Changelog: highlighting for 2.3.1

Suggested-by: Timothy Gu <timothygu99@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Changelog

@@ -1,7 +1,12 @@
 Entries are sorted chronologically from oldest to youngest within each release,
 releases are sorted from youngest to oldest.
 
-version <next>:
+version 2.3.1:
+- public AVDCT API/ABI for DCT functions
+- g2meet: allow size changes within original sizes
+- dv: improved error resilience, fixing Ticket2340 and Ticket2341
+
+version 2.3:
 - AC3 fixed-point decoding
 - shuffleplanes filter
 - subfile protocol

MAINTAINERS

@@ -528,8 +528,8 @@ x86                                     Michael Niedermayer
 Releases
 ========
 
+2.3                                     Michael Niedermayer
 2.2                                     Michael Niedermayer
-2.1                                     Michael Niedermayer
 1.2                                     Michael Niedermayer
 
 If you want to maintain an older release, please contact us

RELEASE

@@ -1 +1 @@
-2.2.git
+2.3.1

RELEASE_NOTES

@@ -7,9 +7,10 @@
    since the release of FFmpeg 2.2.
 
    In this release, there are lots of internal overhauls that make FFmpeg a
-   more accessible project for new developers. Many important new features
+   more accessible project for new developers. Many important new
-   like QTKit and AVFoundation input devices are committed. Contributions done
+   optimizations and features like QTKit and AVFoundation input devices are
-   by Libav such as a new native Opus decoder are also merged.
+   committed. Contributions done by Libav such as a new native Opus decoder
+   are also merged.
 
    Because of the increasing difficulty to maintain and lack of maintainers,
    we are very sorry to say that we have removed all Blackfin and SPARC
@@ -17,8 +18,9 @@
    interested in maintaining optimization for these two architecture, feel
    free to contact us and we will restore the code!
 
-   Since this release, the traditional Changelog file is upgraded to this
+   Oh, and since this release, this modern-looking release note is provided in
-   modern-looking release note. Old changelogs are moved to doc/Changelog.old.
+   addition to the old-style Changelog file, to make it easier for you to
+   focus on the most important features in this release.
 
    Enjoy!
 
@@ -33,9 +35,9 @@
 
        • libavutil      52.92.100
        • libavcodec     55.69.100
-       • libavformat    55.47.100
+       • libavformat    55.48.100
        • libavdevice    55.13.102
-       • libavfilter     4.10.100
+       • libavfilter     4.11.100
        • libswscale      2. 6.100
        • libswresample   0.19.100
        • libpostproc    52. 3.100
@@ -103,7 +105,7 @@
      Other interesting new features including hqx video filter, a pixel art
      scaling filter; a fixed-point AC-3 decoder contributed by Imagination
      Technologies; an On2 TrueMotion VP7 video decoder; an HTML5 WebVTT
-     subtitle decoder that allows creation of WebVTT from any text-based
+     subtitle encoder that allows creation of WebVTT from any text-based
      subtitles; and an 1-bit Direct Stream Digital audio decoder.
 
  ┌────────────────────────────┐

doc/APIchanges

@@ -15,13 +15,13 @@ libavutil:     2012-10-22
 
 API changes, most recent first:
 
-2014-07-14 - xxxxxxx - lavf 55.47.100 - avformat.h
+2014-07-14 - 62227a7 - lavf 55.47.100 - avformat.h
   Add av_stream_get_parser()
 
-2014-07-xx - xxxxxxx - lavu 53.18.0 - display.h
+2014-07-09 - c67690f / a54f03b - lavu 52.92.100 / 53.18.0 - display.h
   Add av_display_matrix_flip() to flip the transformation matrix.
 
-2014-07-xx - xxxxxxx - lavc 55.56.0 - dv_profile.h
+2014-07-09 - 1b58f13 / f6ee61f - lavc 55.69.100 / 55.56.0 - dv_profile.h
   Add a public API for DV profile handling.
 
 2014-06-20 - 0dceefc / 9e500ef - lavu 52.90.100 / 53.17.0 - imgutils.h
@@ -35,6 +35,10 @@ API changes, most recent first:
   is now setting AVStream.time_base, instead of AVStream.codec.time_base as was
   done previously. The old method is now deprecated.
 
