from datetime import datetime

# 基礎(chǔ)格式轉(zhuǎn)換
date_str = "2023-12-15"
date_obj = datetime.strptime(date_str, "%Y-%m-%d")
print(f"轉(zhuǎn)換結(jié)果: {date_obj}")

# 帶時(shí)間轉(zhuǎn)換
datetime_str = "2023-12-15 14:30:45"
datetime_obj = datetime.strptime(datetime_str, "%Y-%m-%d %H:%M:%S")
print(f"帶時(shí)間轉(zhuǎn)換: {datetime_obj}")

# 復(fù)雜格式轉(zhuǎn)換
complex_str = "15/Dec/2023 02:30 PM"
complex_obj = datetime.strptime(complex_str, "%d/%b/%Y %I:%M %p")
print(f"復(fù)雜格式轉(zhuǎn)換: {complex_obj}")

from dateutil import parser

# 自動(dòng)解析多種格式
formats = [
    "2023-12-15",
    "15/12/2023",
    "Dec 15, 2023",
    "2023年12月15日",
    "15-Dec-2023 14:30"
]

for date_str in formats:
    date_obj = parser.parse(date_str)
    print(f"{date_str} => {date_obj}")

# 處理模糊日期
ambiguous = "10-11-12"  # 可能是年月日或月日年
print("默認(rèn)解析:", parser.parse(ambiguous))  # 2023-10-11
print("日優(yōu)先解析:", parser.parse(ambiguous, dayfirst=True))  # 2023-11-10

import pytz

# 帶時(shí)區(qū)解析
tz_str = "2023-12-15T14:30:45+08:00"
tz_obj = parser.parse(tz_str)
print(f"帶時(shí)區(qū)轉(zhuǎn)換: {tz_obj} (時(shí)區(qū): {tz_obj.tzinfo})")

# 添加時(shí)區(qū)
naive_str = "2023-12-15 14:30:45"
naive_obj = datetime.strptime(naive_str, "%Y-%m-%d %H:%M:%S")
shanghai_tz = pytz.timezone('Asia/Shanghai')
aware_obj = shanghai_tz.localize(naive_obj)
print(f"添加時(shí)區(qū)后: {aware_obj}")

# 時(shí)區(qū)轉(zhuǎn)換
new_york_tz = pytz.timezone('America/New_York')
ny_time = aware_obj.astimezone(new_york_tz)
print(f"紐約時(shí)間: {ny_time}")

def parse_log_timestamp(log_line):
    """解析日志時(shí)間戳"""
    # 示例日志格式: [15/Dec/2023:14:30:45 +0800]
    import re
    pattern = r'\[(\d{2}/\w{3}/\d{4}:\d{2}:\d{2}:\d{2} [\+\-]\d{4})\]'
    match = re.search(pattern, log_line)
    if match:
        timestamp_str = match.group(1)
        return parser.parse(timestamp_str)
    return None

# 使用示例
log_line = '127.0.0.1 - - [15/Dec/2023:14:30:45 +0800] "GET / HTTP/1.1" 200 2326'
timestamp = parse_log_timestamp(log_line)
print(f"日志時(shí)間: {timestamp}")

class LogParser:
    """多格式日志時(shí)間解析器"""
    def __init__(self):
        self.formats = [
            # Apache格式
            r'\[(?P<timestamp>\d{2}/\w{3}/\d{4}:\d{2}:\d{2}:\d{2} [\+\-]\d{4})\]',
            # ISO格式
            r'(?P<timestamp>\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}\.\d{3}Z)',
            # 簡(jiǎn)單格式
            r'(?P<timestamp>\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2})'
        ]
    
    def parse(self, log_line):
        """解析日志時(shí)間"""
        for fmt in self.formats:
            match = re.search(fmt, log_line)
            if match:
                try:
                    return parser.parse(match.group('timestamp'))
                except ValueError:
                    continue
        return None

# 使用示例
logs = [
    '127.0.0.1 - - [15/Dec/2023:14:30:45 +0800] "GET / HTTP/1.1" 200 2326',
    '2023-12-15T06:30:45.123Z INFO: System started',
    '2023-12-15 14:30:45 ERROR: Connection failed'
]

parser = LogParser()
for log in logs:
    ts = parser.parse(log)
    print(f"{log[:30]}... => {ts}")

def parse_financial_date(date_str, format_hint=None):
    """解析金融日期"""
    # 常見金融格式
    financial_formats = [
        '%Y-%m-%d',        # ISO格式
        '%m/%d/%Y',        # 美式格式
        '%d-%b-%Y',        # 15-Dec-2023
        '%Y%m%d',          # 20231215
        '%d/%m/%Y %H:%M',  # 英式帶時(shí)間
    ]
    
    if format_hint:
        try:
            return datetime.strptime(date_str, format_hint)
        except ValueError:
            pass
    
    for fmt in financial_formats:
        try:
            return datetime.strptime(date_str, fmt)
        except ValueError:
            continue
    
