常用文档格式转换是被我玩通了，不信你看看

代码：

#coding=utf-8

import os, sys, re, math
import mammoth
import inspect
import pypandoc
import fitz
from win32com import client
from pdf2docx import Converter
from docx import Document
from docx.shared import Inches, Cm
from lxml import etree
from pydocx import PyDocX
from PIL import Image
from win32 import win32api, win32gui, win32print
from win32.lib import win32con
from win32.win32api import GetSystemMetrics

def doc2docx(doc_file, docx_file):
    if not os.path.exists(doc_file):
        print('file[%s] is not exists' % doc_file)
        return False
    
    try:
        word = client.Dispatch("Word.Application")
        doc = word.Documents.Open(doc_file)
        #doc=0, txt=2,3,4,5,7, html=8,10, docx=12,16， pdf=17, 11=xml
        doc.SaveAs(docx_file, 12)
        doc.Close()
        word.Quit()
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    return True

def doc2pdf(doc_file, pdf_file):
    if not os.path.exists(doc_file):
        print('file[%s] is not exists' % doc_file)
        return False
    
    try:
        word = client.Dispatch("Word.Application")
        doc = word.Documents.Open(doc_file)
        #doc=0, txt=2,3,4,5,7, html=8,10, docx=12,16， pdf=17, 11=xml
        doc.SaveAs(pdf_file, 17)
        doc.Close()
        word.Quit()
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    return True

def doc2html(doc_file, html_file, encoding='utf-8'):
    if not os.path.exists(doc_file):
        print('file[%s] is not exists' % doc_file)
        return False
    
    style_map = """
p[style-name='Section Title'] => h1:fresh
p[style-name='Subsection Title'] => h2:fresh
"""
    htmltmp = ""
    try:
        with open(doc_file, "rb") as f:
            res = mammoth.convert_to_html(f, style_map=style_map)
            htmltmp = res.value
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    try:
        htmltmp = htmltmp.decode(encoding='utf-8' ,errors ='ignore')
    except Exception as e:
        pass
    
    try:
        htmltmp = htmltmp.encode(encoding=encoding ,errors ='ignore')
    except Exception as e:
        pass
    
    try:
        h = open(html_file, "wb")
        h.write(htmltmp)
        h.close()
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    return True

def doc2jpg(doc_file, jpg_file):
    pdf_file = doc_file.replace('.doc', '-tmp.pdf')
    if doc2pdf(doc_file, pdf_file):
        if pdf2jpg(pdf_file, jpg_file):
            os.unlink(pdf_file)
            return True
        
    return False

def doc2txt(doc_file, txt_file):
    if not os.path.exists(doc_file):
        print('file[%s] is not exists' % doc_file)
        return False
    
    try:
        word = client.Dispatch("Word.Application")
        doc = word.Documents.Open(doc_file)
        #doc=0, txt=2,3,4,5,7, html=8,10, docx=12,16， pdf=17, 11=xml
        doc.SaveAs(txt_file, 4)
        doc.Close()
        word.Quit()
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    return True

def docx2doc(docx_file, doc_file):
    if not os.path.exists(docx_file):
        print('file[%s] is not exists' % docx_file)
        return False
    
    try:
        word = client.Dispatch("Word.Application")
        doc = word.Documents.Open(docx_file)
        #doc=0, txt=2,3,4,5,7, html=8,10, docx=12,16， pdf=17, 11=xml
        doc.SaveAs(doc_file, 0)
        doc.Close()
        word.Quit()
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    return True

def docx2pdf(docx_file, pdf_file):
    if not os.path.exists(docx_file):
        print('file[%s] is not exists' % docx_file)
        return False
    
    try:
        word = client.Dispatch("Word.Application")
        doc = word.Documents.Open(docx_file)
        #doc=0, txt=2,3,4,5,7, html=8,10, docx=12,16， pdf=17, 11=xml
        doc.SaveAs(pdf_file, 17)
        doc.Close()
        word.Quit()
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    return True

def docx2html(docx_file, html_file, encoding='utf-8'):
    if not os.path.exists(docx_file):
        print('file[%s] is not exists' % docx_file)
        return False
    
    style_map = """
p[style-name='Section Title'] => h1:fresh
p[style-name='Subsection Title'] => h2:fresh
"""
    htmltmp = ""
    try:
        with open(docx_file, "rb") as f:
            res = mammoth.convert_to_html(f, style_map=style_map)
            htmltmp = res.value
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    try:
        htmltmp = htmltmp.decode(encoding='utf-8' ,errors ='ignore')
    except Exception as e:
        pass
    
