Jump to content

User talk:Cyberpower678/Archive 37

Page contents not supported in other languages.
This user uses Twinkle to fight vandalism.
This user has 70% energy left.
This user has administrator privileges on the English Wikipedia.
This user is a global renamer.
This user is an edit filter manager on the English Wikipedia.
This user has interface administrator privileges on the English Wikipedia.
Trout this user
From Wikipedia, the free encyclopedia
Archive 30Archive 35Archive 36Archive 37Archive 38Archive 39Archive 40
Senior Editor II
Senior Editor II

some hints

Webcitation

Please see: Template_talk:WebCite#Date_from_webcite-ID so this code: {{webarchive |url=https://web.archive.org/web/*/http://www.geocities.com/los_angeles_coast/public_transportation.html |date=* |title=Los Angeles Public Transportation }}

produces expand in dewiki to:

Los Angeles Public Transportation (Memento vom 17. Oktober 2009 auf WebCite)

The query-part of the url is for special:linksearch and has no effect for the URL

Archive.is

in dewiki we strictly avoid shortlinks to archiv.is, only long-links are allowed in ns=0 => https://tools.wmflabs.org/giftbot/weblinksuche.fcgi?limit=100&offset=10800&target=https%3A%2F%2Farchive.is%2F%25&namespace=0&associated=0

there are not links to http: we only use https for archive.is: https://tools.wmflabs.org/giftbot/weblinksuche.fcgi?target=http%3A%2F%2Farchive.is%2F%25&namespace=0

If so archive.is dies in fare future, we have the original url. Expand of the short-URLs to long URLs can be done by a bot.

(btw: https://tools.wmflabs.org/giftbot/weblinksuche.fcgi is very cool. It allows Wildcard search and filter of Namespaces ) Example: https://tools.wmflabs.org/giftbot/weblinksuche.fcgi?target=http%3A%2F%2Fwww.webcitation.org%2F%25http%3A%2F%2Fwww.geocities.com%2Flos_angeles_coast%2Fpublic_transportation.html&namespace=0

hth: Boshomi (talk) 17:35, 27 July 2016 (UTC)

See also this RC still open but will likely close to use long form over short on en. -- GreenC 17:42, 27 July 2016 (UTC)
Not sure what's the point in using two different set of rules for WebCite and archive.is, both could have an ID specified in the template. Anyway, if the cite template is properly used, link rot is not a concern with short links as the original URL should be present in the template already. nyuszika7h (talk) 18:04, 27 July 2016 (UTC)
The RfC addresses concerns with the short form and spam. There is not always a template. -- GreenC 18:12, 27 July 2016 (UTC)
I know the RfC also addresses concerns with spam, I was talking about the German Wikipedia. nyuszika7h (talk) 18:14, 27 July 2016 (UTC)
  • Thanks for the link to the RFC.
  • This are hints for enwiki, The extraction of the timestamp from WebciteID is very useful, magic for people who did not know that it is form this 9 Chars/16 digits.
  • I am missing a tool like giftbot/weblinksuche for enwiki. This is very efficient. With the template de:Vorlage:Giftbotweblinksuche, I can search a URL in different web archives and in giftbot found urls on the talk-pages. Boshomi (talk) 20:52, 27 July 2016 (UTC)
    IABot has a huge DB of URLs with respective archives and timestamps. So the timestamp extraction isn't really that big of an issue. IABot is also capable of resolving all formats of webcite, wayback, archiveis, and memento URLs.—cyberpowerChat:Online 08:46, 28 July 2016 (UTC)
    @Boshomi: Memento can help you with searching in multiple archives. nyuszika7h (talk) 13:58, 29 July 2016 (UTC)
@Nyuszika7H: Mementoweb is part of the german {{:de:dead link}}-template (diff dewiki dead link) Boshomi (talk) 14:13, 29 July 2016 (UTC)

My vector page

I asked for help in technical village pump to change the background color of Wikipedia pages. According to their instruction, I was able to change the color background while logged in to Wikipedia. But, the space above the page name is always remaining white. It is not changing colors. Can you tell me how to make the entire page's background colored? This is the screenshot of how I see Wikipedia pages. --Rainbow Archer (talk) 07:10, 29 July 2016 (UTC)

I don't think you can.—cyberpowerChat:Online 07:14, 29 July 2016 (UTC)
@Rainbow Archer: You need to hide the background image at the top. Try the following CSS: #mw-page-base, #mw-head-base { background: transparent } MusikAnimal talk 19:09, 29 July 2016 (UTC)

global rename

Hi, I noticed you are active in renaming users so maybe you can help. I have a global rename request that is listed as on hold until July 28. It has actually been started. The old username is gone. But I can't log in with the new one either; I get a message saying "rename in progress". The status says the rename is complete everywhere except en WP. It has been this way for over 24 hours. I don't know if the process is working or if something got stuck. The old name is Mb66w and the new name is MB. — Preceding unsigned comment added by 174.126.250.222 (talk) 13:30, 29 July 2016 (UTC)

