This page used the Structured Discussions extension to give structured discussions. It has since been converted to wikitext, so the content and history here are only an approximation of what was actually displayed at the time these comments were made.
"Main Page" won't open (PHP error?), all other pages do. Help!
Latest comment: 3 years ago2 comments2 people in discussion
The Poem extension stopped working, so I re-added it. Then Main Page stopped working, with it saying it couldn't open MultiCategorySearch.php. Tried undoing things, but error persists. Oddly, all other pages open fine. Error message as follows:
Fatal error: require_once(): Failed opening required '/home/customer/www/<domain redacted>/public_html/extensions/MultiCategorySearch/MultiCategorySearch.php' (include_path='/home/customer/www/<domain redacted>/public_html/vendor/pear/console_getopt:/home/customer/www/<domain redacted>/public_html/vendor/pear/mail:/home/customer/www/<domain redacted>/public_html/vendor/pear/mail_mime:/home/customer/www/<domain redacted>/public_html/vendor/pear/net_smtp:/home/customer/www/<domain redacted>/public_html/vendor/pear/net_socket:/home/customer/www/<domain redacted>/public_html/vendor/pear/net_url2:/home/customer/www/<domain redacted>/public_html/vendor/pear/pear-core-minimal/src:/home/customer/www/<domain redacted>/public_html/vendor/pear/pear_exception:.:/usr/local/php74/pear') in /home/customer/www/<domain redacted>/public_html/LocalSettings.php on line 155
What's really strange is the reference to MultiCategorySearch.php as I had previously removed it once before (because it didn't work). I've just now tried adding a fake, empty MultiCategorySearch.php file, but that didn't work. I cannot seem to find anything else that references MultiCategorySearch.php (though I cannot find a way to search through file contents).
If other pages open fine, it sounds like you might be experiencing some sort of caching issue with your host. Do you have some sort of html accelerator, CDN or frontend cache enabled (e.g. varnish, cloudflare, cloudfront, etc)? Bawolff (talk) 17:42, 2 November 2022 (UTC)
Where are image files stored?
Latest comment: 3 years ago2 comments2 people in discussion
Question as above. I have been hunting through various folders under the "images" folder but most seem to be empty. Are they simply stored inside the database then? How would I retrieve them? 8420 De Haan (talk) 13:52, 1 November 2022 (UTC)
they are stored in the filesystem (some extensions exist to change that so it depends on if you have those installed). Generally the scheme is take the md5 of the filename (not the file itself, just its name), then the first digit is the first folder and the first 2 are the second.
Latest comment: 3 years ago1 comment1 person in discussion
Currently I'm using Solr search but I was notified that the newest MediaWiki version would not support Solr search anymore, I was wondering if that was the case, and if so, would migrating to ElasticSearch be better, since it is a lot of work I'd rather stay with Solr, just wanted to check if it is going to be supported or not. Georgejobran (talk) 14:52, 1 November 2022 (UTC)
Ask for user's agreement before displaying external content?
Latest comment: 3 years ago1 comment1 person in discussion
Howdy,
is there an extension, or another easy solution, which allows to check for user's agreement before (!) showing external contents, e.g. images from external urls or Youtube videos? This needs to happen before the transcluded content is shown, before the user's browser establishes connection to the other site (GDPR issue).
How to have a user read and agree to a set of rules before creating an account
Latest comment: 3 years ago3 comments2 people in discussion
Not too long ago, I had this question, which was answered by Bawolff. I have since changed my mind on how to allow users to edit on my wiki. I have now decided that instead of making it a whole separate process, I would like to make it easier by trying to make it part of the sign up process that would allow a user to edit. Bawolff told me that there is a way to incorporate a second page into the sign up process, and that he thinks there's some extensions that can do this, but didn't specify. I tried searching around for any such but couldn't find anything. Therefore I wanted to ask if anyone knew any extensions or methods of achieving this goal? Thanks Guillaume Taillefer (talk) 00:01, 2 November 2022 (UTC)
Latest comment: 3 years ago2 comments1 person in discussion
RESOLVED
done; corrected.
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
The discussion above is closed. Please do not modify it. No further edits should be made to this discussion.
There is currently no text in this page - strange things
Latest comment: 3 years ago5 comments2 people in discussion
Hi all,
I'm using mediawiki for +15 years. Sometimes you need newer versions.
After the hosting-installatron updates (start with 1.31.1) of mediawiki, my wiki became "corrupt".
Issue: There is currently no text in this page
Editing page does not work, cause page exists in DB. Also in the text table.
New pages creating goes fine. Putting backup back (v 1.31.1) gives the same issue (that's my strange point). It takes a lot time for me to find apossible issue. Not found yet (for sure I did place also more and older backups - same issue).
I decided to create fresh install (installatron) version 1.38.4
And putted an upgraded DB below.
For new pages everything works fine. The old existing pages which are "There is currently..." aren't available and do have the same issue as earlier (when I putted the backup back). So, the version does not affect issue (it's still same findings)
When I try to edit this page (same as earlier also), the next message becomes:
The revision #0 of the page named "IToon" does not exist.
This is usually caused by following an outdated history link to a page that has been deleted. Details can be found in the deletion log
Also runned: attachLatest.php, findMissingActors.php, findAnomalies.php, migrateActors.php But no results.
Futher analysis: the revisionXX tables are not filled in. New created page they do.
So I tried creating records in the tables of an old existing page.
Just for checking if I can fix the problem with sql update queries.
Now, the correspondending old test page gives after refresh in the browser the message: Main slot of revision not found in database. See T212428. https://phabricator.wikimedia.org/T212428
I don't know if it gives more insight in my problem, but this are some findings..
Table revision_comment_temp , field revcomment_comment_id giving high number, issue still issue.
So I think it should be possible to make with good sql the old pages available, based on table page and table text (perhaps some more). Or do I miss something? Stephankk (talk) 23:45, 2 November 2022 (UTC)
that's probably from manually filling in some of the tables but not others. The error is essentially telling you there are referential integrity issues.
Generally i would reccomend not trying to manually adjust the database but use the updater script where possible. Bawolff (talk) 01:26, 3 November 2022 (UTC)
Yes they were for sure.
Problem found. The intallatron backup mechanism of the hosting skipped some tables.
Latest comment: 3 years ago1 comment1 person in discussion
we have an instance of mediawiki running on another server of an older ver. I have created a new VM (linux) running a newer ver of mediawiki (1.37.6) with nginx and php 7.3. The database is located on a remote host which I can access via CLI using the same credentials I have set in the LocalSettings.php file. When I enter the web address in my local machine browser I get the following. Any ideas as to what may be causing this would be great. As stated previously I am able to log into the DB server remotely using the same credentials as the mediawiki config file. Thank you.
Sorry! This site is experiencing technical difficulties.
Latest comment: 3 years ago2 comments2 people in discussion
Hello I was wondering if there was anyone who set up media wiki with openbsd 7.2 and nginx. I would really like some assistance as after all my work I was unable to get anything working. 92.232.147.196 (talk) 22:19, 2 November 2022 (UTC)
if you explain the problems you encountered, someone might be able to help you. Its unlikely anyone will help with a question this generic. Bawolff (talk) 01:23, 3 November 2022 (UTC)
Trouble with Tabs
Latest comment: 3 years ago6 comments3 people in discussion
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
So, today I wrote this article, which uses an infobox that contains tabbed images.
My first problem is that despite setting the images to the same resolution, because they are all different aspect ratios, these cause the infobox to "jump" up and down depending on the image selected. I would like to know if it's possible to make the bottom of the image "ghost out" beyond a certain number of pixels so that the infobox stays static.
I'm not sure if this is simply a limitation with the way I have implemented tabs in my version of MediaWiki, or something I'm doing wrong, but whenever you load a page that has a tabbed infobox, for about a second (maybe less), you can see all of the tabbed images at the same time before it collapses.
With the second question, I'm going to guess that it's just the way the tabs extension works. Though now I've had a look at the extension page it says "independence from JavaScript results in having no flash of unstyled content when the page loads", so I don't know. Jonathan3 (talk) 00:09, 3 November 2022 (UTC)
note, instead of setting image width you can set height by using syntax [[File:Example.png|x100px]]
For flash of unstyled content. Maybe you could use inline display: none styles, that might load instantly before the main css (not sure) Bawolff (talk) 01:22, 3 November 2022 (UTC)
I think MediaWiki can preload CSS etc (I'm not sure). So maybe the extension's styles, on the page itself, are being used too late in the day. Do you think maybe adding something like
to MediaWiki:Common.css would load earlier than the extension's CSS (as Bawolff suggested is necessary)? Maybe the "body:not(oldbrowser)" part is a red herring but I saw it in the browser console. Jonathan3 (talk) 12:54, 3 November 2022 (UTC)
I could definitely try it. I sort of suspected that it had something to do with the way the Tab extension worked, owing to its age and the fact this doesn't happen with other collapsible elements I've placed into infoboxes.
There seems to be a lot of extensions that are just way beyond deprecation but are still used either out of habit or the fact alternatives don't really exist, even now.
Latest comment: 3 years ago2 comments2 people in discussion
Is it possible to add the Wiki Search bar inside the wikitet area? eg - creating a help page and want to set a search bar on that page with namespace help defaulted.
I have no idea, Im using this config provided by nginx, Since I cannot send links here you can search something like MediaWiki - NGINX 186.80.28.66 (talk) 22:18, 3 November 2022 (UTC)
The error is when it tries to load load.php?lang=es&modules=ext.visualEditor.desktopArticleTarget.noscript%7Cmediawiki.action.styles%7Cmediawiki.interface.helpers.styles%7Cskins.monobook.styles&only=styles&skin=monobook now 186.80.28.66 (talk) 23:38, 3 November 2022 (UTC)
Latest comment: 3 years ago7 comments2 people in discussion
Recently I was trying to test out the limits of UploadWizard on my Wiki (in particular how big of files I could upload), because I don't exactly know what its limits are. The reason is that in LocalSettings.php, I have $wgMaxUploadSize = 1024*1024*1024*10; for 10GB. However in my php.ini I have upload_max_filesize, post_max_size, and memory_limit all set to 1024M (M for megabytes I presume) which seems to me that it would only allow up to 1GB, however I tried uploading a pdf file that was 1.26 GB big, which went through, then I tried a 2.49 GB pdf file as well, which went through, and then when I tried uploading a 5.87 GB pdf file, it took around 45 mins to get through, but then stopped the download, which then gave the "Invalid chunk offset" error. I suspect that this has something to do with my settings, but given from my perspective how that doesn't make they don't make much sense with what was allowed, maybe it is that or its something else. Thanks
So, i think there is an overall 4gb limit to chunked upload (which uploadwizard uses).
The normal upload limits (in php.ini et al) apply to normal upliads via Special:upload. ChunkedUpload splits uploads into multiple chunks so the limit applies to each chunk instead of the whole file. Bawolff (talk) 00:21, 4 November 2022 (UTC)
So I had tried to upload those bigger files through Special:Upload and the problem that I'm getting now (even though its allowing 10Gb), is it is giving me a 502 Bad Gateway message in response after a while. I'm pretty sure that this is a Bluehost and server issue so I'm going to contact them about it, but I wanted to also make sure this isn't a MediaWiki issue whatsoever either. Guillaume Taillefer (talk) 18:49, 5 November 2022 (UTC)
Mediawiki wont internally accept any files bigger than 4gb. The 502 failure is probably just happening before that step.
