed under "preferences" and of course none came. The one adress listed is dead. Any chance to get my old account (Kante4) back? Kante44 (talk) 08:52, 5 May 2018 (UTC)[reply]

Yup, all good now. Kante4 (talk) 20:12, 5 May 2018 (UTC)[reply]

login log check possible?

Is there a log of my login attempts, a log I can see? Or can someone please check that log for me? I was notified that about 5 days ago a login attempt failed. Then I was notified by an email stamped Sunday for 8:01p that someone, probably I, had "recently" logged in. I did log in Saturday night from a network I haven't used in a while and logged off some time around 2 a.m. Sunday, but not at the time the email was sent. But the email says "recently", so maybe you meant my Saturday night login. My Special:Contributions end with a contribution I made; no one contributed under my username since then. I since changed my already-secure password to a still-more secure one. Is there a way to find out the time of the last login under my username last Sunday (if any) (not a login for Monday or since)? Thank you. Nick Levinson (talk) 21:11, 8 May 2018 (UTC) (Correction: Forgot the tildes.)[reply]

Short answer, no. But @Nick Levinson: there is a task being considered to add more information to this, you can follow it here: phab:T174388. — xaosflux Talk 21:16, 8 May 2018 (UTC)[reply]

Linking articles in different languages with different names

I noticed that the English-language article on the Finnish architects Reima and Raili Pietilä was not linked to its respective articles in 10 other languages (incl. Finnish, Korean, etc.). However, in ALL the other languages the site is called just "Reima Pietilä". Reima and Raili Pietilä were an architect couple who nearly always worked together. But when I attempted to add the English-language version to the other language versions, it failed to so. I suppose I will have to change the name of the English-language site to be the same as the others in order for the language links to work? This bothers me, because in including both Reima and Raili Pietilä as an "architect couple" (cf. Charles and Ray Eames) the English-language page comes across as more enlightened.TTKK (talk) 15:52, 4 May 2018 (UTC)[reply]

Three things you could do: The first is to manually add old school local links manually to the article. See WP:ILL § Local links for more info and nb the spirit of the second and third points. However, this will allow linking only to the Reima or the Raili article. The second would be to merge wikidata:Q17496278 into wikidata:Q2582922, but then you would lose "duo"/"married couple" attributes in Wikidata. The third, as you mentioned, would be to rename/split this article like five other Wikipedias do. This would allow linking to both the Reima and the Raili articles.
I will conduct the first option on the English page now. Review and feel free to edit revert. I also look forward to hearing from others who might know more about this topic. —  AjaxSmack  00:30, 5 May 2018 (UTC)[reply]
No, the second option is bad, very bad. Those are different topics, so they shouldn't be merged. About problem in general. I would suggest creating module (that is, copying from other Wikipedia - don't remmeber which), which takes iws from Wikidata item, that you pass as parameter. Pro - then iws are updating, not as in old-iws mode. --Edgars2007 (talk/contribs) 04:06, 5 May 2018 (UTC)[reply]
How do you do that?  AjaxSmack  05:11, 5 May 2018 (UTC)[reply]
Get some Lua guy, that writes you a module to do that (it shouldn't be very hard, but I'm not a Lua programmer). Or find the module on some other Wikipedia (don't remember which). --Edgars2007 (talk/contribs) 06:03, 5 May 2018 (UTC)[reply]
Thanks. It looks like this was copied from Swedish Wikipedia and can do it: Template:Interwiki extra. Not ideal as it cannot allow for links to more than one article (i.e., both Reima and Raili), but better than no IWs at all.  AjaxSmack  05:11, 6 May 2018 (UTC)[reply]

How to find intitle matches for redirects in a category?

If, for instance, I search for intitle:Applied" with incategory:"Physics journals", I get the following. All is cool, that works.


However, if I search for intitle:Appl." with incategory:"Redirects from ISO 4", I get this. Which is no results at all.

How do I make this work? Can it be done?Headbomb {t · c · p · b} 01:06, 5 May 2018 (UTC)[reply]

@Headbomb: I'm fairly certain that search will not return redirects when using a keyword like incategory (this search doesn't work). The documentation at Mediawiki wiki indicates that this is definitely true for some parameters, but doesn't say so about incategory. I did use WP:PETSCAN to get all the pages and then exported it to CSV (I couldn't get petscan's title regex to work), where I filtered the page title column in LibreOffice Calc. That might be your best shot. --Izno (talk) 05:20, 5 May 2018 (UTC)[reply]
@Izno: The idea would be to have something that can search for this live on the wiki, that I can put in a search box on Category:Redirects from ISO 4, similar to (but not quite the same as) the search box in Category:Infobox templates. The CSV file would be mostly pointless here. I tried insource:"R from ISO 4" or hastemplate: "R from ISO 4" to find the templates used to populate the category instead, but that ignores redirects too. Headbomb {t · c · p · b} 05:27, 5 May 2018 (UTC)[reply]
Ah, then you might be SOL. You can try filing a phab task to get the attention of the search guys to see if they know a way to do that. --Izno (talk) 05:39, 5 May 2018 (UTC)[reply]
Yep, you found a rather complex bug. I filed a task for the search folks to take a look at. CKoerner (WMF) (talk) 16:48, 7 May 2018 (UTC)[reply]

Multiple infobox

Hello, what does Wikipedia says about having multiple Infoboxes for separate sections of same biography article? Capankajsmilyo(Talk | Infobox assistance) 02:28, 5 May 2018 (UTC)[reply]

What you wanted to hear about this topic on this page? From technical POV, it is possible. From non-technical POV (which of course would be better suited for other pages), I think it's OK. There are many examples on Wikipedia, person who is WWE and actor, e.g. --Edgars2007 (talk/contribs) 03:59, 5 May 2018 (UTC)[reply]
So in such a case, should |module= be used or two separate Infoboxes? Capankajsmilyo(Talk | Infobox assistance) 04:03, 6 May 2018 (UTC)[reply]

Test for the non-existence of a (possibly empty) third positional parameter?

Resolved

In a template, how do I test for the non-existence of a (possibly empty) third positional parameter?

In this case the template takes only a max of 2 parameters and third indicates a possible error even if empty. So I want to treat any of these as an error

  1. {{Mytemplate|foo||bar}}
  2. {{Mytemplate|foo||bar}}
  3. {{Mytemplate|foo|bar|}}

... but be happy with

  1. {{Mytemplate|foo|bar}}
  2. {{Mytemplate|foo|}}
  3. {{Mytemplate|foo}}
  4. {{Mytemplate||bar}}

I thought I coud take some test which would always throw an error if {{{3}}} was not supplied and use that "error" as my sign of health.

But now I can't find any such test.

I think I have done this before, but my brain seems mushy and I can't recall how I did it.

