Commons:Bots/Work requests

Shortcuts
COM:BR
COM:BWR

Bot policy and list · Requests to operate a bot Requests for work to be done by a bot · Changes to allow localization  · Requests for batch uploads

Archives 1, 2, 3, 4, 5, 6, 7, 8

SpBot archives all sections tagged with {{Section resolved|1=~~~~}} after 5 days.

Display issues of language templates

There are two kinds of display issues of language templates:

1. The text in parameter 1 is not displayed due to a = in the text. The 1= in the beginning that avoids this problem is missing. Examples: Special:Permalink/28178560, Special:Permalink/69118001, Special:Permalink/24951271
2. If there are a certain number of language templates that are used as {{en}}Some sample text, the language select feature effects all texts to be displayed after each other (with the language names in a block above). Examples: Special:Permalink/42105123, Special:Permalink/31255164, Special:Permalink/72387078

We have Category:Template:De with no text displayed and Category:Template:nl with no text displayed where pages with these problems for {{de}} and {{nl}} are in. The vast majority of cases has already been fixed for the former category. A large fraction of pages in these categories do not have one of the two problems described above, since there are not (yet) enough language templates to trigger the language select feature.
I think that there should we maintenance categories for all languages templates, but fixing would not be feasible without getting assistance by bots. The main task would be to move the text after a language template inside it:

{{en}}Some sample text

{{en|Some sample text}}

or

{{en}}Some sample text with a = in it

{{en|1=Some sample text with a = in it}}

--Leyo 10:19, 22 March 2013 (UTC)

Sounds like a good idea. I will look into it. In the mean time I created {{Language templates with no text displayed category header}} to be used in subcategories of Category:Language templates with no text displayed. --Jarekt (talk) 14:51, 22 March 2013 (UTC)

Thanks. I actually had the same idea concerning the template. ;-) --Leyo 14:57, 22 March 2013 (UTC)

I was thinking more about it and many language templates have very few files using them. There is probably no need for maintenance categories for them. May be we should use some threshold and only do the category for templates with more than 1000 (?) transclusions. --Jarekt (talk) 16:49, 22 March 2013 (UTC)

Do you have a good method for finding errors for such templates without a maintenance category? --Leyo 17:10, 22 March 2013 (UTC)

An alternative is to change with a bot {{nl}} to {{nl| }}. This removes those images from the category. Fix the rest manually. Or fix the rest in the category with a bot as well: {{nl|xxxx}} to {{nl|1=xxxx}}. HenkvD (talk) 19:36, 22 March 2013 (UTC)

Your first suggestion would not solve the problem #2 (another example). It would just “hide” it. --Leyo 21:00, 22 March 2013 (UTC)

That is right, but I think it is a quick fix, and as a result all texts are shown (maybe not on better place). Trying to replace {{en}}Some sample text to {{en|Some sample text}} might be a difficult progamming. For instance this where the {{nl}} and {{en}} contains a full information template. HenkvD (talk) 13:13, 23 March 2013 (UTC)

 Done I took a little different approach and added this to {{description}} template which is called by all language templates. So now all the problematic images will be added to Category:Language templates with no text displayed, instead to individual categories for each language. --Jarekt (talk) 01:53, 24 March 2013 (UTC)

I also run a bot run (see here) adding "1=" to language templates with "=" in the text, and replacing blocks of {{xx}}text1{{yy}}text2 with {{xx|text1}}{{yy|text2}}. --Jarekt (talk) 19:30, 24 March 2013 (UTC)

Thank you for your work. I am fine with your modification to {{description}}, but I would not like to miss the subcategory at least for {{de}}. Once the legacy cases are gone, it will provide an easy option to check for problems in files transferred from de.wikipedia to Commons. Furthermore, subcategories tells us more precisely where the error is located.

