blob: ebce796ba52beb40a6a2af7a76ac72eda51a5bb0
1 | /* |
2 | * Copyright (c) 2016 Umair Khan <omerjerk@gmail.com> |
3 | * |
4 | * This file is part of FFmpeg. |
5 | * |
6 | * FFmpeg is free software; you can redistribute it and/or |
7 | * modify it under the terms of the GNU Lesser General Public |
8 | * License as published by the Free Software Foundation; either |
9 | * version 2.1 of the License, or (at your option) any later version. |
10 | * |
11 | * FFmpeg is distributed in the hope that it will be useful, |
12 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
13 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
14 | * Lesser General Public License for more details. |
15 | * |
16 | * You should have received a copy of the GNU Lesser General Public |
17 | * License along with FFmpeg; if not, write to the Free Software |
18 | * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA |
19 | */ |
20 | |
21 | #include "mlz.h" |
22 | |
23 | av_cold void ff_mlz_init_dict(void* context, MLZ *mlz) { |
24 | mlz->dict = av_mallocz_array(TABLE_SIZE, sizeof(*mlz->dict)); |
25 | |
26 | mlz->flush_code = FLUSH_CODE; |
27 | mlz->current_dic_index_max = DIC_INDEX_INIT; |
28 | mlz->dic_code_bit = CODE_BIT_INIT; |
29 | mlz->bump_code = (DIC_INDEX_INIT - 1); |
30 | mlz->next_code = FIRST_CODE; |
31 | mlz->freeze_flag = 0; |
32 | mlz->context = context; |
33 | } |
34 | |
35 | av_cold void ff_mlz_flush_dict(MLZ *mlz) { |
36 | MLZDict *dict = mlz->dict; |
37 | int i; |
38 | for ( i = 0; i < TABLE_SIZE; i++ ) { |
39 | dict[i].string_code = CODE_UNSET; |
40 | dict[i].parent_code = CODE_UNSET; |
41 | dict[i].match_len = 0; |
42 | } |
43 | mlz->current_dic_index_max = DIC_INDEX_INIT; |
44 | mlz->dic_code_bit = CODE_BIT_INIT; // DicCodeBitInit; |
45 | mlz->bump_code = mlz->current_dic_index_max - 1; |
46 | mlz->next_code = FIRST_CODE; |
47 | mlz->freeze_flag = 0; |
48 | } |
49 | |
50 | static void set_new_entry_dict(MLZDict* dict, int string_code, int parent_code, int char_code) { |
51 | dict[string_code].parent_code = parent_code; |
52 | dict[string_code].string_code = string_code; |
53 | dict[string_code].char_code = char_code; |
54 | if (parent_code < FIRST_CODE) { |
55 | dict[string_code].match_len = 2; |
56 | } else { |
57 | dict[string_code].match_len = (dict[parent_code].match_len) + 1; |
58 | } |
59 | } |
60 | |
61 | static int decode_string(MLZ* mlz, unsigned char *buff, int string_code, int *first_char_code, unsigned long bufsize) { |
62 | MLZDict* dict = mlz->dict; |
63 | unsigned long count, offset; |
64 | int current_code, parent_code, tmp_code; |
65 | |
66 | count = 0; |
67 | current_code = string_code; |
68 | *first_char_code = CODE_UNSET; |
69 | |
70 | while (count < bufsize) { |
71 | switch (current_code) { |
72 | case CODE_UNSET: |
73 | return count; |
74 | break; |
75 | default: |
76 | if (current_code < FIRST_CODE) { |
77 | *first_char_code = current_code; |
78 | buff[0] = current_code; |
79 | count++; |
80 | return count; |
81 | } else { |
82 | offset = dict[current_code].match_len - 1; |
83 | tmp_code = dict[current_code].char_code; |
84 | if (offset >= bufsize) { |
85 | av_log(mlz->context, AV_LOG_ERROR, "MLZ offset error.\n"); |
86 | return count; |
87 | } |
88 | buff[offset] = tmp_code; |
89 | count++; |
90 | } |
91 | current_code = dict[current_code].parent_code; |
92 | if ((current_code < 0) || (current_code > (DIC_INDEX_MAX - 1))) { |
93 | av_log(mlz->context, AV_LOG_ERROR, "MLZ dic index error.\n"); |
94 | return count; |
95 | } |
96 | if (current_code > FIRST_CODE) { |
97 | parent_code = dict[current_code].parent_code; |
