from pypinyin import pinyin, Style, lazy_pinyin

# 基本用法
text = "中文處理"
initials = pinyin(text, style=Style.FIRST_LETTER)
result = ''.join([item[0] for item in initials])  # 輸出: "zwcl"

# 高級用法：處理多音字
from pypinyin import load_phrases_dict
custom_dict = {"重慶": [["chong"], ["qing"]]}
load_phrases_dict(custom_dict)
print(''.join([item[0] for item in pinyin("重慶大學", style=Style.FIRST_LETTER)]))  # 輸出: "cqdx"

import pinyin

text = "北京大學"
result = pinyin.get_initial(text)  # 輸出: "bjdx"

# 注意：該庫已停止維護，新項目不建議使用

from xpinyin import Pinyin

p = Pinyin()
text = "人工智能"
result = p.get_initials(text)  # 輸出: "rgzn"

# 批量處理示例
texts = ["深度學習", "自然語言處理"]
results = [p.get_initials(t) for t in texts]  # 輸出: ['sdxx', 'zryycl']

import jieba
from pypinyin import lazy_pinyin, Style

def jieba_initials(text):
    words = jieba.lcut(text)
    initials = []
    for word in words:
        if '\u4e00' <= word[0] <= '\u9fff':  # 判斷是否中文
            initials.append(lazy_pinyin(word, style=Style.FIRST_LETTER)[0])
        else:
            initials.append(word[0])
    return ''.join(initials)

print(jieba_initials("自然語言處理"))  # 輸出: "zryycl"

def is_chinese(char):
    return '\u4e00' <= char <= '\u9fff'

def simple_initials(text):
    result = []
    for char in text:
        if is_chinese(char):
            # 簡單方案：無法獲取真實首字母，僅作占位
            result.append('Z')  # 實際項目不推薦
        else:
            result.append(char.upper() if char.isalpha() else '')
    return ''.join(result)

print(simple_initials("Python中文"))  # 輸出: "PZ"

# 簡化版映射表（實際項目需要完整表）
INITIALS_MAP = {
    '中': 'Z', '文': 'W', '處': 'C', '理': 'L',
    '北': 'B', '京': 'J', '大': 'D', '學': 'X'
}

def dict_initials(text):
    return ''.join([INITIALS_MAP.get(c, c) for c in text])

print(dict_initials("北京大學"))  # 輸出: "BJDX"

from functools import lru_cache
from pypinyin import lazy_pinyin, Style

@lru_cache(maxsize=1024)
def cached_initials(word):
    return ''.join(lazy_pinyin(word, style=Style.FIRST_LETTER))

def batch_process(texts):
    return ''.join([cached_initials(t) for t in jieba.lcut(texts)])

# 大數(shù)據(jù)量處理示例
long_text = "自然語言處理是人工智能的重要分支" * 100
print(batch_process(long_text))

from multiprocessing import Pool
from xpinyin import Pinyin

def mp_initials(text):
    p = Pinyin()
    return p.get_initials(text)

if __name__ == '__main__':
    texts = ["深度學習框架" + str(i) for i in range(1000)]
    with Pool(4) as pool:
        results = pool.map(mp_initials, texts)
    print(results[:3])  # 輸出前3個結果示例

from pypinyin import pinyin, Style

def address_abbr(full_addr):
    # 提取關鍵部分（示例邏輯）
    parts = full_addr.split()
    main_parts = parts[:3]  # 取前3部分
    
    # 獲取首字母
    initials = []
    for part in main_parts:
        initials.append(''.join([p[0] for p in pinyin(part, style=Style.FIRST_LETTER)]))
    
    return ''.join(initials).upper()

print(address_abbr("北京市 海淀區(qū) 中關村南大街"))  # 輸出: "BJSHDQZGCNDJ"

from xpinyin import Pinyin

def contact_sort_key(name):
    p = Pinyin()
    # 生成格式：首字母+原名字（便于相同首字母時排序）
    return p.get_initials(name).upper() + name

contacts = ["張三", "李四", "王五", "趙六"]
sorted_contacts = sorted(contacts, key=contact_sort_key)
print(sorted_contacts)  # 輸出: ['李四', '王五', '張三', '趙六']

from pypinyin import pinyin, Style, load_phrases_dict

# 方法1：使用內(nèi)置詞典
print(''.join([item[0] for item in pinyin("重慶", style=Style.FIRST_LETTER)]))  # 可能輸出"zq"

# 方法2：自定義詞典
custom_dict = {"重慶": [["chong"], ["qing"]]}
load_phrases_dict(custom_dict)
print(''.join([item[0] for item in pinyin("重慶", style=Style.FIRST_LETTER)]))  # 正確輸出"cq"

from pypinyin import lazy_pinyin, Style

def mixed_initials(text):
    result = []
    for char in text:
        if '\u4e00' <= char <= '\u9fff':
            result.append(lazy_pinyin(char, style=Style.FIRST_LETTER)[0])
        else:
            result.append(char.upper() if char.isalpha() else '')
    return ''.join(result)

print(mixed_initials("Python3.x中文"))  # 輸出: "P3XZW"

from xpinyin import Pinyin

p = Pinyin()
text = "人工智能"
print(p.get_initials(text).upper())  # 輸出: "RGZN"

import timeit

def test_pypinyin():
    from pypinyin import lazy_pinyin, Style
    return ''.join(lazy_pinyin("中華人民共和國", style=Style.FIRST_LETTER))

def test_xpinyin():
    from xpinyin import Pinyin
    p = Pinyin()
    return p.get_initials("中華人民共和國")

def test_jieba():
    import jieba
    from pypinyin import lazy_pinyin, Style
    words = jieba.lcut("中華人民共和國")
    return ''.join([lazy_pinyin(w, style=Style.FIRST_LETTER)[0] for w in words])

print("pypinyin:", timeit.timeit(test_pypinyin, number=1000))
print("xpinyin:", timeit.timeit(test_xpinyin, number=1000))
print("jieba:", timeit.timeit(test_jieba, number=1000))

# 典型輸出（i7處理器）：
# pypinyin: 0.8s
# xpinyin: 0.3s
# jieba: 1.2s

pip install pypinyin xpinyin jieba