This topic contains 6 replies, has 0 voices, and was last updated by  PeanutsRevenge 6 years ago.

  • Author
    Posts
  • #7614

    PeanutsRevenge
    Participant

    While playig Vovan and Andrey yesterday, I noticed they were able to swear in game without the filter kicking in.

    They were making no attempt to circumvent the filter, simply typed using latin keys.

    I’m assuming that they’re sending unicode for latin characters, if so, I expect it’s a git to try and filter, but thought I’d post anyway as english should really be filtered.
    I understand not having the time / ability to filter all languages, but english is IMO the web standard, so something should, if possible be done.

    #64648

    Tom.K
    Participant

    Here’s example:

    copy – Normal
    сору – Cyrillic

    Well, no difference, but take a look a bit better… It isn’t the same thing as you’ve expected to be.

    цопѕ – The actual Cyrillic of word copy if it’s roughly converted letter to letter. Not translated, just an example.

    Problem with this is that I’m not sure how that would be properly filtered, as there are many combinations, including decorated words like this: ĈѺ尸丫

    Not sure does it help at all… What I’ve just explained probably caused bigger problem.

    #64649

    Laptops Daddy
    Participant

    further problems relating to unicode. the player name had broken the banner events. that was an easy fix – just added character data sections to the xml files <!CDATA<. here's a problem, though. unicode player names are/were set as html entities. i'd used gavin's nice unicode_to_utf8 function. all very nice, until you apply CDATA and the ents dont get parsed.

    any bright ideas? i suspect i could convert back from html ents to straight characters in php. i guess the trick then would be to fwrite an xml using unicode character encoding. if anyone happens to have a solution for that – please post. im gonna have to do some googling if not.

    #64650

    PeanutsRevenge
    Participant

    Bloody forgeiners causing all this trouble.

    I’ve said it for years and I’ll say it again, all foreigners should be put down @ birth.

    #64651

    Laptops Daddy
    Participant

    ja, that would be kinder. gottverdammten ausländern, mit their accents und funny ways etc

    http://www.youtube.com/watch?v=IRfluaMKoOY

    anyway, after an hours messing about, i found a nice copy paste solution on the php website. htmlspecialchars(html_entity_decode($child, ENT_QUOTES, ‘UTF-8’), ENT_NOQUOTES, ‘UTF-8’); seems to work.

    #64652

    PeanutsRevenge
    Participant

    good ole htmlspecialchars

    #64653

    Laptops Daddy
    Participant

    yeap. who needs unparsed character data when we have htmlspecialchars? probably no one, i say.

    it took me a while to work out/half remember how you do unicode names. turns out we’re using javascript notation in the display.xml, so ɔʇǝ would be u0254u0287u01DD

    lovely code converter here: http://www.rishida.net/tools/conversion/

    #64654

    PeanutsRevenge
    Participant

    @laptops Daddy wrote:

    yeap. who needs unparsed character data when we have htmlspecialchars? probably no one, i say.

    it took me a while to work out/half remember how you do unicode names. turns out we’re using javascript notation in the display.xml, so ɔʇǝ would be u0254u0287u01DD

    lovely code converter here: http://www.rishida.net/tools/conversion/

    Should have asked, I have logs full of it when Vovan, Andrey and …. and … uh, someone else chatting away in complete gibberish 🙂

Viewing 8 posts - 1 through 8 (of 8 total)

You must be logged in to reply to this topic.