Project:Support desk

(Redirected from Support desk)
Jump to: navigation, search
vde   This page is for questions relating to the MediaWiki software.

Welcome to's Support desk, the central on-wiki place to ask MediaWiki questions!

The greater purpose of this page is to make our Manual and other available help so good that you do not have to come here to ask questions, or making them easier to find.

There are other ways for of communication as well (IRC, mailing lists etc.). Read more here.

Before you post

Post a new question

  1. To help us answer your questions, please always indicate which versions you are using:
    • MediaWiki (reported by your wiki's Special:Version page)
    • PHP (likewise)
    • Database (likewise, e.g. MySQL 5.5)
  2. Please include the URL of your wiki unless you absolutely can't. It's often a lot easier for us to identify the source of the problem if we can look for ourselves.
  3. To start a new thread, click "Start a new discussion".

Archiving topics

Topics are automatically archived when they have been inactive for three weeks. If a question you have asked is approaching this limit and still has not been answered, please 'bump' it to prevent it being archived. However do not 'bump' for other reasons.

Start a new discussion
First page
First page
Previous page
Previous page
Last page
Last page

li (or ul) position for category, file and etc.

Not horisontal, but vertical

Dimon4ezzz (talk)14:53, 23 August 2014

Jobs are not executed in MediaWiki 1.23.2

Hi all, this is my first post here, so please forgive me if I somehow break the webiquette of this site.

Affected wiki is a very small installation (roughly 100 pages only) with very few users. The URL can't be postet, it's a private website with password protection and delicate information. I don't have shell access to the web-server.

The environment is as follows:

  • MediaWiki 1.23.2
  • PHP 5.4.26-nmm1 (apache2handler)
  • MySQL 5.5.35-nmm2-log
  • Semantic MediaWiki 2.0
  • $wgEnableParserCache and $wgCachePages are both set to false (for other reasons).

The problem came up after upgrading from MediaWiki 1.22 to the current version and upgrading Semantic MediaWiki as well. When the usual repair and cleanup-jobs of semantic MW are to be run, nothing happened. I had a look at the list of "open jobs", it showed over 100 jobs in the queue. Regardless of the value of $wgJobRunRate, nothing changed after accessing the Wiki a lot.

Is this a known issue? Searching the web didn't reveal any helpful information to me.

I'd appreciate any hints what I could do the get this Wiki to execute the jobs. Thank you in advance for your help!

Best regards,

Tobias, 16 August 2014

You can have more information reading Manual:Job queue.

Try running jobs directly with runJobs.php. If they're run successfully and there are no remaining jobs, then monitor the job queue for a day for example, and see if jobs get stuck there. If yes, then you might try to set $wgRunJobsAsync to false and see if that makes jobs being run again on page requests. If so, you might want to enable debugging while restoring that setting and let us know of any error you might encounter on the log.

Ciencia Al Poder (talk)15:38, 20 August 2014

Thank you for your reply! Since I don't have shell access to the web-server, running jobs directly with runJobs.php is some kind of a challenge. Using the "Maintenance"-extension, I managed to run the jobs. Afterwards, new jobs appeared on the queue and were not run on page requests, just like before.

As suggested, I set $wgRunJobsAsync to false, with success, the queue was processed. Before and after restoring the setting to default, I couldn't identify any error on the log though. Anything in particular to look for? Any suggestions?

Would you recommend to set $wgRunJobsAsync to false again and leave it that way or will this cause any other problems?

Thank you for your help!, 23 August 2014

Image position "right" does not work

Edited by another user.
Last edit: 13:34, 18 August 2014

[[File:image.png|right|descripp]] not working in my site:Галерея

Dimon4ezzz (talk)10:04, 18 August 2014


You might try a tool like Firebug to find out how the positioning is done in wikis where it is working. That will give you an idea, what is currently missing for you., 18 August 2014

That wiki seems to use a custom skin. Right-aligned images are inside a div with CSS class floatright which usually have the CSS rule float:right;. You may to add that CSS class from other skins, like Vector or Monobook. Those styles are defined in the file /skins/common/shared.css, maybe you just need to include it in that skin.

Ciencia Al Poder (talk)16:11, 20 August 2014

oh, tnx!

Dimon4ezzz (talk)12:09, 23 August 2014

Upload link

Edited by another user.
Last edit: 11:16, 17 August 2014

Hi I am trying to add something that detects your website address a then goes to the the special upload page. here is the code.

