Wikipedia:Bots/Requests for approval/BHGbot 8
- The following discussion is an archived debate. Please do not modify it. To request review of this BRFA, please start a new section at Wikipedia:Bots/Noticeboard. The result of the discussion was
Withdrawn by operator.
New to bots on Wikipedia? Read these primers!
- Approval process – How this discussion works
- Overview/Policy – What bots are/What they can (or can't) do
- Dictionary – Explains bot-related jargon
Operator: BrownHairedGirl (talk · contribs · SUL · edit count · logs · page moves · block log · rights log · ANI search)
Time filed: 03:20, Tuesday, August 3, 2021 (UTC)
Function overview: remove all uses of {{Bare URL inline}} from a page, and add {{Cleanup bare URLs}} unless it is already present. This is because both User:Citation bot and WP:reFill skip bare URLs tagged with {{Bare URL inline}}. So the inline tag identifies bare URLs, but impedes a fix. The banner tag {{Cleanup bare URLs}} does not impede the tools.
Automatic, Supervised, or Manual: Supervised
Programming language(s): AWB
Source code available: Wikipedia:Bots/Requests for approval/BHGbot 8/BHGbot-8-AWB-module
Links to relevant discussions (where appropriate):
Edit period(s): Initially a one-off run to clear the backlog. Then periodically until reFill2 and Citationbot are both fixed.
Estimated number of pages affected: {{Bare URL inline}} is currently transcluded in 13,984 pages. I will process all the pages in the article and draft namespaces: current total 13,967.
Namespace(s): article and draft
Exclusion compliant (Yes/No): Yes
Function details: This job will not tag any articles which are not already tagged. Its purpose is to replace one tag with another:
- count the number of {{Bare URL inline}} tags in the page, including aliases
- if the number is zero, skip the page; the following steps don't happen
- remove all the {{Bare URL inline}} tags
- count the number of remaining {{Bare URL inline}} tags in the page, including aliases
- check whether the page already contains a {{Cleanup bare URLs}} tag, including aliases
- if yes, canonicalise the {{Cleanup bare URLs}} tag
- if no, prepend the {{Cleanup bare URLs}} tag (with date)
- Apply WP:GENFIXES
- Save page, with edit summary of the form:
WP:BHGbot 8: removed 4 of 4 {{Template:Bare URL inline}}. Added {{Cleanup bare URLs}}
orWP:BHGbot 8: removed 3 of 3 {{Template:Bare URL inline}}. Already tagged with {{Cleanup bare URLs}}
Discussion
[edit]- Note from bot owner: The primary reason for creating this bot is to cleanup after myself. The TLDR is that while doing mass-tagging with {{Cleanup bare URLs}}, I responded to concerns about its intrusiveness by switching to applying {{Bare URL inline}}. Problems with that tag were drawn to my attention only after over 10,000 pages had been tagged. Here is the full explanation:
- In May 2019, I began using AWB to tag article with the banner tag {{Cleanup bare URLs}}. After an initial run mid-month, I began a big batch in the last three days of the month, in order to facilitate one editor who wanted to be able to clear category. My AWB run triggered some discussion on my talk, with some concern about the intrusiveness of the {{Cleanup bare URLs}} banner template, including from Guliolopez. To resolve this, I tested using {{Bare URL inline}} instead, and Guliolopez agreed it that it seemed to be an improvement.[1]
- I had tagged about 10,000 pages with with the banner tag {{Cleanup bare URLs}}, and then did about 5,000 more with {{Bare URL inline}}. Using the inline tag seemed to largely resolve the concerns which had been expressed.
- There was a further comment about some comment somewhere preferring the banner tag, but the editor who mentioned it couldn't track it down. the tagging has helped a lot with cleanup: see User talk:BrownHairedGirl#Stats_on_bare_URLs_tagged_in_May_2021.
- Then at the end of June, I started another batch, using {{Bare URL inline}}, and after a few hours the objections started. Problem-solving discussion was impeded by one very aggressive editor, but eventually it became clear that a decent number of more reasonable editors objected to the inline tag blocking reFill2. So I stopped the June AWB job after about 8,000 edits.