    try:
        htmltmp = htmltmp.encode(encoding=encoding ,errors ='ignore')
    except Exception as e:
        pass
    
    try:
        h = open(html_file, "wb")
        h.write(htmltmp)
        h.close()
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    return True

def docx2jpg(docx_file, jpg_file):
    pdf_file = docx_file.replace('.docx', '-tmp.pdf')
    if docx2pdf(docx_file, pdf_file):
        if pdf2jpg(pdf_file, jpg_file):
            os.unlink(pdf_file)
            return True
        
    return False

def docx2txt(docx_file, txt_file):
    if not os.path.exists(docx_file):
        print('file[%s] is not exists' % docx_file)
        return False
    
    try:
        word = client.Dispatch("Word.Application")
        doc = word.Documents.Open(docx_file)
        #doc=0, txt=2,3,4,5,7, html=8,10, docx=12,16， pdf=17, 11=xml
        doc.SaveAs(txt_file, 4)
        doc.Close()
        word.Quit()
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    return True

def pdf2doc(pdf_file, doc_file):
    if not os.path.exists(pdf_file):
        print('file[%s] is not exists' % pdf_file)
        return False
    
    try:
        cv = Converter(pdf_file)
        cv.convert(doc_file)
        cv.close()
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    return True

def pdf2docx(pdf_file, docx_file):
    if not os.path.exists(pdf_file):
        print('file[%s] is not exists' % pdf_file)
        return False
    
    try:
        cv = Converter(pdf_file)
        cv.convert(docx_file)
        cv.close()
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    return True

def pdf2html(pdf_file, html_file):
    doc_file = pdf_file.replace('.pdf', '-tmp.doc')
    if pdf2doc(pdf_file, doc_file):
        if doc2html(doc_file, html_file):
            os.unlink(doc_file)
            return True
        
    return False

def pdf2jpg(pdf_file, jpg_file):
    if not os.path.exists(pdf_file):
        print('file[%s] is not exists' % pdf_file)
        return False
    
    dirname = os.path.dirname(jpg_file)
    basename = os.path.basename(jpg_file)
    names = basename.split('.')
    basename = '.'.join(names[0:len(names)-1])
    
    try:
        doc = fitz.open(pdf_file)
        i=0
        tar_h = 0
        for page in doc:
            page = doc.load_page(i) 
            pix = page.get_pixmap()
            tar_h += pix.height
            #jpg_file = dirname+'\\'+basename+'_'+str(i)+'.jpg'
            #pix.save(jpg_file)
            i += 1
        src = doc[0].get_pixmap()
        tar_pix = fitz.Pixmap(src.colorspace, (0, 0, src.width, tar_h), src.alpha)
        
        i = 0
        for page in doc:
            page = doc.load_page(i) 
            pix = page.get_pixmap()
            pix.set_origin(0, pix.height * i)
            tar_pix.copy(pix, pix.irect)
            i += 1
        tar_pix.save(jpg_file)
        
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    return True

def pdf2txt(pdf_file, txt_file):
    if not os.path.exists(pdf_file):
        print('file[%s] is not exists' % pdf_file)
        return False
    
    all_text = ''
    try:
        doc = fitz.open(pdf_file)
        i=0
        for page in doc:
            page = doc.load_page(i) 
            text = page.get_text()
            all_text += "\n\n"+text.strip()
            i += 1
            
        while "\n " in all_text:
            all_text = all_text.replace("\n ", "\n")
            
        while "\n\n\n" in all_text:
            all_text = all_text.replace("\n\n\n", "\n\n")
            
        while all_text.startswith("\n"):
            all_text = all_text[1:len(all_text)]
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    try:
        h = open(txt_file, "w")
        h.write(all_text)
        h.close()
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    return True

def html2doc(html_file, doc_file):
    if not os.path.exists(html_file):
        print('file[%s] is not exists' % html_file)
        return False
    
    try:
        pypandoc.convert_file(html_file, 'docx', outputfile=doc_file)
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    return True

def html2docx(html_file, docx_file):
    if not os.path.exists(html_file):
        print('file[%s] is not exists' % html_file)
        return False
    
    try:
        pypandoc.convert_file(html_file, 'docx', outputfile=docx_file)
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    return True

def html2pdf(html_file, pdf_file):
    doc_file = html_file.replace('.html', '-tmp.doc')
    if html2doc(html_file, doc_file):
        if doc2pdf(doc_file, pdf_file):
            os.unlink(doc_file)
            return True
        
