summaryrefslogtreecommitdiffstats
path: root/libavfilter/vf_scale.c
blob: fab0568d9aaef54622b1ca908b48c6325adda271 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
/*
 * Copyright (c) 2007 Bobby Bingham
 *
 * This file is part of FFmpeg.
 *
 * FFmpeg is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2.1 of the License, or (at your option) any later version.
 *
 * FFmpeg is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
 * License along with FFmpeg; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

/**
 * @file
 * scale video filter
 */

#include "avfilter.h"
#include "formats.h"
#include "video.h"
#include "libavutil/avstring.h"
#include "libavutil/eval.h"
#include "libavutil/mathematics.h"
#include "libavutil/opt.h"
#include "libavutil/pixdesc.h"
#include "libavutil/imgutils.h"
#include "libavutil/avassert.h"
#include "libswscale/swscale.h"

static const char *const var_names[] = {
    "in_w",   "iw",
    "in_h",   "ih",
    "out_w",  "ow",
    "out_h",  "oh",
    "a",
    "sar",
    "dar",
    "hsub",
    "vsub",
    NULL
};

enum var_name {
    VAR_IN_W,   VAR_IW,
    VAR_IN_H,   VAR_IH,
    VAR_OUT_W,  VAR_OW,
    VAR_OUT_H,  VAR_OH,
    VAR_A,
    VAR_SAR,
    VAR_DAR,
    VAR_HSUB,
    VAR_VSUB,
    VARS_NB
};

typedef struct {
    struct SwsContext *sws;     ///< software scaler context
    struct SwsContext *isws[2]; ///< software scaler context for interlaced material

    /**
     * New dimensions. Special values are:
     *   0 = original width/height
     *  -1 = keep original aspect
     */
    int w, h;
    unsigned int flags;         ///sws flags

    int hsub, vsub;             ///< chroma subsampling
    int slice_y;                ///< top of current output slice
    int input_is_pal;           ///< set to 1 if the input format is paletted
    int output_is_pal;          ///< set to 1 if the output format is paletted
    int interlaced;

    char w_expr[256];           ///< width  expression string
    char h_expr[256];           ///< height expression string
} ScaleContext;

static av_cold int init(AVFilterContext *ctx, const char *args, void *opaque)
{
    ScaleContext *scale = ctx->priv;
    const char *p;

    av_strlcpy(scale->w_expr, "iw", sizeof(scale->w_expr));
    av_strlcpy(scale->h_expr, "ih", sizeof(scale->h_expr));

    scale->flags = SWS_BILINEAR;
    if (args) {
        sscanf(args, "%255[^:]:%255[^:]", scale->w_expr, scale->h_expr);
        p = strstr(args,"flags=");
        if (p) {
            const AVClass *class = sws_get_class();
            const AVOption    *o = av_opt_find(&class, "sws_flags", NULL, 0,
                                               AV_OPT_SEARCH_FAKE_OBJ);
            int ret = av_opt_eval_flags(&class, o, p + 6, &scale->flags);

            if (ret < 0)
                return ret;
        }
        if(strstr(args,"interl=1")){
            scale->interlaced=1;
        }else if(strstr(args,"interl=-1"))
            scale->interlaced=-1;
    }

    return 0;
}

static av_cold void uninit(AVFilterContext *ctx)
{
    ScaleContext *scale = ctx->priv;
    sws_freeContext(scale->sws);
    sws_freeContext(scale->isws[0]);
    sws_freeContext(scale->isws[1]);
    scale->sws = NULL;
}

static int query_formats(AVFilterContext *ctx)
{
    AVFilterFormats *formats;
    enum PixelFormat pix_fmt;
    int ret;