Any suggestions? (apart from Lua, which feel like overkill) --BrownHairedGirl (talk) • (contribs) 03:11, 5 May 2018 (UTC)[reply]

Something like this:
{{#if:{{{3}}}|<span class="error">Oh my, you supplied third parameter. Which is the worst thing you could do :)</span>}}
Pay attention to {{{3}}}. If it was {{{3|}}}, then the error would show up only if third param would be non-empty. --Edgars2007 (talk/contribs) 04:03, 5 May 2018 (UTC)[reply]
Thanks, @Edgars2007. I have been trying any options including variants of your suggestion, at User:BrownHairedGirl/sandbox8. The problem is that if {{{3}}} is not supplied its value is "{{{3}}}". See the current version. --BrownHairedGirl (talk) • (contribs) 04:47, 5 May 2018 (UTC)[reply]
But if don't want to show third parameter (only test it and show warning), then it is OK... --Edgars2007 (talk/contribs) 04:57, 5 May 2018 (UTC)[reply]
@Edgars2007: No it isn't OK. The test gives a false positive, so it would trigger an warning when there is no problem. --BrownHairedGirl (talk) • (contribs) 05:25, 5 May 2018 (UTC)[reply]
With the example that I gave you, all of those testcases you mentioned in tour initial post, would pass. If you want to show content of third parameter (if it's not empty) (that you didn't mention in initial post), then you can use additionally: {{#if:{{{3|}}}|Here are contents of third parameter: {{{3|}}}}} What is your use-case? Some real-life problem/template would be good to show. --Edgars2007 (talk/contribs) 05:58, 5 May 2018 (UTC)[reply]
Thanks for trying, Edgars2007 ... but I have already demonstrated[2] how the test you propose gives a false positive.
I will see what other editors suggest. --BrownHairedGirl (talk) • (contribs) 06:32, 5 May 2018 (UTC)[reply]

@BrownHairedGirl: Without using Lua, best solution is probably {{#ifeq:{{{3|¬}}}|¬||<span class="error">Hello, I'm an error message.</span>}}. Jc86035's alternate account (talk) 06:59, 5 May 2018 (UTC)[reply]

Bless you @Jc86035 (1). That does it, as shown in this test: no false positives & no false negatives. --BrownHairedGirl (talk) • (contribs) 07:26, 5 May 2018 (UTC)[reply]
If I understand the original question, Module:Check for unknown parameters may be what you want. You can put {{#invoke:Check for unknown parameters|check|unknown={{main other|[[Category:Pages using TKTKTK with unknown parameters|_VALUE_{{PAGENAME}}]]}}|preview=Page using [[Template:TKTKTK]] with unknown parameter "_VALUE_"|ignoreblank=n| 1 | 2 }} at the end of the template code to categorize any such articles. – Jonesey95 (talk) 14:21, 5 May 2018 (UTC)[reply]
Many thanks @Jonesey95. I used that instead of Jc86035's solufion because Module:Check for unknown parameters covers more cases, such as a named parameter.
See it in use at Template:Category pair where it populates Category:Pages using Template:Category pair with more than 2 parameters. --BrownHairedGirl (talk) • (contribs) 04:02, 6 May 2018 (UTC)[reply]

Tools to comment/uncomment categories?

When you move something from article space to draft/user space, you want to disable the categories with ":" prefixes. And do the reverse in the other direction. Is there some nice bit of automation which makes this easy? -- RoySmith (talk) 17:41, 5 May 2018 (UTC)[reply]

No clue, would just like to say that I'm interested as well.Naraht (talk) 19:17, 5 May 2018 (UTC)[reply]
Erm, a find/replace? "[[Category"->"[[:Category", or "[[:Category"->"[[Category". Seraphimblade Talk to me 19:18, 5 May 2018 (UTC)[reply]
A category with sortkey like [[:Category:Words|X]] would display as X. I have made {{Draft categories}} which avoids this issue, permits lowercase c in "category:", imitates how categories are displayed, and actually adds the categories if the page is in mainspace. Example:
{{Draft categories|
[[Category:Words|X]]
[[category:Animals]]
}}
produces:
Maybe the imitation is too good and should give an indication that the page is not actually in the categories. PrimeHunter (talk) 23:00, 5 May 2018 (UTC)[reply]
I've added a sentence about the categories not yet implying, feel free to improve my wording, and more discussion about the template should probably go to the template's talk page. rchard2scout (talk) 12:12, 6 May 2018 (UTC)[reply]

Autorollbacker userright

What would you think of creating an autorollbacker user right that would automatically revert one's edits for a given time? Maybe it could be limited to rollbacking only non-autoconfirmed users and IPs to prevent abuse. If there is consensus to create this user right, I might start an RfC on it. Your suggestions would be welcome. Best, L293D ( • ) 16:14, 6 May 2018 (UTC)[reply]

You want to create a userright that automatically reverts any edit a user with that right makes? Am I understanding you correctly? ~ Amory (utc) 16:36, 6 May 2018 (UTC)[reply]
No. The right I am proposing would enable a trusted user to automatically rollbacke persistent vandals, so that they don't have to refresh their contribs every minute and revert. (To save time to CVU users) L293D ( • ) 16:38, 6 May 2018 (UTC)[reply]
(edit conflict)Ah okay, well that sounds like a block by another name. WP:AIV is the way to go. ~ Amory (utc) 16:47, 6 May 2018 (UTC)[reply]
No, persistent disrupting users should be blocked, not continuously reverted - and this sounds like you want it to run server side - to continuously monitor someones contributions and undo them. When all you have is rollback...everything is rollback?xaosflux Talk 16:43, 6 May 2018 (UTC)[reply]
@L293D: Re "server side" means this isn't something that can be done today, and either you would need to change mediawiki core, or write and have us adopt an extension for it; with blocking available I don't see either of these methods getting adopted. If you were thinking about something else on how the mechanics would work, please elaborate. — xaosflux Talk 16:51, 6 May 2018 (UTC)[reply]
Well, the idea was to rollback them until they get blocked. Sometimes it takes half an hour, as was the case with Lachlan the editor yesterday. L293D ( • ) 17:03, 6 May 2018 (UTC)[reply]
That does happen; no comment on the wiseness of doing so but one could have a script to do that... Galobtter (pingó mió) 17:06, 6 May 2018 (UTC)[reply]
@L293D: I think there are 2 better options: (1) become an admin, (2) revisit that "vandal fighter" type access with includes rollback, block, delete (but not undelete) - that gets opposed with "be an admin" arguments most of the time.. — xaosflux Talk 17:16, 6 May 2018 (UTC)[reply]
Well, for option 1, I have no chance of passing RfA anytime soon, but the "Vandal Fighter" right might be worth another RfC. If the proposal fails, then it would be nice if someone could write me a script that would rollback automatically, since I have almost no knowledge of Java. L293D ( • ) 17:25, 6 May 2018 (UTC)[reply]

Note: I left a request for an autorollbacking script at Wikipedia:User scripts/Requests. L293D ( • ) 19:16, 6 May 2018 (UTC)[reply]

Please do not edit war. Not even with a troll. And certainly do not use an automaton to edit war. Report to WP:AIV or, for more urgent cases, also report at WP:ANI. Johnuniq (talk) 23:11, 6 May 2018 (UTC)[reply]
I said I would be used against vandals while they await blocking. But as you said, the tool might be abused to edit war, so this is why I suggested that the script would accept autorollbacking only users who are reported to AIV. L293D ( • ) 13:45, 7 May 2018 (UTC)[reply]
Any client side scripts can be fully controlled by clients - this is just not technically feasible to work how you describe it, plus you shouldn't revert war with anyone. — xaosflux Talk 14:00, 7 May 2018 (UTC)[reply]

Recent change breaks scroll in p-cactions menu

A few days ago I added #p-cactions {overflow:scroll} to my Modern.css, as there are times when the links overflow the screen and I'd like to access them. It worked just fine, up to and including this morning (7:30AM EDT). I came back on around 11:30 EDT, and now there is a white bar obscuring most of the p-cactions menu bar. What changed in the intervening four hours? Adding that css code is sufficient to add the bar, and it happens here in safemode as well as other WMF wikis. Theoretically it affects vector, although obviously the dropdown menu means it just affects the appearance. It happens regardless of being logged-in or logged-out, and in Firefox, Chrome, and Safari. I first thought it was changed in the new software, but Special:Version says mw:MediaWiki_1.32/wmf.2 was installed days ago. Can't for the life of me track down what changed, figured I'd check here before going to phabricator. ~ Amory (utc) 16:34, 6 May 2018 (UTC)[reply]

