yes

2026-05-06 23:44:13 +02:00
parent a5f787d80f
commit fd7038b853
1574 changed files with 365439 additions and 0 deletions
@@ -0,0 +1,144 @@
+# -*- coding: utf-8 -*-
+
+#-------------------------------------------------------------------------
+#
+# Copyright (C) 2023 The Khronos Group Inc
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+#-------------------------------------------------------------------------
+#
+# Generate checksums from a sequence of YV12 YUV frames as produced by
+# the Fraunhofer reference decoders.
+#
+#  The reference checksums were generated using the Fluster testing
+#  framework (git@github.com:fluendo/fluster.git @
+#  fd2a3aec596ba5437b4c3364111938531a4b5f92) to build the Fraunhofer
+#  reference decoders,
+#
+#  ~/src/fluster $ make h26[45]_reference_decoder
+#
+#  Then, to generate references for H.264 test content,
+#
+#  $ cd ~/src/vk-gl-cts/external/vulkancts/data/vulkan/video/
+#  $ ~/src/fluster/contrib/JM/bin/umake/gcc-12.2/x86_64/release/ldecod -s -i clip-a.h264 -o clip-a.out
+#  $ python frame_checksums.py  clip-a.out  176 144 ; rm clip-a.out
+#
+#  To generate the H.265 YUV files, change the reference decoder line above to,
+#
+#  $ ~/src/fluster/contrib/HM/bin/umake/gcc-12.2/x86_64/release/TAppDecoder -b jellyfish-250-mbps-4k-uhd-GOB-IPB13.h265 -o jellyfish265.out
+#
+#  The h264_resolution_change example was specially handled, since
+#  it's actually a non-conformant bitstream according to the reference
+#  decoders. There are two streams of different resolution
+#  concatenated together in the test vector. They were manually cut
+#  out and each independent stream was passed through the above
+#  process to generate the checksums.
+#
+
+import os
+import sys
+import hashlib
+
+def print_help():
+    help_text = """Usage: frame_checksums.py [OPTIONS] yuvFile width height
+
+Generate MD5 checksums from YUV frames in C array format.
+
+Arguments:
+  yuvFile              YUV file to process (YV12 format)
+  width                Frame width in pixels
+  height               Frame height in pixels
+
+Options:
+  -a NAME              C array name (default: frameChecksums)
+  -n NUM               Process only first NUM frames
+  -h, --help           Show this help message and exit
+
+Examples:
+  python frame_checksums.py clip-a.out 176 144
+  python frame_checksums.py -n 30 -a clipAChecksums clip-a.out 176 144
+"""
+    print(help_text)
+    sys.exit(0)
+
+# Parse command line arguments
+yuv_filename = None
+width = None
+height = None
+array_name = "frameChecksums"
+frame_limit = None
+
+i = 1
+while i < len(sys.argv):
+    if sys.argv[i] in ('-h', '--help'):
+        print_help()
+    elif sys.argv[i] == '-n' and i + 1 < len(sys.argv):
+        frame_limit = int(sys.argv[i + 1])
+        i += 2
+    elif sys.argv[i] == '-a' and i + 1 < len(sys.argv):
+        array_name = sys.argv[i + 1]
+        i += 2
+    elif yuv_filename is None:
+        yuv_filename = sys.argv[i]
+        i += 1
+    elif width is None:
+        width = int(sys.argv[i])
+        i += 1
+    elif height is None:
+        height = int(sys.argv[i])
+        i += 1
+    else:
+        i += 1
+
+if yuv_filename is None or width is None or height is None:
+    print("Usage: frame_checksums.py [OPTIONS] yuvFile width height")
+    print("Use -h or --help for more information")
+    sys.exit(1)
+
+def checksum(bs: bytes) -> str:
+    md5 = hashlib.md5()
+    md5.update(bs)
+    return md5.hexdigest()
+
+file_size = os.path.getsize(yuv_filename)
+
+# This assumes the YCrCb format is YV12, as produced by the Fraunhofer
+# reference decoders.
+frame_size = int(width * height * 1.5)
+n_frames = file_size // frame_size
+
+# Apply frame limit if specified
+if frame_limit is not None:
+    n_frames = min(n_frames, frame_limit)
+
+md5_hashes = []
+with open(yuv_filename, 'rb') as f:
+    print("Computing checksums for ", n_frames, " frames", file=sys.stderr)
+    for frame in range(n_frames):
+        md5_hashes.append(checksum(f.read(frame_size)))
+
+# Output in C array format (3 hashes per line)
+print(f"static const char *{array_name}[{n_frames}] = {{")
+
+for i in range(0, n_frames, 3):
+    # Get up to 3 hashes for this line
+    line_hashes = md5_hashes[i:i+3]
+    formatted_hashes = ', '.join(f'"{h}"' for h in line_hashes)
+
+    # Add comma after line unless it's the last line
+    line_end = ',' if i + 3 < n_frames else ''
+    print(f"    {formatted_hashes}{line_end}")
+
+print("};")
+print("", file=sys.stderr)