Most webserver setups start to get a bit unreliable after 100mb upload sizes, so while the 502 is a bluehost issue, most hosts would probably have some sort of issue with 10gb upload. Bawolff (talk) 20:23, 5 November 2022 (UTC)
Thats unfortunate, I'm trying to upload interviews and pdfs that could be potentially gigabytes big (although for the time being this might be somewhat rare). Hopefully Bluehost can figure something out for me. Otherwise I'm probably going to have people post big files like that to archive.org and just cite the link on there. Guillaume Taillefer (talk) 20:32, 5 November 2022 (UTC)
Can't get WikibaseImport to work
Latest comment: 3 years ago1 comment1 person in discussion
I am trying to create a local instance of Wikidata. MediaWiki 1.38.4 and Wikibase work fine, but WikibaseImport extension either breaks everything, or simply doesn't work.
There are two github repositories for this extenstion.
to obtain it as stated in README files of both repositories and run
composer update in root mediawiki directory, I get this error:
C:\xampp\htdocs\w>composer update
> ComposerHookHandler::onPreUpdate
Loading composer repositories with package information
Info from https://repo.packagist.org: #StandWithUkraine
Updating dependencies
Your requirements could not be resolved to an installable set of packages.
Problem 1
- Root composer.json requires monolog/monolog 2.2.0, ~1.18, found monolog/monolog[dev-main, 1.0.0-RC1, ..., 1.x-dev, 2.0.0-beta1, ..., 2.x-dev, 3.0.0-RC1, ..., 3.x-dev (alias of dev-main)] but it does not match the constraint.
Problem 2
- Root composer.json requires wikimedia/shellbox 3.0.0 -> satisfiable by wikimedia/shellbox[3.0.0].
- wikimedia/shellbox 3.0.0 requires monolog/monolog ^2.0.0 -> found monolog/monolog[2.0.0, ..., 2.8.0] but it conflicts with your root composer.json require (2.2.0, ~1.18).
Use the option --with-all-dependencies (-W) to allow upgrades, downgrades and removals for packages currently locked to specific versions.
If I'm not mistaken, this kind of errors composer produces when MediaWiki and extension versions are incompatible.
README's in extension's githubs tell I should run composer update in /WikibaseImport/ directory, but this thread and this topic state it shouldn't be done, or root/maintenance/update.php will stop working. If I run composer update in /WikibaseImport/ anyway, update.php indeed stops working:
php C:\xampp\htdocs\w\maintenance\update.php
PHP Fatal error: Wikibase\DataModel\Entity\SerializableEntityId cannot implement Wikibase\DataModel\Entity\EntityId - it is not an interface in C:\xampp\htdocs\w\extensions\Wikibase\lib\packages\wikibase\data-model\src\Entity\SerializableEntityId.php on line 11
Fatal error: Wikibase\DataModel\Entity\SerializableEntityId cannot implement Wikibase\DataModel\Entity\EntityId - it is not an interface in C:\xampp\htdocs\w\extensions\Wikibase\lib\packages\wikibase\data-model\src\Entity\SerializableEntityId.php on line 11
importEntities.php fails with the same output, and instead of pages of MediaWiki in browser there is a "not an interface" Fatal Error message. composer install , composer update and del /s composer.lock in different combinations have no effect - still "not an interface", until I will delete /vendor/ directories in extensions as stated in thread and topic I've mentioned earlier. But deleting them doesn't make WikibaseImport to work.
Please help. I am using MediaWiki 1.38.4, PHP 8.1.10 (apache2handler) and 10.4.25-MariaDB
I've cloned Wikibase from this github branch and followed instructions in it's README file and as stated here - Wikibase/Installation - I have added this to my LocalSettings.php :
and executed git submodule update --init in /Wikibase/ directory.
My w/composer.local.json contains wildcards by default (StackOverflow):
"extensions/*/composer.json",
"skins/*/composer.json" [[User:Podbrushkin|Podbrushkin]] ([[User talk:Podbrushkin|talk]]) 08:13, 4 November 2022 (UTC)
== How to install Wikidiff2 on Windows ==
Please explain how to install [[Wikidiff2]] on Windows (XAMPP).
There is an equivalent question on the [[Talk:Wikidiff2#Windows|discussion page]] for that page, but it has not been answered. [[Special:Contributions/240B:11:2301:FE00:E1AB:4926:B173:F7DA|240B:11:2301:FE00:E1AB:4926:B173:F7DA]] ([[User talk:240B:11:2301:FE00:E1AB:4926:B173:F7DA|talk]]) 09:16, 4 November 2022 (UTC)
:So first off, this will be a lot easier under linux. You should consider using linux.
:On windows you will have to compile it from source (i dont know if anyone has actually tested this on windows). See the manually section of the doc. If you have never done that before you should read a tutorial on compiling php extensions on windows.
:I strongly reccomend just using linux if you need wikidiff2. [[User:Bawolff|Bawolff]] ([[User talk:Bawolff|talk]]) 02:56, 5 November 2022 (UTC)
:Is there a site where I can download the dll for Wikidiff2.
:The .so file compiled on linux (virtual machine) could not be used on windows.
:It is not easy to change the operating system installed on a server. [[Special:Contributions/240B:11:2301:FE00:884E:5FCE:3F9F:AC7B|240B:11:2301:FE00:884E:5FCE:3F9F:AC7B]] ([[User talk:240B:11:2301:FE00:884E:5FCE:3F9F:AC7B|talk]]) 07:31, 5 November 2022 (UTC)
:I am not aware of any precompiled .dll of wikidiff for windows.
:> The .so file compiled on linux (virtual machine) could not be used on windows.
:You have to compile on windows to use on windows (in theory you might be able to cross compile on linux by specifying different compile time options, but that is probably more complicated than compiling on windows) [[User:Bawolff|Bawolff]] ([[User talk:Bawolff|talk]]) 08:22, 5 November 2022 (UTC)
== chmod cannot access files via terminal ==
Whenever I try to run chmod a+x wiki.allan.network/extensions/Scribunto/includes/Engines/LuaStandalone/binaries/lua5_1_5_linux_64_generic/lua it always says that the file was not found. [[Special:Contributions/213.237.94.6|213.237.94.6]] ([[User talk:213.237.94.6|talk]]) 16:03, 4 November 2022 (UTC)
:Maybe use cd to get to the lua5_1_5_linux_64_generic directory then just chmod a+x lua [[User:Jonathan3|Jonathan3]] ([[User talk:Jonathan3|talk]]) 19:19, 4 November 2022 (UTC)
== Upload file error after installation with bitnami ==
I have installed mediawiki with Bitnami, but when I uploaded file using Special:Upload page, it showed the error:
Could not open lock file for "mwstore://local-backend/local-public/b/be/Logohq.png". Make sure your upload directory is configured correctly and your web server has permission to write to that directory.
How to fix this error? [[Special:Contributions/102.47.67.141|102.47.67.141]] ([[User talk:102.47.67.141|talk]]) 19:16, 4 November 2022 (UTC)
:Did you make sure "your web server has permission to write to that directory" as the error suggested? [[User:Bawolff|Bawolff]] ([[User talk:Bawolff|talk]]) 02:45, 5 November 2022 (UTC)
:Do you know how to know if the web server has permission to write to that directory? And how to change the permission to allow it? [[Special:Contributions/102.47.67.141|102.47.67.141]] ([[User talk:102.47.67.141|talk]]) 10:54, 5 November 2022 (UTC)
:It would depend on how your server is configured and what operating system it is using. If you are using linux you can use the chmod and chown command to adjust file system permissions.
:There's some information at [[Manual:Configuring file uploads]] but you would probably be better served by your operating system's documentation. [[User:Bawolff|Bawolff]] ([[User talk:Bawolff|talk]]) 11:45, 5 November 2022 (UTC)
== Editing own content on discussion pages here ==
It seems that I cannot edit the content here once saved:
Next to the title, I'm getting "Edit title" pointing to "..." next to the thread title, but the "..." offers a Permalink only.
So how to edit my own text after saving ? [[Special:Contributions/5.147.163.199|5.147.163.199]] ([[User talk:5.147.163.199|talk]]) 13:22, 5 November 2022 (UTC)
:I meant the "..." next to the content offers a Permalink only. [[Special:Contributions/5.147.163.199|5.147.163.199]] ([[User talk:5.147.163.199|talk]]) 13:29, 5 November 2022 (UTC)
:Anonymous (IP) authors cannot do so. You need to login to edit your own comments. [[User:Clump|Clump]] ([[User talk:Clump|talk]]) 13:53, 5 November 2022 (UTC)
:Thank you. [[Special:Contributions/5.147.163.199|5.147.163.199]] ([[User talk:5.147.163.199|talk]]) 23:59, 12 November 2022 (UTC)
== Editing search results' "Content pages" bar ==
MW 1.31. After running a search a bar appears on top with
'''Content pages''' Multimedia Everything Advanced
and each one of the 3 items linked for more specific site searches. Now, I would like to adapt '''Content pages''' so that it looks like the one below (providing an option to search the query on Google). I tried editing MediaWiki:Searchprofile-articles which contains the '''Content pages''' string as below but it was not parsed.<pre>
Look up on [https://www.google.com/search?q={{urlencode:$1}} Google] | Content pages
Wikidata doesn't precisely support the exact same information as wikipedia, just very similar information in an easier to query fashion.
In particular, commons category doesn't correspond to the category the item is in on wikipedia, but the item that describes the topic of the category. Bawolff (talk) 07:10, 6 November 2022 (UTC)
Is it really not possible to make TimedMediaHandler play other video files like .mov and .wmv?
Latest comment: 3 years ago3 comments2 people in discussion
Hello, I was told a while back that it isn't possible for TimedMediaHandler to play .mov and other video files that aren't mp4, mp3, webm, and other files it allows. However I wanted to ask to make sure if it really isn't possible to somehow make TimedMediaHandler play .mov, .wmv, and other files? Thanks Guillaume Taillefer (talk) 21:29, 5 November 2022 (UTC)
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
The discussion above is closed. Please do not modify it. No further edits should be made to this discussion.
Thumbnail errors w/ PDF file
Latest comment: 3 years ago6 comments2 people in discussion
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
I have pdfhanndler extension installed. 1.38
when uploading a PDF file, I am getting the following error on the File page: (see below)
Do I need to compile/install imagemagick - I prefer not - is there another workaround?