Your rename got stuck. Something went wrong on this Wikipedia.—cyberpowerChat:Online 14:14, 29 July 2016 (UTC)
Is there something I need to do to get it "unstuck" or are you taking care of it and/or notifying someone who can? Thanks. 174.126.250.222 (talk) 14:25, 29 July 2016 (UTC)
Only a sysadmin can fix it, and I have notified them. :/—cyberpowerChat:Online 14:27, 29 July 2016 (UTC)
Still stuck after another 7.5 hours! 174.126.250.222 (talk) 21:53, 29 July 2016 (UTC)
I'm guessing your rename is this one. A sysadmin has been notified, but they may be busy at the moment. Just hang in there; we'll get it unstuck eventually once we know what's going on. —k6ka 🍁 (Talk · Contributions) 23:04, 29 July 2016 (UTC)
Yes, that is the stuck rename. I have been unable to log in for two days now. I have been spending my time reading articles and have about a dozen open tabs now with articles I intend to correct/ improve once I can log in again and edit under my username. Also several AFDs on which I want to comment/vote. I hope this gets fixed soon. 174.126.250.222 (talk) 23:31, 29 July 2016 (UTC)

bot specific |dead-url= keyword

Continues from these discussions: |dead-url=unfit, |dead-url=unfit maintenance category, and alternate |dead-url= keyword.

|dead-url=bot: unknown is now supported in the live cs1|2 module:

{{cite book |title=Title |url=//example.com |archive-url=//example.org |archive-date=2016-07-30 |dead-url=bot: unknown}}
Title. Archived from the original on 2016-07-30. {{cite book}}: Unknown parameter |dead-url= ignored (|url-status= suggested) (help)

Trappist the monk (talk) 10:56, 30 July 2016 (UTC)

Internetarchive bot

Hi, is this bot working ok as it says disabled? Also, does it archive all unarchived links and how long is it from a link in a reference being added to it being archived by the bot? ThanksAtlantic306 (talk) 05:34, 31 July 2016 (UTC)

Hi, see diff, in this case there was an external link with a wikilink inside it, the archive bot then added a wayback link in there inside the wikilink. Please could the bot avoid taking ]] as the end of an external link. Rjwilmsi 07:36, 31 July 2016 (UTC)

Edit and edit summary did not make any sense at all

Hi, I reverted your bot's edit ([1]), as it did not made any sense at all. Your edit summary claims you were "rescuing 1 source", however, all the links in that reference were working fine and the bot turned a live link into an archived link which was already given immediately afterwards. Futher, it attempted to recursively apply archive.org to itself twice. Please fix your bot and check other old edits for the same pattern before reenabling your bot. Thanks. --Matthiaspaul (talk) 19:35, 31 July 2016 (UTC)

All of the listed articles are FA-, FL-, A- or GA-Class, yet the bot's book report lists 715 of them as unassessed. Imzadi 1979  05:03, 1 August 2016 (UTC)

No edit counters

Hi.As X-Toos is down, I think it is an inapprpriate moment to tinker with Supercount which has been working perfecty well for some months now. --Kudpung กุดผึ้ง (talk) 09:37, 1 August 2016 (UTC)

I haven't touched it, and it's working fine for me.—cyberpowerChat:Online 09:48, 1 August 2016 (UTC)
@Kudpung: Xtools has been having some problems with the service watcher on labs. It keeps restarting the webservices for xtools, which is probably what you ran into. If you see an error 503 again, just wait a few minutes. We're working on fixing that problem. ~ Matthewrbowker Drop me a note 17:00, 1 August 2016 (UTC)

Template-protected edit request on 4 August 2016


This article is not correct. The Cowanesque Lake is essentially at the end of the Cowanesque Valley and does not provide flood protection for The Valley. It provides downstream flood protection especially for the City of Corning, New York. What flood protection that is provided to the Cowanesque Valley is provided by small damns constructed in the hills after 1972. An excellent example of this is the North Fork Damn in the Village of North Fork near the New York and Pennsylvania state line.

Former Valley Native - SBR


69.128.228.140 (talk) 00:26, 4 August 2016 (UTC)

Not done: this is the talk page for communicating with the user Cyberpower678 (talk · contribs). Please make your request at the talk page for the article concerned. —k6ka 🍁 (Talk · Contributions) 00:31, 4 August 2016 (UTC)

Any idea why closed listings are hanging around for weeks? Is there anything I can do to help clear it out? czar 05:24, 4 August 2016 (UTC)

Committed Identity

Hello Cyberpower,
Do you have access to edit this page - User:Cyberpower678/Flipper/Hash since it is fully protected but lies in your userspace so I thought i'll ask. Thanks and Regards. VarunFEB2003 I am Online 12:34, 5 August 2016 (UTC)

Nope—cyberpowerChat:Online 12:35, 5 August 2016 (UTC)

All of the listed articles are FA-, FL-, A- or GA-Class, yet the bot's book report lists 715 of them as unassessed. Imzadi 1979  05:03, 1 August 2016 (UTC)

Bumping back from the archive. Imzadi 1979  00:08, 6 August 2016 (UTC)

Typo in bot config

I noticed a typo at User:InternetArchiveBot/Dead-links.js: it says {{WaybacK}} instead of {{Wayback}}. – nyuszika7h (talk) 12:56, 9 August 2016 (UTC)

Thanks I fixed it.—cyberpowerChat:Online 13:09, 9 August 2016 (UTC)

Unblock InternetArchiveBot

This user's unblock request has been reviewed by an administrator, who accepted the request.

