/** * @file vorbisencode.cpp * @brief Vorbis encoding routine routine for Indra. * * $LicenseInfo:firstyear=2000&license=viewergpl$ * * Copyright (c) 2000-2008, Linden Research, Inc. * * Second Life Viewer Source Code * The source code in this file ("Source Code") is provided by Linden Lab * to you under the terms of the GNU General Public License, version 2.0 * ("GPL"), unless you have obtained a separate licensing agreement * ("Other License"), formally executed by you and Linden Lab. Terms of * the GPL can be found in doc/GPL-license.txt in this distribution, or * online at http://secondlifegrid.net/programs/open_source/licensing/gplv2 * * There are special exceptions to the terms and conditions of the GPL as * it is applied to this Source Code. View the full text of the exception * in the file doc/FLOSS-exception.txt in this software distribution, or * online at http://secondlifegrid.net/programs/open_source/licensing/flossexception * * By copying, modifying or distributing this software, you acknowledge * that you have read and understood your obligations described above, * and agree to abide by those obligations. * * ALL LINDEN LAB SOURCE CODE IS PROVIDED "AS IS." LINDEN LAB MAKES NO * WARRANTIES, EXPRESS, IMPLIED OR OTHERWISE, REGARDING ITS ACCURACY, * COMPLETENESS OR PERFORMANCE. * $/LicenseInfo$ */ #include "linden_common.h" #include "vorbisencode.h" #include "vorbis/vorbisenc.h" #include "llerror.h" #include "llrand.h" #include "llmath.h" #include "llapr.h" //#if LL_DARWIN // MBW -- XXX -- Getting rid of SecondLifeVorbis for now -- no fmod means no name collisions. #if 0 #include "VorbisFramework.h" #define vorbis_analysis mac_vorbis_analysis #define vorbis_analysis_headerout mac_vorbis_analysis_headerout #define vorbis_analysis_init mac_vorbis_analysis_init #define vorbis_encode_ctl mac_vorbis_encode_ctl #define vorbis_encode_setup_init mac_vorbis_encode_setup_init #define vorbis_encode_setup_managed mac_vorbis_encode_setup_managed #define vorbis_info_init mac_vorbis_info_init #define vorbis_info_clear mac_vorbis_info_clear #define vorbis_comment_init mac_vorbis_comment_init #define vorbis_comment_clear mac_vorbis_comment_clear #define vorbis_block_init mac_vorbis_block_init #define vorbis_block_clear mac_vorbis_block_clear #define vorbis_dsp_clear mac_vorbis_dsp_clear #define vorbis_analysis_buffer mac_vorbis_analysis_buffer #define vorbis_analysis_wrote mac_vorbis_analysis_wrote #define vorbis_analysis_blockout mac_vorbis_analysis_blockout #define ogg_stream_packetin mac_ogg_stream_packetin #define ogg_stream_init mac_ogg_stream_init #define ogg_stream_flush mac_ogg_stream_flush #define ogg_stream_pageout mac_ogg_stream_pageout #define ogg_page_eos mac_ogg_page_eos #define ogg_stream_clear mac_ogg_stream_clear #endif S32 check_for_invalid_wav_formats(const std::string& in_fname, std::string& error_msg) { U16 num_channels = 0; U32 sample_rate = 0; U32 bits_per_sample = 0; U32 physical_file_size = 0; U32 chunk_length = 0; U32 raw_data_length = 0; U32 bytes_per_sec = 0; BOOL uncompressed_pcm = FALSE; unsigned char wav_header[44]; /*Flawfinder: ignore*/ error_msg.clear(); apr_file_t* infp = ll_apr_file_open(in_fname,LL_APR_RB); if (!infp) { error_msg = "CannotUploadSoundFile"; return(LLVORBISENC_SOURCE_OPEN_ERR); } ll_apr_file_read(infp, wav_header, 44); physical_file_size = ll_apr_file_seek(infp,APR_END,0); if (strncmp((char *)&(wav_header[0]),"RIFF",4)) { error_msg = "SoundFileNotRIFF"; apr_file_close(infp); return(LLVORBISENC_WAV_FORMAT_ERR); } if (strncmp((char *)&(wav_header[8]),"WAVE",4)) { error_msg = "SoundFileNotRIFF"; apr_file_close(infp); return(LLVORBISENC_WAV_FORMAT_ERR); } // parse the chunks U32 file_pos = 12; // start at the first chunk (usually fmt but not always) while ((file_pos + 8)< physical_file_size) { ll_apr_file_seek(infp,APR_SET,file_pos); ll_apr_file_read(infp, wav_header, 44); chunk_length = ((U32) wav_header[7] << 24) + ((U32) wav_header[6] << 16) + ((U32) wav_header[5] << 8) + wav_header[4]; // llinfos << "chunk found: '" << wav_header[0] << wav_header[1] << wav_header[2] << wav_header[3] << "'" << llendl; if (!