- This bot task is to allow reFill2 to be used on these articles, and also to allow Citation bot to be used. Since stopping my AWB tagging, I have fed tens of thousands of bareURL pages through Citation bot, which usually manages on a first pass to fix all bareURL refs in about 25–35% of articles; a second pass will usually clear about 10–15% of the remainder, and in either case some other articles get partial fixes.
- Neither Citation bot nor reFill2 removes {{Cleanup bare URLs}} even after fixing all the tags. reFill2 is now unmaintained, and my request for Citation bot to do so was refused; see User talk:Citation_bot/Archive 26#Cleanup_tag_not_removed_after_problem_fixed. I think I have figured out how to make a bot job which will remove most redundant {{Cleanup bare URLs}} tags in a way that will satisfy the concerns raised there by Headbomb. That will be a separate BRFA.
- I will followup in the morning with more notifications to interested parties.--BrownHairedGirl (talk) • (contribs) 03:20, 3 August 2021 (UTC)[reply]
Arument between editors irrelevant to this request. Please take behavioral issues, accusations of trolling, and complaints of personal attacks to the appropriate venues. !ɘM γɿɘυϘ⅃ϘƧ 22:52, 7 August 2021 (UTC)[reply]
|
---|
|
- So effectively, you're saying {{Bare URL inline}} is useless? If so, why not TfD it? ProcrastinatingReader (talk) 22:33, 7 August 2021 (UTC)[reply]
- @ProcrastinatingReader: TFDing the template would solve nothing. It would leave the articles untagged, which would mean that the tools could fix it, but only theoretically because the pages would no longer be identified. That's why I propose instead to replace these uses with a template which allows both identification and tool-assisted fix. --BrownHairedGirl (talk) • (contribs) 00:42, 8 August 2021 (UTC)[reply]
- Well, in this BRFA you're saying {{Bare URL inline}} shouldn't be used (as an inline template) and instead {{Cleanup bare URLs}} should be (at the top, as a maint template), right? So, isn't that the same as TfDing the inline one, converting all the usages once, and then telling people to use the maint template in the future? Like AFAICS this BRFA is suggesting the inline template never be used in practice, or am I mistaken? ProcrastinatingReader (talk) 01:57, 8 August 2021 (UTC)[reply]
- @ProcrastinatingReader: TFDing the template would solve nothing. It would leave the articles untagged, which would mean that the tools could fix it, but only theoretically because the pages would no longer be identified. That's why I propose instead to replace these uses with a template which allows both identification and tool-assisted fix. --BrownHairedGirl (talk) • (contribs) 00:42, 8 August 2021 (UTC)[reply]
- @ProcrastinatingReader:, this BRFA is based on three core facts:
- as far as I know, every existing use of {{Bare URL inline}} was added by me. My work had unintended ill-effect (due my god faith adoption of a good faith recommendation by another editor), and I want to put things right.
- as of now, the inline template impedes the use of reFiull2 and Citation bot to cleanup the problem;
- neither tool likely to be fixed any time soon, but switching to the banner template will allow a prompt bot-driven cleanup of the bare URLs in probably about 40% of these articles.
- The future of the inline template is a broader question which cannot be decided here. There are several options for its future, one of which is to deprecate its use until the tool are fixed, and another of which is delete it. (Personally, I prefer keep but deprecate, because as both @Enterprisey and L.tak note below, it is in theory a batter way of tagging, if the tools are ever fixed).
- However, if it is deleted or deprecated at TFD, then the functionality of this bot will be needed to convert from {{Bare URL inline}} to {{Cleanup bare URLs}}; plain deletion of the template will not be enough. So as far as I can see, it is best to proceed with replacement, and separately discuss the future of the template.
- All I want is to ensure that the 13,000 pages which I tagged in good faith with {{Bare URL inline}} can be cleaned up ASAP to remove the bare URLs. Please can we get on with that, and sort out the broader questions at a more appropriate venue? --BrownHairedGirl (talk) • (contribs) 02:55, 8 August 2021 (UTC)[reply]
- You're correct on your second paragraph. That TfD would then serve as the consensus for this task and the outcome. I don't really see the point in authorising a task to convert instances of a permissible template, though. I think another venue should be sought to decide whether this template is permissible (TfD). If it is, then this task seems unnecessary. If it isn't, then we can either approve this task or set one of the TfD bots to do it. ProcrastinatingReader (talk) 11:07, 8 August 2021 (UTC)[reply]
- @ProcrastinatingReader: That's a pity. The point of this task is to solve a problem now, without prejudice to any decision on the future of the inline template. The TFD bots don't have the functionality to do the job of this bot, so what you are proposing is another process which may or may not decide to delete {{Bare URL inline}}, and then another BRFA period.
