# -*- coding: utf-8 -*-
import sys
import io

"""
Created on Sun Mar 3 12:22:49 2019

@author: Ben
"""

import importlib
importlib.reload(sys)

from pdfminer.pdfparser import PDFParser,PDFDocument
from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter
from pdfminer.converter import PDFPageAggregator
from pdfminer.layout import LTTextBoxHorizontal,LAParams
from pdfminer.pdfinterp import PDFTextExtractionNotAllowed

#from PyPDF2.pdf import PdfFileReader, PdfFileWriter, ContentStream


import requests
import string
import time
import hashlib
import json


##初始化

api_url = "http://api.fanyi.baidu.com/api/trans/vip/translate"
api_id = "" ##申請(qǐng)的百度翻譯接口的id
cyber = "" ##申請(qǐng)的百度翻譯接口的password

pdffile = "multinet.pdf" ##處理的pdf
ENtextfile = "ENmultinet.txt" ##存儲(chǔ)提取的txt
CNtextfile = "CNmultinet.txt" ##存儲(chǔ)翻譯的結(jié)果
isTranslate = False ##是否將提取的英文翻譯為中文
## 處理PDF
## 讀取PDF的內(nèi)容 filename是待處理的PDF的名字

###使用PDFminer讀取
def getDataUsingPyPDF(filename):
 parser = PDFParser(open(pdffile,'rb')) #以二進(jìn)制打開文件 ,并創(chuàng)建一個(gè)pdf文檔分析器
 doc = PDFDocument() ##創(chuàng)建一個(gè)pdf文檔
 #將文檔對(duì)象和連接分析器連接起來
 parser.set_document(doc) 
 doc.set_parser(parser)
 doc.initialize()
 
 
 #判斷該pdf是否支持txt轉(zhuǎn)換
 
 if doc.is_extractable:
 #創(chuàng)建一個(gè)PDF設(shè)備對(duì)象
 rsrcmgr = PDFResourceManager()
 #創(chuàng)建一個(gè)pdf設(shè)備對(duì)象
 laparamas = LAParams()
 device = PDFPageAggregator(rsrcmgr, laparams=laparamas)
 #創(chuàng)建一個(gè)PDF解釋器對(duì)象
 interpreter = PDFPageInterpreter(rsrcmgr, device)
 contents = "" #保存讀取的text
 
 #依次讀取每個(gè)page的內(nèi)容
 
 for page in doc.get_pages():
 interpreter.process_page(page)
 layout = device.get_result() # 這里layout是一個(gè)LTPage對(duì)象 里面存放著 這個(gè)page解析出的各種對(duì)象 一般包括LTTextBox, LTFigure, LTImage, LTTextBoxHorizontal 等等 想要獲取文本就獲得對(duì)象的text屬性，
 #在windows下，新文件的默認(rèn)編碼是gbk編碼，所以我們?cè)趯懭胛募臅r(shí)候需要設(shè)置一個(gè)編碼格式，如下：
 for x in layout:
 if(isinstance(x,LTTextBoxHorizontal)):
  results = x.get_text()
  results = results.replace("\n","") #去掉換行符 因?yàn)榕虐鎲栴} 有的換行導(dǎo)致句子中斷
  contents += (results)
 ##為了看著舒服，每一句為一行
 saveText(contents.replace(".",".\n"),ENtextfile)
 return contents
## 將讀取的content以txt格式存放到本地
def saveText(content,Textfile):
 with open(Textfile,"w",encoding='utf-8') as f:
 f.write(content)


## 翻譯從pdf提取的content
def translate(content):
 salt = str(time.time())[:10]
 final_sign = str(api_id) + content + salt+ cyber
 final_sign = hashlib.md5(final_sign.encode("utf-8")).hexdigest()
 # from to 代表翻譯的語言 
 paramas = {
 'q':content,
 'from':'en',
 'to':'zh',
 'appid':'%s'%api_id,
 'salt':'%s'%salt,
 'sign':'%s'%final_sign 
 }
 my_url = api_url+'?appid='+str(api_id)+'&q='+content+'&from='+'zh'+'&to='+'en'+'&salt='+salt+'&sign='+final_sign
 response = requests.get(api_url,params = paramas).content
 content = str(response,encoding = "utf-8")
 json_reads = json.loads(content)
 return json_reads['trans_result'][0]['dst']+" " 
###

content = getDataUsingPyPDF(pdffile)
print("讀取pdf成功，將其保存為txt格式")

if(isTranslate):
 clist = content.split(".") #split() 通過指定.將英文分成多個(gè)句子
 i = 0
 chinese = ""
 print("一共有"+str(clist.__len__())+"行需要翻譯")
 print("開始翻譯...請(qǐng)耐心等待")

 while(i<clist.__len__()):
 chinese += (translate(clist[i]).replace("\n","。"))
 #chinese += '\n'
 i+=1
 saveText(chinese,CNtextfile)
 print("翻譯結(jié)束，ok")