(strncmp((char *)&(wav_header[0]),"fmt ",4))) { if ((wav_header[8] == 0x01) && (wav_header[9] == 0x00)) { uncompressed_pcm = TRUE; } num_channels = ((U16) wav_header[11] << 8) + wav_header[10]; sample_rate = ((U32) wav_header[15] << 24) + ((U32) wav_header[14] << 16) + ((U32) wav_header[13] << 8) + wav_header[12]; bits_per_sample = ((U16) wav_header[23] << 8) + wav_header[22]; bytes_per_sec = ((U32) wav_header[19] << 24) + ((U32) wav_header[18] << 16) + ((U32) wav_header[17] << 8) + wav_header[16]; } else if (!(strncmp((char *)&(wav_header[0]),"data",4))) { raw_data_length = chunk_length; } file_pos += (chunk_length + 8); chunk_length = 0; } apr_file_close(infp); if (!uncompressed_pcm) { error_msg = "SoundFileNotPCM"; return(LLVORBISENC_PCM_FORMAT_ERR); } if ((num_channels < 1) || (num_channels > 2)) { error_msg = "SoundFileInvalidChannelCount"; return(LLVORBISENC_MULTICHANNEL_ERR); } if (sample_rate != 44100) { error_msg = "SoundFileInvalidSampleRate"; return(LLVORBISENC_UNSUPPORTED_SAMPLE_RATE); } if ((bits_per_sample != 16) && (bits_per_sample != 8)) { error_msg = "SoundFileInvalidWordSize"; return(LLVORBISENC_UNSUPPORTED_WORD_SIZE); } if (!raw_data_length) { error_msg = "SoundFileInvalidHeader"; return(LLVORBISENC_CLIP_TOO_LONG); } F32 clip_length = (F32)raw_data_length/(F32)bytes_per_sec; if (clip_length > 10.0f) { error_msg = "SoundFileInvalidTooLong"; return(LLVORBISENC_CLIP_TOO_LONG); } return(LLVORBISENC_NOERR); } S32 encode_vorbis_file(const std::string& in_fname, const std::string& out_fname) { #define READ_BUFFER 1024 unsigned char readbuffer[READ_BUFFER*4+44]; /* out of the data segment, not the stack */ /*Flawfinder: ignore*/ ogg_stream_state os; /* take physical pages, weld into a logical stream of packets */ ogg_page og; /* one Ogg bitstream page. Vorbis packets are inside */ ogg_packet op; /* one raw packet of data for decode */ vorbis_info vi; /* struct that stores all the static vorbis bitstream settings */ vorbis_comment vc; /* struct that stores all the user comments */ vorbis_dsp_state vd; /* central working state for the packet->PCM decoder */ vorbis_block vb; /* local working space for packet->PCM decode */ int eos=0; int result; U16 num_channels = 0; U32 sample_rate = 0; U32 bits_per_sample = 0; S32 format_error = 0; std::string error_msg; if ((format_error = check_for_invalid_wav_formats(in_fname, error_msg))) { llwarns << error_msg << ": " << in_fname << llendl; return(format_error); } #if 1 unsigned char wav_header[44]; /*Flawfinder: ignore*/ S32 data_left = 0; apr_file_t* infp = ll_apr_file_open(in_fname,LL_APR_RB); if (!infp) { llwarns << "Couldn't open temporary ogg file for writing: " << in_fname << llendl; return(LLVORBISENC_SOURCE_OPEN_ERR); } apr_file_t* outfp = ll_apr_file_open(out_fname,LL_APR_WPB); if (!outfp) { llwarns << "Couldn't open upload sound file for reading: " << in_fname << llendl; apr_file_close (infp); return(LLVORBISENC_DEST_OPEN_ERR); } // parse the chunks U32 chunk_length = 0; U32 file_pos = 12; // start at the first chunk (usually fmt but not always) while (apr_file_eof(infp) != APR_EOF) { ll_apr_file_seek(infp,APR_SET,file_pos); ll_apr_file_read(infp, wav_header, 44); chunk_length = ((U32) wav_header[7] << 24) + ((U32) wav_header[6] << 16) + ((U32) wav_header[5] << 8) + wav_header[4]; // llinfos << "chunk found: '" << wav_header[0] << wav_header[1] << wav_header[2] << wav_header[3] << "'" << llendl; if (!