- From: Florian Rivoal <florian@rivoal.net>
- Date: Mon, 27 Jul 2015 00:54:42 +0200
- To: "Tab Atkins Jr." <jackalmage@gmail.com>
- Cc: www-style list <www-style@w3.org>
> On 23 Jul 2015, at 02:13, Tab Atkins Jr. <jackalmage@gmail.com> wrote: > > MQ currently requires whitespace around the and/or/not keywords. This > isn't strictly required for parsing; there's no reason to require a > space *before* the keywords, and technically you could put a comment > *after* the keyword to accomplish the separation as well. That is: > > 1. (foo)and (bar) = totes fine > 2. (foo) and/**/(bar) = totes fine > > The one troublesome issue is: > > 3. (foo) and(bar) = whoops, that's an and() function, not a keyword > followed by a parens. > > Because of case 3, and because case 2 is a confusing edge case, we > required whitespace after the keyword, and for symmetry/simplicity, > required it before as well. > > However! Per <https://code.google.com/p/chromium/issues/detail?id=390078>, > Rune once tried to implement this restriction, and broke lots of sites > relying on some MS-produced minifier, which produced code like case 1. > So apparently we can't actually implement this. > > Given that information, can we drop the requirement? Breaking lots of sites is bad, I'm in favor of allowing space before. - Florian
Received on Sunday, 26 July 2015 22:55:11 UTC