convert/service/convert.py

# -*- coding: utf-8 -*-
from demeter.core import *
from datetime import *
import uuid
import os
import os.path

class Convert(object):

	def getKey(self, site_key, file):
		return Demeter.md5(str(site_key) + '_' + file)

	def get(self, site, site_key, file):
		convert = Demeter.model('convert')
		convert.site_id = site
		convert.key = self.getKey(site_key, file)

		data = convert.select(type='fetchone')

		return data

	def update(self, site, site_key, file):

		info = self.getFile(site_key, file)

		convert = Demeter.model('convert')
		convert.site_id = site
		convert.key = info['key']

		data = convert.select(type='fetchone')
		if not data:
			convert.site_id = site
			convert.file = info['file']
			convert.key = info['key']
			convert.name = info['name']
			convert.ext = info['ext']

			convert.local = info['local']
			convert.path = info['path']
			convert.pdf = info['pdf']
			convert.html = info['html']
			convert.url = info['url']

			id = convert.insert()

			info['status'] = 1
			info['id'] = id

		else:
			info['id'] = data['id']
			info['status'] = data['status']

		return info


	def getFile(self, site_key, file):
		info = {}

		(filepath,temp) = os.path.split(file)
		(filename,extension) = os.path.splitext(temp)

		info['file'] = file
		info['key'] = self.getKey(site_key, file)
		info['ext'] = extension
		info['name'] = filename

		info = self.getLocalFile(site_key, file, info)

		return info

	def getLocalFile(self, site_key, file, info):

		day = str(date.today())
		day = day.split('-')

		#filename =  Demeter.md5(str(uuid.uuid5(uuid.uuid1(), info['key'])))
		filename =  info['key']
		filepath = str(site_key) + '/' + day[0] + '/' + day[1] + '/' + day[2]
		filepath = File.mkdirs(os.path.join(Demeter.path, 'runtime','upload', filepath)) + '/' + filename

		local = filepath + info['ext']

		info['local'] = local
		info['pdf'] = filepath + '/' + filename + '.pdf'
		# 这里要增加权限控制 html不能直接访问
		info['html'] = filepath + '/' + filename + '.html'
		info['url'] = info['html'].replace(Demeter.path + 'runtime', '')
		info['path'] = filepath + '/'

		return info
		if File.exists(local):
			return info
		else:
			self.download(file, local);
			return info

	def download(self, file, local):
		if 'http' in file:
			import requests
			r = requests.get(file, stream=True)
			with open(local, 'wb') as up:
				for chunk in r.iter_content(chunk_size=1024):
					if chunk:
						up.write(chunk)

		else:
			import shutil
			shutil.copyfile(file, local)

		if File.exists(local):
			return True
		return False

	def command(self, info):
		File.mkdir(info['path'])
		convert = 'cd ' + info['path'] + ' && '

		if info['ext'] != '.pdf':
			convert = convert + 'libreoffice  --invisible --convert-to pdf ' + info['local']
			convert = convert + ' && '
		else:
			info['pdf'] = info['local']
		convert = convert + 'pdf2htmlEX --zoom 1.3 --no-drm 1 --split-pages 1 '
		convert = convert + '--embed-css 1  --embed-javascript 0 --embed-image 0 --embed-font 1 --process-outline 0 '
		convert = convert + '--embed-external-font 0 --dest-dir '+info['path']+' --page-filename %d.page ' + info['pdf']

		return convert


	def total(self, path):
		page = 0
		for parentdir,dirname,filenames in os.walk(path):  
			for filename in filenames:
				if os.path.splitext(filename)[1]=='.page':
					page = page + 1
		return page

	def handle(self, id):
		model = Demeter.model('convert')
		model.id = id
		info = model.select(type='fetchone')

		if not info:
			return
		status = True
		if info['status'] == 1 or info['status'] == 4:
			status = False

		if info and status == False:
			model.id = id
			update = {}
			update['status'] = 2
			model.update(update)

			if not File.exists(info['local']):
				self.download(info['file'], info['local'])

			if not File.exists(info['html']):
				handle = self.command(info)
				Shell.popen(handle)
				if File.exists(info['html']):

					#self.string_switch(info['html'], "taste", "tasting")
					# 获取有多少页
					page = self.total(info['path'])
					model.id = id
					update = {}
					update['page'] = page
					update['status'] = 3
					model.update(update)
					return

			model.id = id
			update = {}
			update['status'] = 4
			model.update(update)

	def string_switch(self, x,y,z,s=1):
		with open(x, "r", encoding="utf-8") as f:
			#readlines以列表的形式将文件读出
			lines = f.readlines()
	 
		with open(x, "w", encoding="utf-8") as f_w:
			#定义一个数字，用来记录在读取文件时在列表中的位置
			n = 0
			#默认选项，只替换第一次匹配到的行中的字符串
			if s == 1:
				for line in lines:
					if y in line:
						line = line.replace(y,z)
						f_w.write(line)
						n += 1
						break
					f_w.write(line)
					n += 1
				#将剩余的文本内容继续输出
				for i in range(n,len(lines)):
					f_w.write(lines[i])
			#全局匹配替换
			elif s == 'g':
				for line in lines:
					if y in line:
						line = line.replace(y,z)
					f_w.write(line)