However, upon further review, I noticed the link to fr:Utilisateur:NicoV/Wikipedia Cleaner at the top of the page, and followed it. Unfortunately, I don't understand French or what the page is designed to do.
I'm unclear on where the Java program exists, so that is why I removed the section, thinking it had been discontinued. I thought I was updating the page to reflect what was currently happening.
I think I may have acted too hastily and that there was a misunderstanding. Sorry for the confusion. Would you be so kind to explain where the program is to be downloaded / utilized ? Thanks. If you have any questions, please contact me at my talk page. Ian Manka09:56, 15 August 2007 (UTC)Reply
(P.S. I'm going to bed now, so I might not respond for 12-16 hours)
In retrospect, I suppose I was too bold and probably edited when I was too tired to make a logical decision. I probably should have talked to you first before making any stupid edits. Thanks for the explanation, and I look forward to using your program! If you have any questions, please contact me at my talk page. Ian Manka21:05, 15 August 2007 (UTC)Reply
Latest comment: 16 years ago14 comments2 people in discussion
Great program. I've been using it for months and have disambiguated hundreds of pages with it. One recommendation I have, which I assume would be easy, is to be able to fix the redirect links. So when you right-click on one, you can change the link to the correct page. Even better would be a button that automatically corrects all of the redirect links to their correct pages. Having this would allow you to fix two problems at once. Another simple recommendation would be to have a change summary of what you have improved from the previous version, especially if there is a new feature that might not be very noticeable, but very helpful. There have been at least 5 updates since the first version I used, and I haven't noticed any differences in the features. Is there something I haven't noticed, or have they just been performance updates? MrKIA11 (talk) 20:47, 5 January 2008 (UTC)Reply
Yes exactly.
That is perfect, but unfortunately I can't read French, so could you have an English version of it, most likely on the english page.
Any article in the left list that is a redirect page shows as blue in the main text. This is obviously after selecting "Show redirect pages" in the option menu. In the Watergate scandal example, the first few are "18½ minute gap", "abuse of power", and "Assassination of John F. Kennedy". After looking further, I noticed that "other" pages are also shown in blue, so I will update the documentation to reflect that. Having redirect links in red as you said makes sense, so I assume that is just a small programing error. The reason I removed the parts of the documentation that are only on the French wiki version is because I assumed that they will not be used by users reading the English documentation, but I guess they could be editing the French wiki but using the English cleaner version, so I will leave that on the documentation. I do not know how to make an English equivalent to the French {{Page h}}, but I am going to try and see if I can find someone who can. MrKIA11 (talk) 16:11, 6 January 2008 (UTC)Reply
I added {{Page d}}, the equivalent of {{Page h}}. I also updated the disambiguation Templates that are used. Also, for both the "Current Disambiguation List", and the "Watched Pages", what are each of the columns? They only have one letter, and I can't figure out what they stand for. MrKIA11 (talk) 18:28, 6 January 2008 (UTC)Reply
Great. I can understand your reasoning for everything. Thanks for describing what each column is, that should definitely help. MrKIA11 (talk) 23:25, 6 January 2008 (UTC)Reply
When a redirect is fixed, the Send button becomes inactive again.
Redirect links to disambiguation pages show as both a redirect and a disambiguation page, so if the redirect is fixed first, the page can not be fixed as it should be. Not sure how easy it is, but I recommend excluding all redirects to disambiguation pages so that they only show up on the main list.
When the {{dn}} template has been used, the Link to, Replace with and the other top-portion options on the context menu are not available, so someone that is able to fix the link can not chose what page, they have to type it.
I'm not sure what's wrong, but version 0.61 is not downloading. I have restarted the program at least 5 times, but still nothing. MrKIA11 (talk) 23:34, 22 January 2008 (UTC)Reply
I know, I am on holidays and I don't have enough access to update it on my website, it will have to wait until the weekend. --NicoV (talk) 16:46, 23 January 2008 (UTC)Reply
I just updated the documentation. The only thing, is that only internal links have contextual menus, external links do not. This is not a problem, but according to the documentation, they should.
I also have some recommendations for different wording in the program itself:
In the analysis window > Tools menu > Fix redirects, "Link ... by ..." should be "Link ... to ...". I actually wonder whether "Fix redirects" should even be included in the tools menu anymore, as it is easier and better to do it directly in the main text.
Latest comment: 16 years ago2 comments1 person in discussion
Hi. If you are NicoV of Jmol - Folding@Home, I am pleased to meet you again over here! I have just sent you an XYZ file for p4403.
