User talk:Lowercase sigmabot III/Archive HowTo/Archive 1

Archive 1

How to test?

Hi Σ (or anybody), is there an easy way to test whether my new setup will work as expected (or work at all)? Or if not, how frequently does the bot run -- i.e., how long should I wait before concluding that it's not working? (For what it's worth, the page I just set up is WP:POST/S -Pete (talk) 01:17, 22 August 2016 (UTC)

Hm, testing, now there's an idea.

The bot runs every 24 hours at 11:59 PM , UTC. I suspect you'll have to wait til tomorrow.→Σσς. (Sigma) 03:59, 23 August 2016 (UTC)

Pre-existing archives

So, I've already got four existing archives, all done manually. How would I get the bot to archive to Archive 5 instead of Archive 1? — Gestrid (talk) 06:05, 5 November 2016 (UTC)

@Gestrid: When adding the {{User:MiszaBot/config}} template to the talk page in question, make sure that you include the |counter=5 parameter. --Redrose64 (talk) 13:40, 5 November 2016 (UTC)

Tracking invalid archive parameters

User:Lowercase sigmabot III/Archive HowTo#Parameters explained says about the archive parameter: "unless a key is supplied (see below), the target page must be a subpage of the current page". Otherwise no archiving is made but there is no indication of this apart from nothing happening. I suggest adding a tracking category like Category:Pages where archive parameter is not a subpage. The source of User:MiszaBot/config says:

{{#if:{{{key|}}}|[[Category:Pages archived using a key]]|<span></span>}}

This could be inserted in the else part in the span:

{{#ifeq:{{#invoke:String|find|source={{{archive|}}}|target={{FULLPAGENAME}}/}}|1||[[Category:Pages where archive parameter is not a subpage]]}}

@Σ: Can this be done without interfering with the operation of the bot, and does it accurately reflect when the bot omits archiving due to an invalid archive parameter? It tests whether the page name followed by a slash is a prefix of {{{archive}}}. PrimeHunter (talk) 12:55, 7 March 2017 (UTC)

not working

Is there anything wrong with these parameters-

| archive = User talk:Yashovardhan Dhanania/Archive %(counter)d

| algo = Old(7d)
| counter = 1

| maxarchivesize = 100K

| minthreadsleft = 4

| minthreadstoarchive = 1

| archiveheader = {{tan}}

I just can't get it to work on my talk page even days after setting it. There are more than 10 discussions on my talk page and many more than 7 days old. Tried changing the settings multiple times but doesn't work. Yashovardhan (talk) 05:55, 7 April 2017 (UTC)

@Yashovardhan Dhanania: I've changed "Old" to "old", lowercase. I don't know if this will help, though, as I haven't seen the bot's source code. -- John of Reading (talk) 06:36, 7 April 2017 (UTC)

@John of Reading: thanks a lot! I'll have to wait a day to see if it works! Yashovardhan (talk) 06:41, 7 April 2017 (UTC)

Thanks a lot John of Reading! It's working now! Yashovardhan (talk) 09:45, 9 April 2017 (UTC)

Not set up correctly?

Someone set up the bot on Talk:List_of_signature_songs a while back. Everything seems to be OK, except there isn't a box for the archives on the talk page. I've tried to figure it out, but I'm clearly out of my league.

Can anyone help? - SummerPhD^v2.0 03:23, 12 May 2017 (UTC)

@SummerPhDv2.0: The archiving was set up correctly, but then the page was moved to a new name. I've moved the archive to match, and updated the archiving instructions. Should be good now. -- John of Reading (talk) 05:52, 12 May 2017 (UTC)

Thanks! (Apparently it helps to know what you are doing.) - SummerPhD^v2.0 13:59, 12 May 2017 (UTC)

Should we implement my suggestion at #Tracking invalid archive parameters and hope it doesn't break archiving? User:Σ hasn't edited since September but the bot is operational. PrimeHunter (talk) 18:17, 12 May 2017 (UTC)

Help!

I had MiszaBot set up on my original talk page, User_talk:SummerPhD. Eventually, I unrecoverably lost my password and copied everything to User_talk:SummerPhDv2.0. Unfortunately, the bot doesn't like me any more. Clearly I need to do something to get my archiving running again, but I haven't a clue what.