<a href="
    <?php echo $this->data['nav_urls']['upload']['href']; ?>
  <div class="onhoverbg" style="padding-left:0.8em;padding-right:0.8em;float:left;height:40px;font-size:10pt;">
    <img src="" /> 
    <span style="color:#fff;position:relative;top:1px;">
      <?php $this->msg('uploadbtn') ?>
<?php $this->renderNavigation( array( 'NAMESPACES', 'VARIANTS', 'VIEWS', 'ACTIONS' ) ); ?>

The code worked for a few days then stoped working for me., 17 August 2014

Hi I think it is to do with people have to be loged in for it to go to correct link please can I have some help to allow everyone to go to the link., 17 August 2014


The code, which you have there creates a link to the Upload page and then attaches some image in that link. The line $this->renderNavigation() basically does not have anything to do with that.

However, I have not really understood what exactly it is that you want to do:

When the wiki (not the IP of the visitor), but when the wiki has a certain domain name, do you then want to have a redirect?

Second question: Which page should get redirected and what should be the redirect target?

And finally: Where should this be added?, 17 August 2014

Hi well for your two question. is that I would like to allow ip address to view the upload link but when clicked will show that error that you have to login so I would like help to allow ip address to view the link., 17 August 2014

So basically you want IPs to be able to see the link to Special:Upload (which they by default do not(?) see)?, 17 August 2014

Yes please because the link only lets user who are logged in to see it or if you logged out it redirects you to main page., 17 August 2014

Should I use MediaWiki for my project?

Hello guys,

I am involved in an internship. My duty is to create a demo for a wiki which is a repository for process components. I am not sure about involving in such a complicated environment as MediaWiki and I really can use some advice. I have a different skin design than the one MediaWiki uses, also I need to add some rather easy functionalities to wiki which I do not need help to implement. However, I need some of the functionalities of MediaWiki like access restrictions, sign up etc. Ultimately, an experienced programmer is going to write the real wiki with full functionality using MediaWiki yet I have to be able to demonstrate the functionalities in a rather easy way.

For instance, how do I remove all of unnecessary buttons and links from my page and add my own buttons. Where is the code of my Main_Page or any other page I added by simple entering the name after the ~/index.php? So far I only could learn how to create a new skin, which is rather hard because it is not simply writing a css code. If there is a way to code the way I want, how much is it harder than coding the same from scratch using just html, css and js. I didn't install MediaWiki from the git, is this a problem? ANY idea will be appriciated, I am really new to software engineering and just finished my first year of BSCS. I don't want to screw things up in my very first job:)

FurkanIpek (talk)23:46, 22 August 2014


MediaWiki has a broad user base, it is used in many places, has much software built around it and - compared to other wiki systems - it will be rather easy to extend it the way one wants and to get help should that be necessary. So if you say you want a wiki, then you should really use MediaWiki and not something else.

If you want to create an own skin, then you can start from scratch (hard) or take an existing skin (many are provided for download from their maintainers) and change this one the way you like.

As for the functionalities you mention: Sign up and so on are present in MediaWiki, but when it comes to access control you should read Manual:Preventing_access. Bottomline: MediaWiki is not made for access control.

What do you mean with "buttons"? The buttons "Page", "Edit", "History" and so on? Can be hidden with CSS. Own buttons can be added via hooks, here is a list of all available hooks.

The code of a certain page can be viewed using the "Edit" button. If you in contrast mean the template (which is used for every page), then you should have a look at the according skin file - with the default that's skins/Vector.php.

What exactly do you mean with "process components"? Some things and each one should have its own page with information, maybe this information should be displayed based on a template? Something like that?, 23 August 2014
Edited by author.
Last edit: 10:29, 23 August 2014

Thank you for your answer, it answered a lot.

And as for the process components, indeed they are all going to have their own pages, yet should be grouped and listed in separate pages before user reaches the specific page.

Do I have to write the CSS for hiding buttons separately or should I change vector.php? By buttons I meant every link on the menu located at the left or discussion session etc.

Is it possible to show/hide some buttons depending on login status?

FurkanIpek (talk)09:32, 23 August 2014

A grouping or listing of pages is possible, if you put them in Categories. That basically allows you to create a kind of category tree: Main categories, sub-categories, (sub-subcategories if you like, ...) and then the actual pages.

