Talk:Title blacklist
For more information on what the title blacklist is for, and the processes used here, please see Title blacklist/About.
Please post comments to the appropriate section below: Proposed additions, Proposed removals, or Troubleshooting and problems, read the messageboxes at the top of each section for an explanation.
Also, please check back some time after submitting, there could be questions regarding your request. In addition to that, please sign your posts with ~~~~
after your comment.
Completed requests are archived, additions and removal are logged.
- Information
- List of all projects
- Overviews
- Reports
- Wikimedia Embassy
- Project portals
- Country portals
- Tools
- Spam blacklist
- Title blacklist
- Email blacklist
- Rename blacklist
- Closure of wikis
- Interwiki map
- Requests
- Permissions
- Bot flags
- New languages
- New projects
- Username changes
- Translations
- Speedy deletions
SpBot archives all sections tagged with {{Section resolved|1=~~~~}} after 1 day and sections whose most recent comment is older than 30 days.
|
snippet for logging: {{tbl-log|27693124#{{subst:anchorencode:SectionNameHere}}}}
Proposed additions
editThis section is for proposing that a page title be blacklisted; add new entries at the bottom of the section. Provide links demonstrating widespread creation by multiple users or IP's. Completed requests will be marked as done or denied and archived.
|
Nigger escapes on I
edit.*n[1jl/\\]gg(a|er).*
Keywords that are not valid in any language and are used to escape the N-word on blacklists, including . All usernames matching this regex, without antispoof-matching. .*\bn1gger.*
is a strict subset. The base regex of .*\bn[il]gger.* <newaccountonly|antispoof>
is also a strict subset, but this keyword also has antispoof. I haven't tested antispoof on this keyword. LaundryPizza03 (talk) 09:05, 19 October 2024 (UTC)
- Probably, antispoof is too big, since nigger itself matches. LaundryPizza03 (talk) 09:09, 19 October 2024 (UTC)
- Using the Abusefilter debugging tools with:
ccnorm("nigger") irlike ".*n[1jl\/\\\]gg(a|er).*"
, outputs false. So antispoof might be fine. XXBlackburnXx (talk) 15:32, 19 October 2024 (UTC)
- Using the Abusefilter debugging tools with:
Proposed removals
editThis section is for proposing that a title be unlisted; please add new entries at the bottom of the section. Remember to provide the specific title or regex blacklisted, links to the articles they are used in or useful to, and arguments in favour of unlisting. Completed requests will be marked as done or denied and archived. The addition or removal of a title is not a vote, please do not bold the first words in statements. |
Niger
edit.*ni[ğǵģġĠĞ]+er.*
The keyword .*ni[ğǵģĝġĠĞ][ğǵģĝġĠĞ]+er.*
is a strict subset, but this one maight include false positives for Niger (the country) or other non-offensive words including this substring. I'd suggest also abbreviating the smaller regex as .*ni[ğǵģĝġĠĞ]{2,}er.*
. LaundryPizza03 (talk) 08:39, 19 October 2024 (UTC)
Dick usernames
edit.*[8B]\={2,}D.* <newaccountonly>
This regular expression is now useless because = can no longer be used in usernames. A minimal change that could provide slight improvement in performance. Mori Calliope fan (talk) 05:56, 1 November 2024 (UTC)
Troubleshooting and problems
editSpam rules
editI've noticed a few issues with some of the spam rules. The rule .*\b(?:dvantages|benefits) of (?:booking|renting|le(?:tt|as)ing).* <autoconfirmed>
doesn't include the "a" in "advantages", and the rule .*\blimo (p?:rovider|company|service|program|support|assist).* <autoconfirmed>
has the "?:" after the "p" in "provider" instead of before, which causes it to check for "limo p:rovider" and "limo :rovider" instead of "limo provider", which is not correct. I found the diff where these rules were added, and it includes the "a" in "advantages" in the first rule and doesn't include "?:" in either of these rules, indicating that these rules aren't checking for intentional spelling errors, and that someone made a typo when changing these rules. TTWIDEE (talk) 18:52, 17 October 2024 (UTC)
