[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Lexical syntax for boxes

This page is part of the web mail archives of SRFI 111 from before July 7th, 2015. The new archives for SRFI 111 contain all messages, not just those from before July 7th, 2015.



On 05/23/2013 06:26 PM, Alex Shinn wrote:
Well, Scribble is fairly complex too.  But on double checking
SRFI-108 is not as large as I thought - I think it was the whole
set 107, 108, 109 that was intimidating.  I'll take a closer look.

My habit of excessively mentioning every possible alternative
and issue doesn't help any ...

I think using @ for rough Scribble compatibility, or \ for LaTeX
familiarity and backwards compatibility with Scheme would be a
good idea though.

This is discussed here:
http://srfi.schemers.org/srfi-108/srfi-108.html#delimiter-options
as well as the SRFI-10[79] mailing lists.

There was some discussion, and the consensus was &.  We can of course
re-open the issue, especially if the "owners" of Scribble get involved.

There are a number of issues to keep in mind, including:

- I'd like to use a "compatible" syntax for SRFI-109 quasi-strings.
If we use & then it is ok to use &{text} for a string, but if we
use @ then @{text} is incompatible with Scribble, so we presumably
have to consider some alternative - like @'{text} of #'{text}.

- Using @ isn't really an option for SRFI-107 XML-literals, so
we'd get less compatibility there.

- Scribble maps @foo{text} to (foo "text") but SRFI-108 maps
it to ($construct$:foo "text").  I think the latter is more
flexible and thus preferable.  However, this would be an
incompatibility that reduces the advantage of following Scribble.

- The prefix character @ is natural for a splicing operator,
since it is already used for that in quasi-quotation.  Thus
(foo @(list a b c)) would be equivalent to (foo a b c), like
an implicit call to apply.  It seems a shame to foreclose that.
--
	--Per Bothner
per@xxxxxxxxxxx   http://per.bothner.com/