2 * Copyright (c) 2015 Paul B Mahol
4 * This file is part of FFmpeg.
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.
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.
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
21 #include "libavutil/avstring.h"
22 #include "libavutil/channel_layout.h"
23 #include "libavutil/eval.h"
24 #include "libavutil/intreadwrite.h"
25 #include "libavutil/opt.h"
26 #include "libavutil/parseutils.h"
27 #include "libavutil/xga_font_data.h"
34 static const char *const var_names[] = { "VOLUME", "CHANNEL", NULL };
35 enum { VAR_VOLUME, VAR_CHANNEL, VAR_VARS_NB };
37 typedef struct ShowVolumeContext {
42 AVRational frame_rate;
54 #define OFFSET(x) offsetof(ShowVolumeContext, x)
55 #define FLAGS AV_OPT_FLAG_FILTERING_PARAM|AV_OPT_FLAG_VIDEO_PARAM
57 static const AVOption showvolume_options[] = {
58 { "rate", "set video rate", OFFSET(frame_rate), AV_OPT_TYPE_VIDEO_RATE, {.str="25"}, 0, 0, FLAGS },
59 { "r", "set video rate", OFFSET(frame_rate), AV_OPT_TYPE_VIDEO_RATE, {.str="25"}, 0, 0, FLAGS },
60 { "b", "set border width", OFFSET(b), AV_OPT_TYPE_INT, {.i64=1}, 0, 5, FLAGS },
61 { "w", "set channel width", OFFSET(w), AV_OPT_TYPE_INT, {.i64=400}, 80, 8192, FLAGS },
62 { "h", "set channel height", OFFSET(h), AV_OPT_TYPE_INT, {.i64=20}, 1, 900, FLAGS },
63 { "f", "set fade", OFFSET(f), AV_OPT_TYPE_DOUBLE, {.dbl=0.95}, 0.001, 1, FLAGS },
64 { "c", "set volume color expression", OFFSET(color), AV_OPT_TYPE_STRING, {.str="if(gte(VOLUME,-6), if(gte(VOLUME,-2), if(gte(VOLUME,-1), if(gt(VOLUME,0), 0xff0000ff, 0xff0066ff), 0xff00ffff),0xff00ff00),0xffff0000)"}, 0, 0, FLAGS },
65 { "t", "display channel names", OFFSET(draw_text), AV_OPT_TYPE_BOOL, {.i64=1}, 0, 1, FLAGS },
66 { "v", "display volume value", OFFSET(draw_volume), AV_OPT_TYPE_BOOL, {.i64=1}, 0, 1, FLAGS },
67 { "o", "set orientation", OFFSET(orientation), AV_OPT_TYPE_INT, {.i64=0}, 0, 1, FLAGS, "orientation" },
68 { "h", "horizontal", 0, AV_OPT_TYPE_CONST, {.i64=0}, 0, 0, FLAGS, "orientation" },
69 { "v", "vertical", 0, AV_OPT_TYPE_CONST, {.i64=1}, 0, 0, FLAGS, "orientation" },
70 { "s", "set step size", OFFSET(step), AV_OPT_TYPE_INT, {.i64=0}, 0, 5, FLAGS },
74 AVFILTER_DEFINE_CLASS(showvolume);
76 static av_cold int init(AVFilterContext *ctx)
78 ShowVolumeContext *s = ctx->priv;
82 ret = av_expr_parse(&s->c_expr, s->color, var_names,
83 NULL, NULL, NULL, NULL, 0, ctx);
91 static int query_formats(AVFilterContext *ctx)
93 AVFilterFormats *formats = NULL;
94 AVFilterChannelLayouts *layouts = NULL;
95 AVFilterLink *inlink = ctx->inputs[0];
96 AVFilterLink *outlink = ctx->outputs[0];
97 static const enum AVSampleFormat sample_fmts[] = { AV_SAMPLE_FMT_FLTP, AV_SAMPLE_FMT_NONE };
98 static const enum AVPixelFormat pix_fmts[] = { AV_PIX_FMT_RGBA, AV_PIX_FMT_NONE };
101 formats = ff_make_format_list(sample_fmts);
102 if ((ret = ff_formats_ref(formats, &inlink->out_formats)) < 0)
105 layouts = ff_all_channel_counts();
