Commit | Line | Data |
---|---|---|
7215de24 AD |
1 | 2001-12-10 Akim Demaille <akim@epita.fr> |
2 | ||
3 | * src/LR0.h (first_shift, first_reduction): Remove. | |
4 | * src/lalr.c: Don't use first_shift: find shifts through the | |
5 | states. | |
6 | ||
7 | ||
80e25d4d AD |
8 | 2001-12-10 Akim Demaille <akim@epita.fr> |
9 | ||
10 | * src/LR0.c: Attach shifts to states as soon as they are | |
11 | computed. | |
12 | * src/lalr.c (set_state_table): Instead of assigning shifts to | |
13 | state, just assert that the mapping was properly done. | |
14 | ||
15 | ||
0ab3728b AD |
16 | 2001-12-10 Akim Demaille <akim@epita.fr> |
17 | ||
18 | * src/LR0.c (insert_start_shift): Rename as... | |
19 | (insert_start_shifting_state): this. | |
20 | (insert_eof_shifting_state, insert_accepting_state): New. | |
21 | (augment_automaton): Adjust. | |
22 | Better locality of the variables. | |
23 | When looking if the start_symbol is shifted from the initial | |
24 | state, using `while (... symbol != start_symbol ...)' sounds | |
25 | better than `while (... symbol < start_symbol ...)': If fail | |
26 | to see how the order between symbols could be relevant! | |
27 | ||
78af9bbc AD |
28 | 2001-12-10 Akim Demaille <akim@epita.fr> |
29 | ||
30 | * src/getargs.h: Don't declare `spec_name_prefix' and | |
31 | `spec_file_prefix', declared by src/files.h. | |
32 | * src/files.c, src/files.h: Default for spec_name_prefix is "yy". | |
33 | * src/muscle_tab.c (muscle_init): Default prefix to NULL. | |
34 | * src/output.c (prepare): Adjust. | |
35 | * src/reader.c (symbols_output): Likewise. | |
36 | * src/vmsgetargs.c: Vaguely adjust, but who cares? | |
37 | ||
38 | ||
bdef2a41 AD |
39 | 2001-12-10 Akim Demaille <akim@epita.fr> |
40 | ||
41 | * src/muscle_tab.c (muscle_init): NULL is a better default than | |
42 | `"0"'. | |
43 | ||
44 | ||
3735969c AD |
45 | 2001-12-10 Akim Demaille <akim@epita.fr> |
46 | ||
47 | * src/reader.c (reader): Calling symbols_output once is enough. | |
48 | ||
49 | ||
49701457 AD |
50 | 2001-12-10 Akim Demaille <akim@epita.fr> |
51 | ||
52 | Now that states have a complete set of members, the linked list of | |
53 | reductions is useless: just fill directly the state's reductions | |
54 | member. | |
55 | ||
56 | * src/state.h (struct reductions): Remove member `number' and | |
57 | `next'. | |
58 | * src/LR0.c (first_reduction, last_reduction): Remove. | |
59 | (save_reductions): Don't link the new reductions, store them in | |
60 | this_state. | |
61 | * src/lalr.c (set_state_table): No need to attach reductions to | |
62 | states, it's already done. | |
63 | * src/output.c (output_actions): No longer free the shifts, then | |
64 | the reductions, then the states: free all the states and their | |
65 | members. | |
66 | ||
67 | ||
0edad749 AD |
68 | 2001-12-10 Akim Demaille <akim@epita.fr> |
69 | ||
70 | * src/options.c (OPTN, DRTV, BOTH): New. | |
71 | (option_table): Use them. | |
72 | ||
73 | ||
74 | * src/muscle_tab.c: Don't include xalloc.h and string.h: that's | |
75 | the job of system.h. | |
76 | * src/options.c: Don't include stdio.h and xalloc.h for the same | |
77 | reasons. | |
78 | ||
79 | ||
5449dd0f AD |
80 | 2001-12-10 Akim Demaille <akim@epita.fr> |
81 | ||
82 | * src/output.c (output, prepare): Make sure the values of the | |
83 | muscles `action' and `prologue' are 0-terminated. | |
84 | ||
a870c567 AD |
85 | 2001-12-10 Akim Demaille <akim@epita.fr> |
86 | ||
87 | Clean up GCC warnings. | |
88 | ||
89 | * src/reader.c (copy_action): `buf' is not used. | |
90 | (parse_skel_decl): Be static. | |
91 | * src/muscle_tab.c (mhash1, mhash2, muscle_insert): Preserve `const'. | |
92 | * src/options.h (create_long_option_table): Have a real prototype. | |
93 | * lib/hash.c, lib/hash.h (hash_insert, hash_insert_at, hash_delete) | |
94 | (hash_delete_at): Return const void *. | |
95 | Adjust casts to preserve the const. | |
96 | ||
80df8768 AD |
97 | 2001-12-10 Akim Demaille <akim@epita.fr> |
98 | ||
99 | * configure.in: Require 2.52g. | |
100 | M4 is not needed, but AUTOM4TE is. | |
101 | * m4/m4.m4: Remove. | |
102 | * tests/Makefile.am: Adjust. | |
103 | ||
104 | ||
f693ad14 AD |
105 | 2001-12-10 Akim Demaille <akim@epita.fr> |
106 | ||
107 | One structure for states is enough, even though theoretically | |
108 | there are LR(0) states and LALR(1) states. | |
109 | ||
110 | * src/lalr.h (state_t): Remove. | |
111 | (state_table): Be state_t **, not state_t *. | |
112 | * src/state.h (core, CORE_ALLOC): Rename as... | |
113 | (state_t, STATE_ALLOC): this. | |
114 | Add the LALR(1) members: shifts, reductions, errs. | |
115 | * src/LR0.c (state_table): Rename as... | |
116 | (state_hash): this, to avoid name clashes with the global | |
117 | `state_table'. | |
118 | * src/print_graph.c, src/LR0.c, src/LR0.h, src/conflicts.c | |
119 | * src/lalr.c, src/lalr.h, src/output.c, src/print.c: Adjust. | |
120 | ||
121 | ||
74ffbcb6 AD |
122 | 2001-12-10 Akim Demaille <akim@epita.fr> |
123 | ||
124 | Bison dumps core on bash.y. | |
125 | Reported by Pascal Bart. | |
126 | ||
127 | * src/warshall.c (bitmatrix_print): New. | |
128 | (TC): Use it. | |
129 | When performing a transitive closure R(i, j) && R(j, k) => R(i, k), | |
130 | j must be the outer loop. | |
131 | * tests/regression.at (Broken Closure): New. | |
132 | ||
07708e19 AD |
133 | 2001-12-05 Akim Demaille <akim@epita.fr> |
134 | ||
135 | * tests/atlocal.in (CPPFLAGS): Do not leave a space between -I and | |
136 | its argument. | |
137 |