i18n_tools.py•8.53 kB
import argparse
import logging
import os
import subprocess
from pathlib import Path
import polib
logger = logging.getLogger(__name__)
def find_py_files(base_dir: str) -> str:
"""查找所有 .py 文件.
Args:
base_dir: 项目根目录路径
Returns:
list[str]: 所有 .py 文件的路径列表
"""
try:
base_path = Path(base_dir)
if not Path.exists(base_path):
msg = f"目录 {base_dir} 不存在"
raise FileNotFoundError(msg) # noqa: TRY301
py_files = [str(file) for file in base_path.rglob("*.py")]
if not py_files:
msg = f"在 {base_dir} 中未找到任何.py 文件"
raise FileNotFoundError(msg) # noqa: TRY301
logger.info("找到 %s 个 .py 文件", len(py_files))
except Exception as e:
msg = f"在 {base_dir} 中未找到任何.py 文件"
raise FileNotFoundError(msg) from e
return py_files
def extract_messages(py_files: str, output_dir: str, domain: str = "messages") -> str:
"""使用 xgettext 提取翻译字符串."""
if not py_files:
msg = "没有.py 文件可供提取"
raise Exception(msg) # noqa: TRY002
try:
Path.mkdir(output_dir, exist_ok=True)
pot_file = str(Path(output_dir) / f"{domain}.pot")
cmd = [
"xgettext",
"--from-code=UTF-8",
"-L",
"Python",
"-o",
pot_file,
"--package-name=moonshot_tools",
"--msgid-bugs-address=your@email.com",
"--add-comments=TRANSLATORS:",
"--keyword=_",
*py_files,
]
subprocess.run(cmd, check=True, capture_output=True, text=True) # noqa: S603
logger.info("已生成 POT 文件: %s", pot_file)
except subprocess.CalledProcessError as e:
msg = f"xgettext 执行失败: {e}"
raise Exception(msg) from e # noqa: TRY002
except Exception as e:
msg = f"提取翻译字符串时出错: {e}"
raise Exception(msg) from e # noqa: TRY002
return pot_file
def create_po_files(pot_file: str, locale_dir: str, languages: str) -> str:
"""从 POT 文件创建新的 PO 文件.
Args:
pot_file: POT 文件路径
locale_dir: locale 目录路径
languages: 语言代码列表,例如 ['zh_CN', 'en_US']
"""
try:
for lang in languages:
# 创建语言目录
lang_dir = Path(locale_dir) / lang / "LC_MESSAGES"
lang_dir.mkdir(parents=True, exist_ok=True)
# 生成 PO 文件
po_file = Path(locale_dir) / lang / "LC_MESSAGES" / "messages.po"
if not Path.exists(po_file):
cmd = [
"msginit",
"--input=" + pot_file,
"--locale=" + lang,
"--output-file=" + po_file,
"--no-translator",
]
# 移除 text=True,只保留 capture_output=True
# 验证命令参数是否安全
if not all(isinstance(arg, str) for arg in cmd):
msg = "命令参数必须都是字符串类型"
raise ValueError(msg) # noqa: TRY301
subprocess.run(cmd, check=True, capture_output=True, text=True) # noqa: S603
logger.info("已为语言 {%s} 创建新的 PO 文件: {%s}", lang, po_file)
else:
logger.warning("PO 文件已存在: %s", po_file)
except subprocess.CalledProcessError as e:
# 使用 errors='replace' 来处理无法解码的字符
stderr = e.stderr.decode("utf-8", errors="replace")
msg = f"msginit 执行失败: {stderr}"
raise Exception(msg) from e # noqa: TRY002
except Exception as e:
msg = f"创建 PO 文件时出错: {e}"
raise Exception(msg) from e # noqa: TRY002
def merge_po_files(pot_file: str, locale_dir: str) -> None: # noqa: C901
"""使用 polib 合并 .pot 和现有 .po 文件,保留已有翻译."""