** this post was updated to reflect the current error code **
>>>
Error creating thumbnail: convert: no decode delegate for this image format `' @ error/constitute.c/ReadImage/741. convert: no images defined `/tmp/transform_cb79f2dd030d.jpg' @ error/convert.c/ConvertImageCommand/334
I went ahead installed image magick :( I se these parameters in local settings:
$wgUseImageMagick = true;
$wgPdfProcessor = "/usr/bin/gs";
$wgPdfPostProcessor = "/usr/local/bin/convert";
$wgPdfInfo = "/usr/bin/pdfinfo";
$wgPdftoText = "/usr/bin/pdftotext";
I am still getting an error - but different - below:
Error creating thumbnail: convert: no decode delegate for this image format `' @ error/constitute.c/ReadImage/741. convert: no images defined `/tmp/transform_cb79f2dd030d.jpg' @ error/convert.c/ConvertImageCommand/3342 DLl314 (talk) 17:55, 6 November 2022 (UTC)
did you configure image magick to not support jpg files? (I think you can do this at compile time or via a xml config file, altthough its a very unusual thing to do) Bawolff (talk) 20:31, 6 November 2022 (UTC)
I didn't remove jpg support - The GIT discussion suggested I might be missing the priory library, but I am not sure how to fix this.
if you are a user, and not an administrator - search User Styles or Custom User Style - you will be able to update css for your user only DLl314 (talk) 18:30, 6 November 2022 (UTC)
I don't think DLI314 is correct here. A 403 error likely has a different cause.
Do you (or your web host) have apache mod_security or another Web application firewall (WAF) installed? Your host might call it something else like "hacking protection" Bawolff (talk) 20:28, 6 November 2022 (UTC)
@Bawolff this kind of fixed it. whenever a user tries to preview it, it will show a 403 even though I disabled modsecurity, and second of all. is there a way to fix both the issues without disabling security since there is login pages. 213.237.89.238 (talk) 07:33, 7 November 2022 (UTC)
You can adjust your configuration of mod_security. Your apache error log may have more details as to what is being triggered.
only happening on preview suggests some reflected xss filter. I didn't think mod_security supported that but i could be wrong. Maybe you have some sort of additional filter enabled.
The security provided by mod_security is significantly overated in my opinion. I personally don't think its really worth using. Bawolff (talk) 07:47, 7 November 2022 (UTC)
Lua error in package.lua at line 80: module 'strict' not found.
Latest comment: 3 years ago6 comments3 people in discussion
I keep getting the error when uploading new templates. In particular with source templates. When I tried importing them into my wiki, they were successful, but when I click on each template, it gives, or repeats the following error throughout:
Lua error in package.lua at line 80: module 'strict' not found.
I looked this error up and apparently only one person has had the problem:
However I don't want to have to go through trying to find all the older modules that don't have strict, and have already redownloaded Scribunto to the "new version". So is there an actual way to solve this? Thanks Guillaume Taillefer (talk) 00:24, 7 November 2022 (UTC)
> already redownloaded Scribunto to the "new version"
Which version did you download? You would probably need master.
This has been backported now. So it should be included in the version of scribunto for your version of mediawiki if you re-download instead of just master. Bawolff (talk) 06:09, 7 November 2022 (UTC)
Oh wow now it works, Thanks! I had redownloaded the version for 1.38 a couple of hours before I had posted this when I reinstalled it. Now it no longer gives me the error Guillaume Taillefer (talk) 11:31, 7 November 2022 (UTC)
Mediawiki official image(MediaWiki-Docker) works for Azure?
Hi Support Desk,
Currently We're using following mediawiki v1.35.3 with the required software under RHEL7 server(on-premise)
Now, we plan to move our on-premise wiki server to the Azure cloud. Hence we do require your suggestion on migrating same with the mediawiki official image(MediaWiki-Docker). please suggest appropriate image that would contain all required latest software.
Installed software
Latest comment: 3 years ago1 comment1 person in discussion
I'd probably go for 1.35 for now and move to 1.39 after it's been out for a while and once SMW catches up. Both are LTS versions. I don't use SMW so can't comment with any authority on that (I use Cargo). I can speak with authority on having a wiki running on an obsolete MW version :-) I'm on 1.34 and plan to take the upgrade path set out here when I get a chance. Jonathan3 (talk) 09:20, 8 November 2022 (UTC)
Add checkbox in Edit Source / Edit Page
Latest comment: 3 years ago5 comments3 people in discussion
I'm wanting to add a checkbox on both the Edit and Edit Source pages for the purpose of some kind of auditing. I'm not sure which extension or where to modify and I know that I will also have to somehow save whether it was checked or not.
Has anyone done anything like this or have any advice on where to start? Any help would be greatly appreciated. Also, my apologies if this is better posted in the developer hub DrpepperUser (talk) 22:53, 7 November 2022 (UTC)
If you are looking for examples of people who have done similar things:
Thank you for the response! I was looking for examples but more so a direction on where to make changes on existing code, where the edit options code already lives. If I was able to use the FlaggedRevisions to add a checkbox, where would that code go? And I do realize that approach may not be recommended. What do you think?
You should be able to directly upgrade. Your software versions also look fine. PHP 7.4 will soon be out of support from php upstream, but we still support it in MediaWiki 1.39 as that is what wikipedia is using.
Latest comment: 3 years ago1 comment1 person in discussion
我在试图使用Lua脚本创建module时遇到了以下的报错：
Lua error: Cannot create process: proc_open(/dev/null): failed to open stream: Operation not permitted
在我debug时，我始终得到以下的错误提示：
Warning: is_dir(): open_basedir restriction in effect. File(/tmp) is not within the allowed path(s): (/home/admin/web/mywikisite/public_html:/home/admin/tmp) in /home/admin/web/mywikisite/public_html/mywiki/includes/libs/filebackend/fsfile/TempFSFile.php on line 85
Latest comment: 3 years ago1 comment1 person in discussion
The restbase server is not recognizing it and when I open the visual Editor it comes up with this message "Revision IDs returned by the server do not match (document: undefined, metadata: 21)." What is it? And how do I solve it? WikiUser22222 (talk) 14:27, 8 November 2022 (UTC)
Special:Categories - List all on any page
Latest comment: 3 years ago2 comments2 people in discussion
How can I list all Categories at the bottom of a page (eg template:footer page) the same way they are listed on the bottom of the Special:Categories page ???
Is it possible to use the Categorytree extension to do it? I've tried magic word and using <categorytree> but I can only get it to work if I specify one - using "all" returns nothing. DLl314 (talk) 14:38, 8 November 2022 (UTC)
Latest comment: 3 years ago3 comments2 people in discussion
after using nuke to delete spam content from a wiki (created with mediawiki 1.35), deleted content was not moved to the archive table. records are present in the text and content table, but nothing in the revisions or archive table.
maintenance scripts such as deleteArchivedRevisions.php and purgeOldText.php do not find the deleted content (there is nothing in the archives to delete, and purgeOldText only compares the content and text table, without checking whether corresponding entries exist in other tables).
is this the intended behaviour ? most help articles that people will find when trying to clean up a vandalised wiki suggest deleting the pages and then running deleteArchivedRevisions.php, but this no longer works -- at least in my case (not sure if others have experienced this). 192.222.162.46 (talk) 19:44, 8 November 2022 (UTC)
any idea whether my situation is out of the ordinary ?
I'm trying to figure out whether I've made an error somewhere along the way, or if the changes to the database since 1.35 mean the standard spam clean up technique (nuke, then deleteArchivedRevisions, then removeUnusedAccounts) just doesn't work anymore 192.222.162.46 (talk) 17:16, 9 November 2022 (UTC)
Installed MediaWiki Successfully - but keep getting redirected to localhost
Latest comment: 3 years ago3 comments2 people in discussion
Its possible that the installer detected hour server name wrong. Can you check the value of $wgServer in your LocalSettings.php file. Bawolff (talk) 01:41, 9 November 2022 (UTC)
Accidentally deleted an entry from revision table from my database in phpMyAdmin
Latest comment: 3 years ago2 comments1 person in discussion
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Hello.
I accidentally deleted an entry from revision table from my database in phpMyAdmin and now I can no longer view, delete, move, etc. the corresponding page.
When I open the page it behaves like an empty redlink page and then when I try to create it I get an edit conflict displayed and then get caught in an endless loop.
When I try to move the page I get the following message:
Ok, I've got it working on my own now. I simply entered the previous revision in the page table at the corresponding entry under page_latest, now I can view and edit the page again. 178.8.102.218 (talk) 07:57, 9 November 2022 (UTC)
The discussion above is closed. Please do not modify it. No further edits should be made to this discussion.
Where to find shell?
Latest comment: 3 years ago6 comments5 people in discussion
I want to install SemanticResultFormats (after installing Semantic Mediawiki)
It says to run "mediawiki/semantic-result-formats": "~4.0" from the shell.
I could only find shell.php on my server, but i get the following error
That also looks like a composer.json entry not a shell command, are you sure you are reading whatever instructions you are reading correctly? Bawolff (talk) 14:13, 9 November 2022 (UTC)
Instructions are what is reccomended when downloading SemanticResultFormat.
Latest comment: 3 years ago8 comments3 people in discussion
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
I have attempted to edit a single page via Common.css but it seems to work for some things and not others.
The below works when referencing the page and the element I want to target (if I wanted to change all pages, I can remove '.page-Index':
e.g. .page-Index .firstHeading {color:white;!important}
But when I run this, the background doesn't apply to just Index page referenced:
e.g. .page-Index body, html background-color: white;background-image: linear-gradient(to left, transparent, #202b5b), url('exampleimage.jpg'); background-repeat: no-repeat; !important}
Is it possible I have some syntax incorrect here? I'm confused as to why my background can't follow the same page rules I set as other elements. David.bucknor (talk) 14:46, 9 November 2022 (UTC)
I've not looked in detail but are you missing an opening {?
In the copy and paste yes, but I have got the missing bracket opening '{' in the actual thing.
You can change background image via CSS in Common using URL like normal - just unsure why I can only set it for every page despite specifying I only require it for index!
.page-Index body, html { background-color: white;background-image: linear-gradient(to left, transparent, #202b5b), url('exampleimage.jpg'); background-repeat: no-repeat; !important} e.g. adding a {, the reason its applying to everything is you have the , html part, which means all pages. The .page-Index body matches no pages (page-Index is a class on the body element, it will never also have a body element as a child), it should be just .page-IndexBawolff (talk) 02:45, 10 November 2022 (UTC)
This is a brilliant explanation, thank you - code amended and working as intended now - just have to sort out that CM of white space at the top of the page :) David.bucknor (talk) 10:49, 10 November 2022 (UTC)
The discussion above is closed. Please do not modify it. No further edits should be made to this discussion.
How to stop VisualEditor from using images from wiki commons
Latest comment: 3 years ago4 comments3 people in discussion
Whenever I try to insert a file in an article through wiki commons, and try to search for it, it always comes with results from Wikimedia Commons. However I do not was my users to use images from Wiki Commons, nor do I want them to get confused with an image that is also on Wikimedia Commons. How could I disable VisualEditor from doing that, and to only use files from my wiki? Thanks Guillaume Taillefer (talk) 22:40, 9 November 2022 (UTC)
Ok it works now. I did have UseInstantCommons set to false apparently twice but then I had it another time where it was set to true. I deleted all of them except for one false one and now it works Guillaume Taillefer (talk) 11:44, 10 November 2022 (UTC)
Latest comment: 3 years ago2 comments2 people in discussion
I am adding subpages to my MediaWiki and the breadcrumb trail does not display, despite adding the appropriate line to the LocalSettings.php file $wgNamespacesWithSubpages[NS_MAIN] = true;
I'm not any good at this sort of thing! I got my wiki to work based on the "short URLs" page, which is the height of it :-) Jonathan3 (talk) 22:10, 11 November 2022 (UTC)
If you are redirecting to existing short url (and not creating the short url in the first place) then nothing needs to change in LocalSettings.php
So if example.com/wiki/article1 already works,no change would be needed for LocalSettings.php.