It seems that the script of your bot is not yet perfect: see e.g. [1], [2], [3], [4] or other cases. --Leyo 23:39, 24 March 2013 (UTC)

I do not think a perfect one is possible. I was aiming for fixing most files I can, with least chances of breaking things. So in many cases the bot did not finished the job. The rest will likely have to be done in semi-automatic mode. I do not mind leaving Category:Template:De with no text displayed or other specific language categories behind, but I did not wanted 200 of those categories with some files in several dozen of those. --Jarekt (talk) 03:45, 25 March 2013 (UTC)

One possibility to get rid of a probably dozens of cases would be to remove empty templates, i.e. with no text inside or behind, by bot. --Leyo 07:54, 3 April 2013 (UTC)

I run something along those lines, so all cases where that was the only issue should be done. However I was trying to be conservative about how to detect text or behind language template and might have missed some. --Jarekt (talk) 11:35, 3 April 2013 (UTC)

{{Picture of the day}} is also a possible cause of the problem, if a language subpage does not contain a description. Example: Template:Potd/2009-02-13 (pt). BTW: It says

Description missing! Tag this page with {{speedydelete}} if it was created by mistake.

, but the page is not put into a maintenance category. Should this be changed? --Leyo 16:29, 5 April 2013 (UTC)

The image was added to the Category:Language templates with no text displayed maintenance category, and template was added to Category:Potd templates missing descriptions. I think that is correct. The pages in Category:Potd templates missing descriptions should probably be deleted. --Jarekt (talk) 14:47, 10 April 2013 (UTC)

I was working on clearing Category:Language templates with no text displayed, help would be appreciated, since many types of issues are unique enough to write a bot. Another problem is that since this maintenance category was added to a template used on a LOT of pages, the pages were not refreshed afterwards so images in that category are only added after some other event causes the refresh. Past experience with other similar categories and templates suggests that that process can go on for months, so we will not see all the files that should be in Category:Language templates with no text displayed for a while. --Jarekt (talk) 14:55, 10 April 2013 (UTC)

I know. Hence, I am constantly cleaning out Category:Template:De with no text displayed and it is constantly being refilled… --Leyo 14:35, 11 April 2013 (UTC)

I fixed hundreds of category pages with empty language templates created or edited by User:Bernd Schwabe in Hannover.
Could someone please touch all edits to file pages by User:Chumwa? Several of them are not in the maintenance category since they have not been refreshed recently. --Leyo 17:52, 13 April 2013 (UTC)

The maintenance category is again really full. :-( Perhaps, the implementation of Lua (TemplatePar) would help by allowing to distinguish between cases like {{en}} and {{en|Sentence containing a = in the text}}. --Leyo 17:30, 29 April 2013 (UTC)

We could do that but I am not sure how that would be helpful. --Jarekt (talk) 17:51, 30 April 2013 (UTC)

(i) Prioritizing the latter case over the former, (ii) facilitating bot runs.

I guess Lua shouldn't be tested on e.g. Template:en, but rather on a template with less transclusions. --Leyo 17:59, 30 April 2013 (UTC)

Working out the "largest" user pages

Hi,

Faebot recently churned through the list of all active users (the API gives those active in the last 30 days), worked out which had a user page, and returned the length. For example my user page is shown by this call to have a length of 5,702 bytes. So I have an interesting list and can identify the top 10 or top 100 user pages for active users by "length".

Now, here's my problem. I was actually trying to find out which user pages are the "largest" in terms of the size of page the browser has to download. This would include transcluded images, in particular it would highlight those user pages being used for very large galleries and might be so large as to be unusable or unreadable in many browsers or platforms.

