Hey there! I just thought to introduce myself; I’m Alex, the new Wiki Manager assigned to Toaru Majutsu no Index Wiki. A Wiki Manager is a contractor tasked with looking after specific Wikis, helping them grow & assisting their Admins. If you ever need code help, moderation or content assistance, or just have questions, I’m happy to assist you :) Occasionally I’ll drop in to see how things are going, but otherwise you can contact me on my Wall anytime you like. Have a good day!
Just so you know, Ursuul, both are returning after a time away, so I've been compiling an extensive recap to put them in the picture as to what's been going on here in their absence. Annoyingly, I've been interrupted a number of times while doing so, but I intend to complete it as soon as possible. Once we're all up to speed, we should have a clearer idea of where we will need your help the most.
The possibility which I was thinking about was some form of code or mechanism which would automatically search for a specific field in specific templates on specific article, and extract the values inputted for this field in the templates on those pages, to compile a list of the values in the specified field for all of the searched pages (e.g. JAP/ENG Voice in the Character Infobox to gather information to produce a list of which voice actor voices which characters). If possible, this would save a fair amount of time taken in manually going through a large number of pages to compile certain lists for supporting article work - you could set a bot to gather the information while you focus on another task.
I carried out some searches on the Dev wiki to see if there were any pre-existing codes or combinations of them which could be used for this purpose, but I didn't have any success. When I discussed it with OH&S earlier, he wasn't sure how one would be able to implement it on the wiki. Although it would save a fair amount of time, it isn't vital, so if it is not possible or would be too much trouble, I'll abandon the idea and move on. However, I thought I should check first to see whether there was a feasible way to accomplish it or not.
This sounds like something that would have to be accomplished via Semantic MediaWiki, which is a legacy extension not granted anymore, or perhaps via some bot framework hosted with cloud computing to regularly run a routine to grab those values & update the page. Fandom is definitely aware of the issue of a lack of data tools, but I’ll probably have to ask around for the best solution (if any). Will get back to you when I have answers.
What I had originally in mind was something which could be given values and set to run (either by main or bot account), in a manner similar to MassEdit, and then output results. The resulting list would then be used to support the article updates and other work.
In the case of the voice actor example mentioned earlier, what I had in mind was something like this. At present, the list of voice actors is slightly out of date. However, people have added some of them to the Character Infoboxes on the respective character articles, though it isn't necessarily known which and how many. It would take time to go through them and assemble the information, and there are a lot of other tasks which need that time. In this case, with what was originally in mind, the possible procedure may have be something like this:
Set the mechanism to search for fields "JAP Voice" and/or "ENG Voice" from template "Character Infobox". Set to search within the range of character articles (e.g. obtained from the media articles or via Special:WhatLinksHere due to the links formed by the references).
Initiate the mechanism. While you carry out another task, it searches through the designated articles for the designated template (Character Infobox), and searches for the designated field within that template (JAP Voice/ENG Voice). If it finds the field used within the template, it extracts the value inputted into that field and logs the result in some form (e.g. "ENG Voice" for "Esther Rosenthal" is "Amber Lee Connors"), either in the function's output box (which can then be copied and pasted) or deposited in a designated location.
Once the mechanism has completed its operation, all of the associated values would be assembled in a convenient list, which could then be used to update the article. In a similar manner, through error messages or absences from the list, it would be possible to see which articles are lacking values for this particular field in their templates.
Another possible task would be confirming pre-existing conventions so that they can be made clearer in updates to the Manual of Style and similar pages, or to find errors and deviations which may have slipped through the net.
This is a rough description of what I originally had in mind, though I might have missed a few bits. It's getting late over here, so this'll have to do for tonight. Hope it is sufficient enough to help and sorry if it isn't conveyed clearly enough. Thank you for taking the time to look into the possibility, Ursuul.
Got back from Fandom, a script-based method which you describe is probably out of the question due to varied input resulting in malformed output. What we can do is attempt to produce the table purely via DPL on the List of voice actors article as you describe, although I cannot guarantee success & the earliest I could attempt such a thing would be Wednesday next week, possibly later. Is that OK?
I had a feeling that it would not be possible. I'm not too familiar with DPL, but I will read up on it to see what it can do. Given the overall circumstances however, I think it would probably be best to leave the voice actor table as it is for now and move on. Sorry for taking up your time on this minor matter, Ursuul.
One other minor thing I've thought about which I assume would not be possible is a script mechanism for transcribing history logs and the like. Would I be right in assuming this?
Sorry for the delay, Ursuul. I've been having a few problems on my end (computer and others).
With regards to the transcribing script mechanism mentioned earlier, I don't think it is likely to be possible but I thought I should check my assumption just to be sure. It is not urgent or vital and it is perfectly possible to manage without, though it is something which might save a little time on certain activities.
I should explain what I've previously been doing so you have an idea of how such a thing might be used. Although I've kept a number of lists in the past, I started keeping an overall wiki activity log late last year, in order to help me keep track of stuff going on across the wiki, as I was having trouble keeping track and remembering everything. I hoped that keeping a wiki activity log would help me in keeping track of ongoing activity, actions which need adjustment, monitoring or counter-action, progress made etc. It has had somewhat mixed success - while it has helped to keep track of certain things, it has taken a fair amount of time to maintain, and I've had trouble managing it and keeping it up-to-date. I've also compiled other more specialized lists gearing towards supporting specific tasks and activities.
The wiki activity log has been written using extracts from Special:RecentChanges as a starting base (from a set time period, extracted via regular copy/paste), placed beneath the date on the sandbox page. These initial bases are then trimmed down so the unnecessary stuff is removed (usually reducing the stuff for the edits to article, editor, time and any edit summary notes), and then sorted into the form of a bullet point list (with info condensed and notes added regarding the nature of the edit(s) and possible supplementary edits when necessary).