    return False

def html2jpg(html_file, jpg_file):
    doc_file = html_file.replace('.html', '-tmp.doc')
    if html2doc(html_file, doc_file):
        if doc2jpg(doc_file, jpg_file):
            os.unlink(doc_file)
            return True
        
    return False

def html2txt(html_file, txt_file):
    doc_file = html_file.replace('.html', '-tmp.doc')
    if html2doc(html_file, doc_file):
        if doc2txt(doc_file, txt_file):
            os.unlink(doc_file)
            return True
        
    return False

def jpg2doc(jpg_file, doc_file):
    docx_file = jpg_file.replace('.jpg', '-tmp.docx')
    if jpg2docx(jpg_file, docx_file):
        if docx2doc(docx_file, doc_file):
            os.unlink(docx_file)
            return True
        
    return False

def jpg2docx(jpg_file, docx_file):
    if not os.path.exists(jpg_file):
        print('file[%s] is not exists' % jpg_file)
        return False
    
    img = Image.open(jpg_file)
    width = img.width
    height = img.height
    perheight = 800.00
    pagenum = math.ceil(height / perheight)
    
    i = 0
    try:
        document = Document()
        while i < pagenum:
            upper = int(i*perheight)
            if (i+1)*perheight > height:
                lower = int(upper + height - i*perheight)
            else:
                lower = int(upper + perheight)
            
            # 从左上角开始 剪切 200*200的图片
            #(左、上、右、下坐标)
            #(left, upper, right, lower)
            # img2 = img.crop((0, 0, 200, 200))
            jpg_file_tmps = []
            shape = (0, upper, width, lower)
            try:
                jpg_file_tmp = jpg_file.replace('.jpg', '-tmp-%s.jpg' % str(i))
                region = img.crop(shape)
                region.save(jpg_file_tmp)
                jpg_file_tmps.append(jpg_file_tmp)
                
                img_tmp = region
                width_tmp = img_tmp.width  / getdpi() * 2.54
                height_tmp = img_tmp.height  / getdpi() * 2.54
                document.add_picture(jpg_file_tmp, width=Cm(width_tmp), height=Cm(height_tmp))
                os.unlink(jpg_file_tmp)
                i += 1
            except Exception as e:
                print('文件：%s' % __file__)
                print('行：%s' % str(inspect.currentframe().f_lineno))
                print('错误：%s' % e)
                print()
                i += 1
                return False
        
        document.save(docx_file)
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        i += 1
        return False
    
    return True

def jpg2pdf(jpg_file, pdf_file):
    patten = re.compile("[\.|_\d]")
    patten2 = re.compile("[\.|_]")
    dirname = os.path.dirname(jpg_file)
    basename = os.path.basename(jpg_file)
    names = patten.split(basename)
    filename = '.'.join(names[0:len(names)-1])
    ext = names[-1]
    
    nums = []
    for i in os.listdir(dirname):
        tmp_names = patten.split(i)
        tmp_names2 = patten2.split(i)
        tmp_names = [j for j in tmp_names if j != '']
        tmp_filename = '.'.join(tmp_names[0:len(tmp_names)-1])
        tmp_ext = tmp_names[-1]
        if tmp_ext == ext and tmp_filename == filename and len(tmp_names2) >= 3:
            nums.append(int(tmp_names2[-2]))
            
    if len(nums) > 0:
        minnum = min(nums)
        maxnum = max(nums)
    else:
        minnum = 0
        maxnum = 0
    
    filepaths = []
    if os.path.exists(jpg_file):
        filepaths.append(jpg_file)
        
    for i in range(minnum, maxnum+1):
        filepath = os.path.join(dirname, filename+'_'+str(i)+'.'+ext)
        if os.path.exists(filepath):
            filepaths.append(filepath)
    
    try:
        doc = fitz.open()
        for filepath in filepaths:
            img = fitz.open(filepath)
            pdfbytes = img.convert_to_pdf()
            imgpdf = fitz.open("pdf", pdfbytes)
            doc.insert_pdf(imgpdf)
            
        doc.save(pdf_file)
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    return True

def jpg2html(jpg_file, html_file):
    docx_file = jpg_file.replace('.jpg', '-tmp.docx')
    if jpg2docx(jpg_file, docx_file):
        if docx2html(docx_file, html_file):
            os.unlink(docx_file)
            return True
        
    return False

def jpg2txt(jpg_file, txt_file):
    print('Not Supported func jpg2txt')
    return False