Cyberpower678 (block logactive blocksglobal blockscontribsdeleted contribsfilter logcreation logchange block settingsunblockcheckuser (log))


Request reason:

Can an administrator please unblock InternetArchiveBot (talk · contribs). It was blocked due to malfunctions, which have been corrected. Since the bot also has a run page, still disabled, it would be really helpful if the bot were unblocked.—cyberpowerChat:Online 09:28, 9 August 2016 (UTC)

Accept reason:

@Boing! said Zebedee: Can you help me?—cyberpowerChat:Online 11:56, 9 August 2016 (UTC)
Would have been happy to, but someone's beaten me to it ;-) Boing! said Zebedee (talk) 14:39, 9 August 2016 (UTC)
Yea, sorry. When I logged into my bot account, I inadvertently blocked myself, so I fetched myself an admin on IRC. :p. Thanks though. :-)—cyberpowerChat:Online 14:42, 9 August 2016 (UTC)

CB - I would recommend running it for 1000 edits or so then disable. Manually check every edit. Assume there are still problems to be found. Repeat until no errors. I'm concerned about backlash from too many errors, blocks are serious, I want to see this bot succeed. I find this script useful for quickly checking diffs. Cheers. -- GreenC 13:12, 9 August 2016 (UTC)

I'm going to do a slow trial once the patches have been reasonably tested. I did some major overhauling of the core parser, which was the root cause of almost every bug reported.—cyberpowerChat:Online 13:13, 9 August 2016 (UTC)
That's a cool gadget, thanks for suggesting it. I put it in my global JS. :D—cyberpowerChat:Online 13:16, 9 August 2016 (UTC)

Rename

I know. I just do not like to put email. You will rename ?--Νικόλας Παπαποστόλου (talk) 15:29, 9 August 2016 (UTC)

?—cyberpowerChat:Online 15:30, 9 August 2016 (UTC)

Me would rename?--Νικόλας Παπαποστόλου (talk) 15:57, 9 August 2016 (UTC)

Would you rename me? maybe. — RainFall 15:58, 9 August 2016 (UTC)
I'm so confused right now. What email?—cyberpowerChat:Online 15:59, 9 August 2016 (UTC)
This may help. — RainFall 16:02, 9 August 2016 (UTC)

All of the listed articles are FA-, FL-, A- or GA-Class, yet the bot's book report lists 715 of them as unassessed. Imzadi 1979  05:03, 1 August 2016 (UTC)

Bumping back from the archive. Imzadi 1979  00:08, 6 August 2016 (UTC)
Any answer on why the bot is doing this? Imzadi 1979  04:37, 9 August 2016 (UTC)
Sorry for not responding sooner. The short answer is no. I have to figure that out.—cyberpowerChat:Online 09:25, 9 August 2016 (UTC)
I'm not sure why, but it looks like the bot got it sorter and updated all of the missing assessments already. Imzadi 1979  02:17, 11 August 2016 (UTC)
Since I'm a little busy with other matters at the moment, if you figured out what was tripping the bot, please let me know.—cyberpowerChat:Limited Access 05:33, 11 August 2016 (UTC)

Cyberbot I clearing X13-20

Hi Cyberpower678. I just noticed some additional template sandboxes were created 30 July. Is Cyberbot I aware of these and can reset them? — Andy W. (talk ·ctb) 20:30, 10 August 2016 (UTC)

Possibly, but most likely not at the moment.—cyberpowerChat:Limited Access 05:51, 11 August 2016 (UTC)

Can you rescue some sources for the assassination of Martin Luther King article? — Preceding unsigned comment added by 78.146.103.11 (talk) 14:34, 11 August 2016 (UTC)

PC tagging

Hi. I thought I'd notify you that your bot is doing quite a lot of repetitive tagging/de-tagging[2]. -- zzuuzz (talk) 20:39, 11 August 2016 (UTC)

Seems to be having a problem with Boyle's law. Keith D (talk) 20:50, 11 August 2016 (UTC)
Seems to be quite a few articles. I'll disable it if it goes another round. -- zzuuzz (talk) 20:51, 11 August 2016 (UTC)
disabled. -- zzuuzz (talk) 20:54, 11 August 2016 (UTC)
zzuuzz, thanks for this. It caught my attention too. Cyphoidbomb (talk) 21:23, 11 August 2016 (UTC)

necessaryily

Hi. Your bot has placed over 5,000 instances of this misspelled word on various talk pages, setting a bad example. Please replace each instance with the correct spelling, "necessarily". Thank you.   — Jeff G. ツ (talk) 20:34, 10 August 2016 (UTC)

