from newspaper import Article
from docx import Document
from docx.shared import Pt, RGBColor
from docx.enum.style import WD_STYLE_TYPE
from docx.oxml.ns import qn

# tkinter GUI
import tkinter as tk
from tkinter import messagebox


def init_window():
    root = tk.Tk()
    root.title("Url2Word-Tools")
    root.geometry("400x300")

    url_label = tk.Label(root, text="網(wǎng)頁(yè)鏈接", font=("Arial", 16))
    url_label.pack(pady=20)

    global url_input
    url_input = tk.StringVar()
    url_input = tk.Entry(root, textvariable=url_input, font=("Arial", 16))
    url_input.pack(padx=20)
    
    button = tk.Button(root, text="轉(zhuǎn)換", command=on_click, font=("Arial", 16))
    button.pack(pady=20)

    # 運(yùn)行主循環(huán)
    root.mainloop()

def fetch_article_content(url):
    """
    使用 newspaper3k 獲取指定URL頁(yè)面的文章內(nèi)容。
    
    :param url: 要抓取的網(wǎng)頁(yè)URL
    :return: 文章的元數(shù)據(jù)和正文內(nèi)容
    """
    try:
        # 創(chuàng)建Article對(duì)象
        article = Article(url, language='zh')  # 設(shè)置語(yǔ)言為中文
        
        # 下載并解析文章
        article.download()
        article.parse()
        
        # 提取文章信息
        article_info = {
            'title': article.title,
            'authors': article.authors,
            'publish_date': article.publish_date,
            'text': article.text,
            'top_image': article.top_image,
            'images': list(article.images),
            'html': article.html
        }
        
        return article_info
    except Exception as e:
        print(f"Error fetching {url}: {e}")
        return None

def create_style(document, name, font_size=12, font_name='Arial', color=RGBColor(0, 0, 0)):
    """
    創(chuàng)建一個(gè)自定義樣式。
    
    :param document: 當(dāng)前文檔對(duì)象
    :param name: 樣式名稱
    :param font_size: 字體大小 (默認(rèn)12)
    :param font_name: 字體名稱 (默認(rèn)Arial)
    :param color: 字體顏色 (默認(rèn)黑色)
    :return: 新創(chuàng)建的樣式
    """
    style = document.styles.add_style(name, WD_STYLE_TYPE.PARAGRAPH)
    font = style.font
    font.name = font_name
    font.size = Pt(font_size)
    font.color.rgb = color
    return style

def set_run_style(run, font_size=12, font_name='Arial', color=RGBColor(0, 0, 0)):
    run.font.name = font_name
    run._element.rPr.rFonts.set(qn('w:eastAsia'), font_name)
    run.font.size = Pt(font_size)
    run.font.color.rgb = color

def save_to_word(article_info, output_path):
    """
    將文章信息保存為Word文檔。
    
    :param article_info: 包含文章信息的字典
    :param output_path: 輸出Word文檔的路徑
    """
    document = Document()

    # 創(chuàng)建一個(gè)自定義樣式
    normal_style = create_style(document, 'CustomNormalStyle')

    # 添加標(biāo)題
    heading = document.add_heading(article_info['title'], level=1)
    for run in heading.runs:
        # run.font.color.rgb = RGBColor(0, 0, 0)  # 確保標(biāo)題是黑色
        set_run_style(run, font_size=20)

    # 添加作者
    if article_info['authors']:
        authors_str = ', '.join(article_info['authors'].encode('utf-8').decode('utf-8'))
        document.add_paragraph(f"作者: {authors_str}", style=normal_style)

    # 添加發(fā)布日期
    if article_info['publish_date']:
        document.add_paragraph(f"發(fā)布時(shí)間: {article_info['publish_date']}".encode('utf-8').decode('utf-8'), style=normal_style)

    # 添加正文
    document.add_heading('內(nèi)容', level=2).runs[0].font.color.rgb = RGBColor(0, 0, 0)
    paragraphs = article_info['text'].split('\n')
    for paragraph in paragraphs:
        if paragraph.strip():  # 忽略空行
            clean_paragraph = paragraph.encode('utf-8').decode('utf-8')
            p = document.add_paragraph(style=normal_style)
            run = p.add_run(clean_paragraph)
            set_run_style(run)

    # 保存文檔
    document.save(output_path)
    print(f"Document saved to {output_path}")
    messagebox.showinfo('提示','轉(zhuǎn)換成功')

def on_click():
    url = url_input.get()
    print(url)
    article_info = fetch_article_content(f'{url}')
    if article_info:
            # print("Title:", article_info['title'])
            # print("Authors:", article_info['authors'])
            # print("Publish Date:", article_info['publish_date'])
            # print("Text:\n", article_info['text'])
            # print("Top Image:", article_info['top_image'])
            # print("Images:", article_info['images'])
            
            output_path = f"./{article_info['title']}.docx"
            save_to_word(article_info, output_path)

if __name__ == "__main__":
    init_window()