SourceCodeExtractionTool/xpr解析版/xpr解析.py

import os
import shutil
import re
import xml.etree.ElementTree as ET
from pathlib import Path
from datetime import datetime


# =================================================
# 用户配置区（唯一需要用户关心的地方）
# =================================================
def get_project_root():
    """
    获取用户输入的项目根目录，并验证路径是否存在
    """
    while True:
        project_root = input("请输入你的工程根目录路径: ").strip()
        if not project_root:
            print("路径不能为空，请重新输入。")
            continue
        
        # 处理路径中的引号
        if project_root.startswith('"') and project_root.endswith('"'):
            project_root = project_root[1:-1]
        elif project_root.startswith("'") and project_root.endswith("'"):
            project_root = project_root[1:-1]
        
        # 检查路径是否存在
        if os.path.isdir(project_root):
            return project_root
        else:
            print(f"错误: 路径 '{project_root}' 不存在，请重新输入。")



# =================================================
# Testbench 命名规则
# =================================================
TB_PATTERN = re.compile(r'(^tb_)|(_tb_)|(_tb\.)', re.IGNORECASE)


# =================================================
# 通用工具函数
# =================================================

def find_latest_xpr_in_dir(search_dir):
    """
    在指定目录下查找最新修改的 xpr 文件
    """
    files = [
        os.path.join(search_dir, f)
        for f in os.listdir(search_dir)
        if f.lower().endswith('.xpr')
    ]
    if not files:
        raise FileNotFoundError(f"在目录 {search_dir} 下未找到 .xpr 文件")
    return max(files, key=os.path.getmtime)


def load_extensions(config_path):
    """
    从 config.txt 读取需要的文件后缀集合
    """
    exts = set()
    with open(config_path, "r", encoding="utf-8") as f:
        for line in f:
            line = line.strip().lower()
            if not line or line.startswith("#"):
                continue
            if not line.startswith("."):
                line = "." + line
            exts.add(line)
    return exts


def is_testbench_filename(file_path):
    """
    根据文件名判断是否是 testbench
    """
    return bool(TB_PATTERN.search(Path(file_path).name.lower()))


# =================================================
# 模糊路径解析（核心）
# =================================================

def extract_tail_from_psrcdir(path_str):
    """
    从 $PSRCDIR/xxx/yyy 提取 xxx/yyy
    """
    marker = "$PSRCDIR/"
    if marker not in path_str:
        raise ValueError(f"路径中不包含 $PSRCDIR: {path_str}")

    return path_str.split(marker, 1)[1].replace("\\", "/")


def find_file_by_tail(root_dir, tail_path):
    """
    在 root_dir 下递归搜索，
    找到唯一一个以 tail_path 结尾的文件
    """
    matches = []
    tail_path = tail_path.replace("/", os.sep)

    for dirpath, _, filenames in os.walk(root_dir):
        for name in filenames:
            full_path = os.path.join(dirpath, name)
            if full_path.endswith(tail_path):
                matches.append(full_path)

    if not matches:
        # 如果没找到文件，返回 None 表示未找到
        return None

    if len(matches) > 1:
        # 如果找到多个匹配，返回第一个，但打印警告
        print(f"警告: 发现多个匹配文件，将使用第一个:\n" + "\n".join(matches))
        return matches[0]

    return matches[0]


def resolve_xpr_path(path_str, project_root, xpr_file_path):
    """
    将 xpr 中的 $PSRCDIR/... 解析为真实绝对路径
    优先在与xpr文件同名的.srcs文件夹中查找，如果找不到再在根目录下进行全局匹配
    """
    if "$PSRCDIR/" not in path_str:
        # 如果不包含 $PSRCDIR/ 标记，返回 None
        return None
    
    tail = extract_tail_from_psrcdir(path_str)
    
    # 获取xpr文件名（不含扩展名）来构造.srcs目录名
    xpr_filename = os.path.splitext(os.path.basename(xpr_file_path))[0]
    srcs_dir = os.path.join(project_root, xpr_filename + ".srcs")
    
    # 优先在.srcs目录中查找
    if os.path.isdir(srcs_dir):
        real_path = find_file_by_tail(srcs_dir, tail)
        if real_path is not None:
            return real_path
    
    # 如果在.srcs目录中没找到，则在根目录下进行全局匹配
    return find_file_by_tail(project_root, tail)


# =================================================
# 核心逻辑
# =================================================

def extract_from_xpr_and_copy(xpr_path, config_path, project_root):
    """
    从 xpr 中解析 DesignSrcs + synthesis 文件集合，
    使用模糊匹配解析真实路径并复制
    """

    include_exts = load_extensions(config_path)

    tree = ET.parse(xpr_path)
    root = tree.getroot()

    result_files = []
    tb_files = []
    failed_files = []  # 记录未找到的文件

    # 只遍历 DesignSrcs
    for fileset in root.iter("FileSet"):
        if fileset.get("Type") != "DesignSrcs":
            continue

        for file_elem in fileset.findall("File"):
            file_path = file_elem.get("Path")
            if not file_path:
                continue