See also User_talk:Cyberpower678/Archive_21#Misspelling_in_user_message.   — Jeff G. ツ (talk) 20:43, 10 August 2016 (UTC)

No need. The bot has stopped leaving those talk page messages for quite a while now.—cyberpowerChat:Limited Access 05:49, 11 August 2016 (UTC)
Do you want some help fixing the old instances? I can probably do it easily without any coding using AWB. -- GreenC 14:51, 11 August 2016 (UTC)
If you want. I just don't think it's worth the time and effort as it accomplishes nothing. :p—cyberpowerChat:Online 17:32, 11 August 2016 (UTC)
I would appreciate that cleanup, Green Cardamom. That sort of thing hurts my eyes. (It also helps Wikipedia look a little more professional.)— Gorthian (talk) 18:31, 11 August 2016 (UTC)

@Jeff G.: @Gorthian: - I started and fixed around 2500 pages (2954 left), but received a complaint from User:Derek Andrews and User:TAnthony who does not see the point in fixing spelling mistakes on talk pages. Since this is now controversial, it will need to go through the Wikipedia:Bots/Requests for approval which can take 1-2 months. I have no interest in going through that. Cyberpower678, you will likely continue to receive complaints as other editors notice and send you notices about the mistake. Good luck folks I'm done here :) -- GreenC 15:37, 12 August 2016 (UTC)

Hi, I objected before I realized this was a request to correct an earlier bot error. Your edit summary for these changes just reflected the typo correcting, indicating nothing about this request. I see no issue with you continuing if you can just make the edit summary a bit more explanatory. Thanks.— TAnthonyTalk 15:41, 12 August 2016 (UTC)
Ok well your reply was right below the link to this thread so I thought you read it. You are right that since this was a request (or permission) from the person who made the error, it shouldn't need bot approval as it's fixing a comment with permission. I'll add a link to this thread for the remainder then and hopefully it will finish out. -- GreenC 15:47, 12 August 2016 (UTC)

Done - 5705 edits. 18 would need to be done manually, AWB skipped for some reason. -- GreenC 18:12, 12 August 2016 (UTC)

Wow. Thank you, Green Cardamom. I finished off the last 17. (One wasn't made by the bot. :-P) — Gorthian (talk) 02:51, 13 August 2016 (UTC)

Thank you all!   — Jeff G. ツ (talk) 03:51, 13 August 2016 (UTC)

Bot broke syntax

[3]. -- Magioladitis (talk) 20:26, 9 August 2016 (UTC)

Sorry but that edit is stale. If you find one that was made from v1.1, please let me know.—cyberpowerChat:Limited Access 20:27, 9 August 2016 (UTC)

Here's a bug to v1.1 [4]. Can you please review your edits? There might be much more. -- Magioladitis (talk) 23:21, 9 August 2016 (UTC)

This too. But I guess this was a weird case anyway. [5]. -- Magioladitis (talk) 23:25, 9 August 2016 (UTC)

This too. [6]. -- Magioladitis (talk) 23:26, 9 August 2016 (UTC)

These too: [7] and [8]. Same case in fact. -- Magioladitis (talk) 23:27, 9 August 2016 (UTC)

@Magioladitis: I've been seeing and fixing them too. These happened before the bot was shutoff. I reported two cases of these types of problems. 1 2. Both are now fixed. One before the bot was shutdown and one after. I see cyberpower turned the bot on today for testing the new version. Need to report fun-filled errors from today onwards. Bgwhite (talk) 23:48, 9 August 2016 (UTC)
I just ran CheckWiki on articles that InternetArchiveBot edited today. I didn't check talk page. CheckWiki is how we (Magioladitis and I) spotted the errors we have been reporting. CheckWiki reported no errors. Bgwhite (talk) 23:58, 9 August 2016 (UTC)

Yes another old one [9]. But someone has to check all edits to find more errors I guess. -- Magioladitis (talk) 08:06, 10 August 2016 (UTC)