Any help would be appreciated. - SummerPhD^v2.0 01:20, 21 July 2017 (UTC)

@SummerPhDv2.0: I've made an edit, and will keep an eye on it. -- John of Reading (talk) 06:08, 21 July 2017 (UTC)

Thanks. - SummerPhD^v2.0 12:33, 21 July 2017 (UTC)

Bad archive link in Talk:Rock–paper–scissors

Hi. can someone help me understand/debug why the archive link in Talk:Rock–paper–scissors doesn't work? It should go to Talk:Rock-paper-scissors/Archive_1 but somehow the hyphens seem to turn into Unicode U+2013 endashes although I don't see why that should happen. Thanks. jhawkinson (talk) 01:27, 9 October 2017 (UTC)

@Jhawkinson: The article had been moved from Rock-paper-scissors (hyphen) to Rock–paper–scissors (ndash), but the archive page had not. I have moved the archive to Talk:Rock–paper–scissors/Archive 1 so that it shows up as a blue link on the talk page. I've also updated the archiving instructions at the top of the talk page. -- John of Reading (talk) 06:59, 9 October 2017 (UTC)

Oh, gosh, I hadn't noticed the parent article had endashes. That sounds like a...terrible idea? Anyhow, thanks for fixing! jhawkinson (talk) 14:42, 9 October 2017 (UTC)

assistance with Talk:Criticism of the Bahá'í Faith

The archives happen and can be searched but are not listed… help please… Smkolins (talk) 15:52, 6 March 2018 (UTC)

@Smkolins: Talk:Criticism of the Bahá'í Faith uses {{Archives}}, which is fine with archives having progressive numbering, but it doesn't handle monthly archives at all well. Try using {{MonthlyArchive}} instead. --Redrose64 🌹 (talk) 16:47, 6 March 2018 (UTC)

Suggestion for instuctions: Use {{FULLPAGENAME}} ?

Just a thought--in the archive instructions, you might want to use the {{PAGENAME}} or {{FULLPAGENAME}} templates magic words? That way you could give code that people can just copy and paste, since the templates will evaluate properly on the destination talk page. So, f'r'instance, the "incremental archive" example could be:

{{User:MiszaBot/config
| algo                = old(30d)
| archive             = {{subst:FULLPAGENAME}}/Archive %(counter)d
| counter             = 1
| maxarchivesize      = 150K
| archiveheader       = {{Automatic archive navigator}}
| minthreadstoarchive = 1
| minthreadsleft      = 4
}}

I've been using that, since it spares me the risk of mistyping the page name and breaking the archving. Just FWIW! — Narsil (talk) 01:46, 31 March 2018 (UTC)

recent inactivity

My talk page has not been archived since January 29. I wonder whether that might be because I removed a section by hand on February 5. —Tamfang (talk) 17:18, 4 May 2018 (UTC)

@Tamfang: No, that's not the reason. Your archiving template does not set the minthreadsleft or minthreadstoarchive parameters, so the bot is using the default values 5 and 2. This means that nothing can be archived until there are at least 7 threads on the page. Have a look at the "Parameters explained" section. -- John of Reading (talk) 18:06, 4 May 2018 (UTC)

Thanks. —Tamfang (talk) 19:44, 4 May 2018 (UTC)

Hmmm, interesting. I've got a similar problem; my talk page, after years of being archived perfectly, stopped being archived a couple of months ago. I also don't have minthreadsleft or minthreadstoarchive, and my configuration:

{{User:MiszaBot/config
| algo    = old(31d)
| archive = User talk:RoySmith/Archive %(counter)d
| counter = 14
| maxarchivesize = 70K
}}
{{archives|auto=yes|search=yes}}

hasn't changed in years. Apparently something has changed in the system, because it used to work even without those parameters. -- RoySmith (talk) 13:51, 27 May 2018 (UTC)

@RoySmith: I can't see a problem there. The bot set the counter to 15 on 5 April, and the history of User talk:RoySmith/Archive 15 shows threads being archived fairly regularly. The first few threads on your talk page have no timestamps, so I don't think the bot will touch those; the next few are dated 26 April and should be archived fairly soon. -- John of Reading (talk) 14:21, 27 May 2018 (UTC)

Ah, the lack of timestamps on the first few threads is probably what's going on. I hadn't noticed that; all I noticed was that I'd been seeing those for a long time. I'll manually clean those up and watch for that in the future. Thanks for your help. -- RoySmith (talk) 14:27, 27 May 2018 (UTC)

How customizeable is this?