(strncmp((char *)&(wav_header[0]),"fmt ",4))) { num_channels = ((U16) wav_header[11] << 8) + wav_header[10]; sample_rate = ((U32) wav_header[15] << 24) + ((U32) wav_header[14] << 16) + ((U32) wav_header[13] << 8) + wav_header[12]; bits_per_sample = ((U16) wav_header[23] << 8) + wav_header[22]; } else if (!(strncmp((char *)&(wav_header[0]),"data",4))) { ll_apr_file_seek(infp,APR_SET,file_pos+8); // leave the file pointer at the beginning of the data chunk data data_left = chunk_length; break; } file_pos += (chunk_length + 8); chunk_length = 0; } // apr_file_close(infp); /********** Encode setup ************/ /* choose an encoding mode */ /* (mode 0: 44kHz stereo uncoupled, roughly 128kbps VBR) */ vorbis_info_init(&vi); // always encode to mono // SL-52913 & SL-53779 determined this quality level to be our 'good // enough' general-purpose quality level with a nice low bitrate. // Equivalent to oggenc -q0.5 F32 quality = 0.05f; // quality = (bitrate==128000 ? 0.4f : 0.1); // if (vorbis_encode_init(&vi, /* num_channels */ 1 ,sample_rate, -1, bitrate, -1)) if (vorbis_encode_init_vbr(&vi, /* num_channels */ 1 ,sample_rate, quality)) // if (vorbis_encode_setup_managed(&vi,1,sample_rate,-1,bitrate,-1) || // vorbis_encode_ctl(&vi,OV_ECTL_RATEMANAGE_AVG,NULL) || // vorbis_encode_setup_init(&vi)) { llwarns << "unable to initialize vorbis codec at quality " << quality << llendl; // llwarns << "unable to initialize vorbis codec at bitrate " << bitrate << llendl; return(LLVORBISENC_DEST_OPEN_ERR); } /* add a comment */ vorbis_comment_init(&vc); // vorbis_comment_add(&vc,"Linden"); /* set up the analysis state and auxiliary encoding storage */ vorbis_analysis_init(&vd,&vi); vorbis_block_init(&vd,&vb); /* set up our packet->stream encoder */ /* pick a random serial number; that way we can more likely build chained streams just by concatenation */ ogg_stream_init(&os, ll_rand()); /* Vorbis streams begin with three headers; the initial header (with most of the codec setup parameters) which is mandated by the Ogg bitstream spec. The second header holds any comment fields. The third header holds the bitstream codebook. We merely need to make the headers, then pass them to libvorbis one at a time; libvorbis handles the additional Ogg bitstream constraints */ { ogg_packet header; ogg_packet header_comm; ogg_packet header_code; vorbis_analysis_headerout(&vd,&vc,&header,&header_comm,&header_code); ogg_stream_packetin(&os,&header); /* automatically placed in its own page */ ogg_stream_packetin(&os,&header_comm); ogg_stream_packetin(&os,&header_code); /* We don't have to write out here, but doing so makes streaming * much easier, so we do, flushing ALL pages. This ensures the actual * audio data will start on a new page */ while(!eos){ int result=ogg_stream_flush(&os,&og); if(result==0)break; ll_apr_file_write(outfp, og.header, og.header_len); ll_apr_file_write(outfp, og.body, og.body_len); } } while(!eos) { long bytes_per_sample = bits_per_sample/8; long bytes=(long)ll_apr_file_read(infp, readbuffer,llclamp((S32)(READ_BUFFER*num_channels*bytes_per_sample),0,data_left)); /* stereo hardwired here */ if (bytes==0) { /* end of file. this can be done implicitly in the mainline, but it's easier to see here in non-clever fashion. Tell the library we're at end of stream so that it can handle the last frame and mark end of stream in the output properly */ vorbis_analysis_wrote(&vd,0); // eos = 1; } else { long i; long samples; int temp; data_left -= bytes; /* data to encode */ /* expose the buffer to submit data */ float **buffer=vorbis_analysis_buffer(&vd,READ_BUFFER); i = 0; samples = bytes / (num_channels * bytes_per_sample); if (num_channels == 2) { if (bytes_per_sample == 2) { /* uninterleave samples */ for(i=0; i