word_parse_of_eng/get_word_content/filepath2text.py

import os
import re

import textract

from img2text import img2text
from pdf2text import route_pdf
from ppt2text import ppt2text
from word2text import word2text
from wand.image import Image

page_num1 = re.compile("第\s*\d+\s*页")
page_num2 = re.compile("共\s*\d+\s*页")


def post_preprocess(text):
    if not text:
        print('\n+++++++该文件无法解析,请仔细核对文件，如果是PDF和图片请走图像组+++++++\n')
        return None

    res = []
    lines = text.replace("\r", "").split("\n")
    for line in lines:
        if not line.strip():
            continue
        if page_num1.search(line):
            continue
        if page_num2.search(line):
            continue
        res.append(line)
    return "\n".join(res)


def route_filename_old(filename):
    basename, ext = os.path.splitext(filename)
    ext = ext.lower()
    if ext in [".gif", ".jpg", ".jpeg", ".png", ".tiff", ".tif"]:
        try:
            text = img2text(filename)
        except:
            # 有些tif 格式会出现bug, 进行格式转换
            new_filename = 'result.jpg'
            with Image(filename=filename) as img:
                # img.resize(200, 200)
                img.save(filename=new_filename)
                text = img2text(new_filename)

    elif ext in [".doc", ".docx"]:
        text = word2text(filename)
    elif ext in [".pdf"]:
        text = route_pdf(filename)
    elif ext in [".ppt"]:
        text = ppt2text(filename)
    elif ext in [".rtf"]:
        text = textract.process(filename)
        text = text.decode("utf-8")
        text = "\n".join(text.replace("\r", "").split("\n")[7:])
    else:
        text = textract.process(filename)
        text = text.decode("utf-8")

    text = post_preprocess(text)
    return text


def route_filename(filename):
    basename, ext = os.path.splitext(filename)
    ext = ext.lower()

    if ext in [".doc", ".docx"]:
        text = word2text(filename)

    # elif ext in [".ppt"]:
    #     text = ppt2text(filename)
    #
    # elif ext in [".rtf"]:
    #     text = textract.process(filename)
    #     text = text.decode("utf-8")
    #     text = "\n".join(text.replace("\r", "").split("\n")[7:])
    else:
        text = None

    text = post_preprocess(text)

    return text


if __name__ == "__main__":

    filepath = r'result.jpg'
    text = route_filename(filepath)
    print(text)