I currently manually archive my talk page, but I'm wondering if it could be set so the bot archives virtually exactly the way I do it. On the first of each month, Pacific time, I archive the previous month to whatever that year is. Right now, that would be 2018, so I'm archiving things to: User talk:Amaury/2018. I'm sometimes late and have been as late as mid to late month before archiving—for example, I did not archive February until March 23 or May until June 14 and such—which is where the bot could come in handy. You can see how I archive from that link. Could the bot do that, archiving how I do it? Amaury (talk | contribs) 18:23, 19 July 2018 (UTC)

Bot may be broken

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.

Solved

---

Hello, my bot is has not ran in a while and there are month long pages not being archived

{{Archives|auto=yes|search=yes}}
{{User:MiszaBot/config
| algo                = old(30d)
| archive             = User talk:AmericanAir88/Archive %(counter)d
| counter             = 2
| maxarchivesize      = 75K
| archiveheader       = {{Automatic archive navigator}}
| minthreadstoarchive = 2
| minthreadsleft      = 5
}}

@John of Reading: Any problems with this? AmericanAir88 (talk) 17:42, 20 August 2018 (UTC)

Noting resolved on VPT Galobtter (pingó mió) 21:17, 20 August 2018 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Handling threads without timestamps

In #recent inactivity above, I had a thread that wasn't getting archived because it had no timestamp. I've got another one now. Is there some way to force the bot to archive these? -- RoySmith (talk) 14:42, 22 August 2018 (UTC)

@RoySmith: Do you mean this one? What I would do is:

1. add the missing signature and timestamp, you can use {{subst:unsigned2|00:01, 10 August 2018 (UTC)|Roy "Guapo" Lopez}} for that
2. add a proper heading - Roy "Guapo" Lopez (talk · contribs) has clearly added their post at random within the page rather than using the "New section" tab
3. (optionally) move the thread to its proper chronological position between Wikipedia:Articles for deletion/The Thrill of It All Tour closure and Request to reconsider deletion of Alternative names of Crayola crayons
4. manually remove what remains of the References section - its heading and the {{reflist}} template

Talk pages need the occasional tidy up, bots can't handle unusual situations. For instance, that References section was already redundant prior to the edit of Roy "Guapo" Lopez. Instead of being created as a separate section, the thing to do would have been to identify the section where references were being used (it is now archived), and add {{reflist-talk}} to that section. That way, the displayed references are visually part of the same section instead of being detached, which is not a good idea on a page where sections are not associated with one another and may be removed individually without regard for any side-effects. --Redrose64 🌹 (talk) 22:41, 22 August 2018 (UTC)

Question

When does the timer for the bot to archive a thread start? Is it from when it was originally posted or from the last comment on that thread? If it is the former, is there any way to switch it to latter? Puzzledvegetable (talk) 20:10, 22 January 2019 (UTC)

(talk page stalker) @Puzzledvegetable: It's the latter; "counting from newest timestamp" -- John of Reading (talk) 21:12, 22 January 2019 (UTC)

Yearly archive

Hi,

How do you adjust the monthly archive so that it archives discussions only by year, (I want to know for use on User:SSSB/CSD log and User:SSSB/PROD log) Thanks, SSSB (talk) 11:35, 29 April 2019 (UTC)

Just omit the month part. That is, instead of

| archive       = User talk:Example/Archives/%(year)d/%(monthname)s

use

| archive       = User talk:Example/Archives/%(year)d

--Redrose64 🌹 (talk) 19:15, 29 April 2019 (UTC)

"algo" parameter description. Minimum age before archiving.

Previously, the description said:

Algorithm that instructs the bot how to decide whether to archive a thread or not. Complex formulas (including regex matching) will be available in the future, but so far it must be of the form old(...) where ... specifies the maximum age of a thread (in hours or days: 24h, 72h, 5d, 21d, etc.).

I changed the description. If I am correctly understanding what this parameter does, then I think this makes more sense:

Minimum amount of time that must pass after the last reply before the bot can archive the thread. It must be in the form: old(...) where ... specifies hours or days: 24h, 72h, 5d, 21d, etc.. For example; algo=old(30d). Complex formulas (including regex matching) may be available in the future.

