This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.
On multiple "[X] Amendment to the United States Constitution" article talk pages I installed MiszaBot I yesterday. No archiving has occurred on any of them. I know MiszaBot I archives once a day, but I'd still like to know if did the installation correctly. Can someone look at one of those article talk pages (here's a link to one of them). Thanks. SMP0328. (talk) 21:50, 1 August 2009 (UTC)[reply]
It looks ok to me. You only actually installed 1 yesterday, the bulk were done today (1 Aug UTC), MiszaBot started its run on articles talk around 00:30 UTC. Perhaps even by 19:28 when you placed the example you've given, Misza had already grabbed its worklist. I would hazard a guess Misza will hit them in the run about to occur in under 4 hours. –xenotalk21:59, 1 August 2009 (UTC)[reply]
I see why it isn't working on the example from yesterday. You've put the wrong target |archive = Talk:Sixth Amendment to the United States Constitution/Archive %(counter)d , the page is the Fourteenth amendment (Misza has smart failure there). Fixed. The others you tagged look ok. –xenotalk22:02, 1 August 2009 (UTC)[reply]
MiszaBot I has failed to archive material in the Eleventh, Fifteenth, Seventeenth and Twenty-sixth Amendments talk pages. What can be done to fix this failure? SMP0328. (talk) 23:19, 4 August 2009 (UTC)[reply]
MiszaBot will leave 5 threads on the talk page unless you set |minthreadsleft= to less than 5. (Setting to 0 will have MiszaBot completely harvest the page if the threads are old enough... Some editors feel that you should leave at least some talk on the page) –xenotalk23:22, 4 August 2009 (UTC)[reply]
Hi. I've been working on an article that's been generating an enormous amount of posts on the talkpage at Talk:IB Diploma Programme. Spent most of June/July manually archiving but would like to give MiszaBot a try. I've added the markup to the top of the page in "nowiki" tags. Would you mind having a look to see whether I've done it correctly. The biggest concern is the page already has seven archives, so I'm not certain about how to set the counter, and whether the counter increments itself automatically as necessary. Thanks. Truthkeeper88 (talk) 14:18, 5 August 2009 (UTC)[reply]
Couple things: You had the code wrapped in "nowiki" tags - while this probably wouldn't affect the bot, it displays the code on the page instead of hiding it from the viewer. You also had it pointing to User talk, so the bot wouldn't act because it's the wrong location. Fixed. The counter is fine, it will continue to fill archive 7 up to 250k and automagically move onto archive 8. cheers, –xenotalk14:54, 5 August 2009 (UTC)[reply]
Thanks for the help. I put the code in the nowiki tags so the other editors could approve the parameters. Thanks for fixing the destination, I knew I'd get something wrong. Truthkeeper88 (talk) 16:44, 5 August 2009 (UTC)[reply]
No problem. MiszaBot might've actually done its work whether or not the nowiki parmaters were there. The bot looks for the specific words and ignores everything around it. –xenotalk16:47, 5 August 2009 (UTC)[reply]
xeno, I know that Truthkeeper88 did a lot of digging around to help get this set up for us, and that all the rest of us who frequently edit the IB Diploma Programme article really appreciate all of his efforts and your help. Thanks a lot! Regards, • CinchBug • 20:49, 5 August 2009 (UTC)[reply]
This one has me stumped... It *may* be the fake October timestamp that occurs at the top of the page before any headers; that's just my best guess. –xenotalk14:13, 7 August 2009 (UTC)[reply]
It was throwing a ValueError: year out of range error, so my guess is a bogus timestamp with a year like 20009. Dunno what the accepted range is, though; I change the code to simply ignore such timestamps. Миша1315:39, 7 August 2009 (UTC)[reply]
it is difficult to prove whether their leadership actually had a hand or not.--[[User:23prootie|23prootie]] ([[User talk:23prootie|talk]]) 19:<!---->02, 9 July (UTC) ? –xenotalk15:45, 7 August 2009 (UTC)[reply]
'Twas the only non-standard timestamp on the page. Oh, thanks for the barnstar =) Figured I can filter out the easy ones for ya. –xenotalk15:52, 7 August 2009 (UTC)[reply]
Yes... At first I preferred to manually archive because things could be organized more neatly, but now I'm going to let MB have at it. The code can't be on a subpaged /top business page tho. –xenotalk17:41, 7 August 2009 (UTC)[reply]
Hi, over at the Simple English Wikipedia, we regularly have simple:Wikipedia:Simple talk overflowing. Thus, would it be possible to introduce a new configuration option that allows users to specify the maximum size for a page (not archive)? It would override the algo and minthreadsleft settings, so would still archive a bloated page with (relatively) new threads. ChenzwTalk12:49, 8 August 2009 (UTC)[reply]
I don't think so. The algo parameter is thread-based, not page-based. To do what you're asking for would require a considerable change in the program's architecture; something I simply have no time for. Миша1305:17, 12 August 2009 (UTC)[reply]
Your talk page doesn't seem to have any sections dated before 10 August, which is within the 2-day period. You'll have to manually date or archive the earlier sections.
Likewise, 'Paid editing' has only one section older than the 30-day period Miszabot is set for; by default it waits for two sections.
That seems unreasonable. Hey, maybe the bot could report that it had tried to act and why it failed, either on the page or on some centralized page? {{Coord}} templates with various errors show up in Category:Coord template needing repair.