            # config.txt 后缀过滤 - 排除不需要的文件
            suffix = Path(file_path).suffix.lower()
            if include_exts and suffix not in include_exts:
                continue

            # testbench 命名判断
            is_tb = is_testbench_filename(file_path)

            fileinfo = file_elem.find("FileInfo")
            if fileinfo is None:
                continue

            used_in_synth = False
            auto_disabled = False
            user_disabled = False

            for attr in fileinfo.findall("Attr"):
                name = attr.get("Name")
                val = (attr.get("Val") or "").lower()

                if name == "UsedIn" and val == "synthesis":
                    used_in_synth = True
                elif name == "AutoDisabled" and val in ("1", "true"):
                    auto_disabled = True
                elif name == "UserDisabled" and val in ("1", "true"):
                    user_disabled = True

            if used_in_synth and not auto_disabled and not user_disabled:
                real_path = resolve_xpr_path(file_path, project_root, xpr_path)
                if real_path is not None:
                    if is_tb:
                        tb_files.append(real_path)
                    else:
                        result_files.append(real_path)
                else:
                    failed_files.append(file_path)  # 记录未找到的文件路径

    print(f"解析并定位到 {len(result_files)} 个非testbench文件")
    print(f"解析并定位到 {len(tb_files)} 个testbench文件")
    if failed_files:
        print(f"未能找到 {len(failed_files)} 个文件")

    # -------------------------------------------------
    # 创建输出目录（秒级时间戳）
    # -------------------------------------------------
    out_dir = datetime.now().strftime("Copy_Files_%Y%m%d_%H%M%S")
    os.makedirs(out_dir)
    print(f"创建输出目录: {out_dir}")

    # -------------------------------------------------
    # 创建tb文件目录（如果存在tb文件）
    # -------------------------------------------------
    tb_out_dir = None
    if tb_files:
        tb_out_dir = os.path.join(out_dir, "tb文件")
        os.makedirs(tb_out_dir)
        print(f"创建testbench输出目录: {tb_out_dir}")

    # -------------------------------------------------
    # 复制非testbench文件
    # -------------------------------------------------
    copied_normal = 0
    skipped_normal = 0

    for src in result_files:
        if not os.path.exists(src):
            print(f"文件不存在，跳过: {src}")
            skipped_normal += 1
            continue

        base = os.path.basename(src)
        dst = os.path.join(out_dir, base)

        # 防止重名覆盖
        if os.path.exists(dst):
            name, ext = os.path.splitext(base)
            idx = 1
            while True:
                new_dst = os.path.join(out_dir, f"{name}_{idx}{ext}")
                if not os.path.exists(new_dst):
                    dst = new_dst
                    break
                idx += 1

        shutil.copy2(src, dst)
        copied_normal += 1

    # -------------------------------------------------
    # 复制testbench文件
    # -------------------------------------------------
    copied_tb = 0
    skipped_tb = 0

    for src in tb_files:
        if not os.path.exists(src):
            print(f"testbench文件不存在，跳过: {src}")
            skipped_tb += 1
            continue

        base = os.path.basename(src)
        dst = os.path.join(tb_out_dir, base) if tb_out_dir else os.path.join(out_dir, base)

        # 防止重名覆盖
        if os.path.exists(dst):
            name, ext = os.path.splitext(base)
            idx = 1
            while True:
                new_dst = os.path.join(tb_out_dir, f"{name}_{idx}{ext}") if tb_out_dir else os.path.join(out_dir, f"{name}_{idx}{ext}")
                if not os.path.exists(new_dst):
                    dst = new_dst
                    break
                idx += 1

        shutil.copy2(src, dst)
        copied_tb += 1

    print("复制完成")
    print(f"  非testbench文件:")
    print(f"    成功复制: {copied_normal}")
    print(f"    跳过文件: {skipped_normal}")

    print(f"  未找到文件: {len(failed_files)}")
    if failed_files:
        print("  未找到的文件列表:")
        for failed_file in failed_files:
            print(f"    - {failed_file}")
    print(f"  输出目录: {os.path.abspath(out_dir)}")
    if tb_files:
        print("\n！！！！！！！！！！！！！！！！！！")
        print("识别到项目中存在testbench文件,请在vivado中检查是否为源代码")
        print("！！！！！！！！！！！！！！！！！！")
    


# =================================================
# main
# =================================================

if __name__ == "__main__":
    try:
        PROJECT_ROOT = get_project_root()
        if not os.path.isdir(PROJECT_ROOT):
            raise RuntimeError(f"PROJECT_ROOT 不存在: {PROJECT_ROOT}")

        xpr_file = find_latest_xpr_in_dir(PROJECT_ROOT)
        config_file = os.path.join(os.getcwd(), "config.txt")

        if not os.path.exists(config_file):
            raise FileNotFoundError("当前目录下未找到 config.txt")

        print(f"使用 xpr 文件   : {xpr_file}")
        print(f"使用 config 文件: {config_file}")
        print(f"项目根目录     : {PROJECT_ROOT}")

        extract_from_xpr_and_copy(xpr_file, config_file, PROJECT_ROOT)

    except Exception as e:
        print("程序执行失败:")
        print(e)

    finally:
        input("\n按回车键关闭窗口...")