May I suggest that developers in the future use clearer multi-word parameter names, with dashes substituted for spaces. Such as minimum-age instead of algo. Thanks for your efforts. -- Timeshifter (talk) 16:00, 12 October 2019 (UTC)

Everything about this is wrong

Parameter	Default	Description
key		A top-secret key that (if valid) allows archives to not be subpages of the page being archived. To obtain such a key, request it from this editor, and you better have a very good justification for doing so.

Why in the world would anyone need a key?

Also, another question. How does {{User:MiszaBot/config|algo=old(30d)|archive=User talk:Example/Archive %(counter)d|counter=1|maxarchivesize=150K| archiveheader={{Automatic archive navigator}}|minthreadstoarchive=1|minthreadsleft=4}} cause an archive if the only parameter is the key? {{#if:{{{key|}}}|[[Category:Pages archived using a key]]|<span></span>}}<noinclude>{{Documentation|User:MiszaBot/Archive HowTo}}</noinclude> {{replyto}} Can I Log In's (talk) page 13:29, April 23, 2020 (UTC)

A key is only necessary, as the documentation states, when you want an archive to be created somewhere other than a subpage of the page that is being archived. There are very few of these, even fewer when you discount those specifying an unauthorised key. As an example of a page that correctly uses a |key=, see Wikipedia:Administrators' noticeboard/Incidents, which has

{{User:MiszaBot/config
|archiveheader = {{Administrators' noticeboard navbox all}}
|maxarchivesize = 800K
|counter = 1034
|algo = old(72h)
|key = 740a8315fa94aa42eb96fbc48a163504d444ec0297a671adeb246c17b137931c
|archive = Wikipedia:Administrators' noticeboard/IncidentArchive%(counter)d
|headerlevel=2
}}

and therefore archives to subpages of Wikipedia:Administrators' noticeboard.

Your second question doesn't make sense. You provided seven parameters, yet you go on to ask "if the only parameter is the key?" It has no |key= parameter. Besides, a |key= parameter alone is insufficient: there are two parameters that are mandatory, these are |algo= and |archive=; |counter= is also mandatory if |archive= includes %(counter)d. --Redrose64 🌹 (talk) 21:37, 23 April 2020 (UTC)

What justification would there be for archiving on another page? I don't see much.

2nd question: At User:MiszaBot/config, the source code only contains one paramter, key. That's the only parameter, so all another parameters wouldn't make a technical difference, yet all these other parameters somehow make a difference. {{replyto}} Can I Log In's (talk) page 23:58, 23 April 2020 (UTC)

It's not a template, although it looks like one - the MediaWiki template parser does not get involved. The bot treats it as a list of instructions; it identifies the start of the list by the line {{User:MiszaBot/config and the end by the line }}. Everything in between must be laid out in the expected manner. The page source could have been entirely empty, but it needs to test whether a value has been set for |key= so that it can put the page into Category:Pages archived using a key. A bot cannot categorise: only MediaWiki markup can do that. --Redrose64 🌹 (talk) 06:55, 24 April 2020 (UTC)

Moving on to next archive

I have the following active on my talk page:

{{User:MiszaBot/config
| algo                = old(30d)
| archive             = User talk:Reyk/Archive %(counter)d
| counter             = 1
| maxarchivesize      = 150K
| archiveheader       = {{Automatic archive navigator}}
| minthreadstoarchive = 1
| minthreadsleft      = 4
}}

I was under the impression that, once my Archive 1 hit 150K, that the bot would create Archive 2, update the counter, and begin archiving there. However, even though my first archive already exceeds 150K significantly, the bot won't update to the next one. How do I get this to work correctly? Reyk _YO! 07:34, 13 August 2020 (UTC)

@Reyk: The history of User talk:Reyk/Archive 1 shows that your first archive is only 83K so far. -- John of Reading (talk) 08:58, 13 August 2020 (UTC)

My pagesize tool says it's at 168K, so I'm not sure what's going on there. It's already far longer than I want so I'll drop it to 75K and see what happens. Reyk _YO! 09:20, 13 August 2020 (UTC)

@Reyk: What pagesize tool is that? If you look at the page history, you'll see that the most recent edit (that of 02:43, 13 August 2020 by Lowercase sigmabot III) shows "(83,487 bytes) (+1,297‎)" which means that the edit added 1,297‎ bytes giving a total page size of 83,487 bytes. --Redrose64 🌹 (talk) 22:28, 13 August 2020 (UTC)

clarify that algo parameter really is limited to hours and days

Unlike similar-looking parameters it does not understand "w" for weeks or "y" for years for instance.

