Update 视频编码器数据库.vb

[Howard20181]

从 ffmpeg 命令行生成编码器数据库
需要验证

ffmpeg_encoders_info.py

import subprocess
import re
import json

def get_video_encoders():
    result = subprocess.run(['ffmpeg', '-hide_banner', '-encoders'], capture_output=True, text=True)
    encoders = []
    for line in result.stdout.splitlines():
        if line.startswith(' V') and '=' not in line:
            parts = line.split()
            if len(parts) > 1:
                name = parts[1]
                # 排除image/gif/webp/apng等非视频编码器
                lower = name.lower()
                if any(x in lower for x in ["image", "apng", "webp", "gif", "tiff", "png", "jpeg", "jpg", "bmp", "pbm", "pgm", "ppm", "pam", "sgi", "sunrast", "xbm", "xface", "xwd", "pcx", "ico", "exr", "jp2", "j2k", "jpeg2000", "jpegxl", "jxl", "tga", "dpx", "fits", "pic", "qoi", "tiff", "tiff12", "tiff24", "tiff32", "tiff48", "tiff64", "tiff8", "tiff16", "tiff32", "tiff64", "tiffle", "tiffbe", "tiffle16", "tiffbe16", "tiffle32", "tiffbe32", "tiffle64", "tiffbe64"]):
                    continue
                encoders.append(name)
    return encoders

def parse_options(lines, key):
    # 只要有参数行就向下查找缩进行提取可选值
    options = []
    param_pattern = re.compile(rf'-{key}\s+')
    for i, line in enumerate(lines):
        if param_pattern.search(line):
            j = i + 1
            while j < len(lines):
                next_line = lines[j]
                # 如果遇到下一个参数行（以-开头）或空行则停止
                if next_line.strip() == '' or next_line.lstrip().startswith('-'):
                    break
                # 只处理缩进行
                if re.match(r'\s+\S+\s+\S+', next_line):
                    opt = next_line.strip().split()[0]
                    options.append(opt)
                j += 1
            # 如果没有缩进行可选值，且参数行为<int>类型且有(from X to Y)，则自动生成整数序列
            if not options:
                m = re.search(r'<int>.*\(from\s+(-?\d+)\s+to\s+(-?\d+)\)', line)
                if m:
                    start, end = int(m.group(1)), int(m.group(2))
                    if start <= 0:
                        start = 1
                    options = [str(i) for i in range(start, end + 1)]
            break
    return options

def get_encoder_info(encoder):
    cmd = ['ffmpeg', '-hide_banner', '-h', f'encoder={encoder}']
    result = subprocess.run(cmd, capture_output=True, text=True)
    lines = result.stdout.splitlines()
    # 提取 pixel formats
    pix_fmt = []
    for line in lines:
        if 'Supported pixel formats:' in line:
            pix_fmt = line.split(':', 1)[1].strip().split()
            break
    # 提取 profile/preset/tune
    def get_option(key):
        opts = parse_options(lines, key)
        # 排除unknown和default字符串
        opts = [x for x in opts if x not in ('unknown', 'default')]
        # nvenc编码器preset仅保留pX级别
        if encoder.lower().endswith('nvenc') and key == 'preset':
            opts = [x for x in opts if re.fullmatch(r'p\d+', x)]
        if opts:
            return opts
        # 检查是否有参数但无可选值
        for line in lines:
            if f'-{key}' in line:
                return []
        return []
    return {
        'preset': get_option('preset'),
        'profile': get_option('profile'),
        'tune': get_option('tune'),
        'pix_fmt': pix_fmt
    }

def main():
    encoders = get_video_encoders()
    all_info = {}
    for enc in encoders:
        all_info[enc] = get_encoder_info(enc)
    with open('ffmpeg_encoders_info.json', 'w', encoding='utf-8') as f:
        json.dump(all_info, f, ensure_ascii=False, indent=2)
    print('已保存到 ffmpeg_encoders_info.json')

if __name__ == '__main__':
    main()

ffmpeg_encoders_db_gen.py

import json, re


def fmt(lst):
    items = [x for x in lst if x]
    return ", ".join(f'"{x}"' for x in items) if items else '""'


def is_int_seq(lst):
    # 判断是否为int序列（允许负数）
    return all(re.fullmatch(r"-?\d+", x) for x in lst if x)


def fmt_preset(lst, enc=None):
    items = [x for x in lst if x]
    if not items:
        return '""'
    # 指定编码器保持json顺序
    if enc and enc in ["av1_amf", "av1_vaapi", "av1_vulkan"]:
        return ", ".join(f'"{x}"' for x in items)
    if is_int_seq(items):
        nums = [x for x in set(items) if re.fullmatch(r"-?\d+", x)]
        if not nums:
            return '""'
        return ", ".join(f'"{x}"' for x in sorted(nums, key=lambda x: int(x)))
    return ", ".join(f'"{x}"' for x in reversed(items)) if items else '""'