When the bot was deployed on the 1.1, it was thought that all lingering bugs were fixed. So I ran it full speed. This however turned out to not be the case, and the bot has already made more than 100k edits, unfortunately checking all edits is impossible, and it is regrettable that the bot ended up disrupting some articles. All I can do is to move forward and test the bot more extensively, while hoping that the active editors on the affected pages notice the break and click the rollback link, and properly by running the test code on my machine in a controlled manner to supervise the edits, which I did yesterday. I stopped the test because I found some new and unreported bugs needing fixing. I only wish these bugs surfaced sooner during initial testing, but Wikipedia is enourmous, and I am only one man.—cyberpowerChat:Online 12:05, 10 August 2016 (UTC)
Cyberpower678 yes. Sure. I 'll do everything to help. These reports are mainly to give any idea. We detect most of the broken templates and hopefully we will fix everything. -- Magioladitis (talk) 13:49, 10 August 2016 (UTC)

v1.1 bug. -- Magioladitis (talk) 18:33, 11 August 2016 (UTC)

another v.1.1 bug. -- Magioladitis (talk) 18:34, 11 August 2016 (UTC)

yet another v1.1. I hope all this info is useful to you. -- Magioladitis (talk) 18:36, 11 August 2016 (UTC)

yet another one. -- Magioladitis (talk) 18:38, 11 August 2016 (UTC)

All of these bugs have been fixed in the upcoming release I'm testing. There's no need to report more. :p—cyberpowerChat:Limited Access 18:40, 11 August 2016 (UTC)
Haha. OK. These all I found anyway. -- Magioladitis (talk) 19:08, 11 August 2016 (UTC)
Please have a look at this diff. Neither of the url parameters should link to the Wayback Machine (or other archiving service). This resulted in the same URL used for both the url and archiveurl parameters in {{Cite web}}. You should also note that {{Query web archive}} is intended to provide links to three (not counting the Bibalex Wayback Machine) archiving services. Regards. – Allen4names (contributions) 05:27, 14 August 2016 (UTC)

Cyberbot

Hi Cyberpower. With this edit Cyberbot left the article in a state with a malformed linkj, as you can see here; specifically, a CS1 error has been left, explained here. I realise that the edit took place some time ago, so possibly you have changed the bot to reflect this, but in case you haven't, i thought i'd draw it to your attention, so you can consider changing it and not leaving those red CS1 error messages. Thanks. Happy days, LindsayHello 11:27, 16 August 2016 (UTC)

You are correct. The bot has already been updated.—cyberpowerChat:Online 11:30, 16 August 2016 (UTC)
No worries, then. Thanks for your attention. Happy days, LindsayHello 11:45, 16 August 2016 (UTC)

Question Regarding Misleading Article

Could you look into your edits regarding Jack Conger? I posted my knowledge on the talk page but essentially, given Lochte's role in speaking to the solely speaking to the media and filing a false report in conjunction with James Feigen, I feel Conger's page should reflect his being implicated by Lochte and his cooperation in police testimonies. As it currently reads, it implies Conger (and Bentz) were actively involved in crimes, but this is not the case. Further, only Lochte to the media that he was robbed (a gun was drawn but otherwise Lochte was aggressive and clearly fabricated the story, his mother ran with it, and Lochte exacerbated the problem). Conger WAS with him, but there is no concrete evidence that shows Conger did anything illegal or played a personal role in false reports. As of today he is in Miami and headed to a connecting flight following a successful testimony against Ryan Lochte.

Thank you for any consideration. -(just a random person following this story too close) — Preceding unsigned comment added by 162.245.201.193 (talk) 23:26, 19 August 2016 (UTC)

?—cyberpowerChat:Offline 00:08, 20 August 2016 (UTC)

Apparent bot error

I was confused by this edit — the work with the census URL makes sense, but why did it tag the county website as a permanent dead link? http://www.co.warren.oh.us is a working HTTP redirect (any English-speaking human can see that the target URL is the county website); I doubt that the bot's set up to mark redirects as dead, and I can't understand what else would have prompted this tag. Nyttend (talk) 14:20, 19 August 2016 (UTC)

This was probably already fixed.—cyberpowerChat:Offline 00:15, 20 August 2016 (UTC)
Actually, I fixed it just after leaving this note. I've checked the other townships in the county (the "townships" line of {{Warren County, Ohio}}), all of which have an external link to the same URL, but the bot didn't tag the EL on any of the other articles. Ditto for the parent Warren County, Ohio article. Nyttend (talk) 02:22, 20 August 2016 (UTC)
I meant internally. That tagging was a remnant left behind in the DB during the trials.—cyberpowerChat:Absent 11:38, 21 August 2016 (UTC)

Double archive URL

[10] (see archiveurl on left side). This wasn't caused by IABot, but wanted to let you know there are cases out there. Or {{wayback}} with a full wayback URL in the url field creates the same thing. I'm looking out for and logging them while WM2 runs. This particular case would be easy to fix with a bot but others are more hairy so for now just logging and skipping. Probably some in the cache DB. -- GreenC 16:19, 22 August 2016 (UTC)

Indeed. At least the bot can still resolve to the original URL given that link and extract the snapshot time. As for improper usages of the wayback template, IABot can detect those errors and fix the template on the spot. :-)—cyberpowerChat:Absent 16:44, 22 August 2016 (UTC)

A barnstar for you!