I guess it's the personal navigation links from userpage to log out you want full access to in Modern. They wrap in Vector. I don't know how to get scroll to work but can you live with a smaller font instead, e.g. 70%:
#p-personal .pBody {font-size: 70%}
PrimeHunter (talk) 18:24, 6 May 2018 (UTC)[reply]
No, it's the #p-cactions menu (edit, history, delete, twinkle, etc.) It's the x-axis scroll bar blocking things; using overflow-y: scroll solves the issue. overflow: auto also works. I just can't figure out why it changed. ~ Amory (utc) 18:29, 6 May 2018 (UTC)[reply]

Page translation

Could someone please walk me through page translation? I made a small change to a template and now Wikimedia is trying to walk me through some byzantine translation process even though I didn't change anything translation related, and I even clicked "do not invalidate translations". Magog the Ogre (tc) 18:44, 6 May 2018 (UTC)[reply]

@Magog the Ogre: you should ask over at: commons:Commons:Translators' noticeboard. — xaosflux Talk 19:31, 6 May 2018 (UTC)[reply]
In short, you edit the main template, and mark it for translation - you should be done now. Normally a script/bot updates the other languages. Is there a specific one not working? — xaosflux Talk 19:32, 6 May 2018 (UTC)[reply]

"Page previews" doesn't show image

As explained here, I can't see S.L. Benfica's logo when hovering over the link S.L. Benfica. Can you confirm this issue? SLBedit (talk) 22:12, 6 May 2018 (UTC)[reply]

I can confirm it. mw:Page Previews (previously called HoverCards and not the same feature as Wikipedia:Tools/Navigation popups) omits displaying an image for lots of articles which have an apparently suited image, for example most clubs below the top-5 at 2016–17 Primeira Liga#League table. I don't know its algorithm. It doesn't have a ban on non-free images. It's possible it will only show images designated as the page image by mw:Extension:PageImages. Page information for S.L. Benfica shows the logo is the page image so this doesn't explain why it's omitted. Limited testing indicates that Page Previews may prefer larger images and svg images. PrimeHunter (talk) 23:32, 6 May 2018 (UTC)[reply]
Yes, the PageImages extension provides the image for page previews. mw:Extension:PageImages#How are images scored? indicates the scoring algorithm. It's probably being discarded in this case because of its aspect ratio. --Izno (talk) 01:27, 7 May 2018 (UTC)[reply]
If it wasn't picked by PageImages then the ratio might explain it but File:Sport Lisboa e Benfica Logo Reduced res.png is picked by PageImages and then rejected by Page Previews. Can that also be due to the ratio? It's 300×300 png. Page Previews displays the 316×316 File:Associação Académica de Coimbra logo.svg for Associação Académica de Coimbra – O.A.F., but not the same size File:Atletico Sport Clube Reguengos.png for Atlético S.C. Both are picked by PageImages. Observations like this made me suspect svg is preferred but I have limited data and it's not mentioned in the documentation. PrimeHunter (talk) 10:45, 7 May 2018 (UTC)[reply]
@SLBedit, PrimeHunter, and Izno: Per English Wikipedia community request, PagePreviews won't ever show an image categorised as non-free. The images you highlight are all added to said category through {{Non-free logo}}; if you think the community erred in this request, I'd recommend an RfC, but I imagine fellow volunteers will disagree. :-) Jdforrester (WMF) (talk) 19:53, 7 May 2018 (UTC)[reply]
As I mentioned, Page Previews displays File:Associação Académica de Coimbra logo.svg but not File:Atletico Sport Clube Reguengos.png. Both use {{Non-free logo}} and have the same non-free categories, except the displayed image also being in Category:Wikipedia non-free files with redirect backlink but that doesn't seem relevant. PrimeHunter (talk) 20:05, 7 May 2018 (UTC)[reply]
Most club logos are non-free, so your explanation doesn't make sense to me. SLBedit (talk) 21:10, 7 May 2018 (UTC)[reply]
@Jdforrester (WMF), SLBedit, PrimeHunter, and Izno: the information about PagePreviews excluding non-free images was mistaken. The timeline is a bit complicated. Initially the function PageImage would return any image, free or non-free. That resulted in non-free images appearing in inappropriate places. So PageImages was modified to exclude non-free images. Later I suggested to the PagePreview manager that non-free was OK in previews - previews appropriately tie the images to significant related educational text. I opened the RFC Non-free_content/Hovercards, with consensus and an update to Non-free_content#Exemptions policy. Phab T131105 enhanced the PageImage function to default to free images only, and to include non-free images on request. Finally, per Phab T147317, PagePreviews were updated to include non-free images. Alsee (talk) 23:16, 11 May 2018 (UTC)[reply]

Gadget issue

Now, I don't know if it's just because I'm using the black background with green text or what, but I have "Add two new dropdown boxes below the edit summary box with some useful default summaries" selected in gadgets, and the text is very light, so it's hard to see against the white background the same goes with the upper bar at the top of the edit box. screenshot (talk page stalker) CrashUnderride 22:58, 6 May 2018 (UTC)[reply]

AdvancedSearch

Birgit Müller (WMDE) 14:45, 7 May 2018 (UTC)[reply]

Looks nifty! I suppose I'll find out tomorrow, but will this affect how search results are displayed in any way, or will it leave those unaltered? ~ Amory (utc) 15:44, 7 May 2018 (UTC)[reply]
Thanks :-) - no this will not affect how search results are displayed. Basically, AdvancedSearch serves as an interface for some of the special search options that are already provided in CirrusSearch, but are difficult to find or to remember - especially when you want to combine several of them. E.g. you would get the same results if you manually type intitle:foo into the search field or use advancedSearch for that. The main advantage of AdvancedSearch is: The existing options become way more visible, users don't have to know the syntax behind each search field, different parameters can be combined easily, and you can decide if you want to learn more about the search syntax (or just use it). - In case you test it in the next days - we're really interested in hearing how it works for everyone :-) Birgit Müller (WMDE) (talk) 14:24, 8 May 2018 (UTC)[reply]

16:27, 7 May 2018 (UTC)

Possible spambots targeting Wikimedia mailing lists

Hello,

A large number of @AOL.com email addresses just signed up to our usergroup mailing list (seemingly also successfully passing the "confirm email" requirement as well). Just curious if any other mailing list admin faced this?

Suspecting this to be some sort of spambot targeting Wikimedia mailing list(s), I have removed all of them, and changed mailing list rules to require admin approval. I've also posted a notice on our usergroup. About 6 addresses managed to successfully subscribe (before I removed them), while another 30+ was stopped thanks to the rules update.

All of them signed up with only a single name (first name), such as Nicolas, Alfie, Homer, Cody, Hort, etc.

For those running mailing lists with open memberships, please remember to exercise caution. As it could be some sort of a targeted spam/phishing attempt. Whoever is in a mailing list, often has access to participating members' emails, as well as discussion histories, real name and username link references, etc.

Yours truly. Rehman 10:26, 8 May 2018 (UTC)[reply]

This may possibly be the recent password spammer. BWolff (WMF), are you able to check? --Izno (talk) 12:33, 8 May 2018 (UTC)[reply]
This is an issue for multiple lists. See wikimedia-l post. — JJMC89(T·C) 02:28, 9 May 2018 (UTC)[reply]
I've got 3 subscription requests from yahoo emails today, all with similar names. @JJMC89, on a separate note, would you by any chance know how to check why I cannot post anything to Wikimedia-l? I've tried emailing 3 separate times, but just got the moderator alerts only, and nothing thereafter. I could see others posting though. Cheers, Rehman 04:25, 11 May 2018 (UTC)[reply]
@Rehman: I don't know why you can't post on wikimedia-l. I would email the list moderators/admins to ask. The conversation has moved to listadmins. — JJMC89(T·C) 06:06, 11 May 2018 (UTC)[reply]

The spam subscriptions are surging from both AOL and Yahoo. I've got 294 emails banned from our mailing list (and counting). If you want the list (to mass block in any of your mailing lists), let me know. I'll be happy to provide. Cheers, Rehman 09:25, 14 May 2018 (UTC)[reply]

Random picture of the day and {{rand}} flags

Is there a way to display a random picture of the day along with the corresponding caption?