Does anyone have a suggestion on how to go about creating such a list of user pages by final download size, possibly being restricted to a short-list similar to the one I already have generated by length? Thanks --Fæ (talk) 10:28, 10 April 2013 (UTC)

I can not think of any easy solutions, but we could write a simple code to download all user pages and count number of characters in HTML code. --Jarekt (talk) 12:01, 12 April 2013 (UTC)

Why don't you try to simply sum all images' size to the page size? Using images as a generator for imageinfo: for example, this query gets the size for all the first 500 images included in my userpage. --Ricordisamoa 00:24, 12 May 2013 (UTC)

Very good, I like this API based way of solving the problem. I'll consider hacking this when I have a couple of spare hours (in, say June!). Thanks --Fæ (talk) 04:16, 12 May 2013 (UTC)

Category:Images from the Geograph British Isles project needing categories

We need to have all the red categories such as Category:Images from the Geograph British Isles project needing categories in grid NZ2073‏‎ created in Category:Images from the Geograph British Isles project needing categories. If possible, they should contain a self-deletion template, that tags the category as speedy/empty when empty. It would be great if those categories could be associated with a county, provided it doesn't needs a lot of efforts. Thank you. --Foroa (talk) 12:22, 19 April 2013 (UTC)

I don't see what you're talking about.  Hazard-SJ  ✈  02:19, 23 April 2013 (UTC)

These red categories need to be created till the contents is removed and properly categorised. When they are empty, they can be deleted as done in Category:Images from the Geograph British Isles project needing categories in grid SP1579. --Foroa (talk) 09:39, 23 April 2013 (UTC)

Wow, that certainly is a lot, but I'm up for it. All we need on them is a condition for self-requesting deletion, mark it as hidden, and add the category you mentioned? If so, I'll code it and request approval to do it.  Hazard-SJ  ✈  01:00, 25 April 2013 (UTC)

For information as it is related to this category creation request, the plan for Faebot is to continue going through every Geograph image and classifying by county name (see User:Faebot/Geograph). This means that for any image with no categories, it will be added to the county (or equivalent administrative region) category so it at least has a starting point and is more likely to be found by the general public user. Thanks --Fæ (talk) 09:49, 23 April 2013 (UTC)

Ok, in Category:Images from the Geograph British Isles project grid squares by number of files which is not very useful in my opinion as the count decreases permanently; it might be more useful if the content count at creation time is the sort key. On the other hand, you have an idea about the problem with categories that don't autodelete. --Foroa (talk) 16:20, 29 April 2013 (UTC)

Is there a blocking factor ?--Foroa (talk) 07:15, 3 May 2013 (UTC)

Similar request for [5]]: all red Federațiunea categories should be created in Category:Federațiunea. If possible: sort key:Year, month. --Foroa (talk) 10:09, 7 May 2013 (UTC)

And similar fro Category:Foae pentru minte. --Foroa (talk) 10:29, 7 May 2013 (UTC)

Links to renamed accounts

As some people might have noticed, m:Single User Login finalisation announcement will have the result that certain Commons accounts will be renamed automatically at the end of the month and then automatically reassigned to someone else. For example, there might be an account "User:Example" which is automatically renamed to "User:Example~commonswiki" upon which the user name "User:Example" is reassigned to a user on German Wikipedia. Would it be possible (and acceptable) to get a bot which automatically updates all links to the associated user and user talk pages? Otherwise, we will probably end up with lots of links which point to the wrong user because the user name has been reassigned to someone else, and this can be confusing. Or is the Wikimedia Foundation planning to do this for us as part of the renaming task? --Stefan4 (talk) 13:57, 2 May 2013 (UTC)

I think that would be quite easy on user by user basis, but I am not sure how to do it for all accounts. --Jarekt (talk) 11:31, 7 May 2013 (UTC)

Ah, but we are not talking 'all' accounts, only a limited list of these SUL changes. If the list of old/new account names is available somewhere, then updating all links is do-able by pulling the what links here list for each old account name link. --Fæ (talk) 06:29, 15 May 2013 (UTC)

Category:Images from KIT

Hi, could a bot generate a list of all red categories with a name that starts with "Category:Images from KIT". Depending on the outcome, we will see how we can handle those categories that are occupying wanted categories since several years.