+2014-06-11 - 67d29da - lavc 55.66.101 - avcodec.h
+  Increase FF_INPUT_BUFFER_PADDING_SIZE to 32 due to some corner cases needing
+  it
+
 2014-06-10 - xxxxxxx - lavf 55.43.100 - avformat.h
   New field int64_t max_analyze_duration2 instead of deprecated
   int max_analyze_duration.
@@ -42,7 +46,7 @@ API changes, most recent first:
 2014-05-30 - 00759d7 - lavu 52.89.100 - opt.h
   Add av_opt_copy()
 
-2014-04-xx - 03bb99a / 0957b27 - lavc 55.66.100 / 55.54.0 - avcodec.h
+2014-06-01 - 03bb99a / 0957b27 - lavc 55.66.100 / 55.54.0 - avcodec.h
   Add AVCodecContext.side_data_only_packets to allow encoders to output packets
   with only side data. This option may become mandatory in the future, so all
   users are recommended to update their code and enable this option.
@@ -52,7 +56,7 @@ API changes, most recent first:
   AVColorTransferCharacteristic, and AVChromaLocation) inside lavu.
   And add AVFrame fields for them.
 
-2014-04-xx - bdb2e80 / b2d4565 - lavr 1.3.0 - avresample.h
+2014-05-29 - bdb2e80 / b2d4565 - lavr 1.3.0 - avresample.h
   Add avresample_max_output_samples
 
 2014-05-24 - d858ee7 / 6d21259 - lavf 55.42.100 / 55.19.0 - avformat.h

doc/Doxyfile

@@ -31,7 +31,7 @@ PROJECT_NAME           = FFmpeg
 # This could be handy for archiving the generated documentation or
 # if some version control system is used.
 
-PROJECT_NUMBER         =
+PROJECT_NUMBER         = 2.3.1
 
 # With the PROJECT_LOGO tag one can specify a logo or icon that is included
 # in the documentation. The maximum height of the logo should not exceed 55

doc/indevs.texi

@@ -483,6 +483,21 @@ ffplay -f lavfi "movie=test.avi[out0];amovie=test.wav[out1]"
 
 @end itemize
 
+@section libcdio
+
+Audio-CD input device based on cdio.
+
+To enable this input device during configuration you need libcdio
+installed on your system.
+
+This device allows playing and grabbing from an Audio-CD.
+
+For example to copy with @command{ffmpeg} the entire Audio-CD in /dev/sr0,
+you may run the command:
+@example
+ffmpeg -f libcdio -i /dev/sr0 cd.wav
+@end example
+
 @section libdc1394
 
 IIDC1394 input device, based on libdc1394 and libraw1394.

libavcodec/Makefile

@@ -15,6 +15,7 @@ HEADERS = avcodec.h                                                     \
 
 OBJS = allcodecs.o                                                      \
        audioconvert.o                                                   \
+       avdct.o                                                          \
        avpacket.o                                                       \
        avpicture.o                                                      \
        bitstream.o                                                      \

