word_parse_of_eng/get_word_content/toTxt.py

import json
import os
import re

import requests

rootdir = ""
suffixes = ".csv .doc .docx .gif .jpg .jpeg .json .html .htm .pdf .png .pptx .ps .rtf .tiff .tif .txt .wav .xlsx .xls"
suffixes = suffixes.split(" ")


def process(txt):
    txt = re.sub("\n+", "\n", txt)
    txt = re.sub("\n第\s*[0-9]\s*页\n", "", txt)
    return txt


def getDocs(rootdir):
    c = []
    for parent, dirnames, filenames in os.walk(rootdir):
        for filename in filenames:
            suffix = filename.split(".")[-1]
            if "." + suffix in suffixes:
                c.append(os.path.join(parent, filename))
    return c


def ocr_trans(upload_file, save_route):
    url = "http://117.50.17.141/ocr"
    data = {}
    files = {"mydata": open(upload_file, "rb")}
    r = requests.post(url, data, files=files)
    if r.status_code != 200:
        print(r.status_code)
        return False
    try:
        text = json.loads(r.text)
    except:
        print(r.text)
        return False
    if text["status"] == "ERROR":
        print(text)
        return False
    text = text["text"]
    process(text)
    with open(save_route, "w") as fn:
        fn.write(text)
    return True


def ocr_test():
    url = "http://117.50.17.141/ocr"
    data = {}
    rootdir = "/".join(os.path.abspath(__file__).split("/")[:-1])
    filename = "BERKUT金雕雷达测速仪.doc"
    filename = rootdir + "/" + filename
    print(filename)
    files = {"file": open(filename, "rb")}
    r = requests.post(url, data, files=files)
    print(r.text)
    print(r.status_code)


def ocr_transAll(upload_route=None, save_route=None):
    if upload_route == None:
        upload_route = "/".join(os.path.abspath(__file__).split("/")[:-1])
    if save_route == None:
        save_route = "/".join(os.path.abspath(__file__).split("/")[:-1])
    docs = getDocs(upload_route)
    for doc in docs:
        file_name = doc.split("/")[-1]
        file_name = ".".join(file_name.split(".")[:-1])
        save_name = save_route + "/" + file_name + ".txt"
        if not ocr_trans(doc, save_name):
            print(file_name + "转换失败")