This is especially pertinent since templates that explain the bot's action accept a wider range of time units. For example, {{Archives}} can be set up with |age= and |units= to produce messages such as

and

However, the bot itself wouldn't understand this. You can't set |algo=old(4w), you would have to use |algo=old(28d). Likewise, instead of the intuitive |algo=old(1y) you'd have to use |algo=old(365d).

I've clarified the documentation. Regards, CapnZapp (talk) 11:38, 21 October 2020 (UTC)

Unspecified max archive size—define "freeze"

Clarification: Below is in the context of archiving a User talk page on Wikimedia Commons, though some links below direct to Wikipedia.

As of this comment, Example 1 states: "The archive will freeze when it becomes 2000KB large." It goes on to suggest using Example 2 with the |maxarchivesize= parameter set with a value of 150K. This isn't very clear.

What exactly does "freeze" mean in the context above? Does it mean that while the bot had been filling User talk:(user)/Archive %(counter)d with %(counter)d=1, it will automatically create and begin archiving to User talk:(user)/Archive %(counter)d with %(counter)d=2 when the next thread to be archived would exceed 2000KB? Was there a special reason 150KB was given as an example? For comparison, the Wikipedia article Black hole, is 150,361B as of this edit.

I've read of a bug on Wikipedia's equivalent bot in which (example): if the size limit is 100K, the current archive is 90K, and the next group of threads totals 250K, the bot would add the thread, exceeding the 100K limit by 240K. Does this bug exist for the 2000KB limit, thus this "freeze" being potentially encountered?

If the statement at the top is accurate, then "freeze" needs to be defined for users. — Christopher, Sheridan, OR (talk) 19:06, 8 November 2020 (UTC)

Single apostrophe

There is a page Talk:St. Bartholomew's Day massacre with the bot set up:

{{User:MiszaBot/config
| algo=old(90d)
| archive=Talk:St. Bartholomew's Day massacre/Archive %(counter)d
| counter=1
| maxarchivesize=75K
| archiveheader={{Automatic archive navigator}}
| minthreadsleft=5
| minthreadstoarchive=1
}}

which is not archiving. I suspect it is a problem with the single apostrophe. Please take a look and tell me if that is the problem and how to work around it. BTW to display the sequence ' I had to munge it otherwise it appears on this page as a single apostrophe. -- PBS (talk) 12:04, 27 June 2019 (UTC)

@PBS: I think it should work with an ordinary ' character. See for example this bot edit to a page with an apostrophe in its title. -- John of Reading (talk) 14:07, 27 June 2019 (UTC)

Thanks I'll try that and let you know. -- PBS (talk) 16:32, 27 June 2019 (UTC)

The bot now archives the page. So we know that it handles a single quote but not ' persumably it will be the same for other characters. This may need noting somewhere as the template parameter {{SUBST:FULLPAGENAME}} generated ' as a substitute for '. -- PBS (talk) 12:46, 2 July 2019 (UTC)

 Done see Help:Archiving a talk page#FULLPAGENAME -- PBS (talk) 09:01, 3 July 2019 (UTC)

I have added a link to this useful piece of info to the spot where it arguably is needed the most: User:Lowercase sigmabot_III/Archive HowTo#Quick examples. CapnZapp (talk) 12:16, 30 December 2020 (UTC)

how about archiving this page?

Does anyone know if perhaps there is an automated tool or program that could repeatedly check the talk page for old content and make it easier than having to archive each section by hand?

-)

PS. That was an attempt at humor PPS. Setting it up in a couple of days unless anyone objects. CapnZapp (talk) 10:25, 2 January 2021 (UTC)

 Done CapnZapp (talk) 10:49, 4 January 2021 (UTC)

Category:Pages where archive parameter is not a subpage

I tried understanding why a page gets on this category? First question, does Lowercase maintain the category, or is it the domain of another bot (or is it disused)?

I tried looking at the first Article Talk space entries. One was clearly malformed (misspelled) and I get why that one was on the page. But the next one seems to autoarchive just fine, and there I decided I needed to first ask if the category is relevant before wracking my brain further.

