| 1 | /* |
| 2 | * Copyright (c) 2012 Clément Bœsch |
| 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 | #ifndef AVFORMAT_SUBTITLES_H |
| 22 | #define AVFORMAT_SUBTITLES_H |
| 23 | |
| 24 | #include <stdint.h> |
| 25 | #include <stddef.h> |
| 26 | #include "avformat.h" |
| 27 | #include "libavutil/bprint.h" |
| 28 | |
| 29 | enum sub_sort { |
| 30 | SUB_SORT_TS_POS = 0, ///< sort by timestamps, then position |
| 31 | SUB_SORT_POS_TS, ///< sort by position, then timestamps |
| 32 | }; |
| 33 | |
| 34 | enum ff_utf_type { |
| 35 | FF_UTF_8, // or other 8 bit encodings |
| 36 | FF_UTF16LE, |
| 37 | FF_UTF16BE, |
| 38 | }; |
| 39 | |
| 40 | typedef struct { |
| 41 | int type; |
| 42 | AVIOContext *pb; |
| 43 | unsigned char buf[8]; |
| 44 | int buf_pos, buf_len; |
| 45 | AVIOContext buf_pb; |
| 46 | } FFTextReader; |
| 47 | |
| 48 | /** |
| 49 | * Initialize the FFTextReader from the given AVIOContext. This function will |
| 50 | * read some bytes from pb, and test for UTF-8 or UTF-16 BOMs. Further accesses |
| 51 | * to FFTextReader will read more data from pb. |
| 52 | * If s is not NULL, the user will be warned if a UTF-16 conversion takes place. |
| 53 | * |
| 54 | * The purpose of FFTextReader is to transparently convert read data to UTF-8 |
| 55 | * if the stream had a UTF-16 BOM. |
| 56 | * |
| 57 | * @param s Pointer to provide av_log context |
| 58 | * @param r object which will be initialized |
| 59 | * @param pb stream to read from (referenced as long as FFTextReader is in use) |
| 60 | */ |
| 61 | void ff_text_init_avio(void *s, FFTextReader *r, AVIOContext *pb); |
| 62 | |
| 63 | /** |
| 64 | * Similar to ff_text_init_avio(), but sets it up to read from a bounded buffer. |
| 65 | * |
| 66 | * @param r object which will be initialized |
| 67 | * @param buf buffer to read from (referenced as long as FFTextReader is in use) |
| 68 | * @param size size of buf |
| 69 | */ |
| 70 | void ff_text_init_buf(FFTextReader *r, void *buf, size_t size); |
| 71 | |
| 72 | /** |
| 73 | * Return the byte position of the next byte returned by ff_text_r8(). For |
| 74 | * UTF-16 source streams, this will return the original position, but it will |
| 75 | * be incorrect if a codepoint was only partially read with ff_text_r8(). |
| 76 | */ |
| 77 | int64_t ff_text_pos(FFTextReader *r); |
| 78 | |
| 79 | /** |
| 80 | * Return the next byte. The return value is always 0 - 255. Returns 0 on EOF. |
| 81 | * If the source stream is UTF-16, this reads from the stream converted to |
| 82 | * UTF-8. On invalid UTF-16, 0 is returned. |
| 83 | */ |
| 84 | int ff_text_r8(FFTextReader *r); |
| 85 | |
| 86 | /** |
| 87 | * Return non-zero if EOF was reached. |
| 88 | */ |
| 89 | int ff_text_eof(FFTextReader *r); |
| 90 | |
| 91 | /** |
| 92 | * Like ff_text_r8(), but don't remove the byte from the buffer. |
| 93 | */ |
| 94 | int ff_text_peek_r8(FFTextReader *r); |
| 95 | |
| 96 | /** |
| 97 | * Read the given number of bytes (in UTF-8). On error or EOF, \0 bytes are |
| 98 | * written. |
| 99 | */ |
| 100 | void ff_text_read(FFTextReader *r, char *buf, size_t size); |
| 101 | |
| 102 | typedef struct { |
| 103 | AVPacket *subs; ///< array of subtitles packets |
| 104 | int nb_subs; ///< number of subtitles packets |
| 105 | int allocated_size; ///< allocated size for subs |
| 106 | int current_sub_idx; ///< current position for the read packet callback |
| 107 | enum sub_sort sort; ///< sort method to use when finalizing subtitles |
| 108 | } FFDemuxSubtitlesQueue; |
| 109 | |
| 110 | /** |
| 111 | * Insert a new subtitle event. |
| 112 | * |
| 113 | * @param event the subtitle line, may not be zero terminated |
| 114 | * @param len the length of the event (in strlen() sense, so without '\0') |
| 115 | * @param merge set to 1 if the current event should be concatenated with the |