- The alternative I am proposing is to fix a problem which I inadvertently caused, so that thousand of articles can be promptly cleaned up by Citation bot and no longer need either {{Bare URL inline}} or {{Cleanup bare URLs}}. Why add extra bureaucracy and delay? --BrownHairedGirl (talk) • (contribs) 13:38, 8 August 2021 (UTC)[reply]
- TfD bots can definitely do the task described in your "Function details".
- BAG is required to ensure consensus exists for a task. I feel it should also take a 'long view' of the situation rather than approve band aids. So as I say, either the template is fine (and thus this task unnecessary), or the template is not fine (and should be scheduled for deletion, which may be implemented by this bot or another). In any case, what is not really fine is approving a bot to move all usages of one template to another while keeping both templates. This task is basically equivalent to the deletion of the inline template. ProcrastinatingReader (talk) 13:43, 8 August 2021 (UTC)[reply]
- @ProcrastinatingReader the problem here is that your binary choice of {{Bare URL inline}} being either fine or not fine is a false binary. As discussed below, the situation would be better described as "template good, but key tools currently can't handle it", so a likely TFD outcome would be "keep, but deprecate for now"... which doesn't mandate the TFD bots.
- This bot job is a pragmatic attempt to allow the articles to be promptly cleaned so that many of them will need neither tag. I added all these tags based on well-intended but flawed advice, and I want to fix it. So how about we restrict this bot's scope to only the articles which I tagged in May and June? That way, it's simply a cleanup of my work, with wider issues left for other venues? --BrownHairedGirl (talk) • (contribs) 14:32, 8 August 2021 (UTC)[reply]
- If the assumption below is true, in that there's no reasonable prospect for the tools being updated, then that assumption of what TfD would conclude may have to be reconsidered. I don't think BRFA should 'assume' what the community would say, in such a case.
- I appreciate you're trying to be pragmatic here but I do have concerns on both the appropriateness of the task, and the method of assessing consensus for it. I would probably decline it personally, or at least put it on hold pending a TfD, but I will let another BAG make the final call in case someone else sees it differently. ProcrastinatingReader (talk) 14:41, 8 August 2021 (UTC)[reply]
- All I want here is to fix my own tagging run, to put it into the shape that I had originally intended before I accepted flawed advice. Does it really help en.wp to impede an editor from cleaning up after themselves and bringing us closer to an uncontested goal of having fewer articles with bare links? --BrownHairedGirl (talk) • (contribs) 14:54, 8 August 2021 (UTC)[reply]
- I totally see where you're coming from (goodness knows I've had my share of cleaning up after my own mistakes), but I genuinely don't see your tagging runs as "a mistake"; you exposed a flaw in the system, and rather than work towards fixing the flaw you want to just roll everything back to square zero and pretend it didn't happen. If and when a fix does get made, are you going to then go through those 14k pages to re-add the inline tag if necessary? That seems like a huge waste of effort, not to mention the inevitable complaints of watchlist-spamming. If the issue can be resolved by updating the bot with zero (or only a few) extra edits on anyone's part, we all win.
- Obviously, if the template itself is not wanted, then by all means we can approve this bot, but we're not at that stage yet. Primefac (talk) 15:04, 8 August 2021 (UTC)[reply]
- @Primefac, no I don't want to roll everything back, and I don't want to pretend it never happened. Quite the opposite.
- I want to move forward to fixing the bare link problem that I tagged. Instead of reverting my tagging, I am proposing to convert to te tag I was previously using,in order to feed the articles to citation bot where i am confident that the bot will fix about 40% of the articles in a few days. The banner tag will also facilitate editors using other tools which are impeded by the inline tag, so the result of my proposal here will be a huge immediate cleanup of the problem, followed by ongoing further progress.