    # 嘗試智能解析
    return parser.parse(date_str)

# 使用示例
financial_dates = [
    "20231215",          # 交易日期
    "15-Dec-2023",       # 結(jié)算日期
    "12/15/2023 14:30",  # 報(bào)價(jià)時(shí)間
    "2023-12-15"         # 生效日期
]

for date_str in financial_dates:
    date_obj = parse_financial_date(date_str)
    print(f"{date_str} => {date_obj}")

import pandas as pd
from pandas.tseries.offsets import BDay

def parse_and_adjust(date_str, market='NYSE'):
    """解析日期并調(diào)整到最近交易日"""
    # 解析日期
    date_obj = parse_financial_date(date_str)
    
    # 調(diào)整到交易日
    if market == 'NYSE':
        # 使用pandas交易日歷
        return date_obj + BDay(0)  # 最近交易日
    
    # 其他市場(chǎng)處理
    # ...
    return date_obj

# 使用示例
trade_dates = ["2023-12-15", "2023-12-16", "2023-12-17"]  # 周五、周六、周日
for date_str in trade_dates:
    adjusted = parse_and_adjust(date_str)
    print(f"{date_str} => 交易日: {adjusted.date()}")

def parse_device_timestamp(timestamp, epoch=946684800):
    """解析設(shè)備時(shí)間戳（多種格式）"""
    # 類型1: Unix時(shí)間戳（秒）
    if isinstance(timestamp, int) and timestamp > 1e9:
        return datetime.utcfromtimestamp(timestamp)
    
    # 類型2: Unix時(shí)間戳（毫秒）
    if isinstance(timestamp, int) and timestamp > 1e12:
        return datetime.utcfromtimestamp(timestamp / 1000.0)
    
    # 類型3: 自定義紀(jì)元時(shí)間
    if isinstance(timestamp, int):
        return datetime.utcfromtimestamp(epoch + timestamp)
    
    # 類型4: 字符串格式
    if isinstance(timestamp, str):
        try:
            return parser.parse(timestamp)
        except ValueError:
            pass
    
    # 類型5: GPS時(shí)間
    if isinstance(timestamp, float):
        # GPS時(shí)間轉(zhuǎn)UTC（示例）
        return datetime.utcfromtimestamp(timestamp + 315964800)
    
    raise ValueError(f"無(wú)法解析時(shí)間戳: {timestamp}")

# 使用示例
device_timestamps = [
    1702593045,           # Unix時(shí)間戳（秒）
    1702593045123,         # Unix時(shí)間戳（毫秒）
    123456789,             # 自定義紀(jì)元（從2000-01-01開始）
    "2023-12-15T14:30:45Z",# ISO格式
    1296000000.0           # GPS時(shí)間（示例）
]

for ts in device_timestamps:
    dt = parse_device_timestamp(ts)
    print(f"{ts} => {dt}")

def align_sensor_data(data_frame, time_col='timestamp', freq='1S'):
    """對(duì)齊傳感器時(shí)間序列"""
    # 轉(zhuǎn)換時(shí)間列
    data_frame[time_col] = pd.to_datetime(data_frame[time_col])
    
    # 設(shè)置為索引
    data_frame.set_index(time_col, inplace=True)
    
    # 重采樣對(duì)齊
    aligned = data_frame.resample(freq).mean()
    return aligned.reset_index()

# 使用示例
import pandas as pd
data = {
    'timestamp': ['2023-12-15 14:30:45.1', '2023-12-15 14:30:45.3', '2023-12-15 14:30:46.2'],
    'temperature': [25.3, 25.5, 25.8],
    'humidity': [45, 46, 44]
}
df = pd.DataFrame(data)

aligned_df = align_sensor_data(df, freq='1S')
print("對(duì)齊后的數(shù)據(jù):")
print(aligned_df)

# 大型數(shù)據(jù)集轉(zhuǎn)換
def convert_large_dataset(file_path):
    """轉(zhuǎn)換大型CSV數(shù)據(jù)集"""
    # 分塊讀取
    chunk_size = 10000
    chunks = []
    
    for chunk in pd.read_csv(file_path, chunksize=chunk_size):
        # 轉(zhuǎn)換日期列
        chunk['date'] = pd.to_datetime(chunk['date_str'], format='%Y-%m-%d')
        chunks.append(chunk)
    
    return pd.concat(chunks)

# 使用示例
# df = convert_large_dataset('large_dataset.csv')

# 直接轉(zhuǎn)換
df = pd.DataFrame({
    'date_str': ['2023-01-01', '2023-01-02', '2023-01-03'] * 100000
})
df['date'] = pd.to_datetime(df['date_str'])
print(f"轉(zhuǎn)換后內(nèi)存占用: {df.memory_usage().sum() / 1024**2:.2f} MB")

from concurrent.futures import ThreadPoolExecutor
import multiprocessing

def parallel_datetime_convert(strings, format='%Y-%m-%d'):
    """并行日期轉(zhuǎn)換"""
    with ThreadPoolExecutor(max_workers=multiprocessing.cpu_count()) as executor:
        results = list(executor.map(
            lambda s: datetime.strptime(s, format), 
            strings
        ))
    return results

