It is almost impossible to google, hence my understanding is limited to contextual clues from reading through the slime source code: perhaps it is part of the object system in common lisp? Something like 'self'?
snippet:
(cond #+#.(swank-backend::sbcl-with-new-stepper-p)
Perhaps this will make it more googleable : pound plus pound // hash plus hash symbol // octothorp plus octothorp
That's pretty rare to see.
#+clim clim:+red+ #-clim mygraphics:+red+
above means that the reader returns either red symbol and it depends whether there is a symbol with the name CLIM is on the list of features *features*
. That's a built-in mechanism in Common Lisp.
#.(cl:if (cl:zerop (cl:random 2)) :high :low)
Above also is a mechanism of the reader. It allows to do computations at read time. Which btw. is a security problem and in Lisp applications it should be disabled - see the variable *read-eval*
for controlling this. At read time the reader using READ will return either :HIGH or :LOW, randomly.
The combination #+#.(FOO) BAR
means that the function foo returns a symbol at read time and this symbol then is checked by the reader if there is a symbol with this name on the feature list *features*
and if that is the case, then the next item in input is read, otherwise the next item is skipped over.
Trivial example, IF always returns :CAPI in this example:
In LispWorks (where CAPI is on the features list):
CL-USER 41 > (read-from-string "#+#.(cl:if cl:t :capi :clim) a b")
A
31
In SBCL
* (read-from-string "#+#.(cl:if cl:t :capi :clim) a b")
B
32
It's actually Sharpsign Plus followed by Sharpsign Dot.
They're Common Lisp reader macro characters:
a textual notation introduced by
dispatch on one or two characters that
defines special-purpose syntax for use
by the Lisp reader, and that is
implemented by a reader macro
function.
Reader macros should not be confused with regular macros - they have nothing to do with each other.
The set-dispatch-macro-character function can be used to extend the Common Lisp syntax with custom reader macros.
Also, #-foo(code to execute)
will execute the code only if foo is not in *features*
.
Information about #+
, #.
, etc. is difficult to find even in the Common Lisp HyperSpec, but relevant HyperSpec pages can be found by Googling "Sharpsign minus", etc. (Thanks Austin.)
2.4.8.17 Sharpsign Plus
http://www.lispworks.com/documentation/HyperSpec/Body/02_dhq.htm
#+test expression
Readtime macro, if test is true read expression, otherwise read it as white space.
2.4.8.6 Sharpsign Dot
http://www.lispworks.com/documentation/HyperSpec/Body/02_dhf.htm
The #. foo syntax performs a read-time evaluation of foo.
http://www.lispworks.com/documentation/HyperSpec/Body/02_dh.htm
All the #X operators (defined by the Lisp HyperSpec)
Why does the Lisp HyperSpec have to be so hard to read?
Probably because it's done by committee. I say it's one of the reasons Lisp is not popular. Also, Lisp is too academic, and the barriers-to-entry are too high (the learning curve and support community are not so great... there aren't 10 thousand little programs out there to easily get you started (in some arcane implementation you are using)).