    if (ctx->inputs[0]) {
        formats = NULL;
        for (pix_fmt = 0; pix_fmt < PIX_FMT_NB; pix_fmt++)
            if (   sws_isSupportedInput(pix_fmt)
                && (ret = ff_add_format(&formats, pix_fmt)) < 0) {
                ff_formats_unref(&formats);
                return ret;
            }
        ff_formats_ref(formats, &ctx->inputs[0]->out_formats);
    }
    if (ctx->outputs[0]) {
        formats = NULL;
        for (pix_fmt = 0; pix_fmt < PIX_FMT_NB; pix_fmt++)
            if (   (sws_isSupportedOutput(pix_fmt) || pix_fmt == PIX_FMT_PAL8)
                && (ret = ff_add_format(&formats, pix_fmt)) < 0) {
                ff_formats_unref(&formats);
                return ret;
            }
        ff_formats_ref(formats, &ctx->outputs[0]->in_formats);
    }

    return 0;
}

static int config_props(AVFilterLink *outlink)
{
    AVFilterContext *ctx = outlink->src;
    AVFilterLink *inlink = outlink->src->inputs[0];
    enum PixelFormat outfmt = outlink->format;
    ScaleContext *scale = ctx->priv;
    int64_t w, h;
    double var_values[VARS_NB], res;
    char *expr;
    int ret;

    var_values[VAR_IN_W]  = var_values[VAR_IW] = inlink->w;
    var_values[VAR_IN_H]  = var_values[VAR_IH] = inlink->h;
    var_values[VAR_OUT_W] = var_values[VAR_OW] = NAN;
    var_values[VAR_OUT_H] = var_values[VAR_OH] = NAN;
    var_values[VAR_A]     = (float) inlink->w / inlink->h;
    var_values[VAR_SAR]   = inlink->sample_aspect_ratio.num ?
        (float) inlink->sample_aspect_ratio.num / inlink->sample_aspect_ratio.den : 1;
    var_values[VAR_DAR]   = var_values[VAR_A] * var_values[VAR_SAR];
    var_values[VAR_HSUB]  = 1<<av_pix_fmt_descriptors[inlink->format].log2_chroma_w;
    var_values[VAR_VSUB]  = 1<<av_pix_fmt_descriptors[inlink->format].log2_chroma_h;

    /* evaluate width and height */
    av_expr_parse_and_eval(&res, (expr = scale->w_expr),
                           var_names, var_values,
                           NULL, NULL, NULL, NULL, NULL, 0, ctx);
    scale->w = var_values[VAR_OUT_W] = var_values[VAR_OW] = res;
    if ((ret = av_expr_parse_and_eval(&res, (expr = scale->h_expr),
                                      var_names, var_values,
                                      NULL, NULL, NULL, NULL, NULL, 0, ctx)) < 0)
        goto fail;
    scale->h = var_values[VAR_OUT_H] = var_values[VAR_OH] = res;
    /* evaluate again the width, as it may depend on the output height */
    if ((ret = av_expr_parse_and_eval(&res, (expr = scale->w_expr),
                                      var_names, var_values,
                                      NULL, NULL, NULL, NULL, NULL, 0, ctx)) < 0)
        goto fail;
    scale->w = res;

    w = scale->w;
    h = scale->h;

    /* sanity check params */
    if (w <  -1 || h <  -1) {
        av_log(ctx, AV_LOG_ERROR, "Size values less than -1 are not acceptable.\n");
        return AVERROR(EINVAL);
    }
    if (w == -1 && h == -1)
        scale->w = scale->h = 0;

    if (!(w = scale->w))
        w = inlink->w;
    if (!(h = scale->h))
        h = inlink->h;
    if (w == -1)
        w = av_rescale(h, inlink->w, inlink->h);
    if (h == -1)
        h = av_rescale(w, inlink->h, inlink->w);

    if (w > INT_MAX || h > INT_MAX ||
        (h * inlink->w) > INT_MAX  ||
        (w * inlink->h) > INT_MAX)
        av_log(ctx, AV_LOG_ERROR, "Rescaled value for width or height is too big.\n");

    outlink->w = w;
    outlink->h = h;