Any insight welcome. Link Category:Pages where archive parameter is not a subpage CapnZapp (talk) 10:28, 2 January 2021 (UTC)

This category was added following the discussion at Wikipedia:Village_pump_(technical)#Page_not_archiving_and_I_cannot_figure_out_whhhhhyyyy. The code was tweaked yesterday, so some pages may be in the category only because the software hasn't got round to re-thinking them. A null edit of the talk page and a purge of the category page may make them disappear. -- John of Reading (talk) 11:00, 2 January 2021 (UTC)

@CapnZapp: Archiving is requested by transcluding User:MiszaBot/config. The transclusion automatically adds the category. No bots are involved. I made the code to add the category but I did not make the bot code and don't know its details. Please report any false positives if the category is still shown on a page after a null edit or purge of the page. A null edit or purge of the category page is irrelevant. 100 minutes ago I made a tweak to avoid false positives on page names with certain characters including ' " & which are encoded by {{FULLPAGENAME}}. It can take a while before they are automatically removed from the category. Other pages are still being added automatically after the original code was made yesterday. PrimeHunter (talk) 11:22, 2 January 2021 (UTC)

Are you saying we should temporarily not ask the question "why is page X in this category"? If so, fair enuff. When would you say it's again worthwhile to study individual entries (the category holds ~2700 entries as I speak). Regards, CapnZapp (talk) 14:50, 2 January 2021 (UTC)

And re: special characters (such as the apostrophe) - did you see my recent edits to this how to page? If I understand you correctly you're making your code correctly understand '. But the archive bot doesn't understand that! Meaning wouldn't it be best to keep such a page in the category, if its reason for existence is to help users get the signal "the archive parameter isn't a subpage" which makes them go check the Help:Archiving_a_talk_page#FULLPAGENAME instructions. (If I'm even talking about the right thing here...) CapnZapp (talk) 14:55, 2 January 2021 (UTC)

@CapnZapp: Temporarily, don't ask "why is page X shown on this category page", but if you have purged or null-edited the page and still see the category on the page itself then please ask "why does page X display this category". I don't know how long it will take to reflect the latest edits but for a few days I suggest to not spend time examining pages with any of the four characters ' " & = in the name. They are likely to be removed automatically when the servers get to it. You can examine all other pages. The purpose of [1] was to avoid adding pages with one of the special characters and a correct archive= to the category. For example, Talk:Adam's Bridge has the correct archive = Talk:Adam's Bridge/Archive %(counter)d. Before the edit, it was compared to {{FULLPAGENAME}} on that page where it returns Talk:Adam's Bridge, so Talk:Adam's Bridge was incorrectly added to the category. After the edit, that is avoided. I had not noticed the issue you documented in [2]. Such cases were undetected both before and after the mentioned edit [3] and were not added to the category. As a solution to that I have made a new edit [4]. It compares the actual unencoded FULLPAGENAME to archive, instead of comparing the encoded FULLPAGENAME to the encoded archive. I have also made {{FULLPAGENAME decoded}} which uses mw:Help:Extension:ParserFunctions##titleparts as a wrapper to decode the encoding automatically made by {{FULLPAGENAME}}. We could instruct users to use {{subst:FULLPAGENAME decoded}} instead of {{subst:FULLPAGENAME}}. Then the encoded archive parameters will not be created to begin with. PrimeHunter (talk) 17:16, 2 January 2021 (UTC)

Temporarily, don't ask "why is page X shown on this category page" & I don't know how long it will take to reflect the latest edits but for a few days I suggest to not spend time: okay, thanks. CapnZapp (talk) 18:34, 2 January 2021 (UTC)

After the edit, that is avoided. Got it! CapnZapp (talk) 18:34, 2 January 2021 (UTC)

I had not noticed the issue you documented Happens every time you add archival instructions (per the copypasta) to an article with a special character in the name, since the resulting addition to the talk page would be Adam's Bridge in your example - the bot can't understand this is the same as Adam's Bridge, and won't archive. This is kind of the opposite to your case: you're making sure correctly archiving pages aren't added to the category, I wanted to make sure you weren't removing nonfunctional ones :) CapnZapp (talk) 18:41, 2 January 2021 (UTC)

