feat: build licensed desktop image processor

2026-08-26 07:26:35 +08:00
parent 6ff3d71921
commit 1f31424015
32 changed files with 6515 additions and 2 deletions
@@ -0,0 +1,192 @@
+#!/usr/bin/env python3
+"""从“郑州情趣王一手科技.exe”提取的 JPEG 图片处理逻辑。
+
+处理步骤与原程序一致：
+1. 裁掉 JPEG 底部两行并以 quality=95 重新编码；
+2. 把 SOF 段中的图片高度改回原始高度；
+3. 删除最后一个 JPEG EOI（FF D9）结束标志；
+4. 按指定大小追加随机字节，输出为同名 .file 文件。
+
+注意：输出是刻意构造的非标准 JPEG，部分看图软件可能拒绝打开。
+"""
+
+from __future__ import annotations
+
+import argparse
+import random
+import sys
+from pathlib import Path
+from typing import Iterable
+
+try:
+    from PIL import Image
+except ImportError as exc:
+    raise SystemExit("缺少 Pillow，请先执行：python3 -m pip install pillow") from exc
+
+
+JPEG_EXTENSIONS = {".jpg", ".jpeg", ".jpe", ".jfif"}
+SOF_MARKERS = {
+    0xC0,
+    0xC1,
+    0xC2,
+    0xC3,
+    0xC5,
+    0xC6,
+    0xC7,
+    0xC9,
+    0xCA,
+    0xCB,
+    0xCD,
+    0xCE,
+    0xCF,
+}
+
+
+def find_and_modify_height(data: bytearray, original_height: int) -> bytearray | None:
+    """找到第一个 JPEG SOF 段，把高度字段恢复为裁剪前的高度。"""
+    index = 0
+    length = len(data)
+
+    while index < length - 1:
+        if data[index] != 0xFF:
+            index += 1
+            continue
+
+        marker = data[index + 1]
+        if marker in SOF_MARKERS:
+            height_offset = index + 5
+            if height_offset + 1 >= length:
+                return None
+            data[height_offset] = (original_height >> 8) & 0xFF
+            data[height_offset + 1] = original_height & 0xFF
+            return data
+
+        # SOI 和 EOI 标记没有长度字段。
+        if marker in (0xD8, 0xD9):
+            index += 2
+            continue
+
+        if index + 3 < length:
+            segment_length = (data[index + 2] << 8) | data[index + 3]
+            index += 2 + segment_length
+        else:
+            index += 2
+
+    return None
+
+
+def remove_eoi(data: bytearray) -> tuple[bytearray, bool]:
+    """从末尾向前查找并删除最后一个 FF D9 结束标志。"""
+    for index in range(len(data) - 1, 0, -1):
+        if data[index - 1] == 0xFF and data[index] == 0xD9:
+            return data[: index - 1] + data[index + 1 :], True
+    return data, False
+
+
+def process_single_image(
+    input_file: Path | str,
+    output_file: Path | str,
+    junk_size_mb: float = 1.0,
+) -> Path:
+    """按原程序逻辑处理一张图片，并返回输出路径。"""
+    input_path = Path(input_file)
+    output_path = Path(output_file)
+    temp_path = Path(f"{output_path}.tmp")
+
+    if not 0 <= junk_size_mb <= 100:
+        raise ValueError("junk_size_mb 必须在 0 到 100 之间")
+
+    output_path.parent.mkdir(parents=True, exist_ok=True)
+
+    try:
+        with Image.open(input_path) as image:
+            width, height = image.size
+            if height <= 2:
+                raise ValueError(f"图片高度不足 3 行：{input_path}")
+
+            cropped_image = image.crop((0, 0, width, height - 2))
+            try:
+                cropped_image.save(temp_path, "JPEG", quality=95)
+            finally:
+                cropped_image.close()
+
+        image_data = bytearray(temp_path.read_bytes())
+        modified_data = find_and_modify_height(image_data, height) or image_data
+        modified_data, _ = remove_eoi(modified_data)
+
+        junk_size_bytes = int(junk_size_mb * 1024 * 1024)
+        if junk_size_bytes > 0:
+            # 保留原程序的逐字节伪随机生成方式。
+            modified_data += bytearray(
+                random.getrandbits(8) for _ in range(junk_size_bytes)
+            )
+
+        output_path.write_bytes(modified_data)
+        return output_path
+    finally:
+        if temp_path.exists():
+            temp_path.unlink()
+
+
+def collect_images(paths: Iterable[Path]) -> list[Path]:
+    """收集输入文件，并按原程序支持的扩展名递归扫描目录。"""
+    images: list[Path] = []
+    for path in paths:
+        if path.is_file() and path.suffix.lower() in JPEG_EXTENSIONS:
+            images.append(path)
+        elif path.is_dir():
+            images.extend(
+                candidate
+                for candidate in path.rglob("*")
+                if candidate.is_file()
+                and candidate.suffix.lower() in JPEG_EXTENSIONS
+            )
+        else:
+            print(f"跳过无效路径或非 JPEG 文件：{path}", file=sys.stderr)
+    return images
+
+
+def build_output_path(input_path: Path, output_dir: Path | None) -> Path:
+    """原程序使用图片主文件名加 .file 作为输出名。"""
+    directory = output_dir if output_dir is not None else input_path.parent
+    return directory / f"{input_path.stem}.file"
+
+
+def main() -> int:
+    parser = argparse.ArgumentParser(
+        description="裁掉 JPEG 底部两行、伪造原高度、删除 EOI 并追加随机数据。"
+    )
+    parser.add_argument("inputs", nargs="+", type=Path, help="JPEG 文件或目录")
+    parser.add_argument("-o", "--output-dir", type=Path, help="统一输出目录")
+    parser.add_argument(
+        "--junk-mb",
+        type=float,
+        default=1.0,
+        help="每张图片追加的随机数据大小，范围 0–100 MB，默认 1 MB",
+    )
+    args = parser.parse_args()
+
+    if not 0 <= args.junk_mb <= 100:
+        parser.error("--junk-mb 必须在 0 到 100 之间")
+
+    images = collect_images(args.inputs)
+    if not images:
+        print("没有找到可处理的 JPEG 图片。", file=sys.stderr)
+        return 1
+
+    succeeded = 0
+    for index, input_path in enumerate(images, start=1):
+        output_path = build_output_path(input_path, args.output_dir)
+        try:
+            process_single_image(input_path, output_path, args.junk_mb)
+            succeeded += 1
+            print(f"[{index}/{len(images)}] 完成：{input_path} -> {output_path}")
+        except Exception as exc:
+            print(f"[{index}/{len(images)}] 失败：{input_path} - {exc}", file=sys.stderr)
+
+    print(f"处理完成：总计 {len(images)}，成功 {succeeded}，失败 {len(images) - succeeded}")
+    return 0 if succeeded == len(images) else 2
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())