try:
# 加载新生成的 .pot 文件
pot = polib.pofile(pot_file)
locale_path = Path(locale_dir)
po_files = list(locale_path.rglob("*.po"))
if not po_files:
logger.warning("在 %s 中未找到 .po 文件,跳过合并", locale_dir)
return
for po_file_path in po_files:
if Path.exists(po_file_path):
# 加载现有的 .po 文件
po = polib.pofile(str(po_file_path), encoding="utf-8")
logger.info("正在合并现有 PO 文件: %s", po_file_path)
else:
# 如果 .po 文件不存在,创建一个新的
po = polib.POFile()
po.metadata = pot.metadata.copy() # 复制 .pot 的元信息
logger.info("创建新的 PO 文件: %s", po_file_path)
# 合并逻辑:保留已有翻译,添加新条目,标记废弃条目
existing_entries = {entry.msgid: entry for entry in po}
for pot_entry in pot:
if pot_entry.msgid in existing_entries:
# 如果已有翻译,保留它
existing_entry = existing_entries[pot_entry.msgid]
if existing_entry.msgstr:
pot_entry.msgstr = existing_entry.msgstr
# 将新条目添加到 .po 文件(如果之前不存在)
if pot_entry.msgid not in existing_entries:
po.append(pot_entry)
# 标记废弃条目(在 .po 中存在但 .pot 中已删除)
for msgid, entry in existing_entries.items():
if msgid not in {e.msgid for e in pot}:
entry.obsolete = True # 标记为废弃
# 保存更新后的 .po 文件
po.save(str(po_file_path))
logger.info("已更新 PO 文件: %s", po_file_path)
except Exception as e:
msg = f"合并 .po 文件时出错: {e}"
raise Exception(msg) from e # noqa: TRY002
def compile_po_files(locale_dir: str) -> None:
"""编译所有 .po 文件为 .mo 文件."""
try:
locale_path = Path(locale_dir)
if not Path.exists(locale_path):
msg = f"locale 目录 {locale_dir} 不存在,跳过编译"
raise FileNotFoundError(msg) # noqa: TRY301
po_files = list(locale_path.rglob("*.po"))
if not po_files:
msg = f"在 {locale_dir} 中未找到 .po 文件"
raise FileNotFoundError(msg) # noqa: TRY301
for po_file in po_files:
# 直接使用 po 文件所在的目录
mo_dir = po_file.parent
mo_file = mo_dir / (po_file.stem + ".mo")
Path.mkdir(mo_dir, exist_ok=True)
cmd = ["msgfmt", "-o", str(mo_file), str(po_file)]
subprocess.run(cmd, check=True, capture_output=True, text=True) # noqa: S603
logger.info("已生成 MO 文件: %s", mo_file)
except subprocess.CalledProcessError as e:
msg = f"msgfmt 执行失败: {e}"
raise Exception(msg) from e # noqa: TRY002
except Exception as e:
msg = f"编译.po 文件时出错: {e}"
raise Exception(msg) from e # noqa: TRY002
def main() -> None:
parser = argparse.ArgumentParser(description="国际化工具：提取、更新和编译翻译文件")
parser.add_argument(
"--base-dir",
default=os.getenv("MOONSHOT_SRC_DIR", str(Path(__file__).parent.parent)),
help="项目源代码根目录路径，默认为当前脚本的上级目录",
)
parser.add_argument(
"--languages",
default="zh_CN,en_US",
help="需要支持的语言代码列表,用逗号分隔,例如:zh_CN,en_US",
)
args = parser.parse_args()
base_dir = args.base_dir
locale_dir = Path(base_dir) / "locale"
languages = args.languages.split(",")
# 1. 查找所有 .py 文件
py_files = find_py_files(base_dir)
# 2. 提取翻译字符串
pot_file = extract_messages(py_files, locale_dir)
# 3. 创建新的 PO 文件(如果不存在)
create_po_files(pot_file, locale_dir, languages)
# 4. 合并 .pot 和现有 .po 文件
merge_po_files(pot_file, locale_dir)
# 5. 编译 .po 文件为 .mo 文件
compile_po_files(locale_dir)
logger.info("翻译文件处理流程完成")
if __name__ == "__main__":
main()