1/*
2 * SubRip subtitle decoder
3 * Copyright (c) 2010  Aurelien Jacobs <aurel@gnuage.org>
4 *
5 * This file is part of Libav.
6 *
7 * Libav is free software; you can redistribute it and/or
8 * modify it under the terms of the GNU Lesser General Public
9 * License as published by the Free Software Foundation; either
10 * version 2.1 of the License, or (at your option) any later version.
11 *
12 * Libav is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
15 * Lesser General Public License for more details.
16 *
17 * You should have received a copy of the GNU Lesser General Public
18 * License along with Libav; if not, write to the Free Software
19 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
20 */
21
22#include "libavutil/avstring.h"
23#include "libavutil/parseutils.h"
24#include "avcodec.h"
25#include "ass.h"
26
27static int html_color_parse(AVCodecContext *avctx, const char *str)
28{
29    uint8_t rgba[4];
30    if (av_parse_color(rgba, str, strcspn(str, "\" >"), avctx) < 0)
31        return -1;
32    return rgba[0] | rgba[1] << 8 | rgba[2] << 16;
33}
34
35enum {
36    PARAM_UNKNOWN = -1,
37    PARAM_SIZE,
38    PARAM_COLOR,
39    PARAM_FACE,
40    PARAM_NUMBER
41};
42
43typedef struct {
44    char tag[128];
45    char param[PARAM_NUMBER][128];
46} SrtStack;
47
48static const char *srt_to_ass(AVCodecContext *avctx, char *out, char *out_end,
49                              const char *in, int x1, int y1, int x2, int y2)
50{
51    char c, *param, buffer[128], tmp[128];
52    int len, tag_close, sptr = 1, line_start = 1, an = 0, end = 0;
53    SrtStack stack[16];
54
55    stack[0].tag[0] = 0;
56    strcpy(stack[0].param[PARAM_SIZE],  "{\\fs}");
57    strcpy(stack[0].param[PARAM_COLOR], "{\\c}");
58    strcpy(stack[0].param[PARAM_FACE],  "{\\fn}");
59
60    if (x1 >= 0 && y1 >= 0) {
61        if (x2 >= 0 && y2 >= 0 && (x2 != x1 || y2 != y1))
62            out += snprintf(out, out_end-out,
63                            "{\\an1}{\\move(%d,%d,%d,%d)}", x1, y1, x2, y2);
64        else
65            out += snprintf(out, out_end-out, "{\\an1}{\\pos(%d,%d)}", x1, y1);
66    }
67
68    for (; out < out_end && !end && *in; in++) {
69        switch (*in) {
70        case '\r':
71            break;
72        case '\n':
73            if (line_start) {
74                end = 1;
75                break;
76            }
77            while (out[-1] == ' ')
78                out--;
79            out += snprintf(out, out_end-out, "\\N");
80            line_start = 1;
81            break;
82        case ' ':
83            if (!line_start)
84                *out++ = *in;
85            break;
86        case '{':    /* skip all {\xxx} substrings except for {\an%d}
87                        and all microdvd like styles such as {Y:xxx} */
88            an += sscanf(in, "{\\an%*1u}%c", &c) == 1;
89            if ((an != 1 && sscanf(in, "{\\%*[^}]}%n%c", &len, &c) > 0) ||
90                sscanf(in, "{%*1[CcFfoPSsYy]:%*[^}]}%n%c", &len, &c) > 0) {
91                in += len - 1;
92            } else
93                *out++ = *in;
94            break;
95        case '<':
96            tag_close = in[1] == '/';
97            if (sscanf(in+tag_close+1, "%127[^>]>%n%c", buffer, &len,&c) >= 2) {
98                if ((param = strchr(buffer, ' ')))
99                    *param++ = 0;
100                if ((!tag_close && sptr < FF_ARRAY_ELEMS(stack)) ||
101                    ( tag_close && sptr > 0 && !strcmp(stack[sptr-1].tag, buffer))) {
102                    int i, j, unknown = 0;
103                    in += len + tag_close;
104                    if (!tag_close)
105                        memset(stack+sptr, 0, sizeof(*stack));
106                    if (!strcmp(buffer, "font")) {
107                        if (tag_close) {
108                            for (i=PARAM_NUMBER-1; i>=0; i--)
109                                if (stack[sptr-1].param[i][0])
110                                    for (j=sptr-2; j>=0; j--)
111                                        if (stack[j].param[i][0]) {
112                                            out += snprintf(out, out_end-out,
113                                                            "%s", stack[j].param[i]);
114                                            break;
115                                        }
116                        } else {
117                            while (param) {
118                                if (!strncmp(param, "size=", 5)) {
119                                    unsigned font_size;
120                                    param += 5 + (param[5] == '"');
121                                    if (sscanf(param, "%u", &font_size) == 1) {
122                                        snprintf(stack[sptr].param[PARAM_SIZE],
123                                             sizeof(stack[0].param[PARAM_SIZE]),
124                                             "{\\fs%u}", font_size);
