Author: cmcq Date: Fri Jul 31 05:11:53 2009 New Revision: 4857 Log: Remove section comments; the function names should be sufficiently clear Modified: amr/amrnbdec.c Modified: amr/amrnbdec.c ============================================================================== --- amr/amrnbdec.c Fri Jul 31 05:08:18 2009 (r4856) +++ amr/amrnbdec.c Fri Jul 31 05:11:53 2009 (r4857) @@ -1117,7 +1117,6 @@ static int amrnb_decode_frame(AVCodecCon av_log_missing_feature(avctx, "dtx mode", 1); return -1; } -/*** LPC coefficient decoding ***/ if (p->cur_frame_mode == MODE_122) { // decode split-matrix quantized lsf vector indexes to lsp vectors @@ -1131,14 +1130,10 @@ static int amrnb_decode_frame(AVCodecCon for (i = 0; i < 4; i++) lsp2lpc(p->lsp[i], p->lpc[i]); -/*** end of LPC coefficient decoding ***/ - for (subframe = 0; subframe < 4; subframe++) { const AMRNBSubframe *amr_subframe = &p->frame.subframe[subframe]; -/*** adaptive code book (pitch) vector decoding ***/ - decode_pitch_vector(p, amr_subframe, subframe); -/*** end of adaptive code book (pitch) vector decoding ***/ + decode_pitch_vector(p, amr_subframe, subframe); decode_fixed_vector(fixed_vector, amr_subframe->pulses, p->cur_frame_mode, subframe); @@ -1155,8 +1150,6 @@ static int amrnb_decode_frame(AVCodecCon set_fixed_gain(p, p->cur_frame_mode, fixed_gain_factor, fixed_vector); -/*** pre-processing ***/ - // The excitation feedback is calculated without any processing such // as fixed gain smoothing. This isn't mentioned in the specification. ff_weighted_vector_sumf(exc_feedback, p->excitation, fixed_vector, @@ -1172,10 +1165,6 @@ static int amrnb_decode_frame(AVCodecCon synth_fixed_vector = anti_sparseness(p, fixed_vector, synth_fixed_gain, spare_vector); -/*** end of pre-processing ***/ - -/*** synthesis ***/ - if (synthesis(p, p->excitation, p->lpc[subframe], synth_fixed_gain, synth_fixed_vector, &p->samples_in[LP_FILTER_ORDER], 0)) // overflow detected -> rerun synthesis scaling pitch vector down @@ -1184,8 +1173,6 @@ static int amrnb_decode_frame(AVCodecCon synthesis(p, p->excitation, p->lpc[subframe], synth_fixed_gain, synth_fixed_vector, &p->samples_in[LP_FILTER_ORDER], 1); -/*** end of synthesis ***/ - postfilter(p, p->lpc[subframe], buf_out + subframe * AMR_SUBFRAME_SIZE); // update buffers and history