A word of advice - on English Wikipedia, we don't just blank an unsuitable entry, as you did to "bethany erickson" (unless it contains a personal attack); what we do is add an appropriate speedy-deletion tag, in this case {{db-person}}. Regards, JohnCD (talk) 05:37, 28 February 2008 (UTC)Reply
No problem, I am not too awake myself. Your Wikipedia Cleaner looks interesting, I might try it - I have been fixing links to DAB pages by hand occasionally. Regards, JohnCD (talk) 06:22, 28 February 2008 (UTC)Reply
Latest comment: 16 years ago1 comment1 person in discussion
I am concerned by your change to {{ambiguous link}} permitting users to mark ambiguous terms in the form {{dn|Ambiguous term}}. This form is not recognized by disambiguation bots that typically search for [[Ambiguous term]] links, thereby making it harder for other editors to find and fix these links. Before making an issue out of this, I wanted to solicit your views and see if I am overlooking something. --Russ(talk)20:21, 29 May 2008 (UTC)Reply
At least an other user has the same problem on fr. I haven't changed anything in the login process, I suspect a problem with MediaWiki API. I hope I will have time this evening to investigate. --NicoV (talk) 06:47, 7 April 2010 (UTC)Reply
The explanation is here. This is a MediaWiki issue, they have changed the loggin process (including for those using the API), and I'm not even sure they warned about it before :( --NicoV (talk) 07:24, 7 April 2010 (UTC)Reply
Actually I haven't had a chance to look at the new capabilities yet. I will once Wikipedia Cleaner is up and functional for me again. It looks like the "minor change" broke quite a few of the tools. And I'm guessing that the disam challenge page stopped updating for the same reason. --User:WoohookittyDisamming fool!08:11, 7 April 2010 (UTC)Reply
Latest comment: 11 years ago4 comments2 people in discussion
Resolved
Hi, I am struggling a bit with the TD script (thanks, BTW!). Here is a technical explanation I was provided with in order to explain what does not look right. Thanks for your time: The rows
act on the variable params, which seems a string, and they work for single parameters (if you just put one parameter in the visual interface, it is correclty translate into json). When you add more than one parameter at time, the first of the two lines returns "Undefined" and the second throws an error, because params is not an object (i.e. an array or dictionary), and the JSON is not generated.
--Elitre (WMF) (talk) 15:41, 20 July 2013 (UTC)Reply
The problem was that, importing your script, it happened a few times that only one parameter could be added to a template, but not a second one. However, I am not sure how - I was later not able to reproduce the problem anymore, and already added TD to a few templates. Thanks for it :) --Elitre (WMF) (talk) 05:12, 21 July 2013 (UTC)Reply
Yes, it seems so. That probably requires quite some work to make it work with multiple languages, I probably won't have time to make this modification this week. --NicoV(Talk on frwiki)16:16, 21 July 2013 (UTC)Reply
Latest comment: 11 years ago1 comment1 person in discussion
Your comment on my talk page is greatly appreciated. WMF simply doesn't even want to acknowledge that 'agile programming' may have been the wrong approach to introducing a visual editor. In fact, I don't think they even deemed it necessary to ask Wikipedians whether it was desirable and took it upon themselves to assume it would be a clever tactic for enticing new editors/contributors. Hmm... --Iryna Harpy (talk) 10:26, 15 August 2013 (UTC)Reply
Latest comment: 10 years ago3 comments2 people in discussion
Hello NicoV,
Since you are active on fr.wikipedia.org, I thought I would ask you. How does request for bots work on frwiki? Is it like on enwiki where you has to request for every kind of edit the bot will do, or is it like on svwiki where you only need a flag to do what you want?
I would like to make Josvebot active on frwiki fixing CHECKWIKI-errors and update disambiguation warnings.
Hi! Requesting the bot flag in frwiki is done through the procedure described at fr:Wikipédia:Bot/Statut. Usually, you request it for some kinds of tasks, but once you have it, you can do other tasks provided you don't go too far. Scoopbot made a request recently for the bot flag for very similar tasks (or Gratus in the last archive). There are already 2 (or maybe more) bots updating disambiguation warnings (I don't run mine often lately, but I think Orikrin is running it regularly). For CW errors, you have to be careful because several errors are considered really minor (they don't change the appearance of the article), and bots are not too welcome on frwiki when they only do this kind of modifications in one edit. --NicoV(Talk on frwiki)21:22, 27 February 2014 (UTC)Reply
Latest comment: 8 years ago6 comments2 people in discussion
Hello,
I updated WikiCleanerConfiguration but im not sure if i completed it right. Can you please check and tell me what's missing, and what should i do next? Thank you :) --Margott 11:16, 18 January 2016 (UTC)
Hi Margott. I've checked quickly the configuration, here's what I see that could be changed:
general_wpcleaner_tag=WPCleaner: this means that modifications done with WPCleaner will be tagged with "WPCleaner", but this tag needs to be defined first (it doesn't seem to be in the list of tags on sq). Either you should comment out this parameter (put a "#" at the beginning of the line) or ask for the creation of this tag on sqwiki (the procedure is wiki dependent).