# 读取原始数据库，构建查找表
orig_db = {}
try:
    with open("视频编码器数据库.vb", "r", encoding="utf-8") as f:
        vb = f.read()
    for m in re.finditer(
        r'字典\.Add\("([^"]+)", New 视频编码器数据单片结构 With \{(.+?)\}\)', vb, re.S
    ):
        enc = m.group(1)
        block = m.group(2)

        def get_list(field):
            fm = re.search(
                rf"\.{field} = New List\(Of String\) From \{{(.*?)\}}", block, re.S
            )
            if fm:
                return [x.strip().strip('"') for x in fm.group(1).split(",")]
            return [""]

        orig_db[enc] = {
            "preset": get_list("Preset"),
            "profile": get_list("Profile"),
            "tune": get_list("Tune"),
            "pix_fmt": get_list("Pix_fmt"),
        }
except Exception as e:
    orig_db = {}


def fix_field(enc, key, val):
    # 如果json字段为空或不存在，直接用原数据库内容（保持原顺序）
    if (
        not val
        or all(x == "" for x in val)
        or any(x == "(unknown, not listed)" for x in val)
    ):
        if enc in orig_db and orig_db[enc][key]:
            return orig_db[enc][key]
        # 过滤掉unknown
        return [x for x in val if x != "(unknown, not listed)"]
    return val


vb_lines = []
vb_lines.append("Public Class 视频编码器数据库")
vb_lines.append(
    "    Public Shared Property 字典 As New Dictionary(Of String, 视频编码器数据单片结构)"
)
vb_lines.append("")
vb_lines.append("    Public Class 视频编码器数据单片结构")
vb_lines.append("        Public Property Preset As New List(Of String)")
vb_lines.append("        Public Property Profile As New List(Of String)")
vb_lines.append("        Public Property Tune As New List(Of String)")
vb_lines.append("        Public Property Pix_fmt As New List(Of String)")
vb_lines.append("    End Class")
vb_lines.append("")
vb_lines.append("")
vb_lines.append("    Public Shared Sub 初始化()")


# 先加copy项
vb_lines.append('        字典.Add("copy", New 视频编码器数据单片结构 With {')
vb_lines.append('.Preset = New List(Of String) From {""},')
vb_lines.append('.Profile = New List(Of String) From {""},')
vb_lines.append('.Tune = New List(Of String) From {""},')
vb_lines.append('.Pix_fmt = New List(Of String) From {""}')
vb_lines.append("})")

# 读取 ffmpeg_encoders_info.json
with open("ffmpeg_encoders_info.json", "r", encoding="utf-8") as f:
    data = json.load(f)
for enc, info in data.items():
    if enc == "copy":
        continue
    preset = fix_field(enc, "preset", info.get("preset", []))
    profile = fix_field(enc, "profile", info.get("profile", []))
    tune = fix_field(enc, "tune", info.get("tune", []))
    pix_fmt = fix_field(enc, "pix_fmt", info.get("pix_fmt", []))
    vb_lines.append(f'        字典.Add("{enc}", New 视频编码器数据单片结构 With {{')
    vb_lines.append(
        f".Preset = New List(Of String) From {{{fmt_preset(preset, enc)}}},"
    )
    vb_lines.append(f".Profile = New List(Of String) From {{{fmt(profile)}}},")
    vb_lines.append(f".Tune = New List(Of String) From {{{fmt(tune)}}},")
    vb_lines.append(f".Pix_fmt = New List(Of String) From {{{fmt(pix_fmt)}}}")
    vb_lines.append("})")

vb_lines.append("    End Sub")
vb_lines.append("")
vb_lines.append("End Class")

with open("ffmpeg_encoders_db_gen.vb", "w", encoding="utf-8") as f:
    f.write("\n".join(vb_lines))

print("已生成 ffmpeg_encoders_db_gen.vb")

[Howard20181]

不太行，建议关掉
做成查询工具人工补全

[Lake1059]

不太行，建议关掉 做成查询工具人工补全

请继续调整

• 尤其是preset必须是从慢到快
• 未收录的编码器请不要加进来