Also, how can I set the {{rand}} template to restrict the numbers it picks. For instance 5-6? Also how can I set it to generate a leading 0 before any number smaller than 10 i.e. 01 not 1. --Tyw7  (☎ Contact me! • Contributions) 10:42, 8 May 2018 (UTC)[reply]

Wikipedia:Picture of the day#Random POTD says: Use {{User:My Chemistry romantic/Templates/Random POTD}} to display a randomly-selected POTD. PrimeHunter (talk) 11:14, 8 May 2018 (UTC)[reply]
Edit: I saw that after posting this msg. I've gone ahead and done my own version with the code

<center>[[Image:{{POTD/{{#invoke:random|date|2007-01-01|{{CURRENTTIMESTAMP}}|format=Y-m-d|same=yes}}|image}}|250px]]{{clear}} {{POTD/{{#invoke:random|date|2007-01-01|{{CURRENTTIMESTAMP}}|format=Y-m-d|same=yes}}|caption}} </center> {{User:Tyw7/picofday}}

--Tyw7  (☎ Contact me! • Contributions) 11:16, 8 May 2018 (UTC)[reply]

You could use mw:Help:Extension:ParserFunctions##expr to add a value to the result of {{rand}} but it may be easier to use Module:Random#Number. {{padleft:n|2}} from mw:Help:Magic words#Formatting will add a leading 0 if n is one character, e.g. {{padleft:1|2}}: 01. PrimeHunter (talk) 11:26, 8 May 2018 (UTC)[reply]
Thanks. --Tyw7  (☎ Contact me! • Contributions) 11:37, 8 May 2018 (UTC)[reply]

Good Article/Mismatches

After posting here JJMC89 was kind enough to create Wikipedia:Good articles/mismatches. I can update this myself. One issue I am having is redirects. If someone moves a Good Article without changing the link from the lists then it is currently recorded as a mismatch. There are nearly 400 redirects in Wikipedia:Good articles/mismatches#On Wikipedia:Good articles/all but not in Category:Good articles and many of these will actually redirect to a Good Article.

One way to fix this is to remove all redirects from Wikipedia:Good articles/all. Then when I create the mismatch list they will not appear. Is there an automated way to fix redirects at Wikipedia:Good articles/all. I know WP:NOTBROKEN, but in this case it kind of is. Email me if you are worried about WP:Beans. Failing that do you know an easy way (i.e. a tool or AWB function - not something that requires a bot) to remove the redirect from the bottom list if it leads to an article in the top list. The article from the top list should also be removed at the same time. I would rather not use a bot as it means I can't keep it updated (and we have enough trouble with AWOL bots at GA at the moment). AIRcorn (talk) 19:21, 8 May 2018 (UTC)[reply]

@Aircorn: There is at least one gadget, and the possibility exists in your CSS, to display redirects in a different style than the default. That would get you most of the way to knowing which are the redirects, and removing or editing them as appropriate. --Izno (talk) 20:35, 8 May 2018 (UTC)[reply]
Thanks Izno. I have it enabled so the redirects show light green, which is how I came up with the 400 number. It does help, but there are a few issues. Not all redirects match up and some are genuine mismatches that need to be removed. This usually happens when an editor merges or redirects a Good Article to another article. The other problem is that for me it doesn't highlight the target article. I need to identify both so I can remove them from both lists. AIRcorn (talk) 22:47, 8 May 2018 (UTC)[reply]
@Aircorn: I use the CSS to check for redirects in navboxes--the way I know what the actual target article is is with either POPUPs (a second gadget) or Page Previews, which I think are in preferences under the Reading tab now (I use the Page Previews). --Izno (talk) 00:53, 9 May 2018 (UTC)[reply]

Update on page issues on mobile web

Hi everyone. The Readers web team has recently begun working on exposing issue templates on the mobile website. This follows a proposal from 2016. Currently, details about issues with page content are generally hidden on the mobile website. This leaves readers unaware of the reliability of the pages they are reading. The goal of this project is to improve awareness of particular issues within an article on the mobile web. We will do this by changing the visual styling of page issues.

So far, we have drafted a proposal on the design and implementation of the project. We were also able to run user testing on the proposed designs. The tests so far have positive results. Here is a quick summary of what we learned:

  • The new treatment increases awareness of page issues among participants. This is true particularly when they are in a more evaluative/critical mode.
  • Page issues make sense to readers and they understand how they work
  • Readers care about page issues and consider them important
  • Readers had overwhelmingly positive sentiments towards Wikipedia associated with learning about page issues

Our next step would be to start implementing these changes. We wanted to reach out to you for any concerns, thoughts, and suggestions you might have before beginning development. Please visit the project page where we have more information and mockups of how this may look. Please leave feedback on the talk page. CKoerner (WMF) (talk) 19:34, 8 May 2018 (UTC)[reply]

Need Help with variables and conditional logic in templates

Hi, I want to use conditional logic in this template: User:Wikiinger/sandbox/AMD GPU features. However somehow it broke the table. If someone can advice? Thanks! Wikiinger (talk) 19:50, 8 May 2018 (UTC)[reply]

What exactly is your difficulty? --Redrose64 🌹 (talk) 20:31, 8 May 2018 (UTC)[reply]
It's hard to help without knowing what you are trying to do. I guess some of your pipes should say {{!}} instead. That's a way for some features to display or return a pipe without the pipe being interpreted as part of the syntax for the feature itself. See mw:Help:Magic words#Other. PrimeHunter (talk) 22:20, 8 May 2018 (UTC)[reply]
The goal is to use parameters (here "fixed", "TeraScale" and "GCN") to control the table layout. More precisely when the template is called like this {{User:Wikiinger/sandbox/AMD GPU features|fixed=0|TeraScale=0|GCN=1}} only the rows belonging to the GCN instructions set (the 5 most-right) and the left-most row should be displayed.
However it is failing much more early... From my understanding {{{GCN|1}}} should default the GCN parameter to 1. Next {{#ifeq: {{{GCN}}} | 1 |! [[AMD RX Vega series|Vega]]|}}, should "print" ! [[AMD RX Vega series|Vega]] if GCN=1. But one (or both) of these statements seem to fail...Wikiinger (talk) 16:21, 9 May 2018 (UTC)[reply]
Both your statements are correct. I said I guess some of your pipes should say {{!}} instead. I took a guess at which ones.[8] Does that help? It sounds like you are confusing rows and columns. Rows are horizontal and columns are vertical. There is no "left-most row". If you want to hide a whole column then you have to hide each cell in the column. PrimeHunter (talk) 20:39, 9 May 2018 (UTC)[reply]
Take a careful look at https://www.mediawiki.org/wiki/Help:Extension:ParserFunctions, and also Help:Conditional expressions.
Something to note: 'ifeq:' is for comparing string functions, and might get confused with numeric comparisons. Try a simple "if:" instead. E.g.: {{#ifeq: {{{GCN|0}}} | {{!}} Jan 2012}}. Note that logical value of both a supplied string value and a digit "1" are "true". ♦ J. Johnson (JJ) (talk) 21:19, 9 May 2018 (UTC)[reply]
I think you mean {{#if: {{{GCN|}}} | {{!}} Jan 2012}}. #if: is true for any non-empty string including the string "0", so {{#if: {{{GCN|0}}} | {{!}} Jan 2012}} returns {{!}} Jan 2012 if GCN is not set. PrimeHunter (talk) 21:36, 9 May 2018 (UTC)[reply]
Yes. ♦ J. Johnson (JJ) (talk) 23:04, 9 May 2018 (UTC)[reply]

Bidirectional isolation bdi

I wrote a module at Commons that has worked well for 8 months. Now the links that it generates are broken. Like the template that it replaced, the module outputs links for navigation at Commons and it uses <bdi> (Bidirectional Isolation element) to wrap the label of each link so the browser can decide whether right-to-left text is needed. The first line that follows is an example of the output wikitext. The second line is the same with bdi removed.