- (introduced here). I did the fixes. Thanks for reporting. - XXBlackburnXx (talk) 09:28, 18 October 2024 (UTC)
I've noticed some more errors. The rule .*(stage|phase|period|developmment|level)s? of (cystic )?(pimple|zit|acne).* <autoconfirmed>
has "developmment" instead of "development", and the rule .*luxury (?:vehicle|automo(tive|bile)|(?:ar or )?truck) (?:review|evaluation|critiques|rating|opinion).* <autoconfirmed>
has "ar" instead of "car". The diff that originally added the "developmment" error adds it as a new word rather than modifying an existing word (albeit modifying an existing rule), so although it's probably just a typo, I could certainly imagine it being added to check for a common typo. The second typo was created by modifying an existing word rather than adding a completely new word, as shown in this diff where the word was originally added without the typo, and the typo seems to have been introduced by someone trying to add "?:" before "car", like what we've seen with the other spam rule typos, so I'm pretty convinced this is just a typo in the rule rather than checking for a common error. TTWIDEE (talk) 18:23, 23 October 2024 (UTC)
- Doing... - XXBlackburnXx (talk) 18:44, 23 October 2024 (UTC)
- Done - Just to be sure, I looked at previous page creation/abusefilter records and it doesn't seem like the typo in 'developmment' was intentional, same goes for "car". XXBlackburnXx (talk) 18:54, 23 October 2024 (UTC)
I've noticed an issue with Pragmaticplay\d+ <newaccountonly>
and \bpay[0o]neer\b <newaccountonly>
. According to the documentation at mw:Extension:TitleBlacklist, when someone registers a new account, it checks "User:" followed by their account name against the blacklist, rather than the account name itself. So Pragmaticplay\d+ <newaccountonly>
will not match any usernames, because it doesn't start with "User:". We should change this rule to User:Pragmaticplay\d+ <newaccountonly>
. Note that this will only match usernames that are just "Pragmaticplay" followed by some digits and will not match usernames that just contain "Pragmaticplay" followed by some digits; for the latter behaviour, we'd have to use .*Pragmaticplay\d.* <newaccountonly>
. (Note that the "+" after the "\d" is not needed in this case, because any subsequent digits are taken care of by the ".*".) The same also applies to \bpay[0o]neer\b <newaccountonly>
—since it begins and ends with "\b", I'm guessing that the intention was to add .*\bpay[0o]neer\b.* <newaccountonly>
, because User:\bpay[0o]neer\b <newaccountonly>
would be equivalent to User:pay[0o]neer <newaccountonly>
, because ":" is not a word character, and the end of the string counts as a word boundary. It would be a bit odd to have "\b" in there for no reason, so I'm guessing that whoever added the payoneer rule didn't realise that the blacklist doesn't do partial matches by default. TTWIDEE (talk) 21:09, 1 November 2024 (UTC)
"Motherfuck"
editThe rule .*[Mm]otherr* *.?[Ff]uck.* <newaccountonly>
can be simplified to just .*mother+ ?.?fuck.* <newaccountonly>
(which is much easier to read), because:
- The filter rules are case-insensitive by default, so there's no need to have a character class with just a capital letter and its corresponding lowercase letter.
rr*
is equivalent tor+
.- Multiple consecutive spaces are treated the same as one space by the MediaWiki software, so " *" can be simplified to just " ?".
The spacing before and after this rule is a bit weird, but obviously that's not as important.
TTWIDEE (talk) 20:06, 24 October 2024 (UTC)
- We had a mix of spaces and tabs filling the same area which was making the editor act weird. Sorted that out first. XXBlackburnXx (talk) 22:14, 26 October 2024 (UTC)
- Done - the odd casing is probably due to gUser search being case sensitive by default. - XXBlackburnXx (talk) 22:40, 26 October 2024 (UTC)