On 1/5/2011 10:20 PM, Ken Wesson wrote:
On Wed, Jan 5, 2011 at 8:39 PM, Tim Daly<d...@axiom-developer.org> wrote:
On 1/5/2011 8:27 PM, Mark Engelberg wrote:
On Wed, Jan 5, 2011 at 4:44 PM, Eric Schulte<schulte.e...@gmail.com>
wrote:
For the most up-to-date and comprehensive documentation of using
Org-mode to work with code blocks (e.g. Literate Programming or
Reproducible Research) the online manual is also very useful.
In literate programming org-mode, will Clojure code be properly
highlighted and indented?
Is there a keystroke (like Ctrl-c,Ctrl-k) that will evaluate the
Clojure code in the entire file and send it to the swank REPL?
Will stacktraces point at the correct line number?
It seems that your real question is whether Clojure knows about
a literate document. It does not. But it would be possible to
modify the reader behavior when given a pamphlet file. The REPL
uses a line numbering reader. Anything between the last
\end{chunk} and the next \begin{chunk} could be considered
as comments to be ignored but the line numbers for the function
would be correct and therefore the stack traces would be correct.
I suppose it would be reasonably easy (there is no such thing
as a simple job) to write a literate reader for the REPL. All
it would need to know is where to turn-on and turn-off the normal
read semantics.
If we had custom reader macros in Clojure we wouldn't even be having
this discussion; you would probably have already implemented it by
now. :)
LispReader is a class that appears to have a read function that does
Clojure s-expression parsing. Wrapping that around a LiterateReader
stream would seem to do the job. The LiterateReader stream only has
to change any non-chunk line into a Clojure comment by prepending
a semicolon. The code to scan a line that begins with a \begin{chunk}
or \end{chunk} does not seem all that challenging.
In fact, a slightly smarter LiterateReader stream could be given the
particular line as an argument and only call LispReader on that
s-expression. So a LiterateReader with an optional line number
argument would allow an editor to specify where to start .read.
A reader macro would require special syntax. This may be reasonable
but it seems that a simple (load-literate "filename" N) would be
all that is needed, requiring no special syntax.
Alternatively the reader could use the file extension so that
pamphlet files would invoke the LiterateReader stream automatically.
Or if the parse of the first line begins with \ then use LiterateReader.
I'm documenting Bit-partitioned hash tries at the moment. I'll
see if I can document the reader next and get an idea of exactly
how it works and what it would take to change it.
Tim Daly
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to clojure@googlegroups.com
Note that posts from new members are moderated - please be patient with your
first post.
To unsubscribe from this group, send email to
clojure+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en