The Special Barnstar
Thank you very much for completing my name change quickly and without any hassle. It is much appreciated. Enjoy this! Class455 (talk) 15:59, 23 August 2016 (UTC)

New Account

How do I go about? What do I need to take care? And who will do it - can I go create myself? Thanks VarunFEB2003 I am Offline 11:45, 25 August 2016 (UTC)

Why do you need another account? Why not focus on building the encyclopedia instead?—cyberpowerChat:Online 12:55, 25 August 2016 (UTC)

Bug: bot tried to fix example.org URLs in example code

Stale
 – Bug already fixed.

Hi, thanks for InternetArchiveBot.

In this edit the bot saw URLs in some XML code and added template markup; but the markup isn't interpreted and the URLs aren't meant to be real anyway. I think the bot should:

  • ignore content within <source>...</source>, or at least not edit it;
  • ignore URLs under example.org and similar domains.

Thanks. —80.192.178.199 (talk) 21:33, 26 August 2016 (UTC)

User:Cyberbot II Suggestion

Hello. Consider using the oldest archived version, especially for news citations, to avoid archived "404" pages, since news stories are rarely changed, but are often deleted. Non-news story pages are sometimes updated and newer non-"404" archived versions are usually preferred. Thanks. Xb2u7Zjzc32 (talk) 23:41, 27 August 2016 (UTC)

IABot picks the archive closest to the initial access date.—cyberpowerChat:Offline 23:50, 27 August 2016 (UTC)

IABot archived not-dead link?

I'm not sure why it archived, maybe the link was temporarily dead. It picked a pretty old archive date too, is the date it uses influenced by access date (although there wasn't one in this case)? https://en.wikipedia.org/wiki/Talk:NoteTab#External_links_modified ChrstphrChvz (talkcontribs) 08:19, 29 August 2016 (UTC)

URL backslash

A heads up I ran into trouble with URLs like this:

https://web.archive.org/web/20041031053932/http://www.boxoff.com/cgi/getclassic.pl?filename=Classic&where=Name&terms=SUNSET\x20BOULEVARD

Note the "\x20" .. this is a part of the URL not an escape sequence. However the IA API returns a JSON like this:

{"results": [{"url": "http://www.boxoff.com/cgi/getclassic.pl?filename=Classic&where=Name&terms=SUNSET\\x20BOULEVARD", "timestamp": "20041031053932", "archived_snapshots": {"closest": {"status": "200", "available": true, "url": "http://web.archive.org/web/20041031053932/http://www.boxoff.com:80/cgi/getclassic.pl?filename=Classic&where=Name&terms=SUNSET\\x20BOULEVARD", "timestamp": "20041031053932"}}, "tag": "0"}]}

Note the double-backslash. This creates problems with internal language libraries which either interpret it as an escape sequence, or the URL is seen as having literally two backslashes, which breaks it. Anyway, might be worth testing out on the bot to see how it responds. Same problem with any URL containing a backslash not just \x## sequences. -- GreenC 14:15, 30 August 2016 (UTC)

My best guess is the double backslash will be interpreted as an escaped backslash (single backslash) when PHP decodes the JSON.—cyberpowerChat:Online 14:18, 30 August 2016 (UTC)
Ok. I checked the RFC which says "characters that must be escaped: quotation mark, reverse solidus, and the control characters (U+0000 through U+001F)". So the IA API is correct to escape the backslash, and PHP is probably handling it correctly. Looking more closely it may be in the way I'm using the Nim JSON parser. Probably still worth checking out PHP in case it turns \x20 into a <space> character somewhere along the way. -- GreenC 15:08, 30 August 2016 (UTC)

Help Account creation interface

Actually I am interested in participating and help out with Requests for account creation. So I accessed the interface and I was unable to register myself and the error was that the username was already registered. But I don't remember registering myself. So I tried forgot password option. I was unable to reset as the username and password didn't match. So I tried the same with my past email id. And I got the response that instructions were sent to my email. But I have deleted the email long back, and I am no longer using. As you are the manager of the interface, I've come here for a solution. Regards, Krishna Chaitanya Velaga (talk • mail) 07:37, 28 August 2016 (UTC)

I don't manage the interface, but what was the account name you tried registering as?—cyberpowerChat:Online 14:57, 28 August 2016 (UTC)
Any particular reason you impersonated me on IRC?—cyberpowerChat:Online 17:39, 28 August 2016 (UTC)
I tried registering as "Krishna Chaitanya Velaga". I apologize for my mistake on IRC, actually I wanted to talk to you on IRC, so in that intention I typed you username in the block and later understood that it was a mistake. OK, that you don't maintain, but please suggest me a solution. Regards, Krishna Chaitanya Velaga (talk • mail) 14:09, 30 August 2016 (UTC)
So this is you on ACC. Your account exists. A dev can forcibly change the email to what you need it to be. Also you need to appeal your account decline. I suggest you send an email to our mailing list requesting a change of email address for you account, so you can then send yourself a password reset. While you're at it, you can also appeal your decline, but I don't know why you're declined.—cyberpowerChat:Online 15:03, 30 August 2016 (UTC)
I sent a mail. But where should I appeal? Regards, Krishna Chaitanya Velaga (talk • mail) 15:57, 30 August 2016 (UTC)
That's done via mail as well.—cyberpowerChat:Online 16:17, 30 August 2016 (UTC)
I have successfully reset my password, but after I login I am getting a response that my account was declined and the reason field was blank. Regards, Krishna Chaitanya Velaga (talk • mail) 13:55, 31 August 2016 (UTC)
Like I said you need to appeal that decline by sending an email as well.—cyberpowerChat:Online 14:15, 31 August 2016 (UTC)

Hello!