general_nobot_templates: not all templates seem to exist on sqwiki, you should comment out this parameter (put a "#" at the beginning of each line) or update the list: "GOCEinuse" doesn't exist so you should remove it from the list, "Inuse" exists but is a redirect to "Duke punuar" so you should add also "Duke punuar", "In use" doesn't exist, "Nobots" seems ok, ...
general_todo_templates: "todo" template doesn't exist, same for "to do", ...
general_todo_subpage: on enwiki, tasks to be done for an article are put on a sub-page named "to do" of the discussion page. Is it also the case for sqwiki ?
general_suggestions_typo: "Wikipedia:AutoWikiBrowser/Typos" doesn't exist on sqwiki, you should comment out this parameter, or put a correct page name
similar problems for other parameters...
I think the easiest way to configure WPCleaner for a new wiki is to comment out what you're not sure about (see the template where almost everything is commented out): if you want, I can create a minimal configuration for you, just tell me, I will do it. After that you can add configuration step by step for the extra features you want to activate in WPCleaner (the help page for configuration is quite big, but the prerequisite chapter gives indications on how to configure the most important features). I will happily help you add configuration once you have a basic configuration: I can explain what parameters should be configured and how depending on what features you want. I know it's not an easy task to fully configurate WPCleaner (I did most of enwiki and frwiki but only because I can speak both languages). Magioladitis, I don't know if you can help. --NicoV(Talk on frwiki)13:25, 18 January 2016 (UTC)Reply
Margott I checked how WPCleaner is working with the basic configuration I put on sqwiki. It seems to roughly work for fixing Check Wiki errors and disambiguation links, which are the main functions of WPCleaner. I think the next things that you could do to improve the configuration for sqwiki are:
First, decide which errors you want to be detected in sqwiki (for example, some errors don't apply in some languages) and modify the _prio_ values accordingly (0 means disabled, 1 to 3 means activated from top priority to lowest priority)
Second, for errors that you want activated, translate into Albanian the _head_ parameters to describe the type of error: it's also used by WPCleaner to suggest a comment when you fix such error. You can also translate _desc_ parameters but they're just used for display.
Hope this helps, don't hesitate to ask me more things. The best way to configure WPCleaner is probably to start using it and ask me when you see some behaviour that is not what you want, I can then help you to configure WPCleaner more precisely. --NicoV(Talk on frwiki)20:54, 24 January 2016 (UTC)Reply
Hello,
As you said "the best way to configure WPCleaner is probably to start using it" :) so i did and im 7000 edits now. The only problem that im facing now is when i fix an article it doesn't appear as DONE here, or thats ok? Because the article might improve everyday and might have the same problem? Thank you, Hugs ----Margott (talk) 09:39, 5 April 2016 (UTC)Reply
Latest comment: 8 years ago2 comments2 people in discussion
You might want to get the "bot" flag added to this user, as having a user with "Bot" in its name without the flag may concern some editors. Just wanted to leave you a message with the friendly suggestion. :-) ~Oshwah~(talk)(contribs)18:40, 5 March 2016 (UTC)Reply
Hi Oshwah, thanks for the message. This account has the bot flag on frwiki where it doest most of its modifications, and here its main edits are not on the main namespace. With the SUL features, I don't want to multiply account names just to have names matching the fact that I requested the bot flag or not... I've updated the description of the account page to give more information, hope it will be enough. --NicoV(Talk on frwiki)19:10, 5 March 2016 (UTC)Reply
Started it, thanks. Trying to configure now... but I absolutely misunderstood the meaning of bunch of parameters like «general_dab_**_templates»! English and French configuration examples doesn't help as well. Can You please make an example of using it?
Latest comment: 7 years ago1 comment1 person in discussion
Hello, NicoV. Voting in the 2016 Arbitration Committee elections is open from Monday, 00:00, 21 November through Sunday, 23:59, 4 December to all unblocked users who have registered an account before Wednesday, 00:00, 28 October 2016 and have made at least 150 mainspace edits before Sunday, 00:00, 1 November 2016.
The Arbitration Committee is the panel of editors responsible for conducting the Wikipedia arbitration process. It has the authority to impose binding solutions to disputes between editors, primarily for serious conduct disputes the community has been unable to resolve. This includes the authority to impose site bans, topic bans, editing restrictions, and other measures needed to maintain our editing environment. The arbitration policy describes the Committee's roles and responsibilities in greater detail.