106 if ((ret = ff_channel_layouts_ref(layouts, &inlink->out_channel_layouts)) < 0)
109 formats = ff_all_samplerates();
110 if ((ret = ff_formats_ref(formats, &inlink->out_samplerates)) < 0)
113 formats = ff_make_format_list(pix_fmts);
114 if ((ret = ff_formats_ref(formats, &outlink->in_formats)) < 0)
120 static int config_input(AVFilterLink *inlink)
122 AVFilterContext *ctx = inlink->dst;
123 ShowVolumeContext *s = ctx->priv;
126 nb_samples = FFMAX(1024, ((double)inlink->sample_rate / av_q2d(s->frame_rate)) + 0.5);
127 inlink->partial_buf_size =
128 inlink->min_samples =
129 inlink->max_samples = nb_samples;
130 s->values = av_calloc(inlink->channels * VAR_VARS_NB, sizeof(double));
132 return AVERROR(ENOMEM);
136 static int config_output(AVFilterLink *outlink)
138 ShowVolumeContext *s = outlink->src->priv;
139 AVFilterLink *inlink = outlink->src->inputs[0];
141 if (s->orientation) {
143 outlink->w = s->h * inlink->channels + (inlink->channels - 1) * s->b;
146 outlink->h = s->h * inlink->channels + (inlink->channels - 1) * s->b;
149 outlink->sample_aspect_ratio = (AVRational){1,1};
150 outlink->frame_rate = s->frame_rate;
155 static void drawtext(AVFrame *pic, int x, int y, const char *txt, int o)
161 font = avpriv_cga_font, font_height = 8;
163 for (i = 0; txt[i]; i++) {
167 for (char_y = font_height - 1; char_y >= 0; char_y--) {
168 uint8_t *p = pic->data[0] + (y + i * 10) * pic->linesize[0] + x * 4;
169 for (mask = 0x80; mask; mask >>= 1) {
170 if (font[txt[i] * font_height + font_height - 1 - char_y] & mask)
171 AV_WN32(&p[char_y * 4], ~AV_RN32(&p[char_y * 4]));
172 p += pic->linesize[0];
176 uint8_t *p = pic->data[0] + y * pic->linesize[0] + (x + i * 8) * 4;
177 for (char_y = 0; char_y < font_height; char_y++) {
178 for (mask = 0x80; mask; mask >>= 1) {
179 if (font[txt[i] * font_height + char_y] & mask)
180 AV_WN32(p, ~AV_RN32(p));
183 p += pic->linesize[0] - 8 * 4;
189 static int filter_frame(AVFilterLink *inlink, AVFrame *insamples)
191 AVFilterContext *ctx = inlink->dst;
192 AVFilterLink *outlink = ctx->outputs[0];
193 ShowVolumeContext *s = ctx->priv;
194 const int step = s->step;
198 if (!s->out || s->out->width != outlink->w ||
199 s->out->height != outlink->h) {
200 av_frame_free(&s->out);
201 s->out = ff_get_video_buffer(outlink, outlink->w, outlink->h);
203 av_frame_free(&insamples);
204 return AVERROR(ENOMEM);
207 for (i = 0; i < outlink->h; i++)
208 memset(s->out->data[0] + i * s->out->linesize[0], 0, outlink->w * 4);
210 s->out->pts = insamples->pts;
212 for (j = 0; j < outlink->h; j++) {
213 uint8_t *dst = s->out->data[0] + j * s->out->linesize[0];
214 for (k = 0; k < outlink->w; k++) {
215 dst[k * 4 + 0] = FFMAX(dst[k * 4 + 0] * s->f, 0);
216 dst[k * 4 + 1] = FFMAX(dst[k * 4 + 1] * s->f, 0);
217 dst[k * 4 + 2] = FFMAX(dst[k * 4 + 2] * s->f, 0);
218 dst[k * 4 + 3] = FFMAX(dst[k * 4 + 3] * s->f, 0);
222 if (s->orientation) {
223 for (c = 0; c < inlink->channels; c++) {
224 float *src = (float *)insamples->extended_data[c];