libavcodec/avdct.c

@@ -0,0 +1,124 @@
+/*
+ * Copyright (c) 2014 Michael Niedermayer <michaelni@gmx.at>
+ *
+ * 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
+ */
+
+#include "avcodec.h"
+#include "idctdsp.h"
+#include "fdctdsp.h"
+#include "avdct.h"
+
+#define OFFSET(x) offsetof(AVDCT,x)
+#define DEFAULT 0 //should be NAN but it does not work as it is not a constant in glibc as required by ANSI/ISO C
+//these names are too long to be readable
+#define V AV_OPT_FLAG_VIDEO_PARAM
+#define A AV_OPT_FLAG_AUDIO_PARAM
+#define E AV_OPT_FLAG_ENCODING_PARAM
+#define D AV_OPT_FLAG_DECODING_PARAM
+
+static const AVOption avdct_options[] = {
+{"dct", "DCT algorithm", OFFSET(dct_algo), AV_OPT_TYPE_INT, {.i64 = DEFAULT }, 0, INT_MAX, V|E, "dct"},
+{"auto", "autoselect a good one (default)", 0, AV_OPT_TYPE_CONST, {.i64 = FF_DCT_AUTO }, INT_MIN, INT_MAX, V|E, "dct"},
+{"fastint", "fast integer (experimental / for debugging)", 0, AV_OPT_TYPE_CONST, {.i64 = FF_DCT_FASTINT }, INT_MIN, INT_MAX, V|E, "dct"},
+{"int", "accurate integer", 0, AV_OPT_TYPE_CONST, {.i64 = FF_DCT_INT }, INT_MIN, INT_MAX, V|E, "dct"},
+{"mmx", "experimental / for debugging", 0, AV_OPT_TYPE_CONST, {.i64 = FF_DCT_MMX }, INT_MIN, INT_MAX, V|E, "dct"},
+{"altivec", "experimental / for debugging", 0, AV_OPT_TYPE_CONST, {.i64 = FF_DCT_ALTIVEC }, INT_MIN, INT_MAX, V|E, "dct"},
+{"faan", "floating point AAN DCT (experimental / for debugging)", 0, AV_OPT_TYPE_CONST, {.i64 = FF_DCT_FAAN }, INT_MIN, INT_MAX, V|E, "dct"},
+
+{"idct", "select IDCT implementation", OFFSET(idct_algo), AV_OPT_TYPE_INT, {.i64 = DEFAULT }, 0, INT_MAX, V|E|D, "idct"},
+{"auto", "autoselect a good one (default)", 0, AV_OPT_TYPE_CONST, {.i64 = FF_IDCT_AUTO }, INT_MIN, INT_MAX, V|E|D, "idct"},
+{"int", "experimental / for debugging", 0, AV_OPT_TYPE_CONST, {.i64 = FF_IDCT_INT }, INT_MIN, INT_MAX, V|E|D, "idct"},
+{"simple", "experimental / for debugging", 0, AV_OPT_TYPE_CONST, {.i64 = FF_IDCT_SIMPLE }, INT_MIN, INT_MAX, V|E|D, "idct"},
+{"simplemmx", "experimental / for debugging", 0, AV_OPT_TYPE_CONST, {.i64 = FF_IDCT_SIMPLEMMX }, INT_MIN, INT_MAX, V|E|D, "idct"},
+{"arm", "experimental / for debugging", 0, AV_OPT_TYPE_CONST, {.i64 = FF_IDCT_ARM }, INT_MIN, INT_MAX, V|E|D, "idct"},
+{"altivec", "experimental / for debugging", 0, AV_OPT_TYPE_CONST, {.i64 = FF_IDCT_ALTIVEC }, INT_MIN, INT_MAX, V|E|D, "idct"},
+#if FF_API_ARCH_SH4
+{"sh4", "experimental / for debugging", 0, AV_OPT_TYPE_CONST, {.i64 = FF_IDCT_SH4 }, INT_MIN, INT_MAX, V|E|D, "idct"},
+#endif
+{"simplearm", "experimental / for debugging", 0, AV_OPT_TYPE_CONST, {.i64 = FF_IDCT_SIMPLEARM }, INT_MIN, INT_MAX, V|E|D, "idct"},
+{"simplearmv5te", "experimental / for debugging", 0, AV_OPT_TYPE_CONST, {.i64 = FF_IDCT_SIMPLEARMV5TE }, INT_MIN, INT_MAX, V|E|D, "idct"},
+{"simplearmv6", "experimental / for debugging", 0, AV_OPT_TYPE_CONST, {.i64 = FF_IDCT_SIMPLEARMV6 }, INT_MIN, INT_MAX, V|E|D, "idct"},
+{"simpleneon", "experimental / for debugging", 0, AV_OPT_TYPE_CONST, {.i64 = FF_IDCT_SIMPLENEON }, INT_MIN, INT_MAX, V|E|D, "idct"},
+#if FF_API_ARCH_ALPHA
+{"simplealpha", "experimental / for debugging", 0, AV_OPT_TYPE_CONST, {.i64 = FF_IDCT_SIMPLEALPHA }, INT_MIN, INT_MAX, V|E|D, "idct"},
+#endif
+{"ipp", "experimental / for debugging", 0, AV_OPT_TYPE_CONST, {.i64 = FF_IDCT_IPP }, INT_MIN, INT_MAX, V|E|D, "idct"},
+{"xvidmmx", "experimental / for debugging", 0, AV_OPT_TYPE_CONST, {.i64 = FF_IDCT_XVIDMMX }, INT_MIN, INT_MAX, V|E|D, "idct"},
+{"faani", "floating point AAN IDCT (experimental / for debugging)", 0, AV_OPT_TYPE_CONST, {.i64 = FF_IDCT_FAAN }, INT_MIN, INT_MAX, V|D|E, "idct"},
+{"simpleauto", "experimental / for debugging", 0, AV_OPT_TYPE_CONST, {.i64 = FF_IDCT_SIMPLEAUTO }, INT_MIN, INT_MAX, V|E|D, "idct"},
+{NULL},
+};
+
+static const AVClass avdct_class = {
+    .class_name              = "AVDCT",
+    .option                  = avdct_options,
+    .version                 = LIBAVUTIL_VERSION_INT,
+};
+
+const AVClass *avcodec_dct_get_class(void)
+{
+    return &avdct_class;
+}
+
+AVDCT *avcodec_dct_alloc(void)
+{
+    AVDCT *dsp = av_mallocz(sizeof(AVDCT));
+
+    if (!dsp)
+        return NULL;
+
+    dsp->av_class = &avdct_class;
+    av_opt_set_defaults(dsp);
+
+    return dsp;
+}
+
+int avcodec_dct_init(AVDCT *dsp)
+{
+    AVCodecContext *avctx = avcodec_alloc_context3(NULL);
+
+    if (!avctx)
+        return AVERROR(ENOMEM);
+
+    avctx->idct_algo = dsp->idct_algo;
+    avctx->dct_algo  = dsp->dct_algo;
+
+#define COPY(src, name) memcpy(&dsp->name, &src.name, sizeof(dsp->name))
+
+#if CONFIG_IDCTDSP
+    {
+        IDCTDSPContext idsp;
+        ff_idctdsp_init(&idsp, avctx);
+        COPY(idsp, idct);
+        COPY(idsp, idct_permutation);
+    }
+#endif
+
+#if CONFIG_FDCTDSP
+    {
+        FDCTDSPContext fdsp;
+        ff_fdctdsp_init(&fdsp, avctx);
+        COPY(fdsp, fdct);
+    }
+#endif
+
+    avcodec_close(avctx);
+    av_free(avctx);
+
+    return 0;
+}