Hello Cyberpower. I've just read your signature key subpage. If you don't mind me asking, do you use a script to change it, or do you do it manually? I would like to display my online/offline status in superscript within my signature. Thanks in advance, Zerotalk 16:00, 29 August 2016 (UTC)

It uses a very complex template in my userspace I designed. User:Cyberpower678/Statussig is what User:Cyberpower678/SignatureStandard uses. It's very complex and requires perfect syntax less you end up substituting all of the template syntax when you sign. If you need help let me know.—cyberpowerChat:Online 16:05, 29 August 2016 (UTC)
I will probably need help haha! I will let you know once I'm back on my laptop, as I'll give it a go then. Thanks! Zerotalk 16:10, 29 August 2016 (UTC)
Patient Zero Contact me if Cyberpower finds less time to explain it I have the same setup (I too copied it from him!) VarunFEB2003 Offline when signed 15:34, 31 August 2016 (UTC)

Report

  • Noticed it's adding a blank "df=" in templates. [11]
    Per @Trappist the monk:.
  • Nice removing "Added by DASHBot" etc.. (WM does the same)
    Unintentional but nice. :-)
  • Why the conversion to YYYY-MM-DD? [12]
    Per bullet 1.
  • Problems with '$' [13][14]
    This looks like a PHP problem.
  • Problem with deadurl field [15]
    What's wrong with it?

-- GreenC 03:22, 29 August 2016 (UTC)

Responses are inline.—cyberpowerChat:Offline 03:50, 29 August 2016 (UTC)
Ok date doesn't make sense to me but will check with TtM. Re the last one I forgot about that. One other thing is the IA API returns URL with port 80 in the URL, it isn't needed so WM strips it out but preserves if non-80. Any chance IAB could remove the ":80" from incoming API results? -- GreenC 04:05, 29 August 2016 (UTC)
I really don't think that is needed. As long as the URL works and the original URL and snapshot time can be extrapolated from them.—cyberpowerChat:Online 16:45, 29 August 2016 (UTC)
I fixed the $ problem.—cyberpowerChat:Online 17:13, 29 August 2016 (UTC)
It just adds a lot of unnecessary ":80" into articles and its simple to strip at the level of incoming IA API JSON. -- GreenC 13:47, 30 August 2016 (UTC)

Regarding dates and User talk:Cyberpower678/Archive 36#date format.. I see the idea there and agree with it when the situation arises. However that's different from converting every cite template (with an archive) to ISO format which probably exceeds the bots mandate, actually a major change. At least use |df=dmy-all / |df=mdy-all (when mdy/dmy are set at article level) so that access-date displays as intended for the article, otherwise the bot will be modifying existing display formats and will likely run into user trouble. -- GreenC 13:47, 30 August 2016 (UTC)

It's not converting every cite template with an archive. It's regenerating the template internally when it changes an archive URL. And I think you misunderstand the df parameter. Attaching "-all" changes all the date formats instead of just the access and archive dates.—cyberpowerChat:Online 13:51, 30 August 2016 (UTC)
Ok that's good it's not a universal change doesn't exceed bot mandate. I guess your right that setting -all might be disruptive in other ways. What about just preserving the existing format, if access-date is mdy continue to use that (unless dmy/mdy is set at global level) and lacking other guidance use ISO. -- GreenC 14:02, 30 August 2016 (UTC)
This stirs up the whole debacle about date formats again. There is no reliable way, that I can see, to accurately detect date formats. It defines the df parameter when there is a global tag in place, otherwise it defaults to ISO per the MOS. That's the best I can do with the bot.—cyberpowerChat:Online 14:05, 30 August 2016 (UTC)
If the bot is able to extract the existing date value from a citation argument such as accessdate, it should be trivial to determine its dmy/mdy/ymd format. For example, WM does this and if it has trouble it defaults but that is rare. And if the bot is able to change date formatting on a per-argument basis then it would be possible, but I don't know enough details of IABot design. See WP:CITEVAR for info on citations and dates. There is no MOS style for dates in citations, however it does say not to arbitrarily change styles (the indented block quote). If it's a technical issues with the bot that's different. -- GreenC 14:42, 30 August 2016 (UTC)
It's more of technical limitation of the bot at the moment.—cyberpowerChat:Online 14:51, 30 August 2016 (UTC)
I think I came up with an idea on how to implement date format detection. I will use the access date parameters and detect whether they follow mdy, dmy, ymd.—cyberpowerChat:Offline 23:07, 30 August 2016 (UTC)
And implemented.—cyberpowerChat:Offline 01:25, 31 August 2016 (UTC)
Great! That's what WM is doing also, but keying on archivedate (WM only works on cites where an archive already exists). If it can't determine it uses the global setting, and if still nothing defaults to ISO. -- GreenC 17:32, 31 August 2016 (UTC)
It looks for all 4 keys.—cyberpowerChat:Online 17:33, 31 August 2016 (UTC)

