I'm meaning to do a complete re-write of the WakabaMark code. With this re-write, I could support things like nested quotes and lists, and quoted lists, and such. The thing is, though, to accomplish this I'd have to make the code strip out the ">"s in front of quoted text. Now, personally, I like quotes to have ">" in front of them, but I could make a much more flexible text parser without them.
Anyone have any strong opinions on this subject?
">" is a widespread and now very intuitively known and used feature on most image boards. Personally, I could get used to not using it, but then again maybe the majority might not.
Well, you'd still USE it, you just wouldn't see it in the actual text afterwards. This is somewhat less intuitive than the current situation, which is what makes me somewhat reluctant to do it.
As long as the quoted text is still a different color, I don't think it'll be that much of an issue.
By the way, I think lists (as in <ul> and <ol>-style stuff, I presume?) are a rather unnecessary feature. I'd rather see you get code blocks working instead of that. :P
I'm working on code blocks, too. Part of the reason I want lists is that I'm meaning to reuse this code later for things like the wiki, where lists are more useful.
Good gravy, you're gonna write wiki software too?!? You're on a rampage!
Feature request: The ability to not use fucking CamelCase. I abhor CamelCase with a fury and a passion, but all too often it's a standard feature on wiki software. [[Braces]] like MediaWiki (Wikipedia's software) uses are much more elegant and result in a page which is much easier on the eyes.
Incidentally, what are you planning on calling it? And do "Wakaba" and "Kareha" mean anything?
I probably won't do anything from scratch, but I've already got my modified TiddlyWiki for the documentation running at http://wakaba.c3.cx/docs/docs.html. I'm meaning to modify it even more when I get some time. For instance, the base TiddlyWiki does not work at all without Javascript, and I've been meaning to make it at least readable without it.
"Wakaba" means "new leaves", and "Kareha" means "old leaves".
"Wakaba" is a take-off on "Futaba" (Literally "two leaves", means a small sprout - come to think of it, "moe" comes from the verb "to sprout" - I wonder if there's a connection there). "Kareha" is to go with "Wakaba", in two senses: New japanese drivers get a "Wakaba mark" to put on their cars, which I've used as a favicon both here and on iichan, while old drivers get a "Kareha mark", which is used as a favicon for the kareha boards.
Can "*" and "_" please be replaced by something used less often?
The only usage I've seen that collides with the WakabaMark usage is when people use * for emoting, like *looks like an idiot*. Personally I can't help but feel that this is not a big loss.
The parser tries to be intelligent about when to trigger the markup, so you can write 6*9=3*18 all you want, or c_code_function_name(), and it won't get parsed.
> The only usage I've seen that collides with the WakabaMark usage is when people use * for emoting
There's also AA that always breaks with it. :(
Well okay, >>11, with AA getting broken by Wakabamark, how about making a couple of characters in a group used to ignore all others in that section, and when its posted, both the mysterious characters are missing AND the AA is wakabamark-free.
I already suggested to implement something like a function that would get triggered by AA (or maybe sjis) in the EMail/Link field and turn off Wakabamark. This could be combined with sage (as it works right now, if sage appears anywhere in the EMail/Link field, the thread will not be bumped).
I've been meaning to implement this, but I always get distracted by something more shiny.
Partition to NOT have quoting with > as part of Wakabamark.
I would like to disable Wakabamark without having to do without the Futaba Channel-style way of quoting.
If I recall my own code correctly, disabling WakabaMark disables > quoting only in Kareha. This matches the behaviours of 0ch and Futaba, respectively - Futaba will make quote sections based on >, but 0ch won't.
I had an idea for autodetecting AA which worked by converting the post to US-ASCII and seeing if anything was lost. It's not a very robust method, though.
It fails as soon as you use any other language than English, which is sort of common around these parts... I was thinking that a better method would be to check for characters that are often used in AA but seldom elsewhere, but I haven't had the energy to figure out such a list of characters.
Having experienced Wakabamark for quite some time at Wakachan and 4-ch now, I think it's not as useful as it seemed. Most cases where it appears still are people who don't understand it and are trying to do something else (emoting, AA kopipe, etc.)...
The idea of a clean, elegant syntax of Wakabamark that is more or less the same as already widely accepted forms of Markup may be noble, but I really think some more complicated and less elegant approach like bbcode would have provided better results in usability on these kind of boards.