libavcodec/avdct.h

@@ -0,0 +1,78 @@
+/*
+ * 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
+ */
+
+#ifndef AVCODEC_AVDCT_H
+#define AVCODEC_AVDCT_H
+
+#include "libavutil/opt.h"
+
+/**
+ * AVDCT context.
+ * @note function pointers can be NULL if the specific features have been
+ *       disabled at build time.
+ */
+typedef struct AVDCT {
+    const AVClass *av_class;
+
+    void (*idct)(int16_t *block /* align 16 */);
+
+    /**
+     * IDCT input permutation.
+     * Several optimized IDCTs need a permutated input (relative to the
+     * normal order of the reference IDCT).
+     * This permutation must be performed before the idct_put/add.
+     * Note, normally this can be merged with the zigzag/alternate scan<br>
+     * An example to avoid confusion:
+     * - (->decode coeffs -> zigzag reorder -> dequant -> reference IDCT -> ...)
+     * - (x -> reference DCT -> reference IDCT -> x)
+     * - (x -> reference DCT -> simple_mmx_perm = idct_permutation
+     *    -> simple_idct_mmx -> x)
+     * - (-> decode coeffs -> zigzag reorder -> simple_mmx_perm -> dequant
+     *    -> simple_idct_mmx -> ...)
+     */
+    uint8_t idct_permutation[64];
+
+    void (*fdct)(int16_t *block /* align 16 */);
+
+
+    /**
+     * DCT algorithm.
+     * must use AVOptions to set this field.
+     */
+    int dct_algo;
+
+    /**
+     * IDCT algorithm.
+     * must use AVOptions to set this field.
+     */
+    int idct_algo;
+} AVDCT;
+
+/**
+ * Allocates a AVDCT context.
+ * This needs to be initialized with avcodec_dct_init() after optionally
+ * configuring it with AVOptions.
+ *
+ * To free it use av_free()
+ */
+AVDCT *avcodec_dct_alloc(void);
+int avcodec_dct_init(AVDCT *);
+
+const AVClass *avcodec_dct_get_class(void);
+
+#endif /* AVCODEC_AVDCT_H */

libavcodec/g2meet.c

@@ -90,6 +90,7 @@ typedef struct G2MContext {
 
     int        compression;
     int        width, height, bpp;
+    int        orig_width, orig_height;
     int        tile_width, tile_height;
     int        tiles_x, tiles_y, tile_x, tile_y;
 
@@ -712,8 +713,8 @@ static int g2m_decode_frame(AVCodecContext *avctx, void *data,
             }
             c->width  = bytestream2_get_be32(&bc);
             c->height = bytestream2_get_be32(&bc);
-            if (c->width  < 16 || c->width  > avctx->width ||
+            if (c->width  < 16 || c->width  > c->orig_width ||
-                c->height < 16 || c->height > avctx->height) {
+                c->height < 16 || c->height > c->orig_height) {
                 av_log(avctx, AV_LOG_ERROR,
                        "Invalid frame dimensions %dx%d\n",
                        c->width, c->height);
@@ -882,6 +883,10 @@ static av_cold int g2m_decode_init(AVCodecContext *avctx)
 
     avctx->pix_fmt = AV_PIX_FMT_RGB24;
 
