$12256 / $11500
Not sure if this is a bug or intentional:
OGA can't seem to handle using some raw unicode characters/emojis in posts. An example is, I wanted to use the heart character in a comment. But after hitting "save", my browser (Chrome) displayed a page with the following text:
This page isn’t working
opengameart.org is currently unable to handle this request.
HTTP ERROR 500
Personally, I'd like to use some of these special characters in my comments & posts.
Testing using heart character in forum post: (doesn't work)
Testing using heart character in forum post url:
I have this problem all the time when trying to type en (--) and em (---) dashes, which I'm used to entering directly as unicode characters (alt+hyphen alt+shift+hyphen respectively)...
For a while I couldn't figure out why my posts were always getting rejected! Then MedicineStorm helped explain it to me. It'd be great to handle these characters, or at the very least strip them out or give an informative error message.
Indeed. I'll bug some people about it, but I don't currently have the tools to perform such a fix.
See also:
--Medicine Storm
I guess that limits our communications to English. :p
Probably, though I have no idea how I submitted a conversation using japanese kanji: https://opengameart.org/comment/22426#comment-22426
Why do these characters copy-paste, but don't work when I type them directly:あいうえおかがきくこしすずせただちってとどなにのはぴまみめもやよらりるれろわをんアィウエクグゲコサザジスセソタッデトニバパフブポムメユラリルレンー事件会作使出利制合呑問営場対布式当数文時書本条来株機次止禁緒聞能良製許諾買質配間限音2
--Medicine Storm
That's interesting. Why would it make a difference?
No idea, but it seems like copy-pasting symbols from different sources may be translating the format to something the site doesn't like. Copying from one source = ok. Copying the same text from another application = white-screen-of-death. Looks like there are very specific characters the site doesn't like, but there doesn't appear to be a pattern. The parts of the kanji it doesn't accept seem totally arbitrary. Consistent, but arbitrary.
--Medicine Storm
Is it safe to label it as a unicode/UTF-8 compatibility issue?