For CSS it depends: If you say you basically want to use the Vector skin, then you can simply add some lines of CSS to the wiki page MediaWiki:Vector.css in your wiki. However, if you basically want to have another, an own skin, then you should add an own skin with its own files in the filesystem and then you can (you better should) add the CSS in these skin files directly., 23 August 2014

Header / Footer Customised


I try to customise my print layout to have a logo on the top right, the date of last modification on the bottom left and maybe a revision number (if possible). I'm trying to have all this on every page that are printed. I have been struggling for a while with print.css but I can't manage to get a header on all printed pages..

Can anyone shed some light please?

I'm using Mediawiki 1.23.2. Skin: Vector

pivic, 11 August 2014


I think this can help you:

That means: Every element on the normal skin will be printed, unless they have the css class "noprint".

Florianschmidtwelzow (talk)21:29, 12 August 2014


Thanks for your reply.

I have been changing the print.css that overrides the stylesheet of mediawiki. Doing so allows me to show/hide different element of what is on the screen and change the layout/font etc.

What I have been trying to achieve for a while is to create customised header/footer. The information is currently not on the screen but I want it to be printed on every page. Like what Chrome automatically does by adding date, url, page number... etc...

I want to deactivate the header/footer of chrome and have my own personal information instead, repeated on all pages. Page number, Logo, last time edited and company name.

Maybe by creating new classes that would only be "printonly"?

Is it even possible to achieve using CSS?

Thanks, pivic, 15 August 2014

> Is it even possible to achieve using CSS? No, and i would say: Not with the "normal" MediaWiki skins. Unfortunately actually have no idea, how to reach this :/

Florianschmidtwelzow (talk)21:15, 16 August 2014

Thanks for your reply. I have been looking everywhere and I can't find a solution by modifying the mediawiki.... The only thing I found to achieve what I want is to create my own render server by installing mwlib... Unfortunately I can't get it to work.... I'm surprised that mediawiki doesn't have a way to fully customize a print. Businesses would be very keen to have such feature!, 23 August 2014

Infobox problems


I've been trying all sorts of things but still can't get the infobox template the work.

I've copied all the information to Common.css, I've imported all the correct templates from Wikipedia (including changing all incidents of ‘text/plain’ with ‘CONTENT_FORMAT_TEXT’), I've installed Parserfunctions, then, after upgrading to 1.22, Scribunto, but I'm still getting a big red "Script Error" - when clicked "Lua error: Internal error: The interpreter exited with status 2."

See -

What exactly have I missed, or what has gone wrong?


Ig1234 (talk)22:48, 29 December 2013

Do you have command line access to the server? If you do, then what happens when you execute the lua binary in the Scribunto directory?

Otherwise, check the permissions on the lua binary. They need to be set to 755 or +x.

Finally, try the debug log. If you add

 $wgScribuntoEngineConf['luastandalone']['errorFile'] = "$IP/images/temp/lua.log";

to your LocalSettings.php, does anything show up in the file? If so, what is the last bit?

MarkAHershberger(talk)20:19, 4 January 2014

Thanks Mark.

Sorry I only just noticed your reply, I'll look at what you said when I get back from work tomorrow and see what happens.

Ig1234 (talk)00:03, 6 January 2014


You'll have to forgive my ignorance.

I can probably get command line access (I'm using Dreamhost) but I fear my ignorance will mess things up in trying to use Putty etc. so I may have to ignore that.

I remember changing the permissions on the lua binary before. I've changed them for everything within the standalone binaries folder but it doesn't appear to make a difference.

Again... ignorance - how do I access the debug log from that line in LocalSettings.php?

Ig1234 (talk)20:58, 6 January 2014


Ig1234 (talk)20:31, 14 January 2014

See Manual:How_to_debug: Manual:$wgDebugLogFile and Manual:$wgDebugLogGroups.

If with debug log you in fact mean /images/temp/lua.log, then download it with FTP., 14 January 2014
Edited by another user.
Last edit: 17:02, 22 January 2014

I'm an idiot. Thanks.