+    // store original sizes and check against those if resize happens
+    c->orig_width  = avctx->width;
+    c->orig_height = avctx->height;
+
     return 0;
 }
 

libavcodec/x86/hevc_mc.asm

@@ -342,7 +342,7 @@ QPEL_TABLE 10, 4, w, sse4
 
 %macro LOOP_END 4
     lea              %1q, [%1q+2*%2q]            ; dst += dststride
-    lea              %3q, [%3q+  %4q]            ; src += srcstride
+    add              %3q, %4q                    ; src += srcstride
     dec          heightd                         ; cmp height
     jnz               .loop                      ; height loop
 %endmacro

libavdevice/pulse_audio_enc.c

@@ -681,7 +681,7 @@ static int pulse_write_frame(AVFormatContext *h, int stream_index,
                AVERROR(EINVAL) : 0;
 
     pkt.data     = (*frame)->data[0];
-    pkt.size     = (*frame)->nb_samples * av_get_bytes_per_sample((*frame)->format) * (*frame)->channels;
+    pkt.size     = (*frame)->nb_samples * av_get_bytes_per_sample((*frame)->format) * av_frame_get_channels(*frame);
     pkt.dts      = (*frame)->pkt_dts;
     pkt.duration = av_frame_get_pkt_duration(*frame);
     return pulse_write_packet(h, &pkt);

libavfilter/vf_spp.c

@@ -233,9 +233,9 @@ static void filter(SPPContext *p, uint8_t *dst, uint8_t *src,
                 const int y1 = y + offset[i + count - 1][1];
                 const int index = x1 + y1*linesize;
                 p->pdsp.get_pixels(block, p->src + index, linesize);
-                p->fdsp.fdct(block);
+                p->dct->fdct(block);
-                p->requantize(block2, block, qp, p->idsp.idct_permutation);
+                p->requantize(block2, block, qp, p->dct->idct_permutation);
-                p->idsp.idct(block2);
+                p->dct->idct(block2);
                 add_block(p->temp + index, linesize, block2);
             }
         }
@@ -378,11 +378,11 @@ static av_cold int init(AVFilterContext *ctx)
     SPPContext *spp = ctx->priv;
 
     spp->avctx = avcodec_alloc_context3(NULL);
-    if (!spp->avctx)
+    spp->dct = avcodec_dct_alloc();
+    if (!spp->avctx || !spp->dct)
         return AVERROR(ENOMEM);
-    ff_idctdsp_init(&spp->idsp, spp->avctx);
-    ff_fdctdsp_init(&spp->fdsp, spp->avctx);
     ff_pixblockdsp_init(&spp->pdsp, spp->avctx);
+    avcodec_dct_init(spp->dct);
     spp->store_slice = store_slice_c;
     switch (spp->mode) {
     case MODE_HARD: spp->requantize = hardthresh_c; break;
@@ -403,6 +403,7 @@ static av_cold void uninit(AVFilterContext *ctx)
         avcodec_close(spp->avctx);
         av_freep(&spp->avctx);
     }
+    av_freep(&spp->dct);
     av_freep(&spp->non_b_qp_table);
 }
 

libavfilter/vf_spp.h

@@ -24,8 +24,7 @@
 
 #include "libavcodec/avcodec.h"
 #include "libavcodec/pixblockdsp.h"
-#include "libavcodec/idctdsp.h"
+#include "libavcodec/avdct.h"
-#include "libavcodec/fdctdsp.h"
 #include "avfilter.h"
 
 #define MAX_LEVEL 6 /* quality levels */
@@ -41,9 +40,8 @@ typedef struct {
     uint8_t *src;
     int16_t *temp;
     AVCodecContext *avctx;
-    IDCTDSPContext idsp;
-    FDCTDSPContext fdsp;
     PixblockDSPContext pdsp;
+    AVDCT *dct;
     int8_t *non_b_qp_table;
     int non_b_qp_alloc_size;
     int use_bframe_qp;

libavformat/dv.c

@@ -72,30 +72,33 @@ static inline uint16_t dv_audio_12to16(uint16_t sample)
     return result;
 }
 
