1 |
2002-07-30 Akim Demaille <akim@epita.fr> |
2002-07-30 Akim Demaille <akim@epita.fr> |
2 |
|
|
3 |
|
Report rules which are never reduced by the parser: those hidden |
4 |
|
by conflicts. |
5 |
|
|
6 |
|
* src/LR0.c (save_reductions): Don't make the final state too |
7 |
|
different: save its reduction (accept) instead of having a state |
8 |
|
without any action (no shift or goto, no reduce). |
9 |
|
Note: the final state is now a ``regular'' state, i.e., the |
10 |
|
parsers now contain `reduce 0' as default reduction. |
11 |
|
Nevertheless, since they decide to `accept' when yystate = |
12 |
|
final_state, they still will not reduce rule 0. |
13 |
|
* src/print.c (print_actions, print_reduction): Adjust. |
14 |
|
* src/output.c (action_row): Track reduced rules. |
15 |
|
(token_actions): Report rules never reduced. |
16 |
|
* tests/conflicts.at, tests/regression.at: Adjust. |
17 |
|
|
18 |
|
2002-07-30 Akim Demaille <akim@epita.fr> |
19 |
|
|
20 |
`stage' was accidently included in a previous patch. |
`stage' was accidently included in a previous patch. |
21 |
Initiate its autoconfiscation. |
Initiate its autoconfiscation. |
22 |
|
|