Mistake warning

I received this message from the bot, warning not to remove AFD notices. Even though I already closed the AFD (though a non-admin closure since I'm not an administrator) ([16] [17] [18]) as a keep, the bot proceeded to re-add the AFD template ([19]), that I had to remove again ([20]). Just thought I'd bring to the attention of an apparent bug that this bot re-adds the AFD template to even articles whose AFD's have been closed. —Mythdon (talk) 05:10, 1 September 2016 (UTC)

Indeed, that is a bug. I need to overhaul that bot script at some point.—cyberpowerChat:Online 13:30, 1 September 2016 (UTC)

I was on the talk page https://en.wikipedia.org/wiki/Talk:Indigenous_peoples_of_Siberia and tested the links provided there for fixing dead refs in the article. After I tested and set the checked indicator to true, I noticed they are not on the actual article page. Is this the usual practice, or a bug? Do they now get put into the article page? Thanks! Nlight2 (talk) 09:58, 1 September 2016 (UTC)

Yes they are. I'm not sure what your seeing, but I'm see that they were added.—cyberpowerChat:Online 13:30, 1 September 2016 (UTC)

Webcite

So have discovered the Mementoweb API results for Webcite links are unreliable, they are either out of date or wrong in about 10-20% of cases. Suspect Webcite is recycling IDs after an archive is deleted and the Mementoweb database is not updated. Hopefully your using the Webcite xml API which seems accurate. At some point I need to go back and check the Webcite links added by WM1, fortunately not many. -- GreenC bot (talk) 15:47, 1 September 2016 (UTC)

Yes, I'm using the XML API, but webcite is not recycling the IDs. The IDs are timestamps in microseconds the moment of capture, and then encoded in base62. Archive.is's IDs however are base62 sequence IDs. Essentially each new snapshot is part of an auto-incrementing sequence, and those snapshots that get deleted, the IDs there do get re-used.
P.S. Your bot seems to have acquired a consciousness, please make sure it doesn't deem humanity as a threat.—cyberpowerChat:Online 15:52, 1 September 2016 (UTC)
A new bot module under development. Ideally the bot will fix its own bugs someday :) Interesting about archive.is recycling ID's another reason to use long-form URL. -- GreenC 19:05, 1 September 2016 (UTC)

Is a transfer of rights to a new account possible?

Hey,

through happenstance I saw that you can see here who created pages related to each user of Wikipedia. In my past I created pages through requests and I really don't want people to know that I created these. In this time I was young and so I grew up, start to study at university the next days. There, we will work with Wikipedia in some cases. I write to you because you were listed on this page I linked to as admin. Is it possible to take my page generation of some pages I don't want they know it was me? If this won't work, is it possible when I create a new account, so that I get there my rights of publish or control pages without any authorization? I really want to work more on Wikipedia; want to prove my progress in style of writing but also want people to don't get me wrong.

Best regards --Serdar² (talk) 08:41, 4 September 2016 (UTC)

Sorry but the English grammar in your first question is off. As such, I don't quite understand what you are asking. Try asking me in German. :p—cyberpowerChat:Online 15:14, 4 September 2016 (UTC)
@Serdar²: I think I understand what you are talking about, and no, it is not possible unless you completely abandon this account and start a new account with a different name. Pppery 16:53, 5 September 2016 (UTC)
@Cyberpower678: and :@Pppery: Sorry for that grammatical disaster - too early in the morning that day. So, if I abandon my account and create a new account then, how could I get my rights I enjoy back? Thanks for answering me! --Serdar² (talk) 21:11, 5 September 2016 (UTC)
The question is why do you want to abandon your account and start a new one? Why not rename your account?—cyberpowerChat:Limited Access 21:14, 5 September 2016 (UTC)
I wrote some articles I don't want others to know it was me. What could I say on this, too naive, too young? Mostly I wrote them for requests. --Serdar² (talk) 21:44, 6 September 2016 (UTC)

Reminder

Hello Cyberpower678. You were apparently away for a while and didn't respond to the message I left here (not that I needed you to). You may do what you like about it, but I just wanted to ensure you had seen that your bot task was disabled.[21] -- zzuuzz (talk) 08:11, 7 September 2016 (UTC)

I did not notice, so thank you. I have been busy with the continuing development of my new pride and joy, InternetArchiveBot.—cyberpowerChat:Limited Access 19:09, 7 September 2016 (UTC)