- Please remember that the aim of this exercise is not to have articles tagged. It is to have the problem fixed so that the tags are no longer needed, and that is what is am trying to achieve here by running this bot. (Please note that for the last 4 weeks I have been running ~5,000 articles per day through Citation bot, removing all bare links from 20-255 of the on first pass. I want to do that for these articles too, but I cannot do so while the inline tags are in use).
- If at some future point, the tools are fixed and there is a preference for the inline template, then yes, I will happily open a BRFA to convert the remaining tags back to the inline template;my work on this module has shown me how I could do that. But ifand when that happens, the set of such articles will be radically smaller, because most of them will already have been fixed.
- And sure, it would be best to have the tools fixed. But there is no sight at all of that happening. So it seems to me that waiting for that is like Waiting for Godot. --BrownHairedGirl (talk) • (contribs) 16:28, 8 August 2021 (UTC)[reply]
- All I want here is to fix my own tagging run, to put it into the shape that I had originally intended before I accepted flawed advice. Does it really help en.wp to impede an editor from cleaning up after themselves and bringing us closer to an uncontested goal of having fewer articles with bare links? --BrownHairedGirl (talk) • (contribs) 14:54, 8 August 2021 (UTC)[reply]
- You're correct on your second paragraph. That TfD would then serve as the consensus for this task and the outcome. I don't really see the point in authorising a task to convert instances of a permissible template, though. I think another venue should be sought to decide whether this template is permissible (TfD). If it is, then this task seems unnecessary. If it isn't, then we can either approve this task or set one of the TfD bots to do it. ProcrastinatingReader (talk) 11:07, 8 August 2021 (UTC)[reply]
- The inline tag template seems like a useful template, being more specific than the banner; can Citation bot and refill be updated instead? Enterprisey (talk!) 00:16, 8 August 2021 (UTC)[reply]
- I also would prefer the line of action Enterprisey suggests. The banner now also comes up on perfectly referenced pages with a few (or even a single!) bare url. For these purposes {{Bare URL inline}} seems much more useful. L.tak (talk) 00:21, 8 August 2021 (UTC)[reply]
- @Enterprisey and L.tak: in theory, I agree 100% that the inline template is better. That's why I adopted the good faith suggestion by Guliolopez to use it. But in practice, I found that to much displeasure of some editors, the inline template impedes actually a solution.
- Similarly, I agree 100% in theory that fixing the bot and reFil2 would be best. But in practice that's a do-nothing option because reFill 2 has been wholly unmaintained for several years, so in practice no fix is available. (It's so outdated that some of its core functionality is broken, even spewing out unsupported parameters such as
|deadurl=y
) - So we now have three options for these ~13,000 articles:
- Let this bot change the templates. Then, within a few days days Citation bot can complete a run through all 13,000, and probably fix 20%–35% of them on a first pass, and another ~10% on a second pass. The rest will be identifiable for editors who want to fix them, and they can use reFill 2.
- Remove the templates. That leaves the problem unidentified, so in most cases there will be no fix.
- Wait for reFill2 to be fixed. But since there is no active maintenance, that is at best like Waiting for Godot: any cleanup is blocked indefinitely by the broken tools.
- I will happily end this bot task as soon as reFill2 and Citation bit are fixed. But in the meantime, please please please please please don't let unattainable perfection be the enemy of a good step forward which will lead to almost half of the tagged issues being being fixed by a bot within days (or at worst weeks). --BrownHairedGirl (talk) • (contribs) 01:44, 8 August 2021 (UTC)[reply]
- For me there are extra options beyond the ones you presented, as we -as you say- shouldn't in a binary way say the inline template is bad, so it shouldn't in 13000 cases be changed, but it seems to hinder certain automated tools. I see three options (although I can imagine the first one is technically impossible):
- remove the tag, run the citation bot, re-add the tag (that would be my preference, but I am not certain if that can be accomplished in a single edit; I am afraid it is not possible)
- only handle those pages where >4 [optionally also: OR >34%] of the refs are bare. In that case I think the big template at the top is warranted.
