Looked some more at `rule` and `rslot` and I not 100% sure it's wrong, but I can't convince myself it's right.
On Thu, Aug 17, 2017 at 3:23 PM, Jeffrey Kegler < [email protected]> wrote: > Sorry to take so long. > > Ambitious stuff! > > Nothing obvious. If it were my code, I'd first double-check the `rule` > and `rslot` logic, though nothing about it is obviously wrong. > > On Wed, Aug 16, 2017 at 9:45 PM, Andreas Kupries <[email protected]> wrote: > >> >> Inching along in my work on >> https://core.tcl.tk/akupries/marpa/timeline >> >> I have come across a weirdness/bogosity regarding the retrieval of the >> valuation instructions (steps) from a `Marpa_Value`. >> >> First, on the good side, the relevant method (get-parse) in the bocage >> glue class [1] works just fine. Whether for the lexer or the parser, I >> properly get all the steps (tokens, rules, nullings, etc). >> >> On the other hand, in the structurally identical functions for the >> C-level runtime I am now working on (lexer [2], parser [3]) I see only >> 2 types of step types: "token" and "inactive". Rule and nulling steps >> are missing (and I placed the TRACE commands such that I really should >> see everything coming out of `marpa_v_step`). >> >> I have no idea what is going with the Marpa_Value. I actually can see >> where in the list of steps the rules should show up, by looking at the >> stack locations coming with the token steps. I.e. I see token(value)s >> getting pushed on the stack, dst-slot increasing and then a token >> suddenly writes into the middle instead of the top. Implying that at >> least one rule is in between them, doing reductions. But that rule >> instruction was not returned by `marpa_v_step`. >> >> I am pretty sure that the grammar and recognizer are correct, simply >> because I do get the proper lexemes out of the system [4] instead of a >> parse error. >> >> Anybody with an idea what I might be doing wrong ? >> >> [1] https://core.tcl.tk/akupries/marpa/artifact/b840a1e9de9c9e42 >> ?ln=130-274 >> [2] https://core.tcl.tk/akupries/marpa/artifact/b28b552a320a424d >> ?ln=383-464 >> [3] https://core.tcl.tk/akupries/marpa/artifact/ab0d46ff4344c06e >> ?ln=162-288 >> [4] At the lexer level the token steps alone are good enough to at >> least know that. >> >> -- >> See you, >> Andreas Kupries <[email protected]> >> <http://core.tcl.tk/akupries/> >> Developer @ SUSE (MicroFocus Canada LLC) >> <[email protected]> >> >> Tcl'2017, Oct 16-20, Houston, TX, USA. http://www.tcl.tk/community/tc >> l2017/ >> ------------------------------------------------------------ >> ------------------- >> >> >> >> >> -- >> You received this message because you are subscribed to the Google Groups >> "marpa parser" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> For more options, visit https://groups.google.com/d/optout. >> > > -- You received this message because you are subscribed to the Google Groups "marpa parser" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