Yamaha5 I found the problem : the parameter before general_todo_templates was missing a whitespace before END, so it was including general_todo_templates in it, making this parameter missing. This parameter is required for putting warnings on talk pages. I also fixed an other place, but I'm not sure there are no other parameters like that, you should check.
Yamaha5 I also suggest using the "Add warnings on talk page" button (in the main window) on a few pages to check that the configuration gives the expected result before running the bot tools on the 9.000 pages. --NicoV(Talk on frwiki)07:36, 20 October 2017 (UTC)Reply
Latest comment: 6 years ago3 comments2 people in discussion
Hi, when [[category:foo]] and [[category:foo|bela bela]] is at an article. WPCleaner should remove the [[category:foo]] but it can not edit. like this which i removed manually Yamaha5 (talk) 13:59, 5 November 2017 (UTC)Reply
Latest comment: 5 years ago1 comment1 person in discussion
Hello, NicoV. Voting in the 2018 Arbitration Committee elections is now open until 23.59 on Sunday, 3 December. All users who registered an account before Sunday, 28 October 2018, made at least 150 mainspace edits before Thursday, 1 November 2018 and are not currently blocked are eligible to vote. Users with alternate accounts may only vote once.
The Arbitration Committee is the panel of editors responsible for conducting the Wikipedia arbitration process. It has the authority to impose binding solutions to disputes between editors, primarily for serious conduct disputes the community has been unable to resolve. This includes the authority to impose site bans, topic bans, editing restrictions, and other measures needed to maintain our editing environment. The arbitration policy describes the Committee's roles and responsibilities in greater detail.
Hello SQL. What's the interest of this process, as my bot only acts on dedicated pages (list of errors provided by this bot) and not on articles or other pages provided by someone else? If this approval is still needed, I will seek approval, but it seems to be an administrative thing without any interest in these specific situation. --NicoV(Talk on frwiki)11:00, 17 December 2018 (UTC)Reply
Hello NicoV, may you be surrounded by peace, success and happiness on this seasonal occasion. Spread the WikiLove by wishing another user a Merry Christmas and a Happy New Year, whether it be someone you have had disagreements with in the past, a good friend, or just some random person. Sending you heartfelt and warm greetings for Christmas and New Year 2019. Happy editing, Dan Koehl (talk) 09:05, 25 December 2018 (UTC)Reply
Latest comment: 5 years ago59 comments2 people in discussion
I am just interested to know how your bot updates pages like Wikipedia:CHECKWIKI/WPC 108 dump. Which database dump do you download and use and do you use the command line or WPCleaner to update them. I would like my bot to update these dumps daily but before my bot can do that, I need to know how is it done. Pkbwcgs (talk) 19:03, 21 January 2019 (UTC)Reply
Hi Pkbwcgs. I'm not sure daily update is very interesting, because it can take a long time to process the dumps for enwiki... I download the official dumps, on a mirror (usually ftpmirror.your.org). For example, I will soon launch the analysis for the 20191020 dump which has become available (downloading file enwiki-20190120-pages-articles.xml.bz2). I process it on the command line (the task file is available on github). For an update using the same dump as the one the was used to generate the lists, the "ListCheckWiki" command can be replaced by "UpdateCheckWiki". I've a PowerShell script that I run to do this (except the download of the dump file which is manual) if you're interested. --NicoV(Talk on frwiki)19:15, 21 January 2019 (UTC)Reply
Command should be something like "java -jar getdown.jar . bot -credentials credentials.txt en DoTasks ListCheckWiki.txt", after creating credentials.txt and updating ListCheckWiki.txt. --NicoV(Talk on frwiki)19:18, 21 January 2019 (UTC)Reply
I use FileZilla to download it, it can resume the download even if there's an error. It's necessary to have something resilient, it's a huge file (~15G). --NicoV(Talk on frwiki)19:37, 21 January 2019 (UTC)Reply
Hi Pkbwcgs. Given the tasks you want to perform with your bot, I suggest trying to run WPCleaner on the command line when possible (not everything is possible currently) because it's a lot easier to set up scripts dedicated for each task. I've also just added yesterday the ability to add a prefix to the comments (see this diff for example, the prefix being the T3) when running WPCleaner on the command line for bot tasks (it just requires a "-prefix prefix" on the command line). --NicoV(Talk on frwiki)14:56, 22 January 2019 (UTC)Reply