I dont think the flags are right in what Johnathan posted. Having Last flag, but not making it an external redirect will probably prevent the short url from being processed. Maybe try instead.
Calling Loop index inside #if (Parser functions and LoopFunctions extensions)
Latest comment: 3 years ago3 comments2 people in discussion
I wanted to create a condition inside a for loop that used the index of the loop. According to the LoopFunctions documentation, one can call the index with $n$. This works inside the for, but not if it is inside a parser function that is inside a for loop. Is there something I am missing or is it simply not possible to use the index inside another function? Jdm.oliveira (talk) 15:29, 10 November 2022 (UTC)
Latest comment: 3 years ago1 comment1 person in discussion
I have been working with LinkedWiki and have queried some endpoints to list some properties' values. Since the things I have queried have the same property names as my own page, I would like to simultaneously query my properties and the other endpoints' and list what is returned either in a map or a table. Is this by any chance possible? Jdm.oliveira (talk) 15:52, 10 November 2022 (UTC)
Recover old wiki versions
Latest comment: 3 years ago3 comments3 people in discussion
I'd like to know if there is a way to recover old wiki versions.
Latest comment: 3 years ago4 comments3 people in discussion
Hi, I'm trying to find instructions for importing a MySQL database that has been extended with additional tables. I am installing 1.38 replacing a 1.18 installation. Nothing was output when update.php was run.
Since Version 1.36, MediaWiki only commits to supporting upgrades from two LTS releases ago (see phab:T259771). Upgrades from older versions of MediaWiki will have to be performed in multiple steps. This means that if you want to upgrade to 1.36 from 1.23 or earlier, you'll first have to upgrade your 1.23 wiki to 1.27 (or 1.35), and, from 1.27 (or 1.35), you'll be able to upgrade to 1.36.
That said, if nothing was output when update.php was run, that probably means that there is something wrong with how it was run, since normally there would be an error message telling you to upgrade to 1.35 first. Check you have php error reporting enabled. Bawolff (talk) 07:43, 12 November 2022 (UTC)
Ah I see. The issue is to upgrade PHP (at 5.6.40) to 7.3.19 to 7.4.3, respectively to update the schema from 1.18 to 1.27 to 1.35+, running update.php in each of the three PHP environments. Further MySQL 5.0 is necessary to upgrade to 1.27 while MySQL 5.5.8 is used thereafter. ... I believe the error 255 was due the PHP mismatch. Belltowerwikis (talk) 20:24, 12 November 2022 (UTC)
Module not working
Latest comment: 3 years ago5 comments2 people in discussion
Hello. I have Scribunto installed as an extension and I made a page called module:message box and pasted local messageBox = require('Module:Message box'); in it but whenever I type {{ambox}} it gives me a red link. Can anyone help me out? 213.237.90.36 (talk) 11:37, 12 November 2022 (UTC)
Did you copy over Template:Ambox ? Use w:special:export on wikipedia, selecting the "include templates" checkbox, and Special:Import on your wiki, to make sure you have all the neccessary templates.
Additionally making a page named module:message box and just typing local messageBox = require('Module:Message box'); probably won't really work because that's just including itself. Bawolff (talk) 17:02, 12 November 2022 (UTC)
it works now but there's still some broken stuff in the pages but it also imported the documentation from wikipedia. and how to mass delete the pages because its not working. 213.237.90.36 (talk) 08:57, 13 November 2022 (UTC)
Weird category bind glitch
Latest comment: 3 years ago3 comments2 people in discussion
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
After my previous issues were fixed, it's largely been plain sailing from there, except for an annoying problem I've discovered with my revised "Song" template.
I know exactly what's causing the issue, since only songs that have article links are broken in this manner. However, I would like a solution that keeps this article link, since these songs are important enough to warrant their own articles. AlexGRFan97 (talk) 13:10, 12 November 2022 (UTC)
You could separate the songname parameter into multiple parameters - {{{artist}}}, {{{songname}}}, {{{songpage}}} and then do something like
<small>▶️ '''{{{artist}}} - {{#ifexist:{{{songpage}}}|[[{{{songpage}}}|{{{songname}}}]]''': {{{playswhen}}}</small>[[Category:Automotive journalism media which feature the track "{{{songname}}}" in its soundtrack]]<br>Bawolff (talk) 17:00, 12 November 2022 (UTC)
Can confirm it works now, but I did have to make a couple changes. Namely, the square brackets for the article link were in the wrong place, causing "songname" not to print anything if "songpage" was left empty.
The discussion above is closed. Please do not modify it. No further edits should be made to this discussion.
Plain text after new install
Latest comment: 3 years ago12 comments3 people in discussion
Downloaded and installed mediawiki to my Apache24/PHP8 site moments ago. Went through all steps of the config, downloaded LocalSettings, put it in the main directory, went to the site and saw plain text with no formatting or no images whatsoever.
The resource from “https :// www . wiki . local/wiki/load.php?lang=en&modules=skins.vector.styles.legacy&only=styles&skin=vector” was blocked due to MIME type (“text/html”) mismatch (X-Content-Type-Options: nosniff).
I cannot see any PHP errors but the following Apache error is raised for every access to Wiki:
Premature end of script headers: load.php, referer: https://www.mywiki.local/wiki/index.php?title=Main_Page
Googled for it but found mostly nodejs and other strange server issues. What should I do to allow Mediawiki to load properly? Furfarthl (talk) 13:50, 12 November 2022 (UTC)
Try this, and then send the result if there is any
Tried it but it did not write any errors to PHP error log.
I've scanned the debug log but found no error messages. Word Error is not found, Word Err is only found in Deferred or UserRights. See, pal: the problem is that if I see an error in Apache's error.log then PHP did not execute at all, and there will be nothing of value/interest in any of WM or PHP logs.
premature end of script headers makes it sound like either php is hitting some sort of timeout or php is segfaulting.
I would check what type of timeouts you have set. If this is the first access, mediawiki has a cold cache, and is maybe taking longer than a normal request as it has to generate a bunch of stuff. Bawolff (talk) 16:53, 12 November 2022 (UTC)
Error 500 occurs before any possible timeouts, within 1900 msec of request.
The next line in the debug window starts at the same time of 1200 and takes 200 msec longer but succeeds.
It depends how things are configured. Its unlikely that anyone would intentionally set a timeout that low, but if everything was configured correctly you wouldn't have an issue. However at first glance it seems less likely given that info.
> The next line in the debug window starts at the same time of 1200 and takes 200 msec longer but succeeds.
Which debug window?
Other investigative options start to get complicated. You could try to use something like strace to figure out what is going on with the php process. In particular determine if the process is segfaulting or something like that.
> the problem is that if I see an error in Apache's error.log then PHP did not execute at all
I dont think this is true. The error you posted suggests php didn't run to completion, not that it neccesarily didn't run at all. Bawolff (talk) 19:24, 12 November 2022 (UTC)
Browser debug window, F12. This is quickly becoming way too technical. Seeing how Wordpress runs just fine off the same virtual host I doubt this issue has anything to do with Apache or PHP. The root .htaccess could theoretically mess things up, true, but after I temporarily removed it Wiki still repeated the same error. Furfarthl (talk) 21:58, 12 November 2022 (UTC)
You would not be getting the apache error you mentioned for rewrite rules or .htaccess.
The "premature end of script headers" error you mentioned is an error about apache communicating with php. It is way below the level mediawiki operates at. Bawolff (talk) 22:12, 12 November 2022 (UTC)
I moved the whole wiki folder to another virtual host that runs PHP7.4 and has little in the way of .htaccess:
<IfModule mod_headers.c>
Header set X-XSS-Protection "1; mode=block"
</IfModule>
and it does not load there as well. The PHP site on that host also runs perfectly fine but not Wiki. We run webmail, CMSs, admins for MySQL and other such things and never had any issues like this one. So, why can Apache communicate with PHP for Wordpress and a handful of other apps but not for Wiki? What are Wikimedia requirements that my host may not be satisfying?
Also, if this is about Apache not communicating with PHP then why did Wiki installation run perfectly fine but not the site itself? They run off the same vhost and same configuration. Why does load.php generate one CSS perfectly fine but not the other?
Quick question: I can see file l10-en-cache.cdb under /wiki/cache. Is it an SQLite DB? I tried to open it with SQlite driver, to validate if it is not corrupt, but it errored that it is not a valid SQLite database.
Is PHP-CGI mode that we use fine by Wikimedia? I googled about this error and found that quite a few users had issues running Wiki under fastCGI, but I do not have any errors mentioned, like mismatch between #of children between httpd.conf and php.ini etc. Furfarthl (talk) 22:17, 12 November 2022 (UTC)
I am not an expert on Procmon but it looks like there are no file permission issues in the below capture of everything that is not Result=SUCCESS for httpd.exe:
Fast-cgi and normal cgi (not the same thing) should both be supported, with fastcgi used by many mediawiki websites (including this one).
it is unlikely what you are describing would be caused by a file permission error. Its also much more likely to be an error on the php process not the httpd process.
> Also, if this is about Apache not communicating with PHP then why did Wiki installation run perfectly fine but not the site itself? They run off the same vhost and same configuration. Why does load.php generate one CSS perfectly fine but not the other?
you would need to narrow it down much further than that to say anything useful (does the php process exit abnormally? Does it happen on a specific line? Is there something significantly different about the response to that page, e.g. a header that is abnormally long, etc).
if it also fails in the same way for a different version of php, that makes it less likely to be a php bug, and more likely to be a config error, but there is still a question of why is it only happening to you and not other people.
> Quick question: I can see file l10-en-cache.cdb under /wiki/cache. Is it an SQLite DB? I tried to open it with SQlite driver, to validate if it is not corrupt, but it errored that it is not a valid SQLite database.
No, its a CDB file as the extension would indicate. CDB is not as popular as sqlite, but its still a popular format and you should be able to find reader software on the internet if you are curious. Mediawiki can also be configured to use other formats for l10n cache.
In any case i dont think corruption would cause the issue you describe, but its also safe to delete that file in which case it should be regenerated on next load. Bawolff (talk) 06:17, 13 November 2022 (UTC)
Thanks for your insight and tips! Seeing how involved the installation is and how vague requirements are, I'll have no choice than to give up :( Furfarthl (talk) 14:26, 14 November 2022 (UTC)
Weird keyboard on automated testing with Cypress
Latest comment: 3 years ago2 comments1 person in discussion
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
I'm working on code that will test regressions in Polish Dark Theme (plugin/extension) and I have problem to login to Wikipedia. First login works but second login looks weird. It shows some kind of IME keyboard and probably auto submit a form without clicking login button or pressing enter.
Here is the link to discussion on Cypress project on GitHub. Can anyone tell who is to blame? Is it Cypress issue or maybe MediaWiki is auto submitting the form without touching the login button?
I'm testing using Google Chrome.
Any hints how to fix the issue would be great.