    /* TODO: make algorithm configurable */

    scale->input_is_pal = av_pix_fmt_descriptors[inlink->format].flags & PIX_FMT_PAL ||
                          av_pix_fmt_descriptors[inlink->format].flags & PIX_FMT_PSEUDOPAL;
    if (outfmt == PIX_FMT_PAL8) outfmt = PIX_FMT_BGR8;
    scale->output_is_pal = av_pix_fmt_descriptors[outfmt].flags & PIX_FMT_PAL ||
                           av_pix_fmt_descriptors[outfmt].flags & PIX_FMT_PSEUDOPAL;

    if (scale->sws)
        sws_freeContext(scale->sws);
    if (inlink->w == outlink->w && inlink->h == outlink->h &&
        inlink->format == outlink->format)
        scale->sws = NULL;
    else {
        scale->sws = sws_getContext(inlink ->w, inlink ->h, inlink ->format,
                                    outlink->w, outlink->h, outfmt,
                                    scale->flags, NULL, NULL, NULL);
        if (scale->isws[0])
            sws_freeContext(scale->isws[0]);
        scale->isws[0] = sws_getContext(inlink ->w, inlink ->h/2, inlink ->format,
                                        outlink->w, outlink->h/2, outfmt,
                                        scale->flags, NULL, NULL, NULL);
        if (scale->isws[1])
            sws_freeContext(scale->isws[1]);
        scale->isws[1] = sws_getContext(inlink ->w, inlink ->h/2, inlink ->format,
                                        outlink->w, outlink->h/2, outfmt,
                                        scale->flags, NULL, NULL, NULL);
        if (!scale->sws || !scale->isws[0] || !scale->isws[1])
            return AVERROR(EINVAL);
    }

    if (inlink->sample_aspect_ratio.num){
        outlink->sample_aspect_ratio = av_mul_q((AVRational){outlink->h * inlink->w, outlink->w * inlink->h}, inlink->sample_aspect_ratio);
    } else
        outlink->sample_aspect_ratio = inlink->sample_aspect_ratio;

    av_log(ctx, AV_LOG_INFO, "w:%d h:%d fmt:%s sar:%d/%d -> w:%d h:%d fmt:%s sar:%d/%d flags:0x%0x\n",
           inlink ->w, inlink ->h, av_pix_fmt_descriptors[ inlink->format].name,
           inlink->sample_aspect_ratio.num, inlink->sample_aspect_ratio.den,
           outlink->w, outlink->h, av_pix_fmt_descriptors[outlink->format].name,
           outlink->sample_aspect_ratio.num, outlink->sample_aspect_ratio.den,
           scale->flags);
    return 0;

fail:
    av_log(NULL, AV_LOG_ERROR,
           "Error when evaluating the expression '%s'.\n"
           "Maybe the expression for out_w:'%s' or for out_h:'%s' is self-referencing.\n",
           expr, scale->w_expr, scale->h_expr);
    return ret;
}

static void start_frame(AVFilterLink *link, AVFilterBufferRef *picref)
{
    ScaleContext *scale = link->dst->priv;
    AVFilterLink *outlink = link->dst->outputs[0];
    AVFilterBufferRef *outpicref;

    if(   picref->video->w != link->w
       || picref->video->h != link->h
       || picref->format   != link->format) {
        AVFilterLink *out_link;
        int ret;
        snprintf(scale->w_expr, sizeof(scale->w_expr)-1, "%d", outlink->w);
        snprintf(scale->h_expr, sizeof(scale->h_expr)-1, "%d", outlink->h);

        link->dst->inputs[0]->format = picref->format;
        link->dst->inputs[0]->w      = picref->video->w;
        link->dst->inputs[0]->h      = picref->video->h;

        if ((ret = config_props(outlink)) < 0)
            av_assert0(0); //what to do here ?
    }


    if (!scale->sws) {
        ff_start_frame(outlink, avfilter_ref_buffer(picref, ~0));
        return;
    }

    scale->hsub = av_pix_fmt_descriptors[link->format].log2_chroma_w;
    scale->vsub = av_pix_fmt_descriptors[link->format].log2_chroma_h;