Anyhoo... I'll check back in a couple of days to see if everything has magically sorted itself out, or if pages remain in the category where I can't understand why they're included (my original query). Regards, CapnZapp (talk) 18:42, 2 January 2021 (UTC)

@CapnZapp: Here is a search showing the archive parameter for pages in the category. You are welcome to start spending time examining and fixing pages not containing any of ' " & =. I think my latest edit handles these pages correctly, both avoiding them in the category when they don't belong there, and adding them when they do belong, but it may take a while for the category to reflect this. PrimeHunter (talk) 18:51, 2 January 2021 (UTC)

@PrimeHunter: Please note that if a page is miscategorised, a WP:PURGE of that page will not alter the situation, since it will not remake the entries in the link tables: only a WP:NULLEDIT will do this. --Redrose64 🌹 (talk) 15:25, 3 January 2021 (UTC)

Yes, a purge will not alter the category page but it will determine whether the category is shown on the page. That's why I wrote: if you have purged or null-edited the page and still see the category on the page itself then please ask "why does page X display this category". PrimeHunter (talk) 15:52, 3 January 2021 (UTC)

I hope you guys aren't saying nothing will change until editors start making null edits to all the articles currently in the category? (If the null edit only speeds up what will happen eventually anyway, that's fine) CapnZapp (talk) 15:58, 3 January 2021 (UTC)

Regarding Primehunter's recent edit summary You are welcome to start spending time examining and fixing pages not containing any of ' " &. Well, no, I came here to look for the answer to a question (the start of this talk section). So far the response has been "don't worry we just changed the code, let that sort itself out first" which is cool. I am currently holding off long enough that if/when I find category entries I can't explain I won't get the same answer. Cheers, CapnZapp (talk) 16:04, 3 January 2021 (UTC)

Short of reactivating or recreating Joe's Null Bot (talk · contribs), there are two options: (i) null edit each affected page individually; (ii) wait for the pages to reach the front of the job queue and be null edited automatically. The job queue software was altered back in mid-2014, since when it hasn't worked as well as it previously did - processes that formerly took two or three hours now take two or three months, sometimes never actually completing. --Redrose64 🌹 (talk) 19:11, 3 January 2021 (UTC)

@CapnZapp: The null edit only speeds up what will happen eventually anyway. It appears it has happened now. I examined many members with one of ' " & = in the name [5] and they all did really belong in the category. It's possible some pages still haven't been added to the category but I think there are few or no false members currently. {{subst:FULLPAGENAME decoded}} should always produce a valid parameter. A bot or AWB could do it but I think it's best with an editor to check whether archiving is appropriate at all, whether there are old archives from a former name, or whether the current name is the result of a bad move which should be reverted to a name which is still in archive. PrimeHunter (talk) 19:21, 3 January 2021 (UTC)

Apologies for still not fully understanding. I got curious what could make over two thousand pages appear in that category, thinking "are there really that many manual mistakes?" (such as the page having been moved by somebody forgetting to also manually change the bot parameters to match the new page name) But right now the category has grown to over three thousand. It appears my question will still need to wait until you've sorted out the aftershocks of recent changes. (Before bothering to start looking at individual entries, obviously first the possibility that this special character malarkey still influences the size of the category needs to be eliminated.) Cheers CapnZapp (talk) 10:32, 4 January 2021 (UTC)

If the category is still populating, that indicates that the job queue is still processing the template changes. I took a look at the cat, and made these fixes; they may be a representative sample of the kind of problems that the cat is highlighting. --Redrose64 🌹 (talk) 16:27, 4 January 2021 (UTC)

I'm not sure whether the category is still growing but I don't think there are still false entries which don't need any action, so it's not a waste of time to manually examine the category members now. But if it's growing then it may seem like a Sisyphean task where you keep fixing pages only to discover that the category keeps growing anyway. It may be more satisfying to start work when you can see an effect in reducing the member count. If you find pages which don't belong in the category then just report it here, and stop working if you don't want to waste time. PrimeHunter (talk) 21:19, 4 January 2021 (UTC)

We'll see. At least I think I understand that the answer to my original question (why a seemingly perfectly functional page was in the category) was that you guys were still working on it. (Remember for pages that are categories us editors have no simple way of looking at the "page history" to see if the "page" is brand new or newly edited) CapnZapp (talk) 10:58, 5 January 2021 (UTC)

