I like your explanation except that it does not seem to say that before ;&$ is executed, the &:>&;: must be executed, in order for the $ to be applied correctly.
On Sun, Oct 27, 2013 at 2:03 PM, Raul Miller <[email protected]> wrote: > Here's the same sentence with redundant parenthesis added: > > 'this is a test' ((((;&$) coer 1)&:>)&;:) 'and this is yet another test' > > Hopefully this helps? > > Put differently, in http://www.jsoftware.com/help/dictionary/dicte.htm the > parsing rule "4 Conj" has this pattern: > > EDGE+AVN *VERB+NOUN **CONJ **VERB+NOUN* > > And this means that (coer 1 &: >) will not reduce because coer is not an > adverb, not a verb, not a noun and not an "edge". So the parser shifts > another word onto the stack and it's not until after we have reduced ('this > is a test' ; & $) that we can start cleaning that up. > > Eventually, of course (since the sentence is syntactically valid) the > parser gets back to where things were, and the "coer 1" part is replaced by > a slightly involved derived verb which lets ((...) &: >) reduce. > > Does this make sense? > > Thanks, > > -- > Raul > > -- (B=) ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