Also, can I set a time on WPCleaner so that it can automatically update the ISBN errors at a certain time and do this once a hour? That would be a very handy feature if it was available. Pkbwcgs (talk) 16:38, 22 January 2019 (UTC)Reply
Hi Pkbwcgs. I'm not sure about the limit on the number of pages, bot mode is rather for doing the maximum... One of the advantages of running WPCleaner through the command line is that you can automate running it, that's what I do for its daily task: a cron runs it each day at the same time. --NicoV(Talk on frwiki)19:07, 22 January 2019 (UTC)Reply
Hi Pkbwcgs. Yes, you can run the same command line several times a day. "cron" is the classic tool on Unix/Linux to run tasks at given times, but you have a similar tool on Windows, the Task Scheduler. --NicoV(Talk on frwiki)19:14, 22 January 2019 (UTC)Reply
What is the code to run the ISBN warnings and ISSN warnings on Command Prompt? I will save this as a .bat file and use this on the task scheduler and let it run automatically at 21:00 everyday. Pkbwcgs (talk) 19:24, 22 January 2019 (UTC)Reply
I also got an error when running WPCleaner on the command line. "Error: Could not find or load main class org.wikipediacleaner.Bot". Pkbwcgs (talk) 19:43, 22 January 2019 (UTC)Reply
Hi Pkbwcgs. To run this task on the command line, you need the following:
A credentials.txt file containing 2 lines: "user=your username" and "password=your password"
A batch file that will run the following command from the directory where WPCleaner is installed: java -jar getdown.jar . bot -credentials credentials.txt -prefix [[Wikipedia:Bots/Requests_for_approval/PkbwcgsBot_8|T8]] en DoTasks ISBN_ISSN.txt
Be sure to test the batch file directly before running it through the task scheduler (the context in the tasks scheduler is different, so it can lead to some problems). --NicoV(Talk on frwiki)06:58, 23 January 2019 (UTC)Reply
Pkbwcgs If you open getdown.txt in notepad, what do you have exactly ? If it's the version downloaded to install WPCleaner, it should only contain one line with appbase = https://tools.wmflabs.org/wpcleaner/wpcleaner/. If you have already managed to run WPCleaner from this directory, it should contain many many lines... --NicoV(Talk on frwiki)17:50, 23 January 2019 (UTC)Reply
Pkbwcgs: I don't know. Try using a new folder, downloading getdown.txt and getdown.jar in it from Tools (right click on each link and "save link as" in the new folder). Copy your credentials.txt and ISBN_ISSN.txt files in the same folder. Open the cmd.exe in the folder and try running the command. --NicoV(Talk on frwiki)11:07, 24 January 2019 (UTC)Reply
I restarted my computer and there has been no luck. The same error is repeating constantly and it means that I can't do the hourly automatic update to the ISBN and ISSN errors. Had this worked, I would have used the same trick to fix CHECKWIKI errors hourly through the event scheduler but it is clearly not working. Pkbwcgs (talk) 21:57, 24 January 2019 (UTC)Reply
Apparently, I have a really old version of JavaScript. Perhaps, that is what is causing the problem. I am currently getting that installed and replacing it with a new version of JavaScript. Pkbwcgs (talk) 08:53, 25 January 2019 (UTC)Reply
This could be the culprit to why the appbase is not able to be recognised. Here is some of the history from the launcher.txt file that came with the WPCleaner installation.
2019/01/25 08:42:23:816 INFO Using 'appbase' from bootstrap config [appbase=null]
2019/01/25 08:42:23:816 WARNING run() failed.
java.lang.RuntimeException: m.missing_appbase
at com.threerings.getdown.data.Application.init(Unknown Source)
at com.threerings.getdown.launcher.Getdown.detectProxy(Unknown Source)
at com.threerings.getdown.launcher.Getdown.run(Unknown Source)
Pkbwcgs What is the content of getdown.txt just after you download it? What is the folder path in which you have downloaded getdown.txt and getdown.jar? What is the exact command that you're running? --NicoV(Talk on frwiki)09:03, 25 January 2019 (UTC)Reply
The folder path in which I have moved the getdown.txt and getdown.jar is C:/Users/username/Documents/WPCleaner
The exact command I am running is java -jar C:\Users\username\Documents\WPCleaner\getdown.jar . bot -credentials credentials.txt en DoTasks ISBN_ISSN.txt
Any space in the folder path? Try this command instead java -jar C:\Users\username\Documents\WPCleaner\getdown.jar C:\Users\username\Documents\WPCleaner bot -credentials credentials.txt en DoTasks ISBN_ISSN.txt. --NicoV(Talk on frwiki)09:46, 25 January 2019 (UTC)Reply