- add the template to the talk page L.tak (talk) 22:04, 8 August 2021 (UTC)[reply]
- @L.tak:
- As I explained in my opening comments, I have already asked about template removal being added to Citation bot, and the answer was no. So tag addition/removal would have to be a separate task. Removing then re-adding means extra step for the majority of pages, because the majority of page with bare URLs do not get fully fixed by the bot
- If editors want to propose some threshold for tagging bare URLs, that discussion should happen elsewhere (e.g. at Template talk:Cleanup bare URLs), and the template's documentation should be updated with whatever consensus is reached. This BRFA is not the place to propose such restrictions. — BrownHairedGirl 12:18, 14 August 2021 — continues after insertion below
- I would say that turns things around. The template was never meant for single bare refs; and now I;d say you are not making a net positive change by doing these edits. My options are meant to help you select those cases where the effect is a net positive for the project... L.tak (talk) 14:35, 18 August 2021 (UTC)[reply]
- @L.tak: I disagree, because alerting editors that a problem needs to be fixed is to me a net positive.
You are of course entitled to your view about appropriate uses of {{Cleanup bare URLs}}, but you have shown no evidence to support your assertion thatthe template was never meant for single bare refs
. Again, I ask you to seek consensus for your view, and to update the template's documentation if there is a consensus to do so. --BrownHairedGirl (talk) • (contribs) 14:59, 18 August 2021 (UTC)[reply]- I did indicate that that followed from the tags text of the tag "This article uses bare URLs, which may be threatened by link rot." which uses the plural form, and suggests this is a general characteristic of the article. L.tak (talk) 19:34, 18 August 2021 (UTC)[reply]
- And I have already pointed out to you that use of the short form "URLs" instead of the more verbose "1 or more URLs" is more likely due to a desire for brevity. Please take this is to the template's talk page. --BrownHairedGirl (talk) • (contribs) 20:04, 18 August 2021 (UTC)[reply]
- @L.tak: I disagree, because alerting editors that a problem needs to be fixed is to me a net positive.
- I would say that turns things around. The template was never meant for single bare refs; and now I;d say you are not making a net positive change by doing these edits. My options are meant to help you select those cases where the effect is a net positive for the project... L.tak (talk) 14:35, 18 August 2021 (UTC)[reply]
- I am not aware of any other case where a cleanup tag is applied to the talk page, and I see no reason to make this one an exception. --BrownHairedGirl (talk) • (contribs) 12:18, 14 August 2021 (UTC)[reply]
- Neither am I. But these are just options that sound better than
slammingplacing a big template in cases when it is not warranted IMO. L.tak (talk) 14:35, 18 August 2021 (UTC)[reply]- @L.tak:
slamming
is just pejorative terminology which adds nothing to the discussion. Please don't do that. --BrownHairedGirl- You're right, I have stricken that. Civility is of importance indeed. L.tak (talk) 19:34, 18 August 2021 (UTC)[reply]
- @L.tak:
- Neither am I. But these are just options that sound better than
(talk) • (contribs) 14:51, 18 August 2021 (UTC)[reply]
- I find myself agreeing quite heavily with PR and Enterprisey in that we have a perfectly valid template being used. It is the script and bot that are incompatible with this template, which indicate that the script and bot are at fault and should be updated. 14k pages is a drop in the bucket considering how many millions of articles we have, so a week or two waiting for a TFD to resolve itself is a minor inconvenience. If you want to talk about "unattainable perfection [being] the enemy of a good step forward", the (what I suspect to be) small number of pages where "other URLs not in ref tags that the bot doesn't see" (to quote Headbomb) is a smaller issue than the template issue; i.e. making the fix to Citation bot will be a better fix than removing-without-deleting this template.That being said, it is very possible that this template shouldn't exist, with its relative lack of use indicating that a general maintenance banner is more effective than specific-line inline templates. However, that is not our decision to make here, and thus I would strongly advocate for a TFD to determine the fate of the template. If the template is kept, that is a good indication to the bot owners that their code should be updated to respect this template. If you do not want to do so I will file, but I don't think this BRFA is going anywhere until such a step is taken. Primefac (talk) 14:35, 8 August 2021 (UTC)[reply]
- On a related note to my above post, which I will put as its own bullet for ease-of-replying, if there is an issue with Citation bot not removing the {{Cleanup bare URLs}} tag even when there are no bare URLs, I see zero issue with a bot going through and checking these pages and removing it when appropriate until the bot gets updated (up to and including the possibility that it never happens). Primefac (talk) 14:35, 8 August 2021 (UTC)[reply]
- @Primefac, as I noted in the penultimate para of my opening comment at the the top of the discussion, that was going to be my next proposal. But it would take me a few hours to code it and test it and a few more to write it up, and right now I don't feel at all like putting in that time, for two reasons:
- I put a lot of work into this pragmatic, get-things-moving-again-for-now proposal. However, it seems that the mood here is mostly to wait for as long as it takes to achieve perfection in some tools which have long-standing flaws, so my efforts (over a day's work) have been wasted. Other editors are entitled to make that choice, although I disagree with it for reasons already stated. And that leaves me concerned that any time I put into my planned followup bot task will be similarly drowned in the pursuit of an ideal world.
