使用python找到PDF文件的文本位置、字体大小、字体名称和字体颜色

def pdf2html(input_path):
    '''
    将pdf转成html
    :param input_path:
    :return:
    '''
    doc = fitz.open(input_path)
    html_content = ''
    for page in tqdm(doc):
        html_content += page.get_text('html')

    # print('开始输出html文件')html_path
    # #html_content +="</body></html>"
    html_path = "d:/ann/input.html"
    with open(html_path, 'w', encoding='utf-8', newline='')as fp:
        fp.write(html_content)
    return html_content

def decodestylestr(style_attrs, attr):
    '''
    解析style
    :param style_attrs:
    :param attr:
    :return:
    '''
    attrvalue = ""
    styles = style_attrs.split(";")
    for sty in styles:
        k, v = sty.split(":")
        v = v.replace("pt", "")
        if k == attr:
            attrvalue = v
    return attrvalue

def pdf2list(input_path):
    '''
    按照p节点提取pdf文本，按照 [文本,left,top,[(fontname、fongsize,fontcolor),]]   (fontname、fongsize,fontcolor)一个或多个存储。
    :param input_path:
    :return:
    '''
    html_content = pdf2html(input_path)  # pdf转html
    bs_obj = BeautifulSoup(html_content, "html.parser")

    #读取P节点
    ptag = bs_obj.findAll("p")
    contents = []
    # 取P节点下文本以及其对应的left值和font-family和font-size的值。
    for p in ptag:
        ptext = p.text
        ptextnospace = ptext.replace(" ", "")
        #如果当前节点text为空，则下一个
        if len(ptextnospace) == 0:  # 当前文本为空字符串
            continue
        else:
            pass

        '''
        读取P节点下的style属性
        '''
        postioninfo=('','',)
        if 'style' in p.attrs:
            attributes = p.attrs['style']
            leftvalue = decodestylestr(attributes, "left")
            topvalue = decodestylestr(attributes, "top")
            postioninfo=(leftvalue,topvalue)
        else:
            postioninfo=('','',)

            pass

        '''
        获取P节点下的span节点，并读取取style属性，主要包括字体名称、字体大小、字体颜色，是否加粗pdf2html没有提取到。如果有也应该获取
        pspans = p.find_all("span",recursive=False ) recursive=False只获取当前节点下的子节点，不循环其孙子及以下节点
        '''
        pspans = p.find_all("span")
        pspansstyles = []

        for pspan in pspans:
            pspantext = pspan.text
            pspantext=pspantext.replace(" ","")
            if len(pspantext) > 0:#当前span节点不为空。
                if 'style' in pspan.attrs:
                    attributes = pspan.attrs['style']
                    fontfamilyvalue = decodestylestr(attributes, "font-family")
                    fontsizevalue   = decodestylestr(attributes, "font-size")
                    fontcolorvalue  = decodestylestr(attributes, "color")
                    pspansstyle = (fontfamilyvalue, fontsizevalue,fontcolorvalue)
                    if pspansstyle in pspansstyles:#如果字体样式已经存在，则删除，在增加，保持最后的是字体的样子，后续判断要用到字体大小
                        pspansstyles.remove(pspansstyle)
                    pspansstyles.append(pspansstyle)
                else:
                    pass
        ptextattrs = [ptext, postioninfo, pspansstyles]
        contents.append(ptextattrs)
    return contents