I've investigated this problem a bit and it seems that my testing browser has MediaWiki IME keyboard present that somehow redirect to login page without pressing enter.
The discussion above is closed. Please do not modify it. No further edits should be made to this discussion.
MediaWiki login protection
Latest comment: 3 years ago2 comments2 people in discussion
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Does Media Wiki (specifically Wikipedia) have some kind of login protection. When using automated code to login to Wikipedia my second login got blocked by some kind of security protection. Got this message in Polish:
Wydaje się, że wystąpił błąd z Twoją sesją zalogowania; to działanie zostało anulowane, aby uniknąć przechwycenia sesji. Prześlij formularz jeszcze raz.
I'm helping someone to write tests for Dark Mode extension. But I'm not able to login second time. This is specific user created just for tests. His password is protected and will not be exposed anywhere. I plan to add tests to GitLab CI/CD, but the password is added as secret variable so it will not be visible in Code. Jcubic (talk) 23:05, 12 November 2022 (UTC)
No we dont. Most likely there is a bug in your automated login code. That error happens if you send the wrong cookies or post parameters. Bawolff (talk) 05:59, 13 November 2022 (UTC)
The discussion above is closed. Please do not modify it. No further edits should be made to this discussion.
Help with Extension:TreeAndMenu
Latest comment: 3 years ago1 comment1 person in discussion
Hello, I am new to MediaWiki. Trying build selfhosted server. I installed it seems to be working fine. I have also installed the extension (Extension:TreeAndMenu#) . I am trying build sidebar menu system similar to this website (https://wikizilla.org/wiki/Main_Page), but so far failed. I have read the instruction on the extension page, but the instructions assumes i am familiar with MediaWiki, hence so far no success. can anyone please guide me or send a sample menu code where i can copy and paste it? I am happy to learn, but at the moment i dont know where to start. TIA 86.161.1.21 (talk) 15:39, 13 November 2022 (UTC)
Remove Cune Logo
Latest comment: 3 years ago2 comments2 people in discussion
Hello, I will like to remove or update the logo from our company as it has been updated and the one published is not the correct:
Latest comment: 3 years ago7 comments3 people in discussion
In the comments of the file tests/parser/parserTests.txt (Wikimedia) it is written that the file includes context from metawiki pages. The GNU FDL and CC BY-SA 3.0 licenses do not allow inclusion of content into the GNU GPLv2. Dušan Kreheľ (talk) 10:45, 14 November 2022 (UTC)
Do the examples have enough creativity to be eligible for copyright? I would assume they would be public domain for being non-creative.
This might not be the best forum to discuss this. It might be better to file a task on phab (tagging legal i guess?) Or start a thread on wikitech-l. Bawolff (talk) 11:14, 14 November 2022 (UTC)
@Bawolff: Wikimedia Mediawiki does not have an official standard for wiki syntax. This means that everything that the Wikimedia Mediawiki software says, acts or does is under the Copyright of Wikimedia Mediawiki, i.e. GPLv2. The exception may be instructions (not part of the official Wikimedia Mediawiki software package), due to the legal exception for educational purposes or non-profit. The copied text serves as a standard for the Mediawiki parser, so if someone wants to develop a new parser, then if he wants to be sure of the correct implementation, he must test his parse on identical data. So, these few lines play an important role in the following. Dušan Kreheľ (talk) 12:29, 14 November 2022 (UTC)
> This means that everything that the Wikimedia software says, acts or does is under the Copyright of Wikimedia
I strongly doubt this interpretation. In particular, I don't think the copyright of any code contributors of MediaWiki extends to the output of the MediaWiki software package. Additionally, "Wikimedia" (Different from Wikimedia foundation) is not a legal entity and is not one of the copyright holders in question.
Anyways, to the topic at hand. I don't think https://meta.wikimedia.org/wiki/Parser_testing/strip and friends meet the threshold of originality. However even if they did, those pages was written mostly by Tim Starling and Erik Moller, who are also mediawiki contributors, and presumably consented to it being re-licensed when it was submitted to MediaWiki, making it a moot point. Bawolff (talk) 13:20, 14 November 2022 (UTC)
False-positive session save path error from 1.35.8
Latest comment: 3 years ago7 comments3 people in discussion
After having failed to troubleshoot error 500 from load.php of the latest version, I tried the oldest but could not start the setup:
Your session data was lost! Check your php.ini and make sure session.save_path is set to an appropriate directory.
It is not true. The session path is set properly, and I can write files to it, with this simple test script:
<?php
try {
$sp = session_save_path();
echo "<p>$sp</p>";
$fil = fopen($sp."/test", "w");
fwrite($fil, "success");
fclose($fil);
echo "<p>success</p>";
} catch(Exception $ex) {
echo "<p>Error: $ex->getMessage()</p>";
}
?>
It prints 'success', and the file is created. Other applications on the same virtual host use sessions all the time, and I can see their session files in that folder being updated as users log in. Any ideas?
When I open the installation page, 2 files are created:
Great question! It is probably a confusion over terminology. By starting the installer I mean proceeding past its 1st screen, i.e. seeing something after clicking Continue. Furfarthl (talk) 13:28, 15 November 2022 (UTC)
I am just a humble user... I do not know all of this stuff: what is reproducible, what is minimal, what is test case etc. It either works or it does not. So, I've spent obscene amount of time on this task that is supposed to be trivial and got absolutely nowhere. Now I am under pressure to deliver at least some documentation CMS, and I have to go. Sorry! 72.10.138.14 (talk) 13:55, 16 November 2022 (UTC)
Your not obliged to do anything, but a bug report without detailed instructions on how to make the bug appear for the person reading the report, are usually ignored. Bawolff (talk) 17:26, 16 November 2022 (UTC)
TOC table of contents not showing
Latest comment: 3 years ago3 comments2 people in discussion
Hi,
My TOC is not showing on any of my pages. Does anyone have an idea where I can start looking to solve this?
Latest comment: 3 years ago1 comment1 person in discussion
I want to add a banner at the far top of the page, basically behind the "message" and "discussion" and other tabs. Is there a CSS command which will put a picture here? Encyclovideo (talk) 13:56, 15 November 2022 (UTC)
How can I get the same skin as on mediawiki.org?
Latest comment: 3 years ago4 comments3 people in discussion
There is like a legacy vector and a 2022 vector that are separate skins with the same name. Maybe you need the master version of vector skin. Bawolff (talk) 23:08, 17 November 2022 (UTC)
Many thanks for the answer. I'm using the Vector 2022 skin. But if this is MediaWiki's default skin, why does the skin on www.mediawiki.org looks better? Are you really using the same skin?
Vector 2022 is under active development, this site might just have a newer version or a different config (sorry, i havent fully been following the development of vector 2022). Bawolff (talk) 20:32, 18 November 2022 (UTC)
Error contacting the Parsoid/RESTBase server (HTTP 400)
Latest comment: 3 years ago4 comments4 people in discussion
I get a pop up saying “Error contacting the Parsoid/RESTBase server (HTTP 400)” whenever I try to use visual editor. I’m logged in as an administrator account. 185.85.44.153 (talk) 11:13, 16 November 2022 (UTC)
Another cause of this issue is wrong Short Url configuration related to apache.
One more possible is the configuration of /rest.php/?$query_string server block located within your apache available site configuration file, 5.193.68.230 (talk) 08:37, 24 November 2022 (UTC)
Add OOUI icon to wiki page
Latest comment: 3 years ago1 comment1 person in discussion
Latest comment: 3 years ago1 comment1 person in discussion
RESOLVED
Look like compatibility, mediawiki need to be
{
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Hi,
I copied pluggable sso fodler in to /extensions and added this wfLoadExtension( 'PluggableSSO' ); in to localsettings.php. i am unable to locate it in "Special:Version", am i missing something here? it is running on windows Ramvenkey6278 (talk) 03:47, 17 November 2022 (UTC)
The discussion above is closed. Please do not modify it. No further edits should be made to this discussion.
Performance for importing big dumps
Latest comment: 3 years ago2 comments2 people in discussion
Can I run reboottextindex.php, rebootrecentchanges.php and refreshLinks.php in parallel?
Currently we are using above Software, But from security Team we got alert that PHP 7.4 Security Support going to end in 10 days, 28 NOV 2022 .https://www.php.net/supported-versions.php
Also if we want to upgrade PHP version to PHP 8, then MediaWiki is not compatible with PHP 8 clearly mention in Compatibility#PHP.
As we can see MediaWiki (1.38) runs on PHP 7.4.3 and higher. But now PHP 7.4 is going to expire .
Please suggest us what we need to do ? Pooja2425 (talk) 08:23, 17 November 2022 (UTC)
You should be fine using php 8 or 8.1 with mediawiki. Some minor fixes for php 8 were included in 1.38, but even in 1.35 there aren't really any important bugs.
Also if you made unspecified changes to the extension, its much less likely anyone is going to help as we have no idea what you changed or if your issues are the result of the changes you made. Bawolff (talk) 22:51, 25 November 2022 (UTC)
Error contacting the Parsoid/RESTBase server (HTTP 400)
Latest comment: 3 years ago1 comment1 person in discussion
I get a pop up saying “Error contacting the Parsoid/RESTBase server (HTTP 400)” whenever I try to use visual editor. I’m logged in as an administrator account. 185.85.44.154 (talk) 09:27, 18 November 2022 (UTC)
Table do not fit to small images
Latest comment: 3 years ago1 comment1 person in discussion
Hello,
I'm working with last version of MediaWiki, using VisualEditor to edit pages.
I have created a table ("wikitable" when checking wikicode). When inserting images, the table dimension automatically fit to the image size. Except if the image is small.
Regardless of the image size, the width of the column with the image is minimum 200pixel (even for 25px images). What can I do to make the table fit even small images?
Maybe another solution would be not to use images. I would like green check and red cross (for a yes or no) in the table. So I am using images bur maybe there is another solution.
I don't know for sure, but i think mapsnapshot uses the geoshape entry of $wgGraphAllowedDomains, not a mapsnapshot entry. Try adding an entry for geoshape. Bawolff (talk) 18:16, 19 November 2022 (UTC)
mapsnapshot:///?width=__&height=__&zoom=__&lat=__&lon=__ [&style=__] - Makes a link to the map snapshot service, so that a static map can be drawn for a given location/zoom/size. Parameters are converted it into a snapshot image request for Kartotherian -
It's very frustrating, I know the data is there, the SQL database is intact, but the MW install is just not talking to it, or crashing because of this error.
Instead. If for some reason you really wanted the domain there, which is not reccomended,it would need to start with http:// Bawolff (talk) 13:14, 19 November 2022 (UTC)
MediaWiki not loading correctly after installation
Latest comment: 3 years ago10 comments3 people in discussion
Fresh server, mediawiki insalled by decompressing the zip in root
Latest comment: 3 years ago2 comments2 people in discussion
Greetings. Is it possible to edit my wiki's HTML? If no, how do I add new page elements, or, for example, move the navigation on the left to right? Maybe the question is dumb, but I'm new to MediaWiki, so sorry. Ферстикс (talk) 12:30, 20 November 2022 (UTC)
Depends what parts.