Same request for categories that start with "Category:Images from US Navy". --Foroa (talk) 07:14, 3 May 2013 (UTC)

Foroa, below are a few sets:

--Jarekt (talk) 12:02, 7 May 2013 (UTC)

I can create those categories if we agree on what should be in each. --Jarekt (talk) 19:09, 7 May 2013 (UTC)

Thank you. Could you please update the list to contain all red Images from KIT categories ? Hopefully, I can get them all arranged this time. Thank you. --Foroa (talk) 12:34, 15 May 2013 (UTC)

OK, there are still hundreds of such categories hanging there for several years. Would it be possible to create all the red ones in Category:Images from KIT categories to be classified, I will move them later with cat-a-lot in more appropriate categories. --Foroa (talk) 09:45, 20 May 2013 (UTC)

 Done --Jarekt (talk) 02:11, 21 May 2013 (UTC)

Romanian newspapers

Thank you. Maybe first the simple cases: the tree last items, the Romanian newspapers (as their uploader is blocked): just in their parent category, no key, as it is easier then to move them in year cats using cat-a-lot. --Foroa (talk) 18:13, 8 May 2013 (UTC)

Thank you for the ones in Category:Foae pentru minte, but there are still many red ones left over. --Foroa (talk) 06:01, 9 May 2013 (UTC)

It looks as if the added categories have been limited to categories with 4 or more items. You can take them all; one day or another the pour Romanian uploader will be unblocked and those cats will fill up anyway. I noticed that we have no Romanian administrator to discuss the problem, wondering what we could do. --Foroa (talk) 09:29, 9 May 2013 (UTC)

 Done now --Jarekt (talk) 00:35, 10 May 2013 (UTC)

Many thanks, only a bunch left for Category:Foae pentru minte. --Foroa (talk) 06:23, 10 May 2013 (UTC)

It makes no sense not to create red ones that contain only 4 or less items, we better settle the problem ones and for all, and those categories will not go away. We are getting Special:WantedCategories down to a basic level of 5 items, and hopefully within a few months to a level of 4. --Foroa (talk) 08:17, 10 May 2013 (UTC)

I think it is done now. I created my original lists based on Special:WantedCategories page which omits categories with not enough files. --Jarekt (talk) 17:18, 10 May 2013 (UTC)

Category:United States Navy images by location to be classified

All red categories categories that start with "Category:Images from US Navy" should be categorised as Category:Images from US Navy, location Bellingham, Wash. in Category:United States Navy images by location to be classified, no sort key. Those categories become typically a subcat or a redirect as the upload bot becomes active from time to time. --Foroa (talk) 14:48, 14 May 2013 (UTC)

 Done --Jarekt (talk) 12:11, 15 May 2013 (UTC)

Images from AELG website

Hi. See the discussions Commons:Deletion requests/Category:Alfredo Conde in AELG and Category talk:Images from AELG. The category Category:Images from AELG contains 264 subcategories of individual writers from the website of Association of Writers in Galician Language. The category names like Category:Alfredo Conde in AELG are incorrect and confusing. The adverbial should express the source of the photos, not the location nor the membership. The discussion appears to come to the solution:

• All 264 categories should be renamed (from [[:Category:‹name› in AELG]] to [[:Category:Photographs of ‹name› from AELG website]]) and be equiped with appropriate {{DEFAULTSORT:}}.
• For each of the 264 categories, the parent category of the writer should be created if doesn't exist yet. (Create [[:Category:‹name›]] as the parent category for every [[:Category:‹name› in AELG]]). The main category of the person should be categorized into Category:Writers in Galician by default, using appropriate {{DEFAULTSORT:}}.