- There was some unpleasant disruption at the start of this discussion which triggered blowback against me, and it all turns out to have been instigated by an IBAN-breaking long-term troublesome editor who manipulated three of us. The consequences of being targeted in this way are now being weaponised against me elsewhere, and while I want to say little here, I am wary of sticking my neck out again.
- So I am now feeling fed up, and minded to just withdraw this proposal. Does anyone believe that there in any point in keeping it open? --BrownHairedGirl (talk) • (contribs) 22:37, 8 August 2021 (UTC)[reply]
- From my perspective, only to get my question answered below. Once that's done, I will close this as withdrawn, and depending on the reply I will take point if it needs attention. Primefac (talk) 22:40, 8 August 2021 (UTC)[reply]
- Fair enough, I'll leave it to you. If this is going nowhere, there is no point in leaving it open to risk taking more of anyone's time. --BrownHairedGirl (talk) • (contribs) 01:37, 9 August 2021 (UTC)[reply]
- From my perspective, only to get my question answered below. Once that's done, I will close this as withdrawn, and depending on the reply I will take point if it needs attention. Primefac (talk) 22:40, 8 August 2021 (UTC)[reply]
Has a fix even been proposed?
[edit]Okay, so just to ask the stupidly stupid question, have the maintainers of Citation bot even been asked about why the bot skips this template, or if they will consider fixing the issue? I've written a few different draft versions of a proposal to hopefully break our logjam, but I keep coming back to the fact that I cannot find any diffs or indication that the bot ops even know this is an issue. Am I missing a discussion somewhere? If they've been asked and declined, then I'll post my proposal at VPT to get consensus on how to proceed, but I have way too many "according to the BRFA filer
" statements and not nearly enough diffs to back them up. Primefac (talk) 20:20, 8 August 2021 (UTC)[reply]
- The bot does not skip these, doing this is just something that no one has ever asked to be added to the bot code. It would take some time to add and debug. Also, the question remains of "what exactly should the bot do?". AManWithNoPlan (talk) 13:38, 18 August 2021 (UTC)[reply]
- The bot will now do URLs flagged with this template within ref tags. AManWithNoPlan (talk) 14:13, 18 August 2021 (UTC)[reply]
- @AManWithNoPlan: I just tried it on a simple test page at User:BrownHairedGirl/sandbox99, without success. The bot made no change to the page. --BrownHairedGirl (talk) • (contribs) 15:07, 18 August 2021 (UTC)[reply]
- the regex assumes the date parameter is set. is that a bad assumption? AManWithNoPlan (talk) 15:13, 18 August 2021 (UTC)[reply]
- seems to assume that date is some bot specific pattern too. no jokes allowed. will make more generic AManWithNoPlan (talk) 15:18, 18 August 2021 (UTC)[reply]
- @AManWithNoPlan: I see it worked after you added the date field.[3] That's great.
- But yes, it is a bad idea to assume that date is set. My AWB job added the date in all cases, but editors adding tags manually typically do so without the date, and rely on AnomieBot to add it. If the bot handled dateless uses, that would allow it to handle pages which the bot has not yet dated, or which the bot has somehow missed. --BrownHairedGirl (talk) • (contribs) 15:24, 18 August 2021 (UTC)[reply]
- Should work now without the date. Also, {{full}} code no longer assumes a date. AManWithNoPlan (talk) 16:06, 18 August 2021 (UTC)[reply]
- seems to assume that date is some bot specific pattern too. no jokes allowed. will make more generic AManWithNoPlan (talk) 15:18, 18 August 2021 (UTC)[reply]
- the regex assumes the date parameter is set. is that a bad assumption? AManWithNoPlan (talk) 15:13, 18 August 2021 (UTC)[reply]
- @AManWithNoPlan: I just tried it on a simple test page at User:BrownHairedGirl/sandbox99, without success. The bot made no change to the page. --BrownHairedGirl (talk) • (contribs) 15:07, 18 August 2021 (UTC)[reply]
- The bot will now do URLs flagged with this template within ref tags. AManWithNoPlan (talk) 14:13, 18 August 2021 (UTC)[reply]
What now?
