parsergen: improve symbol-discard in error handling.

[ocean] / csrc / parsergen.mdc

diff --git a/csrc/parsergen.mdc b/csrc/parsergen.mdc
index c25a87e1f621fd3ffa95d4a809c2276a235b2476..aacc59810be2e13aad50d1494171609797dcd71d 100644 (file)
--- a/csrc/parsergen.mdc
+++ b/csrc/parsergen.mdc
@@ -2713,9 +2713,32 @@ the most recent start of line.  This is how a newline forcible
 terminates any line-like structure - we try to reduce down to at most
 one symbol for each line where newlines are allowed.
 
+When, during error handling, we discard token read in, we want to keep
+discarding until we see one that is recognised.  If we had a full set
+of LR(1) grammar states, this will mean looking in the look-ahead set,
+but we don't keep a full look-ahead set.  We only record the subset
+that leads to SHIFT.  We can, however, deduce the look-ahead set but
+looking at the SHIFT subsets for all states that we can get to by
+reducing zero or more times.  So we need a little function which
+checks if a given token is in any of these look-ahead sets.
+
 ###### parser includes
        #include "parser.h"
+
 ###### parser_run
+
+       static int in_lookahead(struct token *tk, const struct state *states, int state)
+       {
+               while (state >= 0) {
+                       if (search(&states[state], tk->num) >= 0)
+                               return 1;
+                       if (states[state].reduce_prod < 0)
+                               return 0;
+                       state = search(&states[state], states[state].reduce_sym);
+               }
+               return 0;
+       }
+
        void *parser_run(struct token_state *tokens,
                         const struct state states[],
                         int (*do_reduce)(int, void**, struct token_config*, void*),
@@ -2852,7 +2875,7 @@ one symbol for each line where newlines are allowed.
                                break;
                        }
                        tos = &p.stack[p.tos-1];
-                       while (search(&states[tos->state], tk->num) < 0 &&
+                       while (!in_lookahead(tk, states, tos->state) &&
                               tk->num != TK_eof) {
                                free(tk);
                                tk = tok_copy(token_next(tokens));