]> git.saurik.com Git - bison.git/blame - ChangeLog
* src/LR0.c (augment_automaton): Better variable locality.
[bison.git] / ChangeLog
CommitLineData
2a73b93d
AD
12001-12-10 Akim Demaille <akim@epita.fr>
2
3 * src/LR0.c (augment_automaton): Better variable locality.
4 Remove an impossible branch: if there is a state corresponding to
5 the start symbol being shifted, then there is shift for the start
6 symbol from the initial state.
7
8
74392f6a
AD
92001-12-10 Akim Demaille <akim@epita.fr>
10
11 * src/LR0.c (augment_automaton): Call `insert_eof_shifting_state'
12 only when appropriate: when insert_start_shifting_state' is not
13 invoked.
14 * tests/regression.at (Rule Line Numbers): Adjust.
15
16
37c82725
AD
172001-12-10 Akim Demaille <akim@epita.fr>
18
19 * src/LR0.c (augment_automaton): Now that all states have shifts,
20 merge the two cases addition shifts to the initial state.
21
22
6a164e0c
AD
232001-12-10 Akim Demaille <akim@epita.fr>
24
25 * src/lalr.c (set_state_table): Move to...
26 * src/LR0.c: here.
27 * src/lalr.c (lalr): Don't call it...
28 * src/LR0.c (generate_states): do it.
29 * src/LR0.h (first_state): Remove, only the table is used.
30
31
7215de24
AD
322001-12-10 Akim Demaille <akim@epita.fr>
33
34 * src/LR0.h (first_shift, first_reduction): Remove.
35 * src/lalr.c: Don't use first_shift: find shifts through the
36 states.
37
38
80e25d4d
AD
392001-12-10 Akim Demaille <akim@epita.fr>
40
41 * src/LR0.c: Attach shifts to states as soon as they are
42 computed.
43 * src/lalr.c (set_state_table): Instead of assigning shifts to
44 state, just assert that the mapping was properly done.
45
46
0ab3728b
AD
472001-12-10 Akim Demaille <akim@epita.fr>
48
49 * src/LR0.c (insert_start_shift): Rename as...
50 (insert_start_shifting_state): this.
51 (insert_eof_shifting_state, insert_accepting_state): New.
52 (augment_automaton): Adjust.
53 Better locality of the variables.
54 When looking if the start_symbol is shifted from the initial
55 state, using `while (... symbol != start_symbol ...)' sounds
56 better than `while (... symbol < start_symbol ...)': If fail
57 to see how the order between symbols could be relevant!
58
78af9bbc
AD
592001-12-10 Akim Demaille <akim@epita.fr>
60
61 * src/getargs.h: Don't declare `spec_name_prefix' and
62 `spec_file_prefix', declared by src/files.h.
63 * src/files.c, src/files.h: Default for spec_name_prefix is "yy".
64 * src/muscle_tab.c (muscle_init): Default prefix to NULL.
65 * src/output.c (prepare): Adjust.
66 * src/reader.c (symbols_output): Likewise.
67 * src/vmsgetargs.c: Vaguely adjust, but who cares?
68
69
bdef2a41
AD
702001-12-10 Akim Demaille <akim@epita.fr>
71
72 * src/muscle_tab.c (muscle_init): NULL is a better default than
73 `"0"'.
74
75
3735969c
AD
762001-12-10 Akim Demaille <akim@epita.fr>
77
78 * src/reader.c (reader): Calling symbols_output once is enough.
79
80
49701457
AD
812001-12-10 Akim Demaille <akim@epita.fr>
82
83 Now that states have a complete set of members, the linked list of
84 reductions is useless: just fill directly the state's reductions
85 member.
86
87 * src/state.h (struct reductions): Remove member `number' and
88 `next'.
89 * src/LR0.c (first_reduction, last_reduction): Remove.
90 (save_reductions): Don't link the new reductions, store them in
91 this_state.
92 * src/lalr.c (set_state_table): No need to attach reductions to
93 states, it's already done.
94 * src/output.c (output_actions): No longer free the shifts, then
95 the reductions, then the states: free all the states and their
96 members.
97
98
0edad749
AD
992001-12-10 Akim Demaille <akim@epita.fr>
100
101 * src/options.c (OPTN, DRTV, BOTH): New.
102 (option_table): Use them.
103
104
105 * src/muscle_tab.c: Don't include xalloc.h and string.h: that's
106 the job of system.h.
107 * src/options.c: Don't include stdio.h and xalloc.h for the same
108 reasons.
109
110
5449dd0f
AD
1112001-12-10 Akim Demaille <akim@epita.fr>
112
113 * src/output.c (output, prepare): Make sure the values of the
114 muscles `action' and `prologue' are 0-terminated.
115
a870c567
AD
1162001-12-10 Akim Demaille <akim@epita.fr>
117
118 Clean up GCC warnings.
119
120 * src/reader.c (copy_action): `buf' is not used.
121 (parse_skel_decl): Be static.
122 * src/muscle_tab.c (mhash1, mhash2, muscle_insert): Preserve `const'.
123 * src/options.h (create_long_option_table): Have a real prototype.
124 * lib/hash.c, lib/hash.h (hash_insert, hash_insert_at, hash_delete)
125 (hash_delete_at): Return const void *.
126 Adjust casts to preserve the const.
127
80df8768
AD
1282001-12-10 Akim Demaille <akim@epita.fr>
129
130 * configure.in: Require 2.52g.
131 M4 is not needed, but AUTOM4TE is.
132 * m4/m4.m4: Remove.
133 * tests/Makefile.am: Adjust.
134
135
f693ad14
AD
1362001-12-10 Akim Demaille <akim@epita.fr>
137
138 One structure for states is enough, even though theoretically
139 there are LR(0) states and LALR(1) states.
140
141 * src/lalr.h (state_t): Remove.
142 (state_table): Be state_t **, not state_t *.
143 * src/state.h (core, CORE_ALLOC): Rename as...
144 (state_t, STATE_ALLOC): this.
145 Add the LALR(1) members: shifts, reductions, errs.
146 * src/LR0.c (state_table): Rename as...
147 (state_hash): this, to avoid name clashes with the global
148 `state_table'.
149 * src/print_graph.c, src/LR0.c, src/LR0.h, src/conflicts.c
150 * src/lalr.c, src/lalr.h, src/output.c, src/print.c: Adjust.
151
152
74ffbcb6
AD
1532001-12-10 Akim Demaille <akim@epita.fr>
154
155 Bison dumps core on bash.y.
156 Reported by Pascal Bart.
157
158 * src/warshall.c (bitmatrix_print): New.
159 (TC): Use it.
160 When performing a transitive closure R(i, j) && R(j, k) => R(i, k),
161 j must be the outer loop.
162 * tests/regression.at (Broken Closure): New.
163
07708e19
AD
1642001-12-05 Akim Demaille <akim@epita.fr>
165
166 * tests/atlocal.in (CPPFLAGS): Do not leave a space between -I and
167 its argument.
168