Pkbwcgs It's supposed to install WPCleaner and then run the command, otherwise how can it run WPCleaner? Are you sure you are running the command from the folder into which you downloaded the 2 files? Because the only difference in the 2 commands is that it will look in C:\Users\username\Documents\WPCleaner for getdown.txt instead of the current folder (.). If you're not running the command from the correct folder, you have to specify the path for everything java -jar C:\Users\username\Documents\WPCleaner\getdown.jar C:\Users\username\Documents\WPCleaner bot -credentials C:\Users\username\Documents\WPCleaner\credentials.txt en DoTasks C:\Users\username\Documents\WPCleaner\ISBN_ISSN.txt. --NicoV(Talk on frwiki)09:54, 25 January 2019 (UTC)Reply
Pkbwcgs As I said above "the context in the tasks scheduler is different, so it can lead to some problems". Be sure to check the properties of the scheduled task: which user account is it using (by default, it's the System account, not your user account, so it has less possibilities) ? Have you written a .bat file and tested it from the correct directory and from another one (because the scheduled task won't run in this directory)... --NicoV(Talk on frwiki)12:54, 25 January 2019 (UTC)Reply
I have another major problem. When my computer is on sleep mode, the script will stop and it won't work. How to make the script work even while the computer is on sleep mode. Pkbwcgs (talk) 13:08, 25 January 2019 (UTC)Reply
It looks like I am heading to abandon hourly updates and give up because the task scheduler just won't run the task despite applying the correct settings. Pkbwcgs (talk) 15:00, 25 January 2019 (UTC)Reply
It is constantly crashing and whenever the task scheduler runs the task, it is doing nothing but analysing pages. When it is finished with analysing pages, it is analysing them again and not updating the errors. The task is causing my system to crash awfully. Pkbwcgs (talk) 11:20, 26 January 2019 (UTC)Reply
The Arbitration Committee is the panel of editors responsible for conducting the Wikipedia arbitration process. It has the authority to impose binding solutions to disputes between editors, primarily for serious conduct disputes the community has been unable to resolve. This includes the authority to impose site bans, topic bans, editing restrictions, and other measures needed to maintain our editing environment. The arbitration policy describes the Committee's roles and responsibilities in greater detail.
Hi Paul August. Can you explain to me why it was inappropriate (so that I can understand what needs to be modified in my bot)? Usually, a link text ending with a comma is an error, the comma shouldn't be part of the link, but after. Why is it different for this case? --NicoV(Talk on frwiki)13:56, 18 April 2020 (UTC)Reply
Hi Nick. The comma is not punctuating the sentence it is in, instead it is intended to be the part of the thing being liked to, in this case the LSJ entry for "κεραυνός , ὁ,". See [2]. Paul August☎19:35, 18 April 2020 (UTC)Reply
Ok thanks. Although, as I said above, that trailing comma was meant to be part of the link text, linking to the LSJ entry for "κεραυνός , ὁ," (which means the entry for both of the Greek words spelled κεραυνός and κεραυνὁ), I've decided that, in this case, it is sufficient to specify it as just "LSJ s.v. κεραυνός", thus eliminating the need for the trailing comma. Regards, Paul August☎11:26, 19 April 2020 (UTC)Reply
Thanks Jonesey95. I will try to do something, but I think the problem comes from the template {{Plainlist}} itself : it creates an opening div unconditionally, but creates the closing div only if the first argument is given (the </div> is in the #if, not the <div>)... Don't you think it would be best to fix the template? --NicoV(Talk on frwiki)20:09, 20 April 2020 (UTC)Reply
I checked the template, and now I understand why the div is closed only if there's a parameter. Behavior like that are likely to cause missing end tag errors... but it's probably used in too many places to have something clean now... --NicoV(Talk on frwiki)20:14, 20 April 2020 (UTC)Reply
If the template is empty, you can remove the whole template; that's what I do when I come upon an empty template that causes errors like this. If an editor needs it, they can add it back with content inside it. – Jonesey95 (talk) 23:15, 20 April 2020 (UTC)Reply
Thanks Jonesey95. It's too specific for my taste, so for the moment, I will simply not remove automatically empty list items in templates. I will see after that if I can be less restrictive. --NicoV(Talk on frwiki)07:39, 21 April 2020 (UTC)Reply
Hi Seasider53. Yes, there was, and there is again:
[[Category:Educational institutions established in 1899]]
[[Category:1899 establishments in Georgia (U.S. state)]]
[[Category:Kindergarten]]
[[Category:Education in Savannah, Georgia]]
[[Category:Educational institutions established in 1899]]
--NicoV(Talk on frwiki)11:33, 3 May 2020 (UTC)Reply
I'm not familiar with meta, what's the process for bot approval there? I can probably easily add meta to WPCleaner if needed. Tell me which tasks among the ones I'm already doing for enwiki or frwiki could be useful there and I will check what I can do.