This is the output of the debug log:

/home/danlee68/ xrealloc: ../bash/parse.y:2228: cannot allocate 512 bytes (49152 bytes allocated)
/home/danlee68/ xrealloc: ../bash/parse.y:2228: cannot allocate 512 bytes (49152 bytes allocated)
/home/danlee68/ xmalloc: ../bash/subst.c:7606: cannot allocate 112 bytes (45056 bytes allocated)
/home/danlee68/ xrealloc: ../bash/print_cmd.c:1498: cannot allocate 256 bytes (53248 bytes allocated)
/home/danlee68/ xmalloc: ../bash/subst.c:3503: cannot allocate 519 bytes (49152 bytes allocated)
/home/danlee68/ xrealloc: ../bash/print_cmd.c:1498: cannot allocate 256 bytes (53248 bytes allocated)
/home/danlee68/ xmalloc: ../bash/subst.c:3503: cannot allocate 519 bytes (45056 bytes allocated)
/home/danlee68/ xmalloc: ../bash/subst.c:3503: cannot allocate 519 bytes (45056 bytes allocated)
/home/danlee68/ xmalloc: ../bash/subst.c:3503: cannot allocate 519 bytes (45056 bytes allocated)
/home/danlee68/ xrealloc: ../bash/print_cmd.c:1498: cannot allocate 256 bytes (53248 bytes allocated)
/home/danlee68/ xrealloc: ../bash/parse.y:2228: cannot allocate 512 bytes (49152 bytes allocated)
/home/danlee68/ xrealloc: ../bash/print_cmd.c:1498: cannot allocate 256 bytes (49152 bytes allocated)
Ig1234 (talk)21:12, 16 January 2014

Some documentation needs clarified

Sorry, but couldn't find any but seeming mutually contradictory information on the replace string extension. Is it alive on en.Wikibooks (meaning I have a likely precedence flaw) or not is WAS my 'Need'... but I answered my own question remembering I used it successfully a couple of months ago in another template.  

I'm continuing because...
the one reference page has two box headers which seem to contradict themselves at the very page top, and the bugzilla linked there seems to be open and unresolved--but with a mention that everything there was now incorporated into the wiki magicword system, but that and one other.

If I'm making the right mappings from my history:

  1. Extension:Replace Text probably has the two warring box notices right up top... as I went to Bugzilla from there.
  2. Wikitext parser/Core parser functions - says it should work (but someone ought to reconcile these!)
  3. Extension:String Functions#replace is a very unhelpful section header with an apparently included empty sub-page.

I'd take it on, but making sure of my ground in this realm of computer science geeks is like translating German to English to read how an Electrical geek like me has to use it! best regards,

FrankB23:25, 22 August 2014

Is it correct that you have not asked a question?, 22 August 2014

Unable to download IFC file from mediawiki


The IFC files are sucessfully getting uploaded in mediawiki while not getting downloaded and giving the below error:

404 - File or directory not found. The requested resource has been removed or renamed, or is temporarily unavailable.

The upload and download functionality is working fine for files like doc,xlsx,pdf and pptx while getting error just in case of IFC file download.

MediaWiki : 1.9 PHP  : 5.3 Database : MYSQL 6.1CE

Can any one please suggest why i am getting this error.

Thanks, Azeem, 22 August 2014

Hi Azeem!

If your MediaWiki version really is 1.9, then you should really upgrade! This version is maybe around 5 years old and it is no longer supported. It also has tons of security holes and - and I think this is your problem currently - it is absolutely incompatible with PHP 5.3.

As a first step, you should upgrade MediaWiki! The steps are described on Upgrade!, 22 August 2014

Links to pdf files do not work

I have successfully uploaded a number of pdf files but if I use the syntax:


the page shows the correct text (LABEL NAME), but when I click on the text I am directed to 'Special Page Upload file'.

If I use the syntax:


the text 'File:FILENAME.pdf' is shown and when I click on the text I am directed to the page 'File:FILENAME' (which shows that the file exists) rather like when one clicks on a jpg image.
In either case I am unable to download the files as one would normally expect.

Version 1.23.2 MySQL: 5.6.16 phpMyAdmin: 4.1.12

Can anyone help please?, 22 August 2014

Thumbnail errors

MediaWiki 1.23.2
PHP 5.3.28 (cgi-fcgi)
MySQL 5.5.37-35.1
Lua 5.1.5

Having trouble with message commons I think, on gallery pages like this one Space Gallery, the gallery is of different layouts ie. <gallery></gallery> and in wiki-tables. The image calls are all the same [[File:|200px|thumb|center|<center></center>]] for the wiki-table layout and
File:|<center></center> for the <gallery></gallery> layout. If I upload a local copy of the broken thumb file, it works.


With broken thumbs < >

After local upload < >

Mlpearc (open channel)17:23, 22 August 2014

One Image i found which doesn't work was uploaded on commons and reused in your wiki. The Problem with this image (commons:File:City_Lights_of_the_United_States_2012.jpg): It is larger as the default dimensions of MediaWiki.

Size: 6,646 × 4,430

Mediawiki default max: 3500 × 3500

So, try to set a higher value. I see, that you use much images from commons? So it's a good idea to use the Wikimedia Commons default, which allows 50 million pixel.

Florianschmidtwelzow (talk)17:51, 22 August 2014

How to use the OR operator in queries?

Hi, I'm having a bit of trouble getting my query to work properly.

Basically, I need to use an OR operator - Which seem non-existent in Mediawiki unless you use the query function. The issue is, however, that the query function doesn't recognize shared database tables - Which I need.

Here's how my query looks:

SELECT post_id FROM `blog_posts` WHERE type = "published" AND '.$comp_author.' = "'.$author_get.'" AND ('.$comp_tag_1.' = "'.$tag_get.'" OR '.$comp_tag_2.' = "'.$tag_get.'" OR '.$comp_tag_3.' = "'.$tag_get.'" OR '.$comp_tag_4.' = "'.$tag_get.'" OR '.$comp_tag_5.' = "'.$tag_get.'") ORDER BY post_datetime DESC LIMIT 6'.$f_offet.'

How can I make use of the OR operator using Mediawiki's query system?, 21 August 2014

I think you mean the database abstraction layer, yes? If so, you can simply submit your conditions as a string, instead of an array. If you don't want to build the list manually, you can use the DatabaseBase::makeList method with the second parameter "LIST_OR", see the docs for more documentation of this:

And for the select method itself:

Florianschmidtwelzow (talk)06:50, 21 August 2014

Thank-you for your reply!

It seems that, oddly, makeList doesn't work within special pages, I keep getting:

Call to undefined method SpecialViewPosts::makeList()

Does makeList not work in Special Pages? Is there anyway to make my own 'list' without it?, 22 August 2014

...And just to be clear, if there are any makeList alternatives, I'd still need to OR function to work in some form., 22 August 2014
Edited by another user.
Last edit: 13:53, 22 August 2014

...And I apologize for a third reply, but if makeList CAN work within SpecialPages, then how would that look in a query, exactly? My query + the list look like this at present:

  $conds1 = array($comp_tag_1 => $tag_get,
       $comp_tag_2 => $tag_get,
       $comp_tag_3 => $tag_get,
       $comp_tag_4 => $tag_get,
       $comp_tag_5 => $tag_get);
  $conds_var1 = $this->makeList( $conds1, LIST_OR );
       $dbr = wfGetDB(DB_SLAVE);
	$result = $dbr->select('blog_posts', array(
	) , array(
	'type' => 'published',
	$comp_author => $author_get,
	) , __METHOD__, array(
	'ORDER BY' => 'post_datetime ASC DESC LIMIT 6 $f_offet'

Aside from the fact that makeList isn't working right now, is that query even correct?, 22 August 2014

makeList isn't a method of SpecialPage, it's a method of DatabaseBase. So use $dbr->makeList( $conds, LIST_OR );

And the condition parameter expect one array i think. I don't know, what happens, when you try to give two arrays. o please use a string if you want to combine or and and conditions. E.g. use $conds_and as a string for makeList( $and_conditions, LIST_AND ); and $conds_or as a string for makeList( $or_conditions, LIST_OR );

Florianschmidtwelzow (talk)13:54, 22 August 2014

Installing Extension:PdfHandler

Edited by another user.
Last edit: 19:51, 4 June 2013

I tried to install the pdfhandler:extension to work today but it won't work. this is an win2003 server (64) machine. I installed gs and xpdf. ImageMagick worked already before. I configued as follows:

require_once "$IP/extensions/PdfHandler/PdfHandler.php";
$wgPdfProcessor = 'C:\Program Files\gs\gs9.07\bin\gswin64.exe';
$wgPdfPostProcessor = $wgImageMagickConvertCommand;
$wgPdfInfo = 'C:\Program Files\xpdf\bin64\pdfinfo.exe';
$wgPdftoText = 'C:\Program Files\xpdf\bin64\pdftotext.exe';

All paths have been checked several times. Instead of a nice rendered pdf-pic i get only a malformed File: - page for each pdf file. Example:

Pdfhandler malformed.png

Where is the gap?

Carchaias (talk)12:20, 4 June 2013

This is likely due to a modification made to the interface message file-info-size. Check MediaWiki:File-info-size/de or MediaWiki:File-info-size on your wiki (not on Did you edit this one? If so, edit it again and make sure there it has no extra space in front.

Extra space at the beginning of a message renders as a "<pre>".

Krinkle (talk)19:58, 4 June 2013

These messages dont have been modified. Thus I checked for spaces but there are none. When I comment out the pdfhandler in localsettings.php the above mentioned behavior disappears and all looks fine. So I think it comes from the extension..

Carchaias (talk)07:42, 5 June 2013

I've the same problem... with a debian install and pdf handler

Julien09:44, 3 August 2013

Some time has passed and tody I tried agan with pdf-handler on a fresh 1.23.2 Installed wiki. That malforming has gone - fine.

Well, but was does this extension do? I can't see any reaction to my wiki!

Configuration ist this:

require_once "$IP/extensions/PdfHandler/PdfHandler.php";
# Konfiguration von pdfhandler
$wgPdfProcessor = 'C:\Program Files\gs\gs9.70\bin\gswin64.exe';
$wgPdfPostProcessor = $wgImageMagickConvertCommand; // if defined via ImageMagick
//$wgPdfPostProcessor = 'C:\Program Files (x86)\ImageMagick-6.6.9-Q16\convert.exe'; // if not defined via ImageMagick
$wgPdfInfo   = 'C:\Program Files\xpdf\bin64\pdfinfo.exe';
$wgPdftoText = 'C:\Program Files\xpdf\bin64\pdftotext.exe';

Any ideas are welcome...

Carchaias (talk)13:50, 22 August 2014

Some time has passed and tody I tried agan with pdf-handler on a fresh 1.23.2 Installed wiki. That malforming has gone - fine.

Well, but was does this extension do? I can't see any reaction to my wiki!

Configuration ist this:

require_once "$IP/extensions/PdfHandler/PdfHandler.php";
# Konfiguration von pdfhandler
$wgPdfProcessor = 'C:\Program Files\gs\gs9.70\bin\gswin64.exe';
$wgPdfPostProcessor = $wgImageMagickConvertCommand; // if defined via ImageMagick
//$wgPdfPostProcessor = 'C:\Program Files (x86)\ImageMagick-6.6.9-Q16\convert.exe'; // if not defined via ImageMagick
$wgPdfInfo   = 'C:\Program Files\xpdf\bin64\pdfinfo.exe';
$wgPdftoText = 'C:\Program Files\xpdf\bin64\pdftotext.exe';

Any ideas are welcome...

Carchaias (talk)13:51, 22 August 2014

UNIQ error for GbrowseImage extension in version 1.23.2

A thread, Thread:Project:Support desk/UNIQ error for GbrowseImage extension in version 1.23.2 (2), was moved from here to Extension talk:GbrowseImage. This move was made by Florianschmidtwelzow (talk | contribs) on 22 August 2014 at 13:48.

I see search results page instead of article

I've recently upgraded from MediaWiki 1.22 to 1.23, and I notice now that when I type in a search term, I always see a search results page, even if there is an article that matches the name.

  • Before, if I typed "XYZ" and if there is an article named "XYZ", the system displayed that article right away.
  • Now, I get a results page that says "There is a page named XYZ on this wiki", lists that page and then a bunch of other pages that link to it.

Is there a setting that's changed that I'm not aware of? Can somebody point me in the right direction?


Supasaru (talk)19:09, 25 July 2014

I have the same behaviour, I find it annoying, but since it worked before and there anyway are obviously no efforts to change this, I fear some people might even think that's intended., 25 July 2014

In my opinion that's the best way to handle this (but, like i said, that's my personal opinion). What is, if i want to search for the string "XYZ" in other articles? There is only one button "search", not like in past a button for "full article" and "search".

Florianschmidtwelzow (talk)14:36, 26 July 2014

Click on "containing XYZ" and you will be brought to the search page. The fact that hitting enter opens the search, although a page with exactly this name is present, is really confusing. Afterwards I always have to click the page name as if MediaWiki was asking me, if I really meant my search serious. This change lowers user experience. Or is there an option, which I could set to work around that bug?, 26 July 2014

Wikipedia, the biggest user of the MediaWiki software (if I'm not mistaken) continues to bring you directly to page XYZ, and not a search results page.

