aboutsummaryrefslogtreecommitdiff
path: root/libavcodec/ac3enc.c
diff options
context:
space:
mode:
authorJustin Ruggles2010-12-30 22:13:34 +0000
committerJustin Ruggles2010-12-30 22:13:34 +0000
commitb5293036b99d5aad44cd99cec54d3dde40ae4144 (patch)
tree90a4a61ca0f5f72513abc60335bc528fac36bcc0 /libavcodec/ac3enc.c
parent14166a6d6cad8759642ef50a21b4e259d40fee95 (diff)
cosmetics: alignment, line wrap, and spacing
Originally committed as revision 26171 to svn://svn.ffmpeg.org/ffmpeg/trunk
Diffstat (limited to 'libavcodec/ac3enc.c')
-rw-r--r--libavcodec/ac3enc.c48
1 files changed, 19 insertions, 29 deletions
diff --git a/libavcodec/ac3enc.c b/libavcodec/ac3enc.c
index 9218ca674f..031718afa4 100644
--- a/libavcodec/ac3enc.c
+++ b/libavcodec/ac3enc.c
@@ -216,7 +216,6 @@ static av_cold void mdct_end(AC3MDCTContext *mdct)
}
-
/**
* Initialize FFT tables.
* @param ln log2(FFT size)
@@ -229,10 +228,8 @@ static av_cold int fft_init(AVCodecContext *avctx, AC3MDCTContext *mdct, int ln)
n = 1 << ln;
n2 = n >> 1;
- FF_ALLOC_OR_GOTO(avctx, mdct->costab, n2 * sizeof(*mdct->costab),
- fft_alloc_fail);
- FF_ALLOC_OR_GOTO(avctx, mdct->sintab, n2 * sizeof(*mdct->sintab),
- fft_alloc_fail);
+ FF_ALLOC_OR_GOTO(avctx, mdct->costab, n2 * sizeof(*mdct->costab), fft_alloc_fail);
+ FF_ALLOC_OR_GOTO(avctx, mdct->sintab, n2 * sizeof(*mdct->sintab), fft_alloc_fail);
for (i = 0; i < n2; i++) {
alpha = 2.0 * M_PI * i / n;
@@ -251,7 +248,8 @@ fft_alloc_fail:
* Initialize MDCT tables.
* @param nbits log2(MDCT size)
*/
-static av_cold int mdct_init(AVCodecContext *avctx, AC3MDCTContext *mdct, int nbits)
+static av_cold int mdct_init(AVCodecContext *avctx, AC3MDCTContext *mdct,
+ int nbits)
{
int i, n, n4, ret;
@@ -264,14 +262,10 @@ static av_cold int mdct_init(AVCodecContext *avctx, AC3MDCTContext *mdct, int nb
if (ret)
return ret;
- FF_ALLOC_OR_GOTO(avctx, mdct->xcos1, n4 * sizeof(*mdct->xcos1),
- mdct_alloc_fail);
- FF_ALLOC_OR_GOTO(avctx, mdct->xsin1, n4 * sizeof(*mdct->xsin1),
- mdct_alloc_fail);
- FF_ALLOC_OR_GOTO(avctx, mdct->rot_tmp, n * sizeof(*mdct->rot_tmp),
- mdct_alloc_fail);
- FF_ALLOC_OR_GOTO(avctx, mdct->cplx_tmp, n4 * sizeof(*mdct->cplx_tmp),
- mdct_alloc_fail);
+ FF_ALLOC_OR_GOTO(avctx, mdct->xcos1, n4 * sizeof(*mdct->xcos1), mdct_alloc_fail);
+ FF_ALLOC_OR_GOTO(avctx, mdct->xsin1, n4 * sizeof(*mdct->xsin1), mdct_alloc_fail);
+ FF_ALLOC_OR_GOTO(avctx, mdct->rot_tmp, n * sizeof(*mdct->rot_tmp), mdct_alloc_fail);
+ FF_ALLOC_OR_GOTO(avctx, mdct->cplx_tmp, n4 * sizeof(*mdct->cplx_tmp), mdct_alloc_fail);
for (i = 0; i < n4; i++) {
float alpha = 2.0 * M_PI * (i + 1.0 / 8.0) / n;
@@ -571,7 +565,8 @@ static void extract_exponents(AC3EncodeContext *s)
/**
* Calculate exponent strategies for all blocks in a single channel.
*/
-static void compute_exp_strategy_ch(AC3EncodeContext *s, uint8_t *exp_strategy, uint8_t **exp)
+static void compute_exp_strategy_ch(AC3EncodeContext *s, uint8_t *exp_strategy,
+ uint8_t **exp)
{
int blk, blk1;
int exp_diff;
@@ -654,8 +649,7 @@ static void exponent_min(uint8_t *exp, uint8_t *exp1, int n)
/**
* Update the exponents so that they are the ones the decoder will decode.
*/
-static void encode_exponents_blk_ch(uint8_t *exp,
- int nb_exps, int exp_strategy)
+static void encode_exponents_blk_ch(uint8_t *exp, int nb_exps, int exp_strategy)
{
int nb_groups, i, k;
@@ -1035,8 +1029,7 @@ static void reset_block_bap(AC3EncodeContext *s)
* @return the number of bits needed for mantissas if the given SNR offset is
* is used.
*/
-static int bit_alloc(AC3EncodeContext *s,
- int snr_offset)
+static int bit_alloc(AC3EncodeContext *s, int snr_offset)
{
int blk, ch;
int mantissa_bits;
@@ -1265,10 +1258,9 @@ static inline int asym_quant(int c, int e, int qbits)
/**
* Quantize a set of mantissas for a single channel in a single block.
*/
-static void quantize_mantissas_blk_ch(AC3EncodeContext *s,
- int32_t *mdct_coef, int8_t exp_shift,
- uint8_t *exp, uint8_t *bap,
- uint16_t *qmant, int n)
+static void quantize_mantissas_blk_ch(AC3EncodeContext *s, int32_t *mdct_coef,
+ int8_t exp_shift, uint8_t *exp,
+ uint8_t *bap, uint16_t *qmant, int n)
{
int i;
@@ -1413,8 +1405,7 @@ static void output_frame_header(AC3EncodeContext *s)
/**
* Write one audio block to the output bitstream.
*/
-static void output_audio_block(AC3EncodeContext *s,
- int block_num)
+static void output_audio_block(AC3EncodeContext *s, int block_num)
{
int ch, i, baie, rbnd;
AC3Block *block = &s->blocks[block_num];
@@ -1608,8 +1599,7 @@ static void output_frame_end(AC3EncodeContext *s)
/**
* Write the frame to the output bitstream.
*/
-static void output_frame(AC3EncodeContext *s,
- unsigned char *frame)
+static void output_frame(AC3EncodeContext *s, unsigned char *frame)
{
int blk;
@@ -1627,8 +1617,8 @@ static void output_frame(AC3EncodeContext *s,
/**
* Encode a single AC-3 frame.
*/
-static int ac3_encode_frame(AVCodecContext *avctx,
- unsigned char *frame, int buf_size, void *data)
+static int ac3_encode_frame(AVCodecContext *avctx, unsigned char *frame,
+ int buf_size, void *data)
{
AC3EncodeContext *s = avctx->priv_data;
const int16_t *samples = data;