  • <span style="white-space:nowrap">[[Algeria|<bdi>Algeria</bdi>]]</span>Algeria
  • <span style="white-space:nowrap">[[Algeria|Algeria]]</span>Algeria

The first line used to give a link just like the second line. Any idea what has changed that means it now doesn't? I merely copied the replaced template (which had performance problems) and don't know if bdi is necessary. Should bdi be used? The issue was raised at Commons here. Johnuniq (talk) 23:11, 8 May 2018 (UTC)[reply]

See this discussion above.
Trappist the monk (talk) 23:18, 8 May 2018 (UTC)[reply]
Excellent, thanks. That is now archived here and it mentions phab:T193414. The solution is to wait until the end of June for Tidy to be replaced with Remex. Johnuniq (talk) 03:05, 9 May 2018 (UTC)[reply]
A note re debugging: I tried a template with the problem at c:Special:ExpandTemplates and it worked correctly, whereas previewing the same template in a sandbox had broken links. I guess ExpandTemplates does not use Tidy and knowing that might be helpful if debugging stuff like this. I worked around the problem at Commons by temporarily removing bdi from the output. Johnuniq (talk) 03:38, 9 May 2018 (UTC)[reply]

Has there been a change in the way navboxes behave? I'm sure that previously if I viewed a navbox on a page which is linked in the navbox, the link to the host article was disabled and displayed in bold but now that behaviour doesn't happen and a clickable link remains. Nthep (talk) 14:47, 9 May 2018 (UTC)[reply]

Definitely not a redirect and I realise now it might be a skin issue, I'm using Modern and a check shows no issue on Vector or Monobook. Nthep (talk) 14:59, 9 May 2018 (UTC)[reply]
It is a skin problem due to a change in the core parser. The task DJ links below applies to the same issue in Timeless; there should be a separate patch (and/or task submitted) for Modern. --Izno (talk) 16:33, 9 May 2018 (UTC)[reply]
This is phab:T192033TheDJ (talkcontribs) 16:15, 9 May 2018 (UTC)[reply]

Regex find in string?

Template:Str find/Module:String allows you to search for a string in a string. Is it possible to have them search for multiple strings in a string? Or to code something in LUA that would have an equivalent functionality?

For instance look for /(Medicine|Oncology)/ in 'Medicine journal'? Headbomb {t · c · p · b} 17:49, 9 May 2018 (UTC)[reply]

Looks like find in the module can take a regex. --Izno (talk) 18:15, 9 May 2018 (UTC)[reply]
So how would I structure the module invocation for that? {{#invoke:String|find|SOURCESTRING|REGEX}}? Because I don't see how a regex or (|) would play nice here. Headbomb {t · c · p · b} 18:44, 9 May 2018 (UTC)[reply]
(edit conflict)
All of the Module:String functions will use Lua patterns but Lua patterns don't support the alternation (|) operator. In Lua itself you would 'or' together the several string library functions to get the equivalent:
(string.match (source_variable, '(Medicine)') or string.match (source_variable, '(Oncology)'))
Using the String module:
{{#invoke:String|match|s={{{source}}} |pattern=Medicine |nomatch=}}{{#invoke:String|match|s={{{source}}} |pattern=Oncology |nomatch=}}
Trappist the monk (talk) 18:55, 9 May 2018 (UTC)[reply]

Blergh, that is ugly as sin, and likely above my coding chops. However, I think I can half-ass something if I stack a bunch of finds inside an ifeq.

{{#if:{{#invoke:String|find|SourceString|TargetString1}}{{#invoke:String|find|SourceString|TargetString2}}...{{#invoke:String|find|SourceString|TargetStringN}}|match|no match}}

Headbomb {t · c · p · b} 19:04, 9 May 2018 (UTC)[reply]

An attempt to use the search box on WP:MOS gives the following error:

A warning has occured while searching: The namespace found in the search term prefix: is not compatible with the namespaces requested. You can fix it by prefixing your query with all:

Of course, that is not what is wanted at all. Any idea how to get around this? (And yes, "occured" is misspelt.) Hawkeye7 (discuss) 21:01, 10 May 2018 (UTC)[reply]

It's Thursday again. New MediaWiki version. Per phab:T193392, an inputbox now requires a namespaces parameter to match a namespace in the prefix parameter. This case was fixed by [9] but there must be many other inputboxes needing similar fixes. PrimeHunter (talk) 22:22, 10 May 2018 (UTC)[reply]
It's up to the person implementing the change to do it, not us. And although it's hard to follow the poor English in phab:T193392, it says it should emit a deprecation warning, not error the search off. I suggest pulling the change. Hawkeye7 (discuss) 22:35, 10 May 2018 (UTC)[reply]
A search of insource:inputbox insource:prefix in all namespaces gives 1964 results. It looks like most of them need a fix if the feature is not changed. I guess a bot could be coded to fix most cases but a warning in advance would have been nice. Maybe this should be pulled for now. PrimeHunter (talk) 22:43, 10 May 2018 (UTC)[reply]
The warning is made by MediaWiki:Cirrussearch-keyword-prefix-ns-mismatch which currently says: "The namespace found in the search term prefix: is not compatible with the namespaces requested. You can fix it by prefixing your query with all:." See e.g. italics prefix:Wikipedia:. The warning will probably make little sense to most users who just entered a term in an inputbox and didn't write prefix:. I suggest we change the message to say they can click "Everything" without getting technical. Maybe: 'Click "Everything" above to perform this search. A software issue currently requires this step.' PrimeHunter (talk) 23:06, 10 May 2018 (UTC)[reply]
MediaWiki:Cirrussearch-keyword-prefix-ns-mismatch is also used by mobile search which doesn't have the "Everything" option. PrimeHunter (talk) 23:44, 10 May 2018 (UTC)[reply]
I have created MediaWiki:Cirrussearch-keyword-prefix-ns-mismatch to mention the "Everywhere" option and link this discussion. PrimeHunter (talk) 11:01, 11 May 2018 (UTC)[reply]

Looks like the "fix" was not backward compatible. This is not called a "fix", this is called, "breaking working functionality". The burden is on those making a change to ensure current functionality continues to work transparently, or in the rare case, to provide an automatic migration as rapidly as possible, with notification to affected users of a window of functionality downtime well in advance of the migration. D- .   Mathglot (talk) 01:40, 11 May 2018 (UTC)[reply]

...but there must be many other inputboxes needing similar fixes. — Indeed; like every page using {{Talk header}} to search Talk page archives. About 390,000+ just for that one template, according to Template:Talk header, and there are other headers that use it. @PrimeHunter:. Mathglot (talk) 01:52, 11 May 2018 (UTC) Update: Oh, I see what you mean, you're counting inputboxes, and I was counting affected pages, but that's how I see it as a user, not a fixer, and that's how many will see it. Happy if there's < 2000 pages to fix, though. Mathglot (talk) 02:11, 11 May 2018 (UTC)[reply]
26000 uses of {{Talk header}} have an optional search parameter which generates a search box with {{Search box}}. I have fixed it there [10]. There are still other talk page templates which generate search boxes in other ways, e.g. {{Round in circles}}. Is there an expected timeframe for deploying the patch to revert this change? A template search only finds around 60 affected templates. We could quickly fix those. PrimeHunter (talk) 10:48, 11 May 2018 (UTC)[reply]