228 for (i = 0; i < insamples->nb_samples; i++)
229 max = FFMAX(max, src[i]);
231 s->values[c * VAR_VARS_NB + VAR_VOLUME] = 20.0 * log10(max);
232 max = av_clipf(max, 0, 1);
233 s->values[c * VAR_VARS_NB + VAR_CHANNEL] = c;
234 color = av_expr_eval(s->c_expr, &s->values[c * VAR_VARS_NB], NULL);
236 for (j = outlink->h - outlink->h * max; j < outlink->h; j++) {
237 uint8_t *dst = s->out->data[0] + j * s->out->linesize[0] + c * (s->b + s->h) * 4;
238 for (k = 0; k < s->h; k++) {
239 AV_WN32A(&dst[k * 4], color);
245 if (outlink->h > 40 && s->draw_text) {
246 const char *channel_name = av_get_channel_name(av_channel_layout_extract_channel(insamples->channel_layout, c));
249 drawtext(s->out, c * (s->h + s->b) + (s->h - 10) / 2, outlink->h - 35, channel_name, 1);
253 for (c = 0; c < inlink->channels; c++) {
254 float *src = (float *)insamples->extended_data[c];
258 for (i = 0; i < insamples->nb_samples; i++)
259 max = FFMAX(max, src[i]);
261 s->values[c * VAR_VARS_NB + VAR_VOLUME] = 20.0 * log10(max);
262 max = av_clipf(max, 0, 1);
263 s->values[c * VAR_VARS_NB + VAR_CHANNEL] = c;
264 color = av_expr_eval(s->c_expr, &s->values[c * VAR_VARS_NB], NULL);
266 for (j = 0; j < s->h; j++) {
267 uint8_t *dst = s->out->data[0] + (c * s->h + c * s->b + j) * s->out->linesize[0];
269 for (k = 0; k < s->w * max; k++) {
270 AV_WN32A(dst + k * 4, color);
276 if (s->h >= 8 && s->draw_text) {
277 const char *channel_name = av_get_channel_name(av_channel_layout_extract_channel(insamples->channel_layout, c));
280 drawtext(s->out, 2, c * (s->h + s->b) + (s->h - 8) / 2, channel_name, 0);
285 av_frame_free(&insamples);
286 out = av_frame_clone(s->out);
288 return AVERROR(ENOMEM);
289 av_frame_make_writable(out);
291 for (c = 0; c < inlink->channels && s->draw_volume; c++) {
293 if (s->orientation) {
295 snprintf(buf, sizeof(buf), "%.2f", s->values[c * VAR_VARS_NB + VAR_VOLUME]);
296 drawtext(out, c * (s->h + s->b) + (s->h - 8) / 2, 2, buf, 1);
300 snprintf(buf, sizeof(buf), "%.2f", s->values[c * VAR_VARS_NB + VAR_VOLUME]);
301 drawtext(out, FFMAX(0, s->w - 8 * (int)strlen(buf)), c * (s->h + s->b) + (s->h - 8) / 2, buf, 0);
306 return ff_filter_frame(outlink, out);
309 static av_cold void uninit(AVFilterContext *ctx)
311 ShowVolumeContext *s = ctx->priv;
313 av_frame_free(&s->out);
314 av_expr_free(s->c_expr);
315 av_freep(&s->values);
318 static const AVFilterPad showvolume_inputs[] = {
321 .type = AVMEDIA_TYPE_AUDIO,
322 .config_props = config_input,
323 .filter_frame = filter_frame,
328 static const AVFilterPad showvolume_outputs[] = {
331 .type = AVMEDIA_TYPE_VIDEO,
332 .config_props = config_output,
337 AVFilter ff_avf_showvolume = {
338 .name = "showvolume",
339 .description = NULL_IF_CONFIG_SMALL("Convert input audio volume to video output."),
342 .query_formats = query_formats,
343 .priv_size = sizeof(ShowVolumeContext),
344 .inputs = showvolume_inputs,
345 .outputs = showvolume_outputs,
346 .priv_class = &showvolume_class,