98 | offset = (dict[current_code].match_len) - 1; |
99 | if (parent_code < 0 || parent_code > DIC_INDEX_MAX-1) { |
100 | av_log(mlz->context, AV_LOG_ERROR, "MLZ dic index error.\n"); |
101 | return count; |
102 | } |
103 | if (( offset > (DIC_INDEX_MAX - 1))) { |
104 | av_log(mlz->context, AV_LOG_ERROR, "MLZ dic offset error.\n"); |
105 | return count; |
106 | } |
107 | } |
108 | break; |
109 | } |
110 | } |
111 | return count; |
112 | } |
113 | |
114 | static int input_code(GetBitContext* gb, int len) { |
115 | int tmp_code = 0; |
116 | int i; |
117 | for (i = 0; i < len; ++i) { |
118 | tmp_code |= get_bits1(gb) << i; |
119 | } |
120 | return tmp_code; |
121 | } |
122 | |
123 | int ff_mlz_decompression(MLZ* mlz, GetBitContext* gb, int size, unsigned char *buff) { |
124 | MLZDict *dict = mlz->dict; |
125 | unsigned long output_chars; |
126 | int string_code, last_string_code, char_code; |
127 | |
128 | string_code = 0; |
129 | char_code = -1; |
130 | last_string_code = -1; |
131 | output_chars = 0; |
132 | |
133 | while (output_chars < size) { |
134 | string_code = input_code(gb, mlz->dic_code_bit); |
135 | switch (string_code) { |
136 | case FLUSH_CODE: |
137 | case MAX_CODE: |
138 | ff_mlz_flush_dict(mlz); |
139 | char_code = -1; |
140 | last_string_code = -1; |
141 | break; |
142 | case FREEZE_CODE: |
143 | mlz->freeze_flag = 1; |
144 | break; |
145 | default: |
146 | if (string_code > mlz->current_dic_index_max) { |
147 | av_log(mlz->context, AV_LOG_ERROR, "String code %d exceeds maximum value of %d.\n", string_code, mlz->current_dic_index_max); |
148 | return output_chars; |
149 | } |
150 | if (string_code == (int) mlz->bump_code) { |
151 | ++mlz->dic_code_bit; |
152 | mlz->current_dic_index_max *= 2; |
153 | mlz->bump_code = mlz->current_dic_index_max - 1; |
154 | } else { |
155 | if (string_code >= mlz->next_code) { |
156 | int ret = decode_string(mlz, &buff[output_chars], last_string_code, &char_code, size - output_chars); |
157 | if (ret < 0 || ret > size - output_chars) { |
158 | av_log(mlz->context, AV_LOG_ERROR, "output chars overflow\n"); |
159 | return output_chars; |
160 | } |
161 | output_chars += ret; |
162 | ret = decode_string(mlz, &buff[output_chars], char_code, &char_code, size - output_chars); |
163 | if (ret < 0 || ret > size - output_chars) { |
164 | av_log(mlz->context, AV_LOG_ERROR, "output chars overflow\n"); |
165 | return output_chars; |
166 | } |
167 | output_chars += ret; |
168 | set_new_entry_dict(dict, mlz->next_code, last_string_code, char_code); |
169 | if (mlz->next_code >= TABLE_SIZE - 1) { |
170 | av_log(mlz->context, AV_LOG_ERROR, "Too many MLZ codes\n"); |
171 | return output_chars; |
172 | } |
173 | mlz->next_code++; |
174 | } else { |
175 | int ret = decode_string(mlz, &buff[output_chars], string_code, &char_code, size - output_chars); |
176 | if (ret < 0 || ret > size - output_chars) { |
177 | av_log(mlz->context, AV_LOG_ERROR, "output chars overflow\n"); |
178 | return output_chars; |
179 | } |
180 | output_chars += ret; |
181 | if (output_chars <= size && !mlz->freeze_flag) { |
182 | if (last_string_code != -1) { |
183 | set_new_entry_dict(dict, mlz->next_code, last_string_code, char_code); |
184 | if (mlz->next_code >= TABLE_SIZE - 1) { |
185 | av_log(mlz->context, AV_LOG_ERROR, "Too many MLZ codes\n"); |
186 | return output_chars; |
187 | } |
188 | mlz->next_code++; |
189 | } |
190 | } else { |
191 | break; |
192 | } |
193 | } |
194 | last_string_code = string_code; |
195 | } |
196 | break; |
197 | } |
198 | } |
199 | return output_chars; |
200 | } |
201 |