For other editions (meaning other languages?), I usually let other people run bots there with WPCleaner if they want (to avoid the language issue for me). But I can help on setting up WPCleaner and defining tasks that can be run there. Do you have any particular edition in my mind? --NicoV(Talk on frwiki)11:15, 27 May 2020 (UTC)Reply
My apologies for the delay. You seem to have found it on meta, however. Yes, other editions are other languages. Its lint error fixing capabilities are what interest me the most. Please let me know if you have any Qs and I will do my best to assist. I myself, however, have not actually filed a bot request on meta to date though haha. --TheSandDoctorTalk23:05, 6 June 2020 (UTC)Reply
Hi TheSandDoctor. For meta, I'm in the process of improving WPCleaner to deal with the specific ways of meta. For other languages, WPCleaner is already available on many other wikis, see Wikipedia:WPCleaner/Wikis (adding another wiki is not too complex). With a bit of configuration to activate the errors, it can easily be run in bot mode to fix Linter errors on those wiki. But, I'd prefer to find someone to run it on such wiki, mostly because of the language issue. I can of course help to configure and do the first runs. Do you know editors who would be willing to take such task? --NicoV(Talk on frwiki)07:30, 7 June 2020 (UTC)Reply
Hi TheSandDoctor. I ran a dump analysis on meta for #551, result is here, it took about 20mn to do.
If you have installed WPCleaner, you should have a "tasks" folder in WPCleaner. I've just added tasks for meta there (meta/ListCheckWiki.txt and meta/_Common.txt).
Tasks are script files that describe what the bot should do, so you should define one that matches what you want to do.
For example, meta/ListCheckWiki.txt is the one I just used to run the dump analysis on meta. Explanation on what it does:
DoTasks _Common.txt : execute common tasks for meta (list of algorithms activated, edit throttling of the bot)
Set Prefix ... : add a prefix to the comments for each edit of the bot (replace whitespace characters by _)
Set Namespaces ... : list on which namespaces the other tasks will work on
ListCheckWiki ... : do the dump analysis
You can also have a look at enwiki/ListCheckWiki.txt in which some tasks run automatic fixing.
To run a task file, you can use Bot.sh or Bot.bat depending on your operating system: Bot.bat -credentials [credentials_file] DoTasks [task_file]
"ListCheckWiki" needs a dump file (...-pages-articles.xml.bz2) which can be downloaded (I'm using FTP server ftpmirror.your.org, directory /pub/wikimedia/dumps)
"Bot.bat -credentials credentials.txt DoTasks meta/_Common.txt" just says "Launching 99% complete" and then shuts and the process ends. (Win10) --TheSandDoctorTalk18:30, 7 June 2020 (UTC)Reply
Just looked there. Thanks. Seems to have been running into "[main] WARN org.wikipediacleaner.Bot - Unable to find wiki DoTasks" before actually crashing those times. --TheSandDoctorTalk18:56, 7 June 2020 (UTC)Reply
Hi TheSandDoctor. My bad: you have to specify the wiki to connect to, so probably "Bot.bat -credentials credentials.txt meta DoTasks meta/_Common.txt". If you want to kill it, yes, task manager. The other option is to run WPCleaner directly (instead of running it through getdown, which brings automatic update): java -cp "WPCleaner.jar;libs/*" org.wikipediacleaner.Bot -credentials credentials.txt meta DoTasks meta/_Common.txt. --NicoV(Talk on frwiki)05:55, 8 June 2020 (UTC)Reply
Latest comment: 4 years ago3 comments2 people in discussion
This edit by WikiCleanerBot fixed a missing end tag by turning it into a misnested tag. <small>...</small> tags can't wrap multiple lines. I created {{smalldiv}} to get around this limitation; here's how to apply it.
FWIW, the bot's edit makes it easier for me to replace small tags with the smalldiv template, since my pattern looks for opening and closing tags, so if you are unable to add smalldiv, keep doing it this way, and I'll clean up after you. – Jonesey95 (talk) 17:23, 13 June 2020 (UTC)Reply
Latest comment: 4 years ago4 comments2 people in discussion
Resolved
This edit actually broke what had been a correctly matched italics set. The unbalanced set was in the preceding h2 section. So now although there are an equal number of ital open/close in the article as a whole, there is misnesting of them, and imbalance in two different sections. DMacks (talk) 12:56, 15 June 2020 (UTC)Reply
Hi DMacks. Thanks for the notification. In fact, the incorrect edit is due to a side effect of the internal link inside an external link ([http://www.dundalkeagle.com ''The [[Dundalk Eagle]]'' newspaper]): my bot saw a single italic set inside an external link with only one word in the text ("The"), so it closed the italic... I will add a check in my code for such special conditions. But the external link should be fixed: something like [http://www.dundalkeagle.com ''The Dundalk Eagle'' newspaper] or ''The [[Dundalk Eagle]]'' newspaper. --NicoV(Talk on frwiki)14:14, 15 June 2020 (UTC)Reply
Latest comment: 6 years ago3 comments2 people in discussion
Hi,
Is it possible to add possibility to WPCleaner which do these steps?
