lanu-android/pull_and_fill_translations.py

import os
import shutil
import subprocess
import xml.dom.minidom as minidom
import xml.etree.ElementTree as ET
from collections import OrderedDict
import re
import requests

SUPPORTED_LANGUAGES = [
    "zh", "en", "ar"
]
TOLGEE_PATH = "./tolgee_json"
TOLGEE_API_URL = "http://47.236.31.243:8090/api"
TOLGEE_API_KEY = "tgpak_hbpxgyltnryguy3lonyxiyrtgruta4rroryhiolooe3q"
PROJECT_ID = "8"
HEADERS = {
    "X-Api-Key": TOLGEE_API_KEY,
    "Content-Type": "application/json"
}


def fetch_all_cdata_keys():
    dataList = [] #key是name，value是id
    page = 0
    page_size = 1000
    while True:
        url = f"{TOLGEE_API_URL}/v2/projects/{PROJECT_ID}/keys?page={page}&size={page_size}"
        r = requests.get(url, headers=HEADERS)
        if r.status_code != 200:
            print(f"❌ 拉取失败: {r.status_code} -> {r.text}")
            break

        data = r.json()
        # print(data)

        dataList.extend(data['_embedded']['keys'])


        if data['page']['totalPages'] == page +1 :  # 最后一页
            break
        page += 1
    print(f"✅ 已拉取 {len(dataList)} 个 Tolgee keys")
    # print(dataList)
    # 只保留包含 CDATA 的 keys,'custom': {'_androidWrapWithCdata': True}
    dataList = [item for item in dataList if
                item['custom'] and '_androidWrapWithCdata' in item['custom'] and item['custom'][
                    '_androidWrapWithCdata']]
    # 提取name
    dataList = [item['name'] for item in dataList if 'name' in item]
    print(f"🔍 找到 {len(dataList)} 个包含 CDATA 的 keys")
    print(dataList)
    return dataList

def run_tolgee_pull():
    os.makedirs(TOLGEE_PATH, exist_ok=True)
    cmd = [
              "tolgee", "pull",
              "--format", "ANDROID_XML",
              "--path", TOLGEE_PATH,
              "--languages"
          ] + SUPPORTED_LANGUAGES + [
              "--empty-dir",
          ]
    print(f"📥 拉取 Tolgee 翻译中...")
    subprocess.run(cmd, check=True)
    print(f"✅ Tolgee 拉取完成：{TOLGEE_PATH}")


def find_en_strings_files():
    en_files = []
    for root, dirs, files in os.walk("."):
        normalized_root = root.replace("\\", "/")
        if "lite" in normalized_root or "lite" in files:
            continue
        if "/build/" in normalized_root or normalized_root.endswith("/build"):
            continue
        if "strings.xml" in files and "/res/values" in root.replace("\\", "/"):
            if "values-" not in root:  # 只取英文
                en_files.append(os.path.join(root, "strings.xml"))
    return en_files


def parse_strings_file(file_path):
    try:
        tree = ET.parse(file_path)
        root = tree.getroot()
        result = OrderedDict()
        for e in root.findall("string"):
            if 'name' in e.attrib:
                result[e.attrib['name']] = e.text or ""
        return result
    except Exception as e:
        print(f"❌ 解析失败: {file_path} -> {e}")
        return OrderedDict()


# 解析 Tolgee 拉取的翻译，并合并到现有的 strings.xml 中
def merge_translations(lang, base_dict: dict, pulled_dict: dict, existing_dict: dict,
                       translatable_false_keys):
    merged = OrderedDict()

    # 先添加原有 key，保持顺序
    # 只有en才会有translatable_false_keys的数据
    for key in existing_dict:
        merged[key] = existing_dict[key]
        # if lang == "en" and key in translatable_false_keys:
        #     merged[key] = pulled_dict.get(key) or existing_dict[key] or base_dict.get(key) or ""
        # else:
        #     merged[key] = pulled_dict.get(key) or existing_dict[key] or ""