| 116 | * previous one instead of adding a new entry, 0 otherwise |
| 117 | */ |
| 118 | AVPacket *ff_subtitles_queue_insert(FFDemuxSubtitlesQueue *q, |
| 119 | const uint8_t *event, int len, int merge); |
| 120 | |
| 121 | /** |
| 122 | * Set missing durations and sort subtitles by PTS, and then byte position. |
| 123 | */ |
| 124 | void ff_subtitles_queue_finalize(FFDemuxSubtitlesQueue *q); |
| 125 | |
| 126 | /** |
| 127 | * Generic read_packet() callback for subtitles demuxers using this queue |
| 128 | * system. |
| 129 | */ |
| 130 | int ff_subtitles_queue_read_packet(FFDemuxSubtitlesQueue *q, AVPacket *pkt); |
| 131 | |
| 132 | /** |
| 133 | * Update current_sub_idx to emulate a seek. Except the first parameter, it |
| 134 | * matches AVInputFormat->read_seek2 prototypes. |
| 135 | */ |
| 136 | int ff_subtitles_queue_seek(FFDemuxSubtitlesQueue *q, AVFormatContext *s, int stream_index, |
| 137 | int64_t min_ts, int64_t ts, int64_t max_ts, int flags); |
| 138 | |
| 139 | /** |
| 140 | * Remove and destroy all the subtitles packets. |
| 141 | */ |
| 142 | void ff_subtitles_queue_clean(FFDemuxSubtitlesQueue *q); |
| 143 | |
| 144 | /** |
| 145 | * SMIL helper to load next chunk ("<...>" or untagged content) in buf. |
| 146 | * |
| 147 | * @param c cached character, to avoid a backward seek |
| 148 | */ |
| 149 | int ff_smil_extract_next_text_chunk(FFTextReader *tr, AVBPrint *buf, char *c); |
| 150 | |
| 151 | /** |
| 152 | * SMIL helper to point on the value of an attribute in the given tag. |
| 153 | * |
| 154 | * @param s SMIL tag ("<...>") |
| 155 | * @param attr the attribute to look for |
| 156 | */ |
| 157 | const char *ff_smil_get_attr_ptr(const char *s, const char *attr); |
| 158 | |
| 159 | /** |
| 160 | * @brief Same as ff_subtitles_read_text_chunk(), but read from an AVIOContext. |
| 161 | */ |
| 162 | void ff_subtitles_read_chunk(AVIOContext *pb, AVBPrint *buf); |
| 163 | |
| 164 | /** |
| 165 | * @brief Read a subtitles chunk from FFTextReader. |
| 166 | * |
| 167 | * A chunk is defined by a multiline "event", ending with a second line break. |
| 168 | * The trailing line breaks are trimmed. CRLF are supported. |
| 169 | * Example: "foo\r\nbar\r\n\r\nnext" will print "foo\r\nbar" into buf, and pb |
| 170 | * will focus on the 'n' of the "next" string. |
| 171 | * |
| 172 | * @param tr I/O context |
| 173 | * @param buf an initialized buf where the chunk is written |
| 174 | * |
| 175 | * @note buf is cleared before writing into it. |
| 176 | */ |
| 177 | void ff_subtitles_read_text_chunk(FFTextReader *tr, AVBPrint *buf); |
| 178 | |
| 179 | /** |
| 180 | * Get the number of characters to increment to jump to the next line, or to |
| 181 | * the end of the string. |
| 182 | * The function handles the following line breaks schemes: |
| 183 | * LF, CRLF (MS), or standalone CR (old MacOS). |
| 184 | */ |
| 185 | static av_always_inline int ff_subtitles_next_line(const char *ptr) |
| 186 | { |
| 187 | int n = strcspn(ptr, "\r\n"); |
| 188 | ptr += n; |
| 189 | if (*ptr == '\r') { |
| 190 | ptr++; |
| 191 | n++; |
| 192 | } |
| 193 | if (*ptr == '\n') |
| 194 | n++; |
| 195 | return n; |
| 196 | } |
| 197 | |
| 198 | /** |
| 199 | * Read a line of text. Discards line ending characters. |
| 200 | * The function handles the following line breaks schemes: |
| 201 | * LF, CRLF (MS), or standalone CR (old MacOS). |
| 202 | * |
| 203 | * Returns the number of bytes written to buf. Always writes a terminating 0, |
| 204 | * similar as with snprintf. |
| 205 | * |
| 206 | * @note returns a negative error code if a \0 byte is found |
| 207 | */ |
| 208 | ptrdiff_t ff_subtitles_read_line(FFTextReader *tr, char *buf, size_t size); |
| 209 | |
| 210 | #endif /* AVFORMAT_SUBTITLES_H */ |