跳转至

AI 翻译pdf

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
import fitz
import sqlite3
import pywebio
from pywebio import config
from pywebio.input import file_upload, input_group
import json

import requests
from pywebio.output import put_file


class maxai:
    def __init__(self, auth="application-ad80edba84c30eb1161ae04e71ee7f78"):
        self.base_url = 'http://ai.tech.intra.com'
        self.auth = auth
        self.headers = {'accept': 'application/json', 'AUTHORIZATION': self.auth}
        self.new_chat = False  # 默认是否创建新的会话
        self.chatid = self.create_chat()
        print(f"\r登录成功-{self.chatid}")

    def chat(self, query):
        if self.new_chat:
            chat_id = self.create_chat()
        else:
            chat_id = self.chatid
        url = self.base_url + "/api/application/chat_message/" + chat_id
        data = {
            "message": query,
            "re_chat": False,
            "stream": True,
        }
        response = requests.post(url, headers=self.headers, json=data, stream=True)
        translated_text = ''

        if response.status_code == 200:
            # 设置响应编码为UTF-8
            response.encoding = 'utf-8'
            # 逐行读取响应内容
            for line in response.iter_lines(decode_unicode=True):
                if line:
                    event_data = json.loads(line[5:])
                    if event_data['is_end'] == False:
                        text = event_data['content']
                        if text:
                            translated_text += text
        return translated_text

    def create_chat(self):
        print("创建对话...", end='', flush=True)
        id = self.login_and_get_appinfo()
        url = self.base_url + "/api/application/" + id + "/chat/open"
        response = requests.get(url, headers=self.headers).json()
        chat_id = response['data']
        return chat_id

    def login_and_get_appinfo(self):
        print("正在登录...", end='', flush=True)
        url = self.base_url + '/api/application/profile'
        response = requests.get(url, headers=self.headers).json()
        id = response['data']['id']
        return id


def convert_color(color_int):
    """将颜色整数转换为RGB格式"""
    r = (color_int >> 16) & 0xFF
    g = (color_int >> 8) & 0xFF
    b = color_int & 0xFF
    return (r / 255, g / 255, b / 255)


def init_db():
    conn = sqlite3.connect('translations_cache.db')
    cursor = conn.cursor()
    cursor.execute('''
        CREATE TABLE IF NOT EXISTS translations (
            original_text TEXT PRIMARY KEY,
            translated_text TEXT
        )
    ''')
    conn.commit()
    conn.close()


def get_cached_translation(text):
    conn = sqlite3.connect('translations_cache.db')
    cursor = conn.cursor()
    cursor.execute('SELECT translated_text FROM translations WHERE original_text = ?', (text,))
    result = cursor.fetchone()
    conn.close()
    return result[0] if result else None


def cache_translation(original_text, translated_text):
    conn = sqlite3.connect('translations_cache.db')
    cursor = conn.cursor()
    cursor.execute('INSERT INTO translations (original_text, translated_text) VALUES (?, ?)',
                   (original_text, translated_text))
    conn.commit()
    conn.close()


def translate_text(text):
    cached_translation = get_cached_translation(text)
    if cached_translation:
        return cached_translation

    bbb = ccaam.chat(text)
    return bbb


# 初始化数据库
init_db()
ccaam = maxai()


# 打开 PDF 文件

def trancpdf(pdf_document, filename):
    # 遍历每一页
    for page_num in range(pdf_document.page_count):
        page = pdf_document.load_page(page_num)

        # 获取页面上的所有文本块
        blocks2 = page.get_text("dict", flags=11)["blocks"]

        for block in blocks2:
            for line in block["lines"]:
                for span in line["spans"]:
                    text = span["text"]
                    if text.strip() and len(text) > 10:  # Only translate non-empty text
                        translated_text = translate_text(text)
                        if "抱歉" in translated_text or "翻译" in translated_text:
                            continue
                        font_size = span["size"]
                        color = span["color"]
                        bbox = span["bbox"]
                        rect = fitz.Rect(bbox)
                        page.insert_text((rect.x0, rect.y0), translated_text, fontsize=font_size / 3,
                                         color=convert_color(color), fontname="SimSun", fontfile="SimSun.ttf")
                        print((translated_text, font_size, color, bbox))
                        pdf_document.save("zh_CN/" + filename)
        content = open("zh_CN/" + filename, 'rb').read()
        put_file(filename, content, 'download me')


def sendmd():
    data = input_group("翻译pdf", [
        file_upload(accept=".pdf", placeholder="翻译pdf", name="file"),
    ])
    print(data)
    f = data['file']
    filename = f['filename']
    open(filename, 'wb').write(f['content'])

    if "file" in data:
        pdf_document = fitz.open(filename)
        trancpdf(pdf_document, filename)
        pdf_document.close()


if __name__ == '__main__':
    config(title="工具", theme="yeti", description="yes")  # global configuration
    pywebio.start_server(sendmd, port=1008, cdn=False)