This leads me to believe there's a setting somewhere, but where? Anybody have an idea?

Supasaru (talk)23:48, 26 July 2014

I could not believe that, but you are right: While on Wikipedia this works correctly, it is broken e.g. on a system of mine (running MW 1.23.1). I also want to know, how I can fix this for my system!

There is no fitting documentation at Manual:Configuration_settings#Search, 27 July 2014

Ähm, yeah :) I have tested it now with Vector skin on my private wiki (which is running 1.24wmf12), and if i search for a page which exist, i will be redirected to this, not to the search result page. In my custom skin i will be redirected to the search result page ever, so do you use a custom skin? Then you must change the search button from fulltext to go :)

Florianschmidtwelzow (talk)12:58, 27 July 2014

Yes, I do use a custom skin. I still remember how I put a comment in my skin file at the place, where the MediaWiki devs changed the display of the search buttons, because I couldn't get it to work properly for me: The display of the buttons alwaysgot messed up, so I decided to leave them both for now. But does that (the presence of two instead of one button) also influence what happens when I hit enter (because that is where I see the problem and where it annoys me most)?, 27 July 2014

I'm using Vector with no custom skinning, so I don't understand what's causing a search results page to appear.

Supāsaru14:57, 27 July 2014

Does anybody have any answers to this?

Supāsaru12:03, 3 August 2014

I have also the version 1.23.2 and when I type XYZ in research I have a list with the 2 possibilities if the page exists

first line : XYZ second line : "including XYZ" They apperas when I move the mousse

But don't clic on the glass

Chantoune (talk)18:00, 3 August 2014

→ Bump ←

I continue to experience this "bug." Does anybody know the cause?

Supāsaru12:31, 19 August 2014

Is your wiki publicly accessible so we can look what's doing?

Ciencia Al Poder (talk)16:47, 20 August 2014

Sorry - no, it isn't.

Is there anything you'd like me send show you? Screenshots? Excerpt from LocalSettings.php?

Supāsaru01:24, 22 August 2014

Describe how do you submit the search form:

  • Pressing the enter key when inside the search box
  • Click on the search button
  • Another option?

Post an example search results page URL.

Ciencia Al Poder (talk)09:10, 22 August 2014

Hiding users' permissions

Is there a configuration or extension that would prevent anyone but sysops from seeing what permissions another user has? Would this effectively be accomplished by blacklisting Spacial:ListUsers and setting $wgLogRestrictions to hide all promotion logs? (Using v1.22.3, PHP 5.4.31, MySQL 5.5.37-cll.) Thanks.

Arctic.gnome (talk)22:36, 21 August 2014

MediaWiki Ask Query and Category Pages

I have created two pages in a MediaWiki: John, and Category:Smith (note that the page Smith is a category page). Both pages were part of the Category:Person. When I run an Ask query to return all the pages that are under the category Person, only the page John is returned. {{#ask: Category:Person | ?label}}. It seems to me that the Ask query does not return pages that are category pages.

Any comment or solution to this problem?

#ask is part of SMW, not core MediaWiki. You should go to get support for it.

Jackmcbarn (talk)21:41, 21 August 2014

Trying to install MediaWiki for 2 hours

I downloaded xampp v1.8.3 and downloaded mediawiki via git bash using "git clone" I extracted "core" to htdocs in xampp file directory, opened xampp control and started apache and mysql. I copied the index of DefaultSettings.php and created a LocalSettings.php with the same index as DefaultSettings.php. Then changed DBuser to root and password to two single quotes in a row (i do not have password for root user and wiki doesn't show two blank single quotes). In the installation guide, it says that "If you already have a database server and know the root password for it, the MediaWiki installation script can create a new database for you." yet when I type localhost/core/index.php, I get an error like this:

Sorry! This site is experiencing technical difficulties.

Try waiting a few minutes and reloading.

(Cannot contact the database server)

Additionally, I tried to create db by myself with the following code:

create database wikidb; (I also changed the DBname in LocalSettings.php) grant index, create, select, insert, update, delete, alter, lock tables on wikidb.* to 'root'@'localhost' identified by ; (again there are two single quotes in a row after identified by)

mysql creates db yet returns an empty result set.(I think this means that granting was unsuccessful)

After creating db, localhost/core/index.php gives the following error:

"A database query error has occurred. This may indicate a bug in the software."