FULLPAGENAME decoded

A tangential issue - can I assume if {{FULLPAGENAME decoded}} had worked in the archive bot instructions, and removed the special characters issue altogether, you would have replaced all the archive = {{SUBST:FULLPAGENAME}}/Archive %(counter)d already right? Regards, CapnZapp (talk) 10:43, 5 January 2021 (UTC)

I'm not sure what you are asking. I think {{FULLPAGENAME decoded}} works but I'm waiting to look for possible issues before adding it to the archive instructions. When fixing pages in the category, I think it will always produce a valid parameter to save archive = {{SUBST:FULLPAGENAME decoded}}/... instead of manually fixing a wrong name in the parameter. PrimeHunter (talk) 11:22, 5 January 2021 (UTC)

You answered my (implied) question CapnZapp (talk) 12:17, 5 January 2021 (UTC)

February 2021

This discussion might interest readers of this page (at least the preceding section):

Category talk:Pages where archive parameter is not a subpage#St. Louis CapnZapp (talk) 19:46, 16 February 2021 (UTC)

Missing archive pages

This is kind of a strange one, and really isn't a question about the bot, but I don't know where else to ask.

The archive box at Talk:No Time to Die wasn't showing up. Apparently the reason is that the first three archive pages were missing. They never existed, probably someone copied and pasted the archiving template from some other page and forgot to set the index to 1. So I renamed '4' to '1' and '5' to '2' and that solved the problem.

The new problem is that this left redirects behind for '4' and '5'. I think these should be removed, but they don't fit any of the criteria for speedy deletion. Did I do the right thing? What should I do now?

• Talk:No Time to Die/Archive 1
• Talk:No Time to Die/Archive 2
• Talk:No Time to Die/Archive 3
• Talk:No Time to Die/Archive 4
• Talk:No Time to Die/Archive 5

GA-RT-22 (talk) 14:29, 26 March 2021 (UTC)

That talk page shows archives both in the {{Talk header}} template and separately using {{Archives}}. What I suspect was the case was that the Talk header didn't show any archives-related bits, while the Archives box did show up (with a single red-lined entry). If so, everything worked as intended. Apart, obviously, from the fact archives weren't numbered from 1, and thus weren't automatically detected. As for your question, there are templates to request deletion and you should be able to explain the situation to get the pages deleted even if no speedy deletion criteria is technically fulfilled. CapnZapp (talk) 15:46, 26 March 2021 (UTC)

I have deleted Talk:No Time to Die/Archive 4 and Talk:No Time to Die/Archive 5 per WP:CSD#G6 after checking there were no incoming links. PrimeHunter (talk) 17:23, 26 March 2021 (UTC)

Thank you! I'd been looking at WP:R3 but that didn't seem exactly right. GA-RT-22 (talk) 17:54, 26 March 2021 (UTC)

Feature request

Sometimes I want to close a discussion because the original poster did not come back, but I don't want to reset the archive delay. Can we make it so that I could add something like <!-- ignoreSignatureForArchive --> on the same line as the signature to prevent that particular signature from being considered by the archive bot? HighInBC ^{Need help? Just ask.} 05:08, 26 June 2021 (UTC)

My spontaneous question: why is it a problem the archive delay is reset? (You could always manually archive old-ish content you know) CapnZapp (talk) 12:31, 26 June 2021 (UTC)

AN and ANI are busy places and get cluttered fast. Manual archiving is certainly an option but can sometimes give the appearance of trying to dismiss a topic prematurely. Anyway just an idea, it is not very crucial, just a nice to have. HighInBC ^{Need help? Just ask.} 03:09, 27 June 2021 (UTC)

How does the Talk-page archiving work?

How do the archive bots determine which sections on a given Talk page are stale (and should be archived)? Do the archive bots read the the dates posted by each poster (via the poster's four-tilde signature) on each scanned Talk page? Acwilson9 (talk) 01:48, 18 February 2022 (UTC)

@Acwilson9: You are posting at a talk page of lowercase sigmabot III (talk · contribs). This bot selects threads to archive purely determined by the most recent valid timestamp in the thread. Was it not clear in this post? --Redrose64 🌹 (talk) 21:54, 18 February 2022 (UTC)

I have posted at AC's talk page since I sense you two are discussing from very different vantage points. Cheers CapnZapp (talk) 09:54, 19 February 2022 (UTC)