125                                    }
126                                } else if (!strncmp(param, "color=", 6)) {
127                                    param += 6 + (param[6] == '"');
128                                    snprintf(stack[sptr].param[PARAM_COLOR],
129                                         sizeof(stack[0].param[PARAM_COLOR]),
130                                         "{\\c&H%X&}",
131                                         html_color_parse(avctx, param));
132                                } else if (!strncmp(param, "face=", 5)) {
133                                    param += 5 + (param[5] == '"');
134                                    len = strcspn(param,
135                                                  param[-1] == '"' ? "\"" :" ");
136                                    av_strlcpy(tmp, param,
137                                               FFMIN(sizeof(tmp), len+1));
138                                    param += len;
139                                    snprintf(stack[sptr].param[PARAM_FACE],
140                                             sizeof(stack[0].param[PARAM_FACE]),
141                                             "{\\fn%s}", tmp);
142                                }
143                                if ((param = strchr(param, ' ')))
144                                    param++;
145                            }
146                            for (i=0; i<PARAM_NUMBER; i++)
147                                if (stack[sptr].param[i][0])
148                                    out += snprintf(out, out_end-out,
149                                                    "%s", stack[sptr].param[i]);
150                        }
151                    } else if (!buffer[1] && strspn(buffer, "bisu") == 1) {
152                        out += snprintf(out, out_end-out,
153                                        "{\\%c%d}", buffer[0], !tag_close);
154                    } else {
155                        unknown = 1;
156                        snprintf(tmp, sizeof(tmp), "</%s>", buffer);
157                    }
158                    if (tag_close) {
159                        sptr--;
160                    } else if (unknown && !strstr(in, tmp)) {
161                        in -= len + tag_close;
162                        *out++ = *in;
163                    } else
164                        av_strlcpy(stack[sptr++].tag, buffer,
165                                   sizeof(stack[0].tag));
166                    break;
167                }
168            }
169        default:
170            *out++ = *in;
171            break;
172        }
173        if (*in != ' ' && *in != '\r' && *in != '\n')
174            line_start = 0;
175    }
176
177    out = FFMIN(out, out_end-3);
178    while (!strncmp(out-2, "\\N", 2))
179        out -= 2;
180    while (out[-1] == ' ')
181        out--;
182    out += snprintf(out, out_end-out, "\r\n");
183    return in;
184}
185
186static const char *read_ts(const char *buf, int *ts_start, int *ts_end,
187                           int *x1, int *y1, int *x2, int *y2)
188{
189    int i, hs, ms, ss, he, me, se;
190
191    for (i=0; i<2; i++) {
192        /* try to read timestamps in either the first or second line */
193        int c = sscanf(buf, "%d:%2d:%2d%*1[,.]%3d --> %d:%2d:%2d%*1[,.]%3d"
194                       "%*[ ]X1:%u X2:%u Y1:%u Y2:%u",
195                       &hs, &ms, &ss, ts_start, &he, &me, &se, ts_end,
196                       x1, x2, y1, y2);
197        buf += strcspn(buf, "\n") + 1;
198        if (c >= 8) {
199            *ts_start = 100*(ss + 60*(ms + 60*hs)) + *ts_start/10;
200            *ts_end   = 100*(se + 60*(me + 60*he)) + *ts_end  /10;
201            return buf;
202        }
203    }
204    return NULL;
205}
206
207static int srt_decode_frame(AVCodecContext *avctx,
208                            void *data, int *got_sub_ptr, AVPacket *avpkt)
209{
210    AVSubtitle *sub = data;
211    int ts_start, ts_end, x1 = -1, y1 = -1, x2 = -1, y2 = -1;
212    char buffer[2048];
213    const char *ptr = avpkt->data;
214    const char *end = avpkt->data + avpkt->size;
215
216    if (avpkt->size <= 0)
217        return avpkt->size;
218
219    ff_ass_init(sub);
220
221    while (ptr < end && *ptr) {
222        ptr = read_ts(ptr, &ts_start, &ts_end, &x1, &y1, &x2, &y2);
223        if (!ptr)
224            break;
225        ptr = srt_to_ass(avctx, buffer, buffer+sizeof(buffer), ptr,
226                         x1, y1, x2, y2);
227        ff_ass_add_rect(sub, buffer, ts_start, ts_end, 0);
228    }
229
230    *got_sub_ptr = sub->num_rects > 0;
231    return avpkt->size;
232}
233
234AVCodec ff_srt_decoder = {
235    .name         = "srt",
236    .long_name    = NULL_IF_CONFIG_SMALL("SubRip subtitle"),
237    .type         = AVMEDIA_TYPE_SUBTITLE,
238    .id           = CODEC_ID_SRT,
239    .init         = ff_ass_subtitle_header_default,
240    .decode       = srt_decode_frame,
241};
242