import requests
from lxml import html
import pandas as pd
import numpy as np

def html_to_df_lxml(target_div):
    # 获取字段名
    headers = target_div.xpath('.//div[1]/table//th//text()')
    headers = [header.strip() for header in headers if header.strip() != '']

    # 获取数据
    table_data = []
    rows = target_div.xpath('.//div[2]/table//tr')
    for row in rows:
        # 忽略 style="height:0;font-size:0" 的行
        if 'style' in row.attrib and row.attrib['style'] == "height:0;font-size:0":
            continue

        columns = row.xpath('.//td')
        output_row = {}
        for i, column in enumerate(columns):
            if headers[i] == "重要性":
                img_src = column.xpath('./div/img/@src')
                if 'oneStar' in img_src[0]:
                    output_row[headers[i]] = 1
                elif 'twoStar' in img_src[0]:
                    output_row[headers[i]] = 2
                elif 'threeStar' in img_src[0]:
                    output_row[headers[i]] = 3
                else:
                    output_row[headers[i]] = np.nan
            elif headers[i] == "指标":
                spans = column.xpath('./div/span/text()')
                output_row[headers[i]] = ' '.join(spans)
            else:
                text = column.xpath('.//text()')
                text = [t for t in text if t.strip() != '']
                if text:
                    text = text[0].strip()
                else:
                    text = np.nan
                output_row[headers[i]] = text
        table_data.append(output_row)

    df = pd.DataFrame(table_data)

    # forward fill NaN values in the '时间' column
    df['时间'].fillna(method='ffill', inplace=True)

    return df

# 目标网页
url = "https://www.fx168news.com/calendar"

# 使用requests库发送GET请求
response = requests.get(url)

# 解析HTML内容
tree = html.fromstring(response.content)

# 获取特定的div内容
target_div = tree.xpath("/html/body/div[1]/div[1]/div[3]/div[2]/div[4]/div[2]/div[2]/div/div/div/div/div")

# 你的xpath可能得到一个列表，我们只取第一个元素
if target_div:
    df = html_to_df_lxml(target_div[0])
    print(df)