--ŠJů (talk) 11:27, 14 May 2013 (UTC)

Exuse me. You say we come up with the solution. But I disagree. I dont think such difficult system you are proposing is good. I would propose better system:

1. Add to all files in categories of Category:Images from AELG source template, which will add images to Category:Images from Association of Writers in Galician Language. Possibly it can be done just by setting all those images to category.
2. Hide Category:Images from Association of Writers in Galician Language.
3. Rename Category:‹name› in AELG to Category:‹name›
4. Add all new Category:‹name› to Category:Members of Association of Writers in Galician Language
5. Posibly delete all categories:‹name› in AELG. Category:Images from AELG might be redirected to Category:Images from Association of Writers in Galician Language

This system is nothing unusual. It is based on similar systems which already exists on Commons, like previously mentioned files from Bundesarchive (in English German federal archive), where is the following system:

Bundesarchive – all images donated by German Federal Archive are placed in Category:Images from the German Federal Archive. But this category is hidden. It means technical category telling us the origin of files. As far as there is a field "source" in the description, there is no nead to have a valid category for source. Than every each image is set to category like subject. So e.g. File:-Bundesarchiv B 145 Bild-F055060-0007A, Köln, SPD-Parteitag, Börner-2.jpg, which depicts Holger Börner somewhere is in Category:Holger Börner and by templates its set to technical categories: Images from the German Federal Archive, CC-BY-SA-3.0-DE, Fotowerkstatt, Bilderwerkstatt, Edited images.

So I am proposing to have here categories with valid English names (i.e. no abreviations). Category:Images from Association of Writers in Galician Language will than include all images donated by AELG and Category:Members of Association of Writers in Galician Language will than include all categories with AELG members (i.e. Category:‹name›). Category:Images from Association of Writers in Galician Language will be hidden, because it is technical category. Than there will be 264 basic categories: Category:‹name› including pictures donated by AELG, but also coming form other sources. So if we have a look on one example:

EXAMPLE:

File:Alfredo Conde (AELG)-2.jpg will be in Category:Alfredo Conde, Category:name of place (where the image was taken) and hidden Category:Images from Association of Writers in Galician Language. Category:Alfredo Conde will be set in Category:Members of Association of Writers in Galician Language.--Juandev (talk) 14:15, 14 May 2013 (UTC)

I agree with ŠJů proposal because with this categories we have subcategorized more than 800 photos. If all the photos were from different writers the proposal of Juandev would be correct but in this case we have more than one photo of many of the writers.

Only a note to ŠJů proposal, I think that all the parent category of the writers exists. Bye, --Elisardojm (talk) 22:11, 14 May 2013 (UTC)

The categorization structure proposed here by Juandev is also acceptable, his argument by Bundesarchive is very appropriate. Wikimedia Commons use commonly not so detailed categorization within a category by source. Would you willing to accept such solution, Elisardojm?

As regards a dilemma between original abbreviation and translated full name of the organization, I would prefer the English full name for the main category of the organization and original abbreviation in names of subcategories and templates.

Btw, Elisardojm, you wrote: „I think that all the parent category of the writers exists.“ I checked all the 264 existing subcategories of Category:Images from AELG. Only 3/264 of the subcategories (Alfredo Conde, Darío Xohan Cabana, Pilar García Negro) have their parent category of the writer. I think, Juandev's solution is more simply and useful than to create 261 new categories as the second categorization layer. --ŠJů (talk) 12:30, 15 May 2013 (UTC)

Thus, I try to summarize the proposed process modified according to the discussion:

1. Label all included images with {{AELG}}.  Done a long time ago.
2. Categorize all images with {{AELG}} to the hidden source category Category:Images from AELG website through the template.  Done right now.
3. Move (rename, merge) all 264 subcategories of Category:Images from AELG from the current name [[:Category:‹name› in AELG]] to the new name [[:Category:‹name›]]. Together with this step for every category:
   1. Add appropriate {{DEFAULTSORT:<last name>, <first (and second) name>}}
   2. Remove the category from Category:Images from AELG (delete the emptied category)
   3. Add the category as a subcategory into Category:Writers in Galician by default, possibly also (or alternatively) into Category:Members of the Association of Writers in Galician Language

Some improvements or objections? --ŠJů (talk) 12:30, 15 May 2013 (UTC)

Sorry, I thought that all categories had its parent category of writer, but if the parents category don't exists we can create them.