-/*
- * This is the dumbest implementation of all -- it simply looks at
- * a fixed offset and if pack isn't there -- fails. We might want
- * to have a fallback mechanism for complete search of missing packs.
- */
 static const uint8_t *dv_extract_pack(uint8_t *frame, enum dv_pack_type t)
 {
     int offs;
+    int c;
 
-    switch (t) {
+    for (c = 0; c < 10; c++) {
-    case dv_audio_source:
+        switch (t) {
-        offs = (80 * 6 + 80 * 16 * 3 + 3);
+        case dv_audio_source:
-        break;
+            if (c&1)    offs = (80 * 6 + 80 * 16 * 0 + 3 + c*12000);
-    case dv_audio_control:
+            else        offs = (80 * 6 + 80 * 16 * 3 + 3 + c*12000);
-        offs = (80 * 6 + 80 * 16 * 4 + 3);
+            break;
-        break;
+        case dv_audio_control:
-    case dv_video_control:
+            if (c&1)    offs = (80 * 6 + 80 * 16 * 1 + 3 + c*12000);
-        offs = (80 * 5 + 48 + 5);
+            else        offs = (80 * 6 + 80 * 16 * 4 + 3 + c*12000);
-        break;
+            break;
-    case dv_timecode:
+        case dv_video_control:
-        offs = (80*1 + 3 + 3);
+            if (c&1)    offs = (80 * 3 + 8      + c*12000);
-        break;
+            else        offs = (80 * 5 + 48 + 5 + c*12000);
-    default:
+            break;
-        return NULL;
+        case dv_timecode:
+            offs = (80*1 + 3 + 3);
+            break;
+        default:
+            return NULL;
+        }
+        if (frame[offs] == t)
+            break;
     }
 
     return frame[offs] == t ? &frame[offs] : NULL;

libavformat/img2dec.c

@@ -666,6 +666,7 @@ static const AVClass imgname ## _class = {\
 };\
 AVInputFormat ff_image_ ## imgname ## _pipe_demuxer = {\
     .name           = AV_STRINGIFY(imgname) "_pipe",\
+    .long_name      = NULL_IF_CONFIG_SMALL("piped " AV_STRINGIFY(imgname) " sequence"),\
     .priv_data_size = sizeof(VideoDemuxData),\
     .read_probe     = imgname ## _probe,\
     .read_header    = ff_img_read_header,\

tests/fate.sh

@@ -19,6 +19,8 @@ test -n "$slot"    || die "slot not specified"
 test -n "$repo"    || die "repo not specified"
 test -d "$samples" || die "samples location not specified"
 
+: ${branch:=master}
+
 lock(){
     lock=$1/fate.lock
     (set -C; exec >$lock) 2>/dev/null || return
@@ -28,14 +30,14 @@ lock(){
 checkout(){
     case "$repo" in
         file:*|/*) src="${repo#file:}"      ;;
-        git:*)     git clone --quiet "$repo" "$src" ;;
+        git:*)     git clone --quiet --branch "$branch" "$repo" "$src" ;;
     esac
 }
 
 update()(
     cd ${src} || return
     case "$repo" in
-        git:*) git fetch --force && git reset --hard FETCH_HEAD ;;
+        git:*) git fetch --force && git reset --hard "origin/$branch" ;;
     esac
 )
 
@@ -82,7 +84,9 @@ clean(){
 report(){
     date=$(date -u +%Y%m%d%H%M%S)
     echo "fate:0:${date}:${slot}:${version}:$1:$2:${comment}" >report
-    cat ${build}/config.fate ${build}/tests/data/fate/*.rep >>report
+#    echo "fate:1:${date}:${slot}:${version}:$1:$2:${branch}:${comment}" >report
+    cat ${build}/config.fate >>report
+    cat ${build}/tests/data/fate/*.rep >>report || for i in ${build}/tests/data/fate/*.rep ; do cat "$i" >>report ; done
     test -n "$fate_recv" && $tar report *.log | gzip | $fate_recv
 }
 

version.sh

@@ -4,7 +4,11 @@
 
 # check for git short hash
 if ! test "$revision"; then
-    revision=$(cd "$1" && git describe --tags --match N 2> /dev/null)
+    if (cd "$1" && grep git RELEASE 2> /dev/null >/dev/null) ; then
+        revision=$(cd "$1" && git describe --tags --match N 2> /dev/null)
+    else
+        revision=$(cd "$1" && git describe --tags --always 2> /dev/null)
+    fi
 fi
 
 # Shallow Git clones (--depth) do not have the N tag: