Re: The beauty of LATEX
Joel de Guzman <[email protected]>
| Newsgroups | gmane.comp.lib.boost.documentation |
|---|---|
| Message-ID | <[email protected]> |
On 10/30/2011 8:29 AM, Dave Abrahams wrote:
> "text-to-text into (a subset of) the IR" doesn't sound compatible with
> the idea that the "IR **IS A TREE**." Text is typically unstructured
> data.
Not necessarily. XML is structured text. So is the template IR language
that is the subject of this long discussion. This, for example:
[orderedlist
[One]
[Two]
[Three
[itemizedlist
[Three.a]
[Three.b]
[Three.c]
]
]
[Four]
]
is a structured tree, but it is *still* textual. The wiki preprocessor
can be a lexical processor that generates the code above from this
wiki text:
# One
# Two
# Three
* Three.a
* Three.b
* Three.c
# Four
So, in a sense, the wiki parser should know a bit about
(a subset of) the IR language, enough to generate what it
needs to deal with.
Regards,
--
Joel de Guzman
http://www.boostpro.com
http://boost-spirit.com