[edit]I have just tested the Bot's handling of {{Bare URL inline}} on two pages, with success:[4],[5]. Many thanks to @AManWithNoPlan for the prompt fix, and to @Primefac for raising the issue.
However, that leaves unresolved the second part of the reason why I proposed replacing these inline templates with the banner template: WP:reFill, which doesn't support {{Bare URL inline}}. The tool has been unmaintained for some years, and has a number of big glitches. At the end of June, Curb Safe Charmer kindly announced that they would have a go[6], but so far there has been no further news. Obviously, many thanks to CCS for even trying to have a look at what I assume is a lot of complex code, but for the purpose of decision-making here I don't think we can assume that a fix is imminent.
There are currently 13,430 transclusions of {{Bare URL inline}}. From my close monitoring of Citation bot's progress on the batches I feed it, I estimate that when the bot is fed lists of the articles tagged with {{Bare URL inline}}, it will will clear about 30% of pages on a first pass, and that a second pass on the remainder will clear about 10% of the rest. So that's bot clearance of a bit less than 40% of the tagged pages.
That will leave about 8,000 articles still tagged with {{Bare URL inline}}, which blocks repairs by WP:reFill. That still seems to me to be undesirable. --BrownHairedGirl (talk) • (contribs) 16:50, 18 August 2021 (UTC)[reply]
- What's the issue with refill? It works if the bare template isn't there, but does nothing if it is there? Do you have a sample page with this issue? I'm finding it hard to visualise why that would cause a bot/tool to skip. ProcrastinatingReader (talk) 17:24, 18 August 2021 (UTC)[reply]
- @ProcrastinatingReader: The complaint was indeed that
It works if the bare template isn't there, but does nothing if it is there
. I accepted those assertionsby other editors, and did not test them. - In response to your question, I tweaked User:BrownHairedGirl/sandbox99,and tried reFill2 on it...and found that it worked perfectly, both on the link with a bareuRL tag and on the one without: see diff[7].
- So it seems that I may have been misled,and I am kicking myself for not having tested this more thoroughly at the time; I assumed that the objections were fell-founded, and replying to the storm kept me busy. Sorry
Facepalm. --BrownHairedGirl (talk) • (contribs) 18:21, 18 August 2021 (UTC)[reply]
- white space matters https://en.wikipedia.org/w/index.php?title=User%3AAManWithNoPlan%2Fsandbox2&type=revision&diff=1039441007&oldid=1039440919 AManWithNoPlan (talk) 18:39, 18 August 2021 (UTC)[reply]
- @AManWithNoPlan: that's a pity, but the {{Bare URL inline}} tags I added using AWB had a space before them, e.g. [8]. And nearly all the {{Bare URL inline}} tags in use were added by me.
- It seem to me that with Citation bot and reFill2 now both handling the tags, that the reasons for the bot Job I proposed here have all been resolved. --BrownHairedGirl (talk) • (contribs) 19:02, 18 August 2021 (UTC)[reply]
Withdrawn by operator. Good to hear it's working. I'll mark this as withdrawn then, but feel free to open a new task (or update here) if there remain outstanding issues for which a bot may be helpful. ProcrastinatingReader (talk) 20:19, 18 August 2021 (UTC)[reply]
- white space matters https://en.wikipedia.org/w/index.php?title=User%3AAManWithNoPlan%2Fsandbox2&type=revision&diff=1039441007&oldid=1039440919 AManWithNoPlan (talk) 18:39, 18 August 2021 (UTC)[reply]
- @ProcrastinatingReader: The complaint was indeed that
- The above discussion is preserved as an archive of the debate. Please do not modify it. To request review of this BRFA, please start a new section at Wikipedia:Bots/Noticeboard.