# 性能對(duì)比
large_strings = ['2023-12-15'] * 1000000

%timeit [datetime.strptime(s, '%Y-%m-%d') for s in large_strings]  # 單線程
%timeit parallel_datetime_convert(large_strings)  # 多線程

import parsedatetime as pdt

def parse_natural_language(text):
    """解析自然語(yǔ)言日期"""
    cal = pdt.Calendar()
    result, status = cal.parse(text)
    
    if status:
        return datetime(*result[:6])
    else:
        raise ValueError(f"無(wú)法解析日期: {text}")

# 使用示例
phrases = [
    "tomorrow",
    "next Monday",
    "3 days ago",
    "December 15, 2023",
    "two weeks from now"
]

for phrase in phrases:
    date_obj = parse_natural_language(phrase)
    print(f"{phrase} => {date_obj.date()}")

from dateparser import parse

def advanced_nlp_parse(text, languages=['en']):
    """高級(jí)自然語(yǔ)言日期解析"""
    result = parse(text, languages=languages)
    if result:
        return result
    else:
        raise ValueError(f"無(wú)法解析日期: {text}")

# 使用示例
complex_phrases = [
    "The deadline is next Friday at 5pm",
    "會(huì)議定于下周三下午兩點(diǎn)",
    "Report due in 3 working days",
    "Q4 closing: Dec 31, 2023"
]

for phrase in complex_phrases:
    lang = 'en' if ' ' in phrase else 'zh'
    date_obj = advanced_nlp_parse(phrase, languages=[lang])
    print(f"{phrase} => {date_obj}")

def robust_date_parse(date_str, default=None, formats=None):
    """健壯的日期解析函數(shù)"""
    # 預(yù)定義格式列表
    default_formats = [
        '%Y-%m-%d',
        '%Y/%m/%d',
        '%d-%b-%Y',
        '%d/%m/%Y',
        '%m/%d/%Y',
        '%Y%m%d',
        '%Y-%m-%d %H:%M:%S',
        '%Y-%m-%dT%H:%M:%S%z'
    ]
    
    formats = formats or default_formats
    
    # 嘗試格式匹配
    for fmt in formats:
        try:
            return datetime.strptime(date_str, fmt)
        except ValueError:
            continue
    
    # 嘗試智能解析
    try:
        return parser.parse(date_str)
    except (ValueError, OverflowError):
        pass
    
    # 嘗試自然語(yǔ)言解析
    try:
        return parse_natural_language(date_str)
    except ValueError:
        pass
    
    # 返回默認(rèn)值
    return default

# 使用示例
problematic_dates = [
    "2023-02-30",  # 無(wú)效日期
    "15/12/23",    # 年份簡(jiǎn)寫
    "December 32", # 無(wú)效日
    "unknown"      # 無(wú)法解析
]

for date_str in problematic_dates:
    result = robust_date_parse(date_str, default=datetime.now())
    print(f"{date_str} => {result.date()}")

# 明確指定格式
date_str = "15/12/2023"
date_obj = datetime.strptime(date_str, "%d/%m/%Y")  # 明確日/月/年

# 始終存儲(chǔ)UTC時(shí)間
utc_time = datetime.utcnow()
# 顯示時(shí)轉(zhuǎn)換本地時(shí)間
local_time = utc_time.astimezone(pytz.timezone('Asia/Shanghai'))

# 大型數(shù)據(jù)集使用pandas
df['date'] = pd.to_datetime(df['date_str'])

try:
    date_obj = datetime.strptime(date_str, fmt)
except ValueError as e:
    logger.error(f"日期解析失敗: {date_str} - {str(e)}")
    date_obj = datetime.now()

def parse_with_logging(date_str):
    try:
        return parser.parse(date_str)
    except Exception as e:
        logger.warning(f"無(wú)法解析日期: {date_str}, 使用默認(rèn)值")
        return datetime.now()

import unittest

class TestDateParsing(unittest.TestCase):
    def test_valid_formats(self):
        test_cases = [
            ("2023-12-15", datetime(2023, 12, 15)),
            ("15/12/2023", datetime(2023, 12, 15)),
            ("Dec 15, 2023", datetime(2023, 12, 15))
        ]
        for date_str, expected in test_cases:
            self.assertEqual(robust_date_parse(date_str), expected)

    def test_invalid_dates(self):
        self.assertEqual(robust_date_parse("2023-02-30"), datetime(2023, 2, 28))
        self.assertEqual(robust_date_parse("invalid"), datetime.today().date())