check article base on the checkwiki errors.
check the history from the last revision and find which revision caused that error
send a warning text to the editor/user's talk page to correct his/her wrong edit.
if the revision is the last one. have a possibility to revert automatically that revision (base on the number of the error which is defined on the local wiki)
Now other users do wrong edits and users like us should clean it! with this feature we can force them to helpYamaha5 (talk)
Sure Yamaha5, it's possible, but it will require some work. I don't know when I'll be able to do that, I'm starting a new job, and I'm currently trying to work on the integration with Linter. --NicoV(Talk on frwiki)15:10, 5 November 2017 (UTC)Reply
For recognizing the article which is corrected or not. do you have any code or api? (which part of this code check errors?) i want to integrate it with pywikibot codes and developed a bot which checks the revisions and do warning jobs.Yamaha5 (talk) 15:13, 5 November 2017 (UTC)Reply
Reading the page describing the tool, we see that this is clearly a powerful tool with various applications – too much to take in all at once. A busy editor might think: "Too hard!", and so miss out on the value it offers. (I'm feeling somewhat daunted, and I've been in computing since the 1960s.) So a modular course in using WPCleaner would be more effective. The most important module would be an introductory tutorial, giving a non-technical overview and emphasising both its value to the editor and its ease of use. And a separate tutorial for installing and running WPCleaner would be a good idea - unless it's dead easy. The WPCleaner page's Usage section has a list of 13 help pages covering different functions. A tutorial might cover one to three related functions from this list.
Visual teaching helps many people learn faster. So well-designed videos, hosted on Vimeo or Youtube, introducing and demonstrating different aspects of WPCleaner's use, would be effective.
Clearly, this tool is popular, as it's highly ranked by number of edits on Special:Tags. So there are probably many editors already familiar with its use, of whom some would be able, and if willing, could make such tutorials.
Having such tutorials could encourage many more editors to use WPCleaner, thus helping improve Wikipedia more effectively and efficiently – perhaps contributing to radical improvements in quality. Using a popular video hosting site would also help increase the visibility of the WPCleaner tool. (Until I checked the Special:Tags page today, I hadn't even known it existed ….)
Hi yoyo. Thanks for the suggestion. Short answer : I haven't made any tutorials and I don't know any available, but sure this is a good idea. I never made any such tutorial, and I don't have much free time lately (I don't see that change in the near future). So I don't think I will be able to do such things in a timely manner, but contributions are welcome : if you feel that you can make such tutorial, I will be much happy to include them. Regarding installing and running WPCleaner, the process to do so is currently undergoing major changes, see Wikipedia talk:WPCleaner#Need help to test new installation (as you can see, it took me 1 year to change the release process to be able to make a new release, not much free time...). What do you think ? --NicoV(Talk on frwiki)05:08, 6 August 2018 (UTC)Reply
@NicoV: having read Wikipedia talk:WPCleaner#Need help to test new installation, I think … Ouch! Sounds like a difficult beast to deploy, for the average user who's used to running InstallShield to setup and configure a Windows app – and these days I'm pretty much an average user. I don't have the spare grey cells (or time) to tweak command-line stuff any more. I was hoping it might be a bit easier than this to get started! So I won't be in a position to make any tutorials any time soon. But maybe others can?
I see you've used the WebM format for the video. This downloaded quickly and ran well using the VLC media player. Was this difficult to make? If you tell me it was dead easy and tell me how to, I could try to make a tutorial for one specific and popular function (you name which one); then we could compare notes.
It seems to me the installation would be similar on both Windows 7 (SP1) and 8.1, which still have many users.
Advice? Well, you did a good thing in highlighting the text (from the install page) that describes the steps you are about to show. Would it be clearer, do you think, to have a separate "slide" that simply lists these steps, with numbers (perhaps revealed one step at a time), rather like a slideshow presentation? e.g.
Hi yoyo. Yes, we can move the discussion to the dedicated talk page, maybe other people will be interested.
Technically, it's easy to do because I used a virtual machine in VirtualBox and only used the built-in feature to record the VM display. Microsoft provides free virtual machines with various Windows OSes that are perfect for this kind of test. The main problem is that I did the entire tutorial in one pass, so I had to do several trials before not making mistakes... I didn't try any editing, any voice recording, subtitles are managed directly by MediaWiki
The installation should be similar in other Windows flavors.
I think the most popular function is "Full analysis": put a page name to analyze in the first window, click "Full analysis", fix the page (disambiguation, check wiki...), save the page. Otherwise, there's the "Project Check Wikipedia" : click "Project Check Wikipedia", select errors to analyze, load errors, load a page, fix the page, save the page. And after that, all the functions that return a list of pages from which you can call the "Full analysis" on selected pages.