You might try these: require'trace' trace'+/"1''abc''=/''abcbdefbbcbaa''' trace'(+/"1''abc''=/])''abcbdefbbcbaa'''
This does not show everything, but does take you through the parsing process up to where the error happens in the second statement. (Also, beware of email agents which change quote marks - those should all be ascii single quotes, even when two or three of them occur in sequence, and not some other unicode quotes...) To see beyond that point, if you have jqt running (and all the addons installed), you might try this: require'debug/dissect' dissect '+/"1''abc''=/''abcbdefbbcbaa''' dissect'(+/"1''abc''=/])''abcbdefbbcbaa''' This will try to show you the execution process and also the specific verb which is throwing the error along with its arguments. I hope this helps. -- Raul On Tue, Sep 13, 2016 at 7:48 PM, Skip Cave <s...@caveconsulting.com> wrote: > Bob, > > Drat! I was hoping I could just take the expression: > +/"1'abc'=/'abcbdefbbcbaa' > Which works fine, and just turn it into a tacit verb definition without > fiddling with it: > ts1 =. +/"1'abc'=/] > Or better yet: > ts1 =. +/"1'abc'=/ > Apparently, when I removed the noun argument, the phrase turned into a > 'hook'. > > I have no idea why this happened. Seems like you should be able to get a > verb phrase working with an argument, and then remove the noun and add an > assignment to make a tacit verb. Apparently it doesn't work that way > though. > > Skip > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
