User talk:JL-Bot

Not sure this is configured correctly

On this page:

• Portal:Scouting/Recognized_content

It looks like a couple of topics are repeated, and on (the DYK section) isn't list at all. How can I fix this? --evrik ^(talk) 16:25, 7 June 2022 (UTC)

@Evrik: Do you have examples of things that are repeated/not listed? Headbomb {t · c · p · b} 22:30, 7 June 2022 (UTC)

For the DYK section, there was an extra equals sign on the end of the line. I removed that and re-ran. The DYK section is now present. I'm not seeing any duplicate topics. Let me know what you are questioning and I will take a look at it. -- JLaTondre (talk) 01:06, 8 June 2022 (UTC)

• @JLaTondre:, thank you for the fix. @Headbomb:, "Picture of the day pictures" seems to duplicate "Featured pictures", "Featured lists" mirrors "Main page featured lists" are the two with the most overlap. --evrik ^(talk) 01:22, 8 June 2022 (UTC)

  Well that's because they're different things.

  I suppose some condensed output could be something like

  • Neil Armstrong ( 2019-07-21)

  Could also be done. Headbomb {t · c · p · b} 03:00, 8 June 2022 (UTC)

• That looks good. How would I code that on the page? --evrik ^(talk) 03:08, 8 June 2022 (UTC)

• JLaTondre would first have to code support for that, and then the instruction would be at WP:RECOG if/when that's implemented. Headbomb {t · c · p · b} 05:04, 8 June 2022 (UTC)

The bot currently outputs by category. There is overlap in the categories (for example, I would assume all "Main page featured lists" are "Featured lists"), but they are not the same (not all featured content appears on the main page). So what is the ask here?

1. When outputting a featured type (ex. "Featured lists"), provide an option to add a Wikipedia icon if it appeared on the main page?
   1. It would then be up to specifier to only include the one category type and the option; or
   2. The bot would only display the larger category if the option is set even if both categories are specified
2. Provide an option to consolidate by page type (article, list, picture, sound) where it would show a different icon for each recognized type? So for example, the new section could be "Recognized articles" and you would get a different icon and date (where applicable) for each of "Featured article", "Former featured article", "DYK", etc.

-- JLaTondre (talk) 11:55, 8 June 2022 (UTC)

Well, TFAs should be either current or former FAs. So a condensed option would, IMO, required that both FA and FFA are covered. And then the TFAs could be 'merged' into FA/FFA. Likewise for TFLs, which would be merged into both FL and FFL sections. So if you have

 |content-featured-articles
 |content-former-featured-articles
 |content-mainpage-featured
 

the output would be as is, but if you had something like

|content-featured-articles
|content-former-featured-articles
|content-mainpage-featured=condensed (or something equivalent)

then the output would be merged as above (or similar, depending on whether or not icons were desired) Headbomb {t · c · p · b} 12:38, 8 June 2022 (UTC)

@JLaTondre: if you run the bot on Portal:Scouting/Recognized_content I can do a full mockup. Headbomb {t · c · p · b} 01:55, 12 June 2022 (UTC)

Done. -- JLaTondre (talk) 15:40, 12 June 2022 (UTC)

Mockup. POTD icons will be supported once Template_talk:Icon#POTD_support is enacted. Headbomb {t · c · p · b} 16:32, 12 June 2022 (UTC)

@JLaTondre and Headbomb: Thank you both! --evrik ^(talk) 02:33, 13 June 2022 (UTC)

FM captions

Is there a reason why the "caption" option actually displays the media's title rather than its caption? The titles are so rarely helpful, while the captions would definitely be! MeegsC (talk) 17:38, 7 August 2022 (UTC)

Primarily performance, but also the lack of standard formats. Captions are on the pages that use the images and it would be add significant time to go pull them. The task already takes most of a day to run. Images can also appear on multiple pages with significantly different captions. Captions are not always in a standard format which makes pulling them from the page text problematic. -- JLaTondre (talk) 12:00, 9 August 2022 (UTC)

Could we not use the captions that are in the picture file, and default to the title only if the picture file doesn't have an English caption? Most file captions are better than the title! MeegsC (talk) 13:14, 9 August 2022 (UTC)

The description field? Yes, that might work. I will look into it. -- JLaTondre (talk) 16:36, 13 August 2022 (UTC)

Wikipedia:WikiProject_Academic_Journals/Journals_cited_by_Wikipedia/Maintenance/Invalid_DOI_prefixes

DOI prefixes now go up to the 80000s. Cutoff should be 90000s. Headbomb {t · c · p · b} 07:00, 9 July 2026 (UTC)

Since it's pulling the DOIs from CrossRef, I should be able to automate the upper boundary. -- JLaTondre (talk) 10:05, 9 July 2026 (UTC)

Looking at the code, the upper boundary is already set based on CrossRef. CrossRef's highest result is 10.71936. The first three DOIs on the Invalid page do not resolve in CrossRef, but they do work when linked through doi.org. I'd prefer to not re-hardcode the limit (as it will eventually change again). Do you know any other references to query other than CrossRef? -- JLaTondre (talk) 22:21, 12 July 2026 (UTC)

This is the list of registration agencies: https://www.doi.org/the-community/existing-registration-agencies

I don't know any other that are query-able, it's been a while since I looked, so things may be different now. I also didn't have a list of agencies either, so the above will greatly simplify a search for APIs. Headbomb {t · c · p · b} 23:13, 12 July 2026 (UTC)

This may have some way to query it

• https://support.datacite.org/docs/api

Maybe useful links

• https://support.datacite.org/docs/how-do-i-query-the-rest-api-and-whats-in-the-response
• https://support.datacite.org/docs/how-can-i-use-the-datacite-api-to-search-and-retrieve-the-dois-and-metadata-for-my-organization-or-a-specific-repository

Headbomb {t · c · p · b} 23:42, 12 July 2026 (UTC)

I didn't see a way to get prefixes (or publishers) from that API. It seems more DOI focused. My approach to solving this issue for now is: If the DOI prefix is greater than the largest returned by CrossRef, and the DOI is of a valid format, look up the DOI via doi.org (for example https://doi.org/10.82042/jba.2.2960). Treat it as invalid if that doesn't resolve. Since there is only a few, that approach works. Hopefully CrossRef will catch up over time. If instead the gap grows, I can look at alternatives. -- JLaTondre (talk) 16:48, 19 July 2026 (UTC)

multiple DYKs break listings

See causing

Headbomb {t · c · p · b} 07:04, 12 July 2026 (UTC)

The current behavior is to group multiple DYKs for the same page - see last few entries on Wikipedia:WikiProject Women in Green/DYK/2026 DYK Blurbs. For most uses (combined listings), this is the best behavior. I will look at updating the dyk-blurb-paged processing to leave them separate. -- JLaTondre (talk) 22:50, 12 July 2026 (UTC)

Done. When in dyk-blurb-paged mode, DYK blurbs will no longer be grouped. -- JLaTondre (talk) 15:10, 19 July 2026 (UTC)