So when is an actual fix coming? It's affecting the Reference Desk search box too. Narutolovehinata5 tccsdnew 02:56, 11 May 2018 (UTC)[reply]

Maybe fix mw:Extension:InputBox to automatically specify "all:" when using prefix? I was looking at it, and there doesn't even seem a way in that to make it specify "all:" and fix the issue Galobtter (pingó mió) 05:28, 11 May 2018 (UTC)[reply]

Nvm can use "searchfilter= all:" Galobtter (pingó mió) 05:30, 11 May 2018 (UTC)[reply]
Nvm that nvm, doesn't work. I could be missing a parameter but it doesn't even seem like inputbox allows fixing this; which just shows terrible foresight in making this change. I see JJMC89 has tried something but It doesn't seem to fix it - the namespaces option seems to show buttons to specify namespace, not for specifying namespace in search automatically.. Galobtter (pingó mió) 05:40, 11 May 2018 (UTC)[reply]

Please excuse me for the annoyance this has caused, I prepared a patch to revert this change. I'll make sure that the prefix feature offered by InputBox is not affected again. DCausse (WMF) (talk) 09:49, 11 May 2018 (UTC)[reply]

While we wait on that, I've fixed the admin boards search, although I'd be interested in why that particular feature was deprecated. Bellezzasolo Discuss 12:33, 13 May 2018 (UTC)[reply]

Archive search is broken

The following boxed message appears:

A warning has occured while searching: The namespace found in the search term prefix: is not compatible with the namespaces requested. You can fix it by prefixing your query with all:.

Reproduce: Search any talk page wiht a searchable talk header. Example: Talk:Psychokinesis, search 'foo'. Mathglot (talk) 01:35, 11 May 2018 (UTC)[reply]

Duplicate of #MOS search above. Sorry, I searched 'archive' and didn't see this one at first. Mathglot (talk) 01:36, 11 May 2018 (UTC)[reply]
This search fiasco is arguably worthy of a spot in the stocks. The Blade of the Northern Lights (話して下さい) 01:53, 14 May 2018 (UTC)[reply]

Page performance study on other wikis

This should not be considered evidence of malware in the user's computing environment.

Just a quick heads-up that Technology's Performance team is planning to start a little on-page microsurvey later this month. It won't be running on enwiki, but somehow, all roads lead to VPT when people are surprised by what's on their screens... Feel free to ping me if someone brings questions to you about this. Whatamidoing (WMF) (talk) 21:44, 10 May 2018 (UTC)[reply]

I'm adding searchable text so people can find this. "Did this page load fast enough?" "For more information, see this survey's privacy statement." PrimeHunter (talk) 22:07, 10 May 2018 (UTC)[reply]

Something's wrong with 'unsortable' column headers

Hi,

Since maybe 24 or 48 hours, the headers of the columns with the 'unsortable' attribute are sized just like a sortable column, without the sorting button.

I take this example from Wikipedia:Sortable.

Numbers Alphabet Dates Currency Unsortable
1 Z 02-02-2004 5.00 This
2 y 13-apr-2005 Column
3 X 17.aug.2006 6.50 Is
4 w 01.Jan.2005 4.20 Unsortable
5 V 05/12/2006 7.15 See?
Total: 15 Total: 22.85

A better example is Opinion polling for the next Dutch general election#Seats. The table has 19 columns, of which 16 are unsortable. Now look at the headers. On each of these 16 headers, the text has moved to the left (when it should be centered), and there is a large unfilled space on the right, probably meant for the button, despite it doesn't have to be there. As a result, the table is twice as high as before because the rows' height has doubled (except, I suppose, on very large screens).

It does not seem to be browser-related, but maybe it is Firefox 60's fault.

Kahlores (talk) 01:57, 11 May 2018 (UTC)[reply]

This is a deliberate change, to avoid the browser having to re measure everything when the sorting system is applied. I've created phab:T194451 to improve the case of unsortable columns. —TheDJ (talkcontribs) 07:32, 11 May 2018 (UTC)[reply]
The columns are also definitively not the correct width, either; now they're about 67px when they're supposed to be just 40px wide. Mélencron (talk) 14:13, 11 May 2018 (UTC)[reply]
Very true. So we'll have to manually reduce the width of every unsortable column until we find something correct? I don't know the extent of the attribute's use on Wikipedia as a whole, but there are certainly hundreds of polling pages affected. On the Dutch page, we must change from "width:40px;" to "width:22px;" or smaller, so that it fits (at least on my screen). In fact, removing the width property altogether is the best solution. But we can kiss goodbye to columns of equal width, when there are plenty like on the Dutch page. Kahlores (talk) 15:14, 11 May 2018 (UTC)[reply]
I think it's a bug/unintended behavior – as far as I can tell the new column widths seem to be dependent on the width of the text within the columns, which is... annoying, to say the least. I hope this can be resolved soon – a temporary "solution" would just be to drop the sortability of the tables entirely to restore the normal appearance. Mélencron (talk) 15:26, 11 May 2018 (UTC)[reply]
Sorry. It's a side effect of a deliberate change. It will be fixed for unsortable columns. I don't think anything changed for other columns. However.... I would say that if you need to tweak widths of columns, then maybe you are trying to squeeze too much data into a table. Wikipedia is not a spreadsheet. It's not a database. Good luck to the reader with a table like that on a mobile device... Good prose, good graphs etc are way more useful than big tables of election results. —TheDJ (talkcontribs) 15:38, 11 May 2018 (UTC)[reply]
It's never a good idea to force table column widths, particularly if you're doing it to make it look nice and neat on your own screen. You have no idea how wide anybody else's device is, particularly in this era of ultra-wide desktop monitors coexisting with android phones held vertically. The only sensible thing to do is omit all width specifiers, and let the browser work it out based upon the column content. --Redrose64 🌹 (talk) 07:11, 12 May 2018 (UTC)[reply]

can I check my contributions across all Wikimedia sites for hacked login?

There was a login possibly not by me (after which I changed my password). I checked contributions under my username at en-Wiki and there were none not by me, so that's good, but can I check for contributions across all Wikipedia projects and sites and all languages under my username within a time period of a few days between two legitimate logins? That's too many sites to check by hand, because I think there are hundreds of sites (including languages). Nick Levinson (talk) 02:44, 11 May 2018 (UTC)[reply]

@Nick Levinson: you can use Special:CentralAuth/Nick Levinson to see all the sites you have made ANY edits on very quickly, as well as your approx first use date on a wiki. — xaosflux Talk 02:48, 11 May 2018 (UTC)[reply]
You could also use a tool like this one but it is much slower. — xaosflux Talk 02:50, 11 May 2018 (UTC)[reply]
@Nick Levinson: Did it say there was an actual login, or a "failed attempt"? If it was a failed attempt, there is probably no need to worry -- there is a wide-ranging login attack affecting a huge number of editors that has been going on for several days now. If you want to check your global contributions, you can do so at https://tools.wmflabs.org/guc/?by=date&user=Nick+Levinson --Ahecht (TALK
PAGE
) 02:51, 11 May 2018 (UTC)[reply]
It was an actual login, preceded by a few days by a failed one. The actual login may have been by me, but since the email did not say when it was (only that it was "recently") and I did not log in at the date-time of the email's time stamp, I decided to be cautious (to my annoyance) by changing my password.
The slow method is good, although it can't be used too soon after the period in question (someone has to wait, and maybe make a later edit to be sure of completeness) (whether the 20-edit limit matters I don't know). The generic URL for it is https://tools.wmflabs.org/guc/. The other (faster) page does not show edits within a time period, so it does not work for this purpose.
Nick Levinson (talk) 20:51, 12 May 2018 (UTC)[reply]

if login doubtful or fails, send email at that moment, not later

If a login fails or is spurious (e.g., Wikimedia is notifying us in case it was not by us) and Wikimedia sends an email to the user represented by that login, the user should know immediately and, if the user does not see the email immediately, at least the user should see a date-time stamp that is within a minute of when the login was attempted. There seems to be no other way to find out the date and time of the attempt, e.g., no log to check later. Please send the email right away, even if other kinds of emails go out once a day. Nick Levinson (talk) 02:49, 11 May 2018 (UTC)[reply]

