If I understand well, we have to interpret the code as we compile it (as a macro's body is executed when a function is defined/compiled, not when executed) ? To me, macros were something translating a piece of code into another one, period. But that's not always the case as your example shows. If someone writes
(def bar (n)
(foo) (* n 2))
That means we have to display "blah" on compile time, right ? Hmm, that's much harder than I thought... It really means implementing an arc interpreter in arc (the way the official one is written in mzscheme), with for example prepending all the interpreted names with _ to distinguish them from the compiler's names...
Ok, I think I'm starting to realy get it ; or am I still missing an even trickier issue ?
That is about it. ^^ However my current implementation uses tables as environments for the function. Keys in the table are symbols for the variable names in that environment. A "parent" string key accesses the environment's parent.
The global environment is shadowed by a table; when this table is accessed, if a table entry does not exist, we look it up in the real environment using 'eval (!!) and retain a copy. This allows the macro to use stuff like 'map.
However there is still a leak here: if the real global-level environment has a function that modifies a global, it will modify a different global variable from the one that is visible to the macro. This shouldn't happen anyway: we should really avoid using globals in utility functions ^^.