That's quite reasonable - the bot is not an admin and cannot force its will upon the MediaWiki software. As for logging archive errors for everyone to see, I thought about it but I don't have time to implement it. It's not even as easy as it sounds. Миша1305:16, 12 August 2009 (UTC)[reply]
Please see this diff : I set the bot config to archive threads older than three days, but it archived the thread "Strike out votes" which had a comment from yesterday (directly above the next archived section, "== The split vote problem =="). Is there something you can see in the diff that caused this? Every other thread archived was in fact older than three days. Sswonk (talk) 15:34, 13 August 2009 (UTC)[reply]
Talk:Human/Archive 29, created by MiszaBot several days ago, contains material related to an RFC that is still open. The only material in Archive 29 is a single section. I want to restore the content to Talk:Human so that the discussion, which had moved on to a new section, will be all in one place, but I'm worried about pitfalls. Should I:
copy and paste, leaving duplicate content on Archive 29 or
copy and paste and remove the content on Archive 29, in effect making a blank archive page, or
copy and paste, then request speedy deletion of Archive 29?
Concerns [have been raised] about screwing up future archiving, which apparently happened before. Thanks for any advice. Also, for future reference, is there any way (without entirely stopping archiving a page) to make MiszaBot ignore specific sections? Null edits? Secret code :) ?Rivertorch (talk) 07:03, 16 August 2009 (UTC)[reply]
Choose option 2 and be advised that the bot searches for the newest timestamp in a section, so just add one that will not become old until the RfC is closed. Миша1308:28, 16 August 2009 (UTC)[reply]
It would amuse me terribly if we started the practice of periodically adding "bump!" to the end of sections we want to keep around. :D -Silence (talk) 16:39, 16 August 2009 (UTC)[reply]
First, the maxarchivesize parameter, properly used it should read 128K, 256K, 512K, etc... With the capitol "K", correct? I've tried to use that param a couple of times now and have run into problems where MiszaBot would create a new page for every individual section archived (suggesting that it thought that maxarchivesize was immediately being passed).
Also, is there a list available of what articles each MiszaBot is archiving, somewhere? I'm guessing that it just uses a hidden category based on the template or something, but being able to locate that might be helpful occasionally. The actual reason that I'm asking though is that I was just trying to figure out if MiszaBot I skipped an article which was recently added to be archived. The page might be moved soon (probably will), so I tried using the {{TALKPAGE}} magic word for the archive param. I think that it skipped over the page because of this instead of archiving it though, since the text name differed from the page name and there's no "key" given for the page (and none is actually needed), but somehow getting confirmation on that would be nice. (PS.:Feel free to reply here, as I have your page on my watchlist for now. I prefer to keep conversations in one place.) — V = I * R (talk) 07:38, 19 August 2009 (UTC)[reply]
Yes, it should be capital "K", with no space after the number. Can you point out a page where it created many small archives with that configuration?
You can browse the bots' contributions as well as Special:Whatlinkshere/User:MiszaBot/config. The bots are responsible for different namespaces. "I" does article talk, "III" does user talk and "II" does everything else. If you give me a specific page I can inspect execution logs to figure out why exactly did it fail to work there. Миша1320:32, 27 August 2009 (UTC)[reply]
Sorry, I missed this reply for a couple of days. Let's see... Here is the history page for Talk:Ares I. Take a look at August 14. It's been cleaned up, but it's there at least. I can go ahead and edit the config documentation to clarify this, if you don't mind. — V = I * R (talk to Ω) 00:29, 30 August 2009 (UTC)[reply]
Page structure changes causing problems
Would you be able to advise on how best to sort out what happened at Template:Arbitration talk archive navbox? When the arbitration page structure was changed in May, I don't think the archive bot instructions were considered or changed (but I might be wrong). The move took place here. Before the move, as best I can make out, the bot was happily archiving to pages 22 and 23 of the "Wikipedia talk:Requests for arbitration" archive subpages (apart from a few squeals about full archives). It then switched to archiving to archive subpages 21 and 23 of the new "Wikipedia talk:Arbitration/Requests" page structure. My first question (though not that important) is why it seems to have skipped page 22? The other point is that Kirill move the arbitration enforcement archives to archive subpages pages 25 and 26 of the new page structure. Will this mess the bot up when it gets to those pages, or will it skip them? I'll drop Kirill a note as well and see if he knows what to do, as this may have happened on other archive pages as well. Carcharoth (talk) 15:13, 23 August 2009 (UTC)[reply]
See also earlier discussion here. From the looks of that, the clerks know what is going on so they (and Kirill) can probably sort it out, though the bit about the AE archives is new. Carcharoth (talk) 15:20, 23 August 2009 (UTC)[reply]
Archiving new comments
Could you tell me why MiszaBot archived my new comments, here? I've seen the bot do this several times before, and I would like to know why. Thanks. Viriditas (talk) 22:38, 29 August 2009 (UTC)[reply]
This is another of those two timestamps per line things. MiszaBot read the first timestamp that occurred in your prose. –xenotalk22:47, 29 August 2009 (UTC)[reply]
Ah, thanks for explaining. Since I tend to start paragraphs with timestamps, I'll bet that's exactly what happened last time! :) Viriditas (talk) 22:49, 29 August 2009 (UTC)[reply]
If you take off the "UTC" or otherwise disrupt the time<!-- like so -->stamp somehow, that'll be a workaround until Misza can look at this bug. Putting your sig on a newline would probably help too. –xenotalk14:48, 30 August 2009 (UTC)[reply]