If you want to adjust the navigation interface, you can add a new Skin
Latest comment: 3 years ago8 comments5 people in discussion
I attempted upgrade (actually fresh install) to 1.38.4 [PHP 8.14] and call to "SpecialPages" (to verify update version) got me...
[Y3qbAtsZ37nOZRri8h6B9wAAAPA] /wiki/index.php/Special:SpecialPages ArgumentCountError: Too few arguments to function MovePageForm::__construct(), 0 passed in /home4/pedsane1/public_html/wiki/extensions/Translate/tag/SpecialPageTranslationMovePage.php on line 31 and exactly 11 expected
Backtrace:
from /home4/pedsane1/public_html/wiki/includes/specials/SpecialMovepage.php(116)
[Y3rTAxVaEs1V1iVzJncf6AAAAXU] /wiki/index.php/Special:SpecialPages ArgumentCountError: Too few arguments to function MovePageForm::__construct(), 0 passed in /home4/pedsane1/public_html/wiki/extensions/Translate/tag/SpecialPageTranslationMovePage.php on line 31 and exactly 11 expected
Backtrace:
from /home4/pedsane1/public_html/wiki/includes/specials/SpecialMovepage.php(116)
Generally reccomend upload to a separate directory and then transfer over, since if there is a mix of different versions it can cause weird issues. Bawolff (talk) 14:07, 21 November 2022 (UTC)
I understand....thought I did that....but I guess I pushed the extensions (and images, LocalSettings.php) to the new install without adequately checking. Totally my fail.
We generally discourage links (and other formatting) as it often makes things harder to translate, however they are allowed if it makes sense.
My larger concern is that its using an interwiki link, but there is no garuntee what the interwiki map of a third party wiki would be. Bawolff (talk) 00:50, 23 November 2022 (UTC)
Does it make sense? It depends, Wikimedia projects usually adopt the template:Soft redirect to link to other project help pages or use specific graphic formats to avoid confusion in users. So if you changed your policy at least a related mass message should be sent to the Wikimedia wikis. Pierpao (talk) 10:06, 23 November 2022 (UTC)
How to put a Page into a Namespace in own wiki?
Latest comment: 3 years ago2 comments2 people in discussion
Latest comment: 3 years ago6 comments4 people in discussion
I've recently started a new wiki and found some extensions that would provide some much-needed functionality to my wiki. Trouble is, they're for an older version. I would be willing to update them myself with a little bit of help. These extensions both use magic words and implement custom parser functions, and I've been having a bit of trouble trying to figure out what exactly I would need to modify and how to get them up and running.
I did manage to get them connected briefly, but they return magic word errors when I load up the site, so I guess that's a good starting point in trying to figure out how to update all of this. Any help or pointers to similar projects that have the changes that need to be made laid out clearly would be really helpful. 2600:1700:1AD9:180:FC50:73D2:EB2:6C81 (talk) 22:40, 22 November 2022 (UTC)
Thankfully they're pretty small and basic (I think). They are used to create flexible navigation between article pages. The first one creates a parser function that correlates to a magic word. This word can then be used to read through a list and spit out certain values based on that list. The other takes the actual item being pointed at in the list and converts it to a usable URL portion that you can add to the end of a link to point to a specific page on or off the wiki.
They're used over on the HR Wiki (under Help:Parser functions) and one of the admins was kind enough to let me look at the code itself to try to figure out how to adapt it for my own project. They are #navlist and #urlformat, respectively and the documentation does a really great job explaining how they work.
I've been looking through some of the older and still maintained extensions available, and I think the closest (at least for #urlformat) is NumberFormat. I'm not sure what the closest would be for #navlist.
If its really old, the way magic words are registered has changed. I think 1.15 was new enough to support the new way, but lots of extensions from that era were still using the old ways.
languageGetMagic hook is no longer a thing, instead you have to use an i18n file.
Global $wgParser should no longer be used. Register parser function in ParserFirstCallInit hook instead of an extension function.
I did some work fixing up the code (at least for URLFormat, see previous GitHub link and "Issues" section for error logs), though it looks like now I'm just having syntax issues? I think the problem is on line 18 in the file, if I'm reading things correctly. Any feedback would be helpful. In the meantime, I'm going to try to do some more reading on what syntax I may have gotten wrong. 75.45.55.110 (talk) 17:09, 23 November 2022 (UTC)
Installation stops at "install"; Is there an error log?
Latest comment: 3 years ago12 comments3 people in discussion
RESOLVED
Resolved
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
I am trying to set up mediawiki 1.38.4 on RedHat Linux, and get quite far.
I have the whole LAMP stack installed, and can get through most of the installation of Mediawiki.
But when I get to the "Install" section of https://<internalIP>/mediawiki/mw-config/index.php?page=Install, the page seems to stop.
when I start the installation of Mediawiki via https://<ipaddress>/mediawiki/, I get "The environment has been checked. You can install MediaWiki.", can follow through all the steps:
Welcome to MediaWiki!
Connect to database
Upgrade existing installation
Database settings
Name
Options
But then I get the prompt [By pressing "Continue →", you will begin the installation of MediaWiki. If you still want to make changes, press "← Back".], and when I acknowledge that, All I get is a screen that says "MediaWiki 1.38.4 installation", the word "Install" underneath it, and the steps I went through so far on the right.
I *would* expect the space under the word "Install" to fill up with steps taken to install the different tables and other bits. But I don't get that.
I'll have a look at the php error log if that helps me any further.
I've found the log directory /var/log/php-fpm/ and the file /var/log/php-fpm/www-error.log-20221127 contains the following
[23-Nov-2022 13:05:27 UTC] PHP Notice: Uncommitted DB writes (transaction from PopulateArchiveRevId::makeDummyRevisionRow) in /var/www/html/mediawiki-1.38.4/includes/libs/rdbms/database/TransactionManager.php on line 540
[23-Nov-2022 13:05:27 UTC] PHP Notice: DB transaction writes or callbacks still pending (User::insertNewUser, MediaWiki\User\ActorStore::acquireSystemActorId, MediaWiki\User\ActorStore::setUpRollbackHandler) in /var/www/html/mediawiki-1.38.4/includes/libs/rdbms/database/Database.php on line 5592
The last few entries in the mysqld log file are
2022-11-23T15:48:48.320420Z 0 [Warning] [MY-010068] [Server] CA certificate ca.pem is self signed.
2022-11-23T15:48:48.320468Z 0 [System] [MY-013602] [Server] Channel mysql_main configured to support TLS. Encrypted connections are now supported for this channel.
2022-11-23T15:48:48.361247Z 0 [System] [MY-011323] [Server] X Plugin ready for connections. Bind-address: '::' port: 33060, socket: /var/lib/mysql/mysqlx.sock
looking in the /var/log directory, I find three logs that may be important:
/var/log/mysql/mysqld.log
/var/log/php-fpm/error.log
/var/log/php-fpm/www-error.log
Only in the www-error.log I find interesting data:
[01-Dec-2022 08:57:42 UTC] PHP Notice: Uncommitted DB writes (transaction from PopulateArchiveRevId::makeDummyRevisionRow) in /var/www/html/mediawiki-1.39.0/includes/libs/rdbms/database/TransactionManager.php on line 583
[01-Dec-2022 08:57:42 UTC] PHP Notice: DB transaction writes or callbacks still pending (User::insertNewUser, MediaWiki\User\ActorStore::acquireSystemActorId, MediaWiki\User\ActorStore::setUpRollbackHandler) in /var/www/html/mediawiki-1.39.0/includes/libs/rdbms/database/Database.php on line 3757
[01-Dec-2022 09:26:18 UTC] PHP Notice: Uncommitted DB writes (transaction from PopulateArchiveRevId::makeDummyRevisionRow) in /var/www/html/mediawiki-1.39.0/includes/libs/rdbms/database/TransactionManager.php on line 583
[01-Dec-2022 09:26:18 UTC] PHP Notice: DB transaction writes or callbacks still pending (User::insertNewUser, MediaWiki\User\ActorStore::acquireSystemActorId, MediaWiki\User\ActorStore::setUpRollbackHandler) in /var/www/html/mediawiki-1.39.0/includes/libs/rdbms/database/Database.php on line 3757
[01-Dec-2022 09:42:26 UTC] PHP Notice: Uncommitted DB writes (transaction from PopulateArchiveRevId::makeDummyRevisionRow) in /var/www/html/mediawiki-1.39.0/includes/libs/rdbms/database/TransactionManager.php on line 583
[01-Dec-2022 09:42:26 UTC] PHP Notice: DB transaction writes or callbacks still pending (User::insertNewUser, MediaWiki\User\ActorStore::acquireSystemActorId, MediaWiki\User\ActorStore::setUpRollbackHandler) in /var/www/html/mediawiki-1.39.0/includes/libs/rdbms/database/Database.php on line 3757
The discussion above is closed. Please do not modify it. No further edits should be made to this discussion.
Extension support: WindowsAzureStorage
Latest comment: 3 years ago2 comments2 people in discussion
I am in the process of creating a wiki which is hosted in Azure using a docker container and app service. I would like to use the WindowsAzureStorage extension to persist wiki storage. I have followed the install instructions, however I am being thrown the following error when I navigate to "Special Pages":
[bf49979ead08c3a03544c173] /index.php/Special:SpecialPages Error: Class 'WindowsAzureFileBackend' not found
Backtrace:
from /var/www/html/includes/filebackend/FileBackendGroup.php(200)
Double check that all files from the extension are installed correctly. Maybe try redownloading and reinstalling the extension. Bawolff (talk) 18:48, 24 November 2022 (UTC)
After a certain number of edits using the VisualEditor's LaTeX editor, my server hangs
Latest comment: 3 years ago5 comments2 people in discussion
Whenever I make a serious number of edits, or even just small ones sometimes, my server will hang for hours. I have no idea why this happens. MediaWiki does not respond at all, and there is nothing in the error logs. 73.191.214.102 (talk) 22:31, 23 November 2022 (UTC)
I have tried multiple solutions, it is not ModSecurity, nor is it an error contacting Parsoid. I have tried loading the Parsoid module, and configuring PHP to access more memory. Nothing, absolutely nothing, works. I am using Apache, and it also crashes my Nginx server. 73.191.214.102 (talk) 22:32, 23 November 2022 (UTC)
Rather, this happens solely when I use the VisualEditor to edit math. Otherwise, it is perfectly fine. What always happen is:
It is by far one of the most irritating things I have encountered in my entire life. The issue is solved temporarily by restarting the server and continuing, but eventually my server will hang yet again. I cannot continue like this. 73.191.214.102 (talk) 22:39, 23 November 2022 (UTC)
Nginx hangs?
What do the nginx logs say? Are appropriate timeouts configured? Is the nginx process still running?
workaround: Only if I additionally do a manual logon in Wikidata, then I can access to the translated section of the FR API Wikidata page correctly.
Question 1 : i understand that if there is no logon, the default user language is EN, so that there is a break from FR page landing to an EN page, but why WikiData is not awaked by the global connection mechanism to assure continuity from FR page to FR page ?
Latest comment: 1 year ago5 comments3 people in discussion
* MediaWiki version: 1.35.3
* PHP version: 7.4
* Database type and version: mysql 8.0.31
We're getting the below message whenever we run "php maintenance/runJobs.php". The effect being, while the mediawiki page still loads, the contents are not updated. We've tried searching for possible clues to the issues online and did not find anything. Would appreciate any leads, please. Thank you!
Is ElasticSearch running and configured properly? (Or if you dont want to run elastic search, disable the CirrusSearch extension) Bawolff (talk) 18:43, 24 November 2022 (UTC)
Yes, elasticsearch service is running (version 6.8.22, on ubuntu 20.04). We ran a check on the health and the status shows "green" Timtaytango (talk) 06:57, 25 November 2022 (UTC)
I also noticed the following lines in elasticsearch_deprecation.log, consistently being added whenever we run the runJobs php script (not sure if it can be considered a possible cause for the ElasticaWrite job failed messages):
This may be due to Elasticsearch having run out of disk space for the indexes at one point, and set them to read-only.
I'm not an Elasticsearch expert, but I did run into this before with some other software that uses it, so this was familiar ground once I realized I was having the same issue with CirrusSearch.
The solution in both cases was to free disk space and/or alter the "high water" or "flood-stage" settings in Elasticsearch, then unblock the indexes as prescribed in this Support desk topic. This seems like something that CirrusSearch could detect, and give better error messages about, or even something that Elasticsearch could give better error messages about, but alas. Here we are, googling the error message.
On the topic of the deprecation warnings in elasticsearch_deprecation.log, I wouldn't worry too much about those. That is a warning to the developers of the CirrusSearch extension (that they need to update their code), not you.
I'm using Python requests, what cookies should I set? Also, where in settings would I look for cache issues? I don't have a cache in front of the server. TheManInTheBlackHat (talk) 00:16, 25 November 2022 (UTC)
You should ensure the bot sends back any cookies that are set on it.
If $wgSessionCacheType is set to something that doesn't work, it can cause problems (e.g. if you set to CACHE_ACCEL, using apcu but apcu.shm_size was set to 0, it would cause problems). If nothing is set, it defaults to the DB, which should be fine. Bawolff (talk) 00:22, 25 November 2022 (UTC)
Genius, that fixed it, thank you. I passed my cookies from the previous API response to the login API call. This is my code if anyone reading this in the future wants it.
Latest comment: 1 year ago4 comments3 people in discussion
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Product Version
MediaWiki 1.35.3
PHP 8.0.25 (apache2handler)
MySQL 8.0.26
Lua 5.1.5
Elasticsearch 6.5.4
We are using above & Mpdf 6.1 Extension:Mpdf as erlier we re using php v7.4 . But Now we have upgraded the Php to 8.0 and PDF is not working, done so many changes i.e () to [] ,replaced each() with new function in mpdf file but still facing issues. pls suggest which mpdf version we need to use for pdf .
also when trying to update mpdf to another version facing issues with elastic search
Problem 1
- elasticsearch/elasticsearch v6.5.0 requires php ^7.0 -> your PHP version (8.0.25) does not satisfy that requirement.
- elasticsearch/elasticsearch v6.5.0 requires php ^7.0 -> your PHP version (8.0.25) does not satisfy that requirement.
- Installation request for elasticsearch/elasticsearch 6.5 -> satisfiable by elasticsearch/elasticsearch[v6.5.0].
As elastic server 6.5.4 is suitable for MW 1.35.3 & ElasticSearch Php Client 6.5 for Elastic Server 6.5.4.
Warning: Undefined variable $footer in /data/www/html/wiki/extensions/Mpdf/MpdfAction.php on line 39
[Y4BnM7MF2q2VgmHCDaL-kQAAAAE] /wiki/...../&action=mpdf MpdfException from line 297 of /data/www/html/wiki/extensions/Mpdf/vendor/mpdf/mpdf/classes/otl.php: Can't open file /data/www/html/wiki/extensions/Mpdf/vendor/mpdf/mpdf/ttfontdata/dejavusansmono.GSUBGPOStables.dat
also header is not coming in pdf when generating pdf in some pages, alignment issue also there.
As in php 7 supporting int+float+string , but not php 8. i.e ( 10px+27px +auto ) so many issue in mpdf 6.1 libary code in php 8. Pooja2425 (talk) 08:03, 25 November 2022 (UTC)
Neither of those errors you posted sound like php8 issues.
The dumps contain *-pagelinks.sql.gz, *-imagelinks.sql.gz, etc. Do I need to do a refreshlinks.php if the dumps already have these datasets? 31.44.94.162 (talk) 14:50, 25 November 2022 (UTC)
I dont think the order matters as long as the importer is not refreshing links (if it was, the .sql files are pointless).
If you import the .sql link tables, running refreshLinks.php should be unnessary. Also refreshLinks is very slow, running it on an enwikipedia import will probably take the rest of your life. Bawolff (talk) 15:58, 25 November 2022 (UTC)
LinkedWiki problems (SMW)
Latest comment: 3 years ago1 comment1 person in discussion
Is it possible to query multiple endpoints at once (for example, Wikidata, Europeana and our own pages) and then represent the returned properties in a map or data table, as it is possible to do with a single endpoint? I wanted to be able to show common properties between these endpoints and each of their entries. Jdm.oliveira (talk) 14:55, 25 November 2022 (UTC)
How to exclude noindex pages from sitemap?
Latest comment: 3 years ago2 comments1 person in discussion
I have some pages that should not be indexed. How do I exclude them from the script generateSitemap.php?
Latest comment: 3 years ago4 comments3 people in discussion
I installed everything correctly as I have in the past a thousand times, and for some reason there is no skin showing at all. Please, if anybody can help me get this sorted, I'd be forever grateful.
I have working wikis that have the same internal server error. But those wikis still show up fine. This is strange. I've done a complete fresh install again with entirely new databases. I also made sure PHP is defaulted to 8.1. I'm not seeing a single reason why this should be happening. Hydrofox78 (talk) 00:31, 28 November 2022 (UTC)
look in your web server or php log. See How to debug.
> I have working wikis that have the same internal server error
5. Not logged automatically (no 'reload' popup, ..strange) ; i hit manually ... Login on the Incubator page ;
result: ok incubator awakes, connects and reloads the EN page automatically
6. Question 1: why incubator doent present the popup 'Reload the page' to be connected ?
7. Remark 1: I have lost continuity because I request the /fr page and I get the EN one when connected
10. nok, wikidata is not awaked; hit '...' and Logon have no effect ;
Must redo a manual logon in wikidata (global userID + pw) to be connected in wikidata
11. Question 2: why my declared Central connexion on Meta does not operate/propagate onto wikidata ?
The same problem occurs from Meta; example for translation of image of the day alert received in MW: MW=> Meta is ok to reach the article (connect with reload), META => Wikidata is not ok to update the elements (does not propagate and you must do manually).
Which manual action should i do (or have forgotten) for Wikidata no longer bother me with password ?
Latest comment: 3 years ago3 comments2 people in discussion
After upgrading Ubuntu from 22.04 to 22.10 my local wiki with version 1.36.2 failed because Apache2 did not find /etc/apache2/mods-enabled/php7.4.load /usr/lib/apache2/modules/libphp7.4.so. I installed PHP8.1 and enabled it for Apache2. Now mediawiki 1.36.2 started but with a number of messages like mediawiki 1.36.2 ... deprecated. Therefore I started to upgrade to 1.38.4. I downloaded the tar and extracted it into a new directory. Then, according to the documentation I moved my old extensions directory into the new directory. A lot of files where thus overwritten.
Then I run php update.php
../maintenance$ sudo php update.php
[sudo] password for tpollak:
MediaWiki 1.38.4 Updater
Your composer.lock file is up to date with current dependencies!
InvalidArgumentException from line 491 of /var/www/html/mediawiki-1.38.4/includes/libs/rdbms/database/Database.php: Wikimedia\Rdbms\Database::getClass no viable database extension found for type 'mysql'
As it worked before the upgrade, I do not understand what error there is (other apps using MySQL work).
Php uses separate config files for cli and web server. It probably means that mysql is disabled in the cli (i.e. the php.ini file is missing the line to load the mysql extension). Web stuff might still work due to using a different config file. Bawolff (talk) 17:18, 28 November 2022 (UTC)
The problem is solved. However uncommenting extension=mysql in phph.ini did not solve it. As a comparison showed that my settings for PHP7.4 included the so but not 8.1 I installed the mysql extension with "sudo apt install php8.1-mysql". Now I could execute "php update.php" and restart apache2 service. The wiki is now avalable. TPollak (talk) 15:38, 5 December 2022 (UTC)
The revision #0 of the page named "x" does not exist.
Latest comment: 3 years ago5 comments2 people in discussion
I get this error for a specific page (MW 1.31) without following a history link.
The revision #0 of the page named "x" does not exist.
This is usually caused by following an outdated history link to a page that has been deleted. Details can be found in the deletion log.
When I try to edit and save, I get the same error and/or
Someone else has changed this page since you started editing it. The upper text area contains the page text as it currently exists. Your changes are shown in the lower text area. You will have to merge your changes into the existing text. Only the text in the upper text area will be saved when you press "Save changes".
The migration was many years ago... Is there any solution to this? Nuking the page? Not sure what parameters to include in Manual:CleanupUsersWithNoId.php. I tried php maintenance/cleanupUsersWithNoId.php --prefix '*' --table user
i mean, if it was many years ago, but just happening now, the cause is probably different.
You need to figure out what is wrong with db. Basically find the entry in the page table for that page and the corresponding revision entry, and see if anything it references is 0 or missing. Bawolff (talk) 16:15, 29 November 2022 (UTC)
It may have been there all along, but just now I discovered it.
The first screen from a working page, second from dodgy one
I can see in a working revision table entry the rev_parent_id has a value, whereas in the non-working one it is 0 but I guess this only means that there are no other revisions for that page. Spiros71 (talk) 17:05, 29 November 2022 (UTC)
Wiki family and redirect issue
Latest comment: 3 years ago2 comments2 people in discussion
Hi. I was trying to setup a wiki farm (family) on my hosting. I've created two subdomain, placed installation files to one, and set other one's Document Root to the subdomain where installation is located.
But when I try to enter second subdomain, it redirects to installation subdomain instead of showing second wiki.
Is it possible to have templates parse individual characters from a parameter?
Latest comment: 3 years ago5 comments2 people in discussion
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
To try and explain what I want, in the clearest manner possible:
I wish to alter the template so it will automatically take the first letter of the songname parameter (i.e. if the song's name is "Jessica", then it will take the "J", and that will allow me to create a sub-parameter that will then link to a page called "List of music featured in automotive journalism media/J", which will then have (#songname) appended to the end so it automatically jumps to the relevant song in that article.
My only question how would I then use the substring as a parameter of its own once it's been generated?
So I'd imagine it would be set up like:
{{#sub:{{{songname}}}|1}} and that would get me the substring I seek without modifying the data already input. So if the songname variable was "Jessica", the "1" would denote it to sample only the J.
I would then append a link to [[List of music featured in automotive journalism media/{{{substringvarname}}}#{{{songname}}}]] in the template. If I'm right, it should just be a case of putting the "sub" template in the place of "substringvarname", but I could be wrong. AlexGRFan97 (talk) 13:07, 29 November 2022 (UTC)
For future reference, in case anyone wants to do what I did, you need to use {{#sub:{{{songname}}}|0|1}}so it takes only the first letter of the phrase, as opposed to just "1" on its own, which takes away the first letter and leaves the rest of the string intact.