And what made you think that it is currently not like that ? —TheDJ (talkcontribs) 07:21, 11 May 2018 (UTC)[reply]
If you have your prefs set to send e-mail less frequently, then it probably doesn't override. Whatamidoing (WMF) (talk) 17:07, 11 May 2018 (UTC)[reply]
What made me think that is that the email said the login event was "recently" and the email was timestamped for when I did not log in, but I had logged in within the preceding 24 hours, so, since I'm set to get emails daily, it could have been my own login and therefore nothing to worry about. But since the email was vague, I replaced the password, which means, since it's new, it's harder to remember. I don't want to have to replace it every time because of a vague recentness of a login.
I do have emails set to daily, which is more efficient for my work. However, a security email that is sent in the interest of both the user and Wikimedia probably should be allowed to override, so, if that's the problem, I can propose adding an override. Do you think there could be a problem with that?
Nick Levinson (talk) 20:39, 12 May 2018 (UTC)[reply]

Control level of section auto collapse in mobile

In mobile views, only the first level of headings is auto-collapsed, when it's expanded, all subsections are already expanded.

Visit WP:ITNC on a mobile device (browser, not app, I don't use the app so I don't know) and you'll see how crazy huge it gets. I'm wondering if this can be controlled somehow on a per page basis. I'd be thrilled if all the date specific subsections under "Suggestions" were collapsed. --LaserLegs (talk) 12:15, 11 May 2018 (UTC)[reply]

nope, this cannot be controlled right now. —TheDJ (talkcontribs) 12:44, 11 May 2018 (UTC)[reply]

Hidden text

Let's say I want to add something that makes editing articles easier to an infobox, but only display it to editors that choose to see this, rather than to everyone. How would I do that (with CSS or something).

e.g.

Readers Opt-in editors
Publications details
Publications details
Search for {{{1}}}

Headbomb {t · c · p · b} 19:53, 11 May 2018 (UTC)[reply]

  • In general editor notes are done with HTML comments, only visible in the source. Anything else would be a hack to "hide" from everyone (like with a display=none directive) except for people that knew to make their own directive to override it. I don't think this is a good idea for articles - what are your use cases? — xaosflux Talk 20:05, 11 May 2018 (UTC)[reply]
    @Xaosflux: Basically, in {{infobox journal}}, I'd want a sort of "look for sources" series of links that cross checks the ISSN/publication title with several databases/National Libraries. This would allow us to fill other infobox parameters more easily and expand the articles. It's not something that should be displayed to readers, since it'd be an editor help, not reader help. The display=none thing sounds like exactly what I need, I just don't know how to set it up so I can override it in my .css/.js settings. Hidding the text is easy <span style="display:none;">...</span>. The problem is making it display. Headbomb {t · c · p · b} 20:13, 11 May 2018 (UTC)[reply]
    You can make your own, pointed to the same class and include a !important; designation to override with a new setting. Making a change to all infoboxes just for YOU is a horrible idea though! — xaosflux Talk 20:54, 11 May 2018 (UTC)[reply]
    I tried that, but it seemed to just not work. And it wouldn't be just for me, but rather for a handful of editors that heavily deal with that infobox. It's the difference between being able to improve 100s articles per day because you've got everything you need at the end of a click, or 5 articles per day because you have to do a dozen queries by copy pasting from the infobox into a variety of databases. Headbomb {t · c · p · b} 21:03, 11 May 2018 (UTC)[reply]
    Well whatdoyouknow, now it works. Probably been a cache issue. (Edit: Turns out the issue was a stray ;).Headbomb {t · c · p · b} 21:10, 11 May 2018 (UTC)[reply]
    @Headbomb: I think project-wise a better solution would be to just add a unique class to the box, then use user javascript to add the special links. — xaosflux Talk 00:02, 12 May 2018 (UTC)[reply]
    Well, people are free to code that, but that's beyond my skills. Headbomb {t · c · p · b} 00:03, 12 May 2018 (UTC)[reply]
    I agree that it is coding, but you might be able to puzzle it out. The help text at CS1 errors might help you figure it how the CS1 templates do it. – Jonesey95 (talk) 03:22, 12 May 2018 (UTC)[reply]
    We already have cases of links that are normally hidden but which may be revealed with some user CSS. For instance, any WikiProject banner template that is built around {{WPBannerMeta}} has three "v-t-e" links which are normally hidden; these may be revealed with the CSS rule
    .wpb .navbar {
      display: inline !important;
    }
    
    Similarly, stub templates built around {{asbox}} also have three "v-t-e" links which are normally hidden, and these may be revealed with a very similar CSS rule
    .stub .navbar {
      display: inline !important;
    }
    
    These use the !important annotation (which is normally a cop-out since it's usually better to increase the specificity of the selector) because the display:none; declaration which we are overriding is part of the element's style="..." attribute. --Redrose64 🌹 (talk) 19:41, 12 May 2018 (UTC)[reply]
@Redrose64: those banners are not meant for pages normally used by readers that could get hit with a FOUCed up display. — xaosflux Talk 19:58, 12 May 2018 (UTC)[reply]
In general, we tend to avoid this technique in article space. As noted, it is used in some areas outside of article space and it is also in use on elements of transitory nature (notices). But making it part of the article content is general frowned upon. Especially, when it is used in the top of the article, where search engines etc tend to pick it up and possibly accidentally include it in their previews. Other transformed views could also be affected, think for instance of a machine learning system being fed this accidentally. This is also my problem with the current implementation of {{short description}}. xaoxflux suggestion of using a class and gadget seems like an interesting approach. —TheDJ (talkcontribs) 20:10, 12 May 2018 (UTC)[reply]
I don't see how FOUC is an issue in my two examples. Both templates use style="display:none;" so unless the browser ignores style attributes when first displaying the page (and I don't know of any that do), the first that the user sees of the v-t-e links is when the style sheets are processed. They then appear, and do not disappear again. --Redrose64 🌹 (talk) 20:20, 12 May 2018 (UTC)[reply]

Issue with blackscreen gadget and Notices

I use the blackscreen gadget with monobook on Edge, Win 10. TOday I have noticed a problem when trying to view Notices (eg thanks, reverts of my edits). The problem is that the text displays in black on a black background. It is not possible to read black text on a black background. Help would be appreciated! I would upload a screenshot but it appears to be illegal and contrary to Wikipedia's terms and conditions to upload a screenshot of Wikipedia. DuncanHill (talk) 00:08, 12 May 2018 (UTC)[reply]

@DuncanHill: screens shots of wikipedia placed here for technical support are generally OK, mostly so long as you are not also including non-free media in the screenshot. — xaosflux Talk 01:36, 12 May 2018 (UTC)[reply]
See Wikipedia:Screenshots of Wikipedia. — xaosflux Talk 01:38, 12 May 2018 (UTC)[reply]
See, now I just got a notice, but I can't tell from whom or for what. I think it might be for Xaosflux above. I did try uploading a picture, but the instructions made no sense whatsoever for what I needed to do. DuncanHill (talk) 02:23, 12 May 2018 (UTC)[reply]
Apart from actually getting the screenshot and saving it as a file on your computer/smartphone/device/etc., it's really no different from uploading a photo or other image to Commons. --Redrose64 🌹 (talk) 20:10, 12 May 2018 (UTC)[reply]
Also see phab:T193959#4186101. --AKlapper (WMF) (talk) 12:00, 13 May 2018 (UTC)[reply]

TOC broken on my watchlist

Over the last few days, I’ve noticed two things on my Special:EditWatchlist page. First, the checkboxes have changed from the little grayish 3D-looking buttons to blank white squares. At the same time, I’ve found that clicking on the links in the table of contents has no effect anymore; I end up having to manually scroll down the page.