What should I do? Thank you guys in advance:)

FurkanIpek (talk)14:43, 21 August 2014


If you are not a developer, then you should probably not use the Git checkout, but the regular tarballs. Get the file from here: and extract that one into your htdocs directory.

After that make sure your servers (Apache and MySQL) are running and visit the wiki with your webbrowser. You will then be guided through the rest of the installation process. See Manual:Installation_guide for more information on how to install MediaWiki!, 21 August 2014

[RESOLVED] Passing {{filepath:filename.ext}} to an extension

I'm trying to write a simple extension. I want to pass it the name of an uploaded media file but I can't seem to get it to the extension. Specifically, I don't want to have to put the path to the file in the call, I'd like to just put {{filepath:filename.ext}} and have the location found for me. I've stripped the extension down to just a few lines:

$wgHooks['ParserFirstCallInit'][] = 'henryinit';
function henryinit( $parser ) {
  $parser->setHook( 'Henry', 'henryrender' );
  return true;
function henryrender($input, $args) {
  global $wgScriptPath;
  return "File: ".$args['file']."<br>" ;

Then, in a page, I put the following (where GH-Scouts.mp4 is a previously uploaded video):

<henry width="400" height="300" file="{{filepath:GH-Scouts.mp4}}" />

Filepath: {{filepath:GH-Scouts.mp4}}

This returns the following:

File: {{filepath:GH-Scouts.mp4}}


As you can see, the second line correctly displays the full path to the file but the first, which calls the extension, reports the file path as it was entered. I feel like I need to have my henryinit associated with a different hook but can't figure out which one. If, for instance, I use 'ParserBeforeTidy' it finds the path, but then it doesn't run my extension. Can someone point me to an example or discussion of how I can do this? Thanks.

Henryhartley (talk)19:45, 20 August 2014

I think you missunderstood something :) setHook() is a method to add a tag (like <example></example>), that's called a tag extensions. If you try to add a parser function, please read the documentation of it :)

Florianschmidtwelzow (talk)06:56, 21 August 2014

I'm not trying to add a parser function. I'm simply trying to use one (that already exists) in my tag extension. Is that not possible? At least that's what I think I'm trying to do.

Henryhartley (talk)13:24, 21 August 2014

That is to say, if I put this in a page, my extension works just fine and I can do whatever I need with the video.

<henry width="400" height="300" file="/images/a/ad/GH-Scouts.mp4" />

But, I don't want the user to have to find the path to a file. I want them to be able to enter the tag like this and have that translated by the filepath parser function into the full path.

<henry width="400" height="300" file="{{filepath:GH-Scouts.mp4}}" />

But the file attribute gets passed as a string (including the curly braces) to my tag extension. I want {{filepath:GH-Scouts.mp4}} to return the path first and pass that to my extension. It seems that my tag extension is parsed first, before the parser function (filepath) has a chance to get processed. I thought making my tag get processed later would do the trick, but apparently that was wrong. Does that make sense?

Note, I also tried it like this with no better results:

<henry width="400" height="300">{{filepath:GH-Scouts.mp4}}</henry>
Henryhartley (talk)13:37, 21 August 2014

Content that is passed to a tag extension is never parsed, iirc. So the easiest way should be:

Let the users pass the filename ("name.mp4") to your extension and you do the following:

$file = wfFindFile( $filename );
$fileUrl = $file->getFullUrl(); // this will contain a fully qualified url to the file


Florianschmidtwelzow (talk)14:15, 21 August 2014

That's exactly what I needed. Thanks.

Henryhartley (talk)16:40, 21 August 2014

Can a MediaWiki site running under IIS include a link to a local .asp file running on that site?

We at are considering MediaWiki to handle most of our millions of pages. We would generate such MediaWiki records from code that read and formatted the information in our existing database, and from then on allow MediaWiki users to edit these pages. But we have a few, such as, that use vbscript to invoke mysql queries and format results, all tailored to user input in form fields. Such external pages would never be edited by a user, and need not be stored in any MediawWiki table. Is there a way to do this?

- David Stang webmaster, 21 August 2014

I don't know if I've understood correctly, but you can add links to any page just like the link you posted on your message, even if it points to the same server where MediaWiki is hosted.

Ciencia Al Poder (talk)20:24, 21 August 2014
First page
First page
Previous page
Previous page
Last page
Last page