    # 再补充新 key（存在于 base_dict，但不在原文件中）
    for key in base_dict:
        #如果 key 不在 pulled_dict 中，跳过
        if key not in pulled_dict:
            continue
        if key in translatable_false_keys and lang != "en":
            # 如果是非英文语言且在translatable_false_keys中，则不添加，并且移除掉
            merged.pop(key, None)
            continue
        if pulled_dict.get(key) is not None:
            merged[key] = pulled_dict.get(key)

    return merged


def find_translatable_false_keys():
    result = []
    for root, dirs, files in os.walk("."):
        normalized_root = root.replace("\\", "/")
        if normalized_root.endswith("/res/values") and "strings.xml" in files:
            file_path = os.path.join(root, "strings.xml")
            try:
                tree = ET.parse(file_path)
                root_elem = tree.getroot()
                for elem in root_elem.findall("string"):
                    if elem.attrib.get("translatable") == "false":
                        result.append(elem.attrib["name"])
            except Exception as e:
                print(f"⚠️ 解析失败：{file_path} -> {e}")
    print(f"🔍 找到 {len(result)} 个 translatable=false 的 key")
    # print(result)
    return result

def write_strings_xml(file_path, data: dict,translatable_false_keys,cdata_keys):
    if data is None or not data:
        print(f"⚠️ 跳过写入：{file_path}，没有数据")
        return

    os.makedirs(os.path.dirname(file_path), exist_ok=True)
    cdata_map = {}

    # 创建 <resources> 根节点
    resources = ET.Element("resources")

    # 遍历 key，构建 string 元素
    for key, value in data.items():
        attributes = {"name": key}
        text = value

        if key in translatable_false_keys:
            attributes["translatable"] = "false"

        if isinstance(value, dict):
            text = value.get("value", "")
            for attr in ["translatable", "formatted", "product"]:
                if attr in value:
                    attributes[attr] = value[attr]

        string_elem = ET.SubElement(resources, "string", attrib=attributes)
        if key in cdata_keys:
            placeholder = f"__CDATA__{key}__"
            string_elem.text = placeholder
            cdata_map[key] = text
        else:
            string_elem.text = text

    # 格式化 + 替换 CDATA
    rough_string = ET.tostring(resources, encoding="utf-8")
    pretty_xml = minidom.parseString(rough_string).toprettyxml(indent="   ")
    pretty_xml = "\n".join([line for line in pretty_xml.split("\n") if line.strip()])

    # 替换默认声明为标准的 XML 头部
    if pretty_xml.startswith('<?xml version="1.0" ?>'):
        pretty_xml = pretty_xml.replace(
            '<?xml version="1.0" ?>',
            '<?xml version="1.0" encoding="utf-8"?>',
            1
        )

    for key in cdata_map:
        placeholder = f"__CDATA__{key}__"
        cdata_value = cdata_map[key]
        pretty_xml = pretty_xml.replace(placeholder, f"<![CDATA[{cdata_value}]]>")

    with open(file_path, "w", encoding="utf-8") as f:
        f.write(pretty_xml + "\n")


def get_module_base_path(en_file_path):
    return os.path.dirname(os.path.dirname(en_file_path))  # 到 src/main/res



def main():
    run_tolgee_pull()
    en_files = find_en_strings_files()
    translatable_false_keys = find_translatable_false_keys()
    cdata_keys = fetch_all_cdata_keys()

    for en_path in en_files:
        print(en_path)
        en_strings = parse_strings_file(en_path)
        if not en_strings:
            continue

        module_res_path = get_module_base_path(en_path)
        print(f"处理模块：{module_res_path}")
        for lang in SUPPORTED_LANGUAGES:
            if lang == "en":
                lang_dir = "values"
            elif lang == "id":
                lang_dir = "values-in"
            else:
                lang_dir = f"values-{lang}"
            pulled_path = os.path.join(TOLGEE_PATH, f"values-{lang}", "strings.xml")
            out_path = os.path.join(module_res_path, lang_dir, "strings.xml")
            print(f"处理语言：{lang} → {out_path}, 源：{pulled_path}")

            pulled_translations = parse_strings_file(pulled_path)

            if not pulled_translations:
                print(f"⚠️ 跳过：{pulled_path}，没有可用翻译")
                continue
            # print(f"已拉取翻译：{pulled_translations}")

            existing_translations = parse_strings_file(out_path)
            # print(f"现有翻译：{existing_translations}")
            merged = merge_translations(
                lang=lang,
                base_dict=en_strings,
                pulled_dict=pulled_translations,
                existing_dict=existing_translations,
                translatable_false_keys=translatable_false_keys
            )
            write_strings_xml(out_path, merged,translatable_false_keys,cdata_keys)
    # ✅ 删除中间产物
    if os.path.exists(TOLGEE_PATH):
        shutil.rmtree(TOLGEE_PATH)
        print(f"🧹 已删除中间目录：{TOLGEE_PATH}")


if __name__ == "__main__":
    main()