def txt2doc(txt_file, doc_file):
    docx_file = jpg_file.replace('.txt', '-tmp.docx')
    if txt2docx(txt_file, docx_file):
        if docx2doc(docx_file, doc_file):
            os.unlink(docx_file)
            return True
        
    return False

def txt2docx(txt_file, docx_file):
    if not os.path.exists(txt_file):
        print('file[%s] is not exists' % txt_file)
        return False
    
    try:
        text = ''
        with open(txt_file, 'r') as f:
            text = f.read()
        document = Document()
        document.add_paragraph(text)
        document.save(docx_file)
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    return True

def txt2pdf(txt_file, pdf_file):
    if not os.path.exists(txt_file):
        print('file[%s] is not exists' % txt_file)
        return False
    
    try:
        word = client.Dispatch("Word.Application")
        doc = word.Documents.Open(txt_file)
        #doc=0, txt=2,3,4,5,7, html=8,10, docx=12,16， pdf=17, 11=xml
        doc.SaveAs(pdf_file, 17)
        doc.Close()
        word.Quit()
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    return True

def txt2html(txt_file, html_file):
    if not os.path.exists(txt_file):
        print('file[%s] is not exists' % txt_file)
        return False
    
    try:
        word = client.Dispatch("Word.Application")
        doc = word.Documents.Open(txt_file)
        #doc=0, txt=2,3,4,5,7, html=8,10, docx=12,16， pdf=17, 11=xml
        doc.SaveAs(html_file, 8)
        doc.Close()
        word.Quit()
    except Exception as e:
        print('文件：%s' % __file__)
        print('行：%s' % str(inspect.currentframe().f_lineno))
        print('错误：%s' % e)
        print()
        return False
    
    return True

def txt2jpg(txt_file, jpg_file):
    pdf_file = txt_file.replace('.txt', '-tmp.pdf')
    if txt2pdf(txt_file, pdf_file):
        if pdf2jpg(pdf_file, jpg_file):
            os.unlink(pdf_file)
            return True
        
    return False

def ext2ext(src_file, desc_file):
    patten = re.compile("[\.|_\d]")
    dirname = os.path.dirname(src_file)
    basename = os.path.basename(src_file)
    names = patten.split(basename)
    filename = '.'.join(names[0:len(names)-1])
    ext = names[-1]
    
    existsflag = False
    if os.path.exists(src_file):
        existsflag = True
    else:
        for i in os.listdir(dirname):
            tmp_names = patten.split(i)
            tmp_names = [j for j in tmp_names if j != '']
            tmp_filename = '.'.join(tmp_names[0:len(tmp_names)-1])
            tmp_ext = tmp_names[-1]
            if tmp_filename == filename and tmp_ext == ext:
                existsflag = True
    
    if existsflag == False:
        print('file[%s] is not exists' % src_file)
        return False
    
    src_ext = src_file.split('.')[-1]
    desc_ext = desc_file.split('.')[-1]
    func = '%s2%s' % (src_ext, desc_ext)
    if not func in globals().keys():
        print('func[%s] is not defined' % func)
        return False
    
    res = eval(func+"(r'"+src_file+"', r'"+desc_file+"')")
    
    return res

def get_real_resolution():
    """获取真实的分辨率"""
    hDC = win32gui.GetDC(0)
    # 横向分辨率
    w = win32print.GetDeviceCaps(hDC, win32con.DESKTOPHORZRES)
    # 纵向分辨率
    h = win32print.GetDeviceCaps(hDC, win32con.DESKTOPVERTRES)
    
    return w, h

def get_screen_size():
    """获取缩放后的分辨率"""
    w = GetSystemMetrics (0)
    h = GetSystemMetrics (1)
    return w, h

def getdpi():
    real_resolution = get_real_resolution()
    screen_size = get_screen_size()
    screen_scale_rate = round(real_resolution[0] / screen_size[0], 2)
    screen_scale_rate = screen_scale_rate * 100
    
    return screen_scale_rate

if __name__ == '__main__':
    doc_file = os.getcwd()+'\\'+'test.doc'
    docx_file = os.getcwd()+'\\'+'test.docx'
    html_file = os.getcwd()+'\\'+'test.html'
    pdf_file = os.getcwd()+'\\'+'test.pdf'
    jpg_file = os.getcwd()+'\\'+'test.jpg'
    txt_file = os.getcwd()+'\\'+'test.txt'
    
    ext2ext(jpg_file, txt_file)

 

执行结果：