The main history log which is being transcribed as a base is Special:RecentChanges (within a set time/edit number frame), though the page histories (action?history) and other Special pages (e.g. Special:NewPages, Special:ListFiles, Special:Log/X, Special:WhatLinksHere, Special:Unused/UncategorizedFiles and others) have been used on occasion for other specific lists.
The main thing I wished to check was whether my assumption, that a script mechanism for transcribing logs probably wasn't possible, was right.
It might very well be possible, for example my old bot was used to make logs of Special:Chat for a couple years, but it required external hosting to maintain & the use of someone else’s pre-made Ruby framework that I didn’t fully understand. What you’re asking for seems even more thorough than that, stripping down content for only select items you want for your log, so I think it’d require a prohibitive amount of work to set such a thing up (assuming that it definitely is possible, which it potentially might not be). Additionally, with the MediaWiki upgrade imminent, the radical changes to RecentChanges’ coding would probably break the mechanism shortly after it was created.
The operation which I've been planning to carry out is part of something I'm calling the Image Information Survey.
Basically, there are currently a fair number of images which are lacking information (e.g. Description, Source, Licensing etc) at the moment, due to people in the past neglecting to add the information. This is something which I've been intending to tackle for a while and there are a number of potential plans which will rely on the relevant images having this information. However, there isn't a list of the images which are lacking them and it would take a considerable amount of time to go through over 9000 of them. Therefore, I've been planning to carry out a bot operation to go through the images and identify the ones which need work done.
I'm not all that good at coding and external bot operation programs, so what I've devised is a series of sweeps using the MassEdit function, to be carried out by the bot account on a secondary unit. By following this series of improvised steps, I'm hoping to have the bot go through batches of images, get the existing information into a common framework, identify the areas where information is lacking and tag them so that a convenient list is produced for subsequent work, while I carry out work in other areas at the same time. My hope is that this will save a bit of time and make subsequent work a little easier.
I had the bot carry out work on a test batch (while monitoring) in early November to test the procedure and see if there were any problems with it. It seemed to work well for the most part, though there were a few oversights and I identified various areas which needed improvement. Unfortunately, I was interrupted several times while doing so and I did not make a particularly clear note of my observations/thoughts, so I ended up losing track of some of the information when I came back to the task later and had to find it all over again.
Once I've cleaned things up a bit and made the relevant adjustments to the procedure, I intend to have the bot work through the remaining twenty-seven or so batches.
Some of them are not going particularly smoothly at the moment. The two month delay for the anime has bought a little more time, but there is still a lot to be done and some of the ongoing projects (of which there are a lot) have stalled or derailed due to various circumstances.
I'm currently trying to recover, get a clear grasp of where everything is, and get things back on track. I'm also preparing for a cleaning session on some of the older articles, not a full overhaul but enough for a stopgap measure. I'm planning to start Operation Interim on July 1st.
I am sorry to hear of your troubles, if I could give you some advice I would say that you may not actually have to complete everything by the time the anime comes. They usually portend a lot of users arriving to the Wiki which you may be able to harness to help get things done. Special:Community/MediaWiki:Community-to-do-list is a good place to list things that users can do to help, & when new folks arrive you can point them there as a springboard to get new editors to dive in to current & future projects.
Thanks. The basic gist of the problem is this; I'm unable to access my bot account (something happened last night with both accounts, though it has been fixed for the main one), the password reset email did not arrive at the expected destination, and I don't know where it has gone. What I was wondering was whether it was possible to check the set email destination for the bot account without logging onto it (though it would not be too problematic if it wasn't).
For privacy sake, the email associated with your bot account begins with “palm” — that mean anything to you? If you need the full email then I recommend sending a Zendesk request so that they can release it privately, if at all.
With regards to circumstances here, since our last discussion on June 28th there have been a number of new releases (including a new light novel volume), with plenty of information for processing. A couple of the manga spinoffs concluded at the end of July. The anime has resumed and its fourth episode since resuming will be airing tomorrow.
There have been some new users, only a few of which have contributed to the articles. A few of those, who have been making additions with the episode airings, look quite promising (though some of their edits need a little few grammatical tweaks). There haven't been many deliberate troublemakers on the articles, though there have been a few well-meaning headaches (and some messes along with them).
There's still quite a lot to be done and things are not fully back on track yet. I've had to extend the time frame for the stopgap overhaul plan Operation Interim, with several articles under temporary protection. I've been considering making some form of request for volunteers to help out with certain tasks.
Would you be interested in requesting help from Fandom Content Team for this purpose? Might not be able to grant it, but if you're absolutely overwhelmed & not able to get the content out then we might be able to assist. It's not a big issue if it just takes you a while to finish however, there's no rush.
The gist of it is that Special:RecentChanges seems to be cutting off sooner than anticipated. Even with the number and day parameters set to maximum, it is only displaying items from about a day or so back and fewer than expected. I was wondering if this might be the result of the large-scale bot operations that I've been running over the past couple of weeks (Note: The problem is the same regardless of whether bot edits are displayed or hidden). Would you say that is the case?
Hm, I'm getting 500 edits in both views in RecentChanges — RC always cuts off at a certain edit amount, no matter how many days are set for. That's just the maximum amount of days that can potentially be displayed.
I just ran a quick calculation for the September 20th display on my end. With 250 items set for display in Recent Changes and bot edits hidden, 79 items from the September 20th were displayed. With 500 set and bot edits also set to hidden, 121 items from the September 20th were displayed. An increase of 42 rather than 250.