The main problem that I see to the proposal of Juandev is that it delete the subcategories of the writers of the AELG category and this category will have more than 600 photos. These photos can be grouped because there is a lot of them that are of the same writer, and therefore I think that would have to be in subcategories so that the main category was orderly. The same category of the images of Bundesarchive has subcategories, by year, by author, etc.

I agree to rename the categories that have an unsuitable name. Bye, --Elisardojm (talk) 11:42, 20 May 2013 (UTC)

Category:Pages with broken file links

I need a bot for running null-edits on all content in this cat and subcats. Running it once should be sufficient. For the subcat Category:Files with broken file links I need a regular bot run at least twice a week for null-edits. These null-edits will result in cat-removal from all images where a thumbnail image was missing upon upload, all vandalized images or images with missing link targets would stay in the matching cats. --Denniss (talk) 23:42, 19 May 2013 (UTC)

I'll have this done as soon as I can.  Hazard-SJ  ✈  03:52, 22 May 2013 (UTC)

 Partially done. I've run a script through Category:Pages with broken file links non-recursively (it null-edited about 1617 pages in all), taking the category down from 1,504 pages to 43, but in turn, increasing the other categories' numbers of pages. I've updated the code to recurse whenever it goes through that category, so in the future it should end up going through all the categories.  Hazard-SJ  ✈  05:05, 23 May 2013 (UTC)

It looks as if the automatic background category page update on Commons has changed. Many template driven categories work now with a delay of days or even weeks. (non empty and broken category redirects, disambiguation pages, ...). --Foroa (talk) 06:02, 23 May 2013 (UTC)

Thanks for running the script, as stated in the first post a recurring run is only required for the files subcat as it's usually a problem with the thumbs not present on image upload (damn user-specific templates displaying a thumb of the image itself, cat already grown back to 350+ files). I could not really see an automatic background category page update - most content of the main cat stayed there for months unless a page was actually edited. And I must have (null)edited thousands of them. BTW is it possible to run this script through the files subcat on specific days, say Monday and Thursday at 0:00 UTC? --Denniss (talk) 11:20, 27 May 2013 (UTC)

A bot

Hi!

My name is Sophie Österberg and I'm working as an education manager at Wikimedia Sverige. I'm trying to help teachers here to set up a portal for them on Commons as they want a place to put, organise and see their flipped-classroom movies. But I would need this (as specified by a colleague who knows much more than I about this)

Specifications: A bot which once every 24 hours:

1) checks a given maintenance category Y, and for each movie A in this category:

• • visits the movie description and identifies the categories it is in.
  • check if any of these categories correspond to a pre-specified list Z of keywords. If so associate A with the corresponding value (z)
  • remove category Y from the movie

2) Visits a given page containing one gallery associated which each keyword in Z.

• For each of the keywords in Z add any associated movie (from step 1) to the top of the relevant gallery.
• prune any galleries so that they are no longer than W movies.

I would be so happy if anyone would find this challenging and intriguing and thus help teachers in Sweden with their portal by creating this. If so, I would be very grateful. Sophie Österberg (WMSE) (talk) 11:59, 21 May 2013 (UTC)

For your gallery bot to be even more community driven, I suggest a project page somewhere contains a list of "maintenance categories" (there may be several) and a list with "key words" to be monitored, possibly each with the associated link to its gallery page/sub-page. Using on-wiki lists that the community can adapt and see the bot respond to within 24 hours, will avoid any need to ask for the bot script to be changed for any of these configuration details. There are bot scripts that do very similar things, so this seems very do-able. --Fæ (talk) 12:16, 21 May 2013 (UTC)

Ok, sounds like a great idea! --Sophie Österberg (WMSE) (talk) 12:24, 22 May 2013 (UTC)

Broken redirects

Could a bot, on a regular base, delete all the broken file redirects in Special:BrokenRedirects since Bernsteinbot no longer reports those. --Foroa (talk) 09:05, 25 May 2013 (UTC)