I don’t care about the checkbox appearance, but I wish the links in the TOC would work. [running Mobile Safari in iOS 11.3.1 on iPad; viewed as desktop version] NotARabbit (talk) 02:55, 12 May 2018 (UTC)[reply]

The page was recently updated to use OOUI. I've reported the bug. — JJMC89(T·C) 05:53, 12 May 2018 (UTC)[reply]

Hi, I returned after a very long hiatus. I have completely forgotten how I had edited the Quick links on the left Margin of this page. (The list of links below the Wikipedia logo on the left). I had placed a few customized links here and cant seem to remember how I had done it. or may be the newer Wikimedia software has made it harder to do so. In any case I thought may be someone here knows how to edit it. --DBigXray 19:26, 12 May 2018 (UTC)[reply]

You did this using javascript in your User:DBigXray/common.js. Ruslik_Zero 20:33, 12 May 2018 (UTC)[reply]
(e/c) I cleaned that up a bit for you, because you had some older and/or non-working stuff in your User:DBigXray/common.js. —TheDJ (talkcontribs) 20:38, 12 May 2018 (UTC)[reply]

Odd templating issue

Can someone take a look in the BASIC-PLUS article? About half way through there is the text code|MAT A=ZER, which is being rendered as {{{1}}} on-screen. None of the other code templates is having a problem, why this one? Maury Markowitz (talk) 10:44, 13 May 2018 (UTC)[reply]

Wait, now LOTS of them are broken. I assume a missing { or } somewhere. Maury Markowitz (talk) 10:47, 13 May 2018 (UTC)[reply]
Or is it the equals sign? Maury Markowitz (talk) 10:50, 13 May 2018 (UTC)[reply]
It is the equals sign, Maury Markowitz; {{code|MAT A=ZER}} breaks as mediawiki thinks there's an argument |MAT A= being passed with the value ZER instead of the whole "MAT A = ZER" as the first argument. I've fixed it as in hereGalobtter (pingó mió) 10:56, 13 May 2018 (UTC)[reply]

Search string

Can a search be refined to show results for articles with two or more images?--John Cline (talk) 14:39, 13 May 2018 (UTC)[reply]

Adding insource:File insource:/File:.*File:/ to a search will find pages where the wiki source says "File:" at least twice. That misses cases with images added in other ways, and the search is slow and may time out if there are many results on the part before the insource code. PrimeHunter (talk) 14:59, 13 May 2018 (UTC)[reply]
Thank you. It works just as you said.--John Cline (talk) 15:38, 13 May 2018 (UTC)[reply]

Can someone produce a report listing all disambiguation pages with external links? I tried a search for insource:http deepcat:"all disambiguation pages", but this found a bunch of false positives, e.g. IN because a hidden comment includes the URL http://en.wikipedia.org/w/index.php?title=IN&oldid=300398091. Searching for insource:[http, with a bracket before the http, produced identical results. Nyttend (talk) 22:42, 13 May 2018 (UTC)[reply]

How about insource:http insource:/\[http/ deepcategory:"All disambiguation pages"?
(Note: the first "insource:http" is to limit the number of matching pages that will be searched with the regular expression, see Help:Searching#insource: and Help:Searching/Regex#Regular expressions.) --Pipetricker (talk) 23:23, 13 May 2018 (UTC)[reply]
quarry:query/18417 — JJMC89(T·C) 23:58, 13 May 2018 (UTC)[reply]

jquery to move rollback button

Wondering if someone could provide a line of jquery that would move "[rollback]" (class "mw-rollback-link") to before the user name, on each line (ideally in whatever page context it occurs in). I'm not sure about that location yet, but I think if given the form of the jquery I could adjust it myself, as I understand basic DOM stuff. Thank you! Outriggr (talk) 01:35, 14 May 2018 (UTC)[reply]

Outriggr, there's a project going on to make a fundamental change to rollback: users who opt in will be given a confirmation screen. I don't know how much change that will entail, but it's a big enough deal that any changes related to rollback links may end up being useless rather soon. Nyttend (talk) 02:21, 14 May 2018 (UTC)[reply]

Feedback round: A proposal for referencing sections of the same work more easily

A mockup of how referencing sections of the same work more easily could work. Please leave your feedback here

Referencing multiple sections of the same work in an article is currently cumbersome. Editors have asked for an easier way to do this for more than ten years. In 2013 and 2015 a wish to change this made it into Wikimedia Germany’s Technical Wishlist and it was wish #24 in the international Community Wishlist survey 2015.

WMDE’s Technical Wishes team conceptualized an idea how the problem could be solved: A generic solution that can be used for any refinement, such as pages, chapters, verses etc., and that could be used as a voluntary option, not forcing the users who don’t want to change their working mode to use it.

In order to find out if we can start working on this solution, we’re inviting editors from all wikis to have a look at it and tell us what they think in a feedback round from May 14th to May 27th. Thanks to everyone who participates and helps spreading the word!

-- Thank you, Johanna Strodt (WMDE) (talk) 07:42, 14 May 2018 (UTC)[reply]

Simple way to extract DEFAULTSORT?

Is there an easy way to extract the DEFAULTSORT value from an article (using AWB or other methods)? As {{Sortname}} is deprecated, I was hoping to quickly get the DEFAULTSORT value from biographies to use for table sorting. S.A. Julio (talk) 09:31, 14 May 2018 (UTC)[reply]

A technical question

Wilt the Stilt is a redirect with the following content:
#REDIRECT [[Wilt Chamberlain]]
{{R from alternative name}}
It categorizes to:
Category:Redirects from alternative names

{{R from nickname}} is a template redirect to:
{{R from alternative name}}

As such, Wilt the Stilt with the following content:
#REDIRECT [[Wilt Chamberlain]]
{{R from nickname}}
Will also categorizes to:
Category:Redirects from alternative names

Is it possible for the template redirect to influence categorization such that the second page could categorize to:
Category:Redirects from alternative names and
Category:Nicknames as well?--John Cline (talk) 10:26, 14 May 2018 (UTC)[reply]

{{R from nickname}} is currently a redirect to {{R from alternative name}}. As long as that is the case, it is not possible to behave differently. {{R from nickname}} could be changed to get its own code and do anything, e.g. add a new Category:Redirects from nicknames. It shouldn't add Category:Nicknames. It includes many articles, and redirect templates should not automatically add article categories. Individual redirects can be categorized. See Wikipedia:Categorizing redirects. Category:Nicknames seems to be for articles and disambiguation pages about nicknames so I wouldn't add the category to a redirect to a specific biography. PrimeHunter (talk) 11:32, 14 May 2018 (UTC)[reply]
Thank you, I understand.--John Cline (talk) 12:11, 14 May 2018 (UTC)[reply]