    outpicref = avfilter_get_video_buffer(outlink, AV_PERM_WRITE|AV_PERM_ALIGN, outlink->w, outlink->h);
    avfilter_copy_buffer_ref_props(outpicref, picref);
    outpicref->video->w = outlink->w;
    outpicref->video->h = outlink->h;

    outlink->out_buf = outpicref;
    if(scale->output_is_pal)
        ff_set_systematic_pal2(outpicref->data[1], outlink->format == PIX_FMT_PAL8 ? PIX_FMT_BGR8 : outlink->format);

    av_reduce(&outpicref->video->sample_aspect_ratio.num, &outpicref->video->sample_aspect_ratio.den,
              (int64_t)picref->video->sample_aspect_ratio.num * outlink->h * link->w,
              (int64_t)picref->video->sample_aspect_ratio.den * outlink->w * link->h,
              INT_MAX);

    scale->slice_y = 0;
    ff_start_frame(outlink, avfilter_ref_buffer(outpicref, ~0));
}

static int scale_slice(AVFilterLink *link, struct SwsContext *sws, int y, int h, int mul, int field)
{
    ScaleContext *scale = link->dst->priv;
    AVFilterBufferRef *cur_pic = link->cur_buf;
    AVFilterBufferRef *out_buf = link->dst->outputs[0]->out_buf;
    const uint8_t *in[4];
    uint8_t *out[4];
    int in_stride[4],out_stride[4];
    int i;

    for(i=0; i<4; i++){
        int vsub= ((i+1)&2) ? scale->vsub : 0;
         in_stride[i] = cur_pic->linesize[i] * mul;
        out_stride[i] = out_buf->linesize[i] * mul;
         in[i] = cur_pic->data[i] + ((y>>vsub)+field) * cur_pic->linesize[i];
        out[i] = out_buf->data[i] +            field  * out_buf->linesize[i];
    }
    if(scale->input_is_pal)
         in[1] = cur_pic->data[1];
    if(scale->output_is_pal)
        out[1] = out_buf->data[1];

    return sws_scale(sws, in, in_stride, y/mul, h,
                         out,out_stride);
}

static void draw_slice(AVFilterLink *link, int y, int h, int slice_dir)
{
    ScaleContext *scale = link->dst->priv;
    int out_h;

    if (!scale->sws) {
        ff_draw_slice(link->dst->outputs[0], y, h, slice_dir);
        return;
    }

    if (scale->slice_y == 0 && slice_dir == -1)
        scale->slice_y = link->dst->outputs[0]->h;

    if(scale->interlaced>0 || (scale->interlaced<0 && link->cur_buf->video->interlaced)){
        av_assert0(y%(2<<scale->vsub) == 0);
        out_h = scale_slice(link, scale->isws[0], y, (h+1)/2, 2, 0);
        out_h+= scale_slice(link, scale->isws[1], y,  h   /2, 2, 1);
    }else{
        out_h = scale_slice(link, scale->sws, y, h, 1, 0);
    }

    if (slice_dir == -1)
        scale->slice_y -= out_h;
    ff_draw_slice(link->dst->outputs[0], scale->slice_y, out_h, slice_dir);
    if (slice_dir == 1)
        scale->slice_y += out_h;
}

AVFilter avfilter_vf_scale = {
    .name      = "scale",
    .description = NULL_IF_CONFIG_SMALL("Scale the input video to width:height size and/or convert the image format."),

    .init      = init,
    .uninit    = uninit,

    .query_formats = query_formats,

    .priv_size = sizeof(ScaleContext),

    .inputs    = (const AVFilterPad[]) {{ .name       = "default",
                                    .type             = AVMEDIA_TYPE_VIDEO,
                                    .start_frame      = start_frame,
                                    .draw_slice       = draw_slice,
                                    .min_perms        = AV_PERM_READ, },
                                  { .name = NULL}},
    .outputs   = (const AVFilterPad[]) {{ .name       = "default",
                                    .type             = AVMEDIA_TYPE_VIDEO,
                                    .config_props     = config_props, },
                                  { .name = NULL}},
};
OpenPOWER on IntegriCloud