The discussion above is closed. Please do not modify it. No further edits should be made to this discussion.
Wiki farm recommendation?
Latest comment: 3 years ago3 comments3 people in discussion
Might be the wrong place to ask but I'm finding it difficult to find a wiki farm that suits my needs. Preferably, I'd want something run with MediaWiki, with access to the Vector skin (Wikipedia style) and all the basic functions and options for extensions. Miraheze seems to be the closest to what I'm looking for but I really hate their hands-on approach and would rather have something moderation-free. ShoutWiki was another good suggestion but it's currently down and has been for a while. Could anyone help me out here? Thanks 31.205.38.26 (talk) 00:39, 29 November 2022 (UTC)
Is self-hosting not an option in your current situation? I'd consider a web service that specialises in MW like A2, pay the extra premium for your first year, and then you can relax and pay comparatively pennies for however long you plan to run your website for. Plus, no moderation and no rules.
Latest comment: 3 years ago10 comments5 people in discussion
I have notice that MW sites (including this one) has a button in the bottom-right corner. Could some one let me know how to have it at my wiki? Fokebox (talk) 14:53, 29 November 2022 (UTC)
It seems daft that hiding the sidebar here doesn't move the left margin across to the left, so this is an improvement! From the link above it looks like it's not fully ready yet. Jonathan3 (talk) 20:30, 30 November 2022 (UTC)
Latest comment: 3 years ago3 comments3 people in discussion
For an extension I need to obtain the parsed and fully rendered, formatted HTML from a subsection.
I have a content object in `$content` and can get sections via `$content->getSection("3")` and can proceed with $`getText()` but what I actually need is something like `$content>getSection("3.1")`
Unfortunately this returns `false`.
However, I do get the text of the entire section via `$content->getSection("3.0")->getText()` so I am a bit confused...
I'm not 100% sure, but i believe this uses just incrementing section numbers not hierarical.
So for example:
==A==
1
===B===
2
===C===
3
==D==
4
Section 1 includes all of section A including B and C (but not D), where section 2 is just B, section 3 is just C and section 4 is just D. Bawolff (talk) 23:52, 4 December 2022 (UTC)
Yes. And that's the perfect explanation. I had incorrectly assumed that the sectional structure is reflected in this API call. Thanx. Very helpful 87.163.207.87 (talk) 16:53, 5 December 2022 (UTC)
Variable number of parameters in template?
Latest comment: 3 years ago6 comments2 people in discussion
How do I make a template that can accept any number of arguments? I want to make a template that will call another template for every argument provided, but I'm not sure if this is possible to do without dropping down to Lua. RPI2026F1 (talk) 16:32, 29 November 2022 (UTC)
Latest comment: 3 years ago6 comments2 people in discussion
RESOLVED
Problems of Extension:DynamicPageList (Wikimedia) are discussed (compare phab:T287380). The extension might cause trouble when intersecting large categories. The pure listing of categories is fine. An example is used to illustrate this and is explained in some depth.
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Hello, due to a lack of community-Contributors (content is more or less fine) I would have liked to work on some automatic structuring-"tools" based on categories and DPL on dewikibooks. Now I read, DPL has problems (and possibly might be abandoned in the future? Compare phab:T287380) and I found
w:en:Category:Overpopulated_categories
. So I'm worried, if my ideas will be feasible at all.
What I would like to do: via categorization of pages sort the structure of Wikibooks more or less automatically, so that content contributors need only to worry about categorization and not editing community-pages. Hence less edits, less errors, less archive-work. To do this and to have some kind of control of the visual representation I would have liked to use DPL. Because of the above mentioned problem I don't think this might be a good idea. Now, if I discard the idea of having some kind of control on the visual representation, I could use category-pages itself. But if those are limited in some way it might be a bad idea for the future. Not that I expect to run into problems in the near future, but there might come the time, when somebody will. Restructuring then might become a pretty big task. So, if my ideas do not have any structural benefit for the future (aka "exploding" categories instead of "exploding" other pages) it might be a good idea to retire the idea and think about something else.
So two short questions:
can someone elaborate in short where the problem is with dpl, because I'm not quite sure if I understand it correctly? And as a follow up: Would you recommend not using it? The answer here is probably not really crucial, because I already think it's a good idea not to use it extensively. But any explanation here will be highliy appreciated.
is there a physical limit on how many pages a category can contain, and if so, which is it and where can I read up on this topic?
Honestly, i would just use DPL for now. There is unease with the extension, but as of right now there doesn't seem to be any positive plan to get rid of it.
So the problem with the extension: If you combine (intersect) multiple categories it does that inefficiently for large categories. i.e. It works by loading all the entries in the first category, and checking each one to see if it is in all the other categories[1]. If the first category has 200 entries, this is fine. If instead it has 2 million entries, this is not so fine.
However if you are only using DPL with 1 category, then none of this applies. DPL just looks at the entries of the category it needs and displays them. It takes the same amount of time regardless of how big the category is. [And the sorting method is categorysortkey or categoryadd. Sorting by lastedit, length, created may require loading the whole category]
Anyways, de.wikibooks.org is really small, so things should be fine no matter what you do.
In conclusion, I would suggest using DPL if its useful. Keep in mind that should de.wikibooks.org really take off and increase in size by 10 or 20 times, you might have to change your templates if they use DPL with multiple categories intersecting.
As far as the drama goes with ru.wikinews.org - A lot of that was due to the community not being very cooperative with developers. In the very unlikely event that something happens and developers tell you you need to change something due to performance concerns, just change whatever needs to be changed. As they say on en wikipedia at w:Wikipedia:Don't worry about performance - generally speaking you shouldn't worry too much, but if something does happen, just cooperate with the people who are trying to fix whatever it is.
For question 2: No limit. de.wikibooks only has about 70,000 pages, so i suppose at most you could have a category with 70,000 pages in it (And currently the biggest is much smaller [2]). Most of MediaWiki is designed so that if you look at a category, mediawiki only has to load the first 200 entries not the entire thing, so it doesn't matter much how big they get (DPL when intersecting multiple categories is an exception to this).
[1] - Please note, I'm oversimplifying a lot here, and strictly speaking this is not quite true, but its mostly close enough. Bawolff (talk) 20:08, 29 November 2022 (UTC)
Thank you very much for the explanation, that's exactly the "simplified" version I was looking for. Regarding "oversimplification", just one follow-up for question 2:
So I'm getting it right, there's no limit at all? Not only "no limit, because de.wb is small"? Is this in any way server-cost relevant if categories are large? Can I read up on this somewhere?
That said, another thing regarding oversimplification, being curious and seeing that you might have deeper insights, please feel free to elaborate further, if you'd like. I'm electrical engineer by training can program a little and might get it, if I'm not forced to look at a lot of pages figuring out the causal relations by myself. Use any of my talk pages, if you have the time and would like to.
Another side-note: I didn't follow the mentioned "drama". It might not be a question of "wanting to cooperate", but of availability, activity in the wikimedia-universe, enough spare time and ability (regarding rights and/or knowledge). Sure, I'd like to cooperate, but I might not be able to guarantee it because of "real-life-constraints" ;-). Thanks again, regards, HirnSpuk (talk) 12:06, 30 November 2022 (UTC)
There are eventually some limits - servers would eventually run out of disk space and that sort of thing. However, we're not really very close to those limits. There's also some softer limits - as the dataset gets bigger, less of it is going to be cached in ram at any given time, which can have a performance impact. Similarly, that's why large wikis like English Wikipedia get their own DB server, and smaller wikis share a DB server.
Currently the largest category is on commons with 34 million entries [3].
The way databases work, is you have the data, and then a bunch of "indexes" of the data. The index is just a sorted list of all the data based on some criteria. In the categorylinks table there are a bunch of indexes but the important one for viewing a category page is the one on (cl_to, cl_type, cl_sortkey, cl_from) which basically sorts all the entries in order based on the name of the category, followed by the page type (media, subcategory or "normal" page) and the sortkey/name of the page in the category. So if you want to display the first 200 normal page entries of a category, the DB basically just finds where that category begins in the index (Taking O(log N) time), and then starts reading 200 entries on that list in order starting at that point. Since its already in sorted order, the database can just look at the first 200 entries and stop, instead of looking through the whole category. If you're interested in the nitty gritty details on how this works see w:B-tree.
However, for DPL where multiple categories are specified, you can't really do it like that since you have to find the pages that are in all specified categories. The best case scenario is still relatively cheap, but the worst case might involve looking through all the entries in the category.
The situation on russian wikinews, was that they mass imported a lot (Like half a million) pages from a freely-licensed russian news source. All of these pages had an infobox on them with a DPL. The increasing size of the categories made these DPLs slow. Additionally, importing all these articles quickly, meant they all had to be rendered at the same time. Most of MediaWiki assumes database queries are very quick, and that most of the time rendering is spent in the CPU doing non-database stuff. This wasn't really true in this situation, as a result the DB started to get backed up, and requests to it piled up overwhelming it, making everything even slower.
Anyways, after that was dealt with, a few months later russian wikinews did another similar import. Although there may possibly have been language barrier issues, i don't know, at the time it seemed like they did it very uncautiously, doing the import as fast as possible without any consideration of possible risks. This caused the same DB problem, but also because the DB was so slowed down, normal requests to projects using that DB sort of hanged. As a result all the PHP servers were stuck waiting on the DB to respond, and web requests started to pile up, which caused things to spill out of control even further, triggering downtime not just for sites using that DB server, but all wikimedia websites. Anyways, when russian wikinews was told they had to stop, they got very angry, called all the devs incompetent, and wrote a "news" article about how WMF devs are screwing over ru wikinews. This didn't really endear them to the developers involved, who were not too happy to have to deal with a second outage caused by the same people doing the same thing. End result was that DPL was removed from ruwikinews (After all, it was only designed for small wikis, and ruwikinews was no longer small after the mass import). Additionally some changes were made to DPL that was hoped would improve performance. Recently made DPL queries were cached for a short time, as a big part of the problem was doing the same DPL query over and over for templates used on many pages. There was an attempt to use [PoolCounter]] to limit concurrency, but that wasn't enabled due to a bug that we couldn't figure out. A timeout to DPL queries was also added.
Anyhow, the moral of the story, if you do something that takes down Wikimedia websites, be really careful before doing it a second time, and don't get angry when someone tells you to stop. Bawolff (talk) 04:50, 1 December 2022 (UTC)
You Sir are a steely eyed wikimedia man :-)! Thank you so much for the perfect answer and explanation, I really sincerely appreciate your time and effort you put into answering my question! Have a nice christmas time of year! Best regards HirnSpuk (talk) 14:46, 1 December 2022 (UTC)
The discussion above is closed. Please do not modify it. No further edits should be made to this discussion.
[[::]]
Latest comment: 3 years ago2 comments2 people in discussion
We have table that have cell which reference data from another page by using [[::]]. I'm having trouble implementing this at another part of the site. Is there documentation on this? 73.74.209.90 (talk) 22:09, 30 November 2022 (UTC)
