文档转换之PDF转换为HTML

我研究的是Aspose.PDF，感觉效果挺好的，具体使用官网上介绍的都有，但是都比较简单，主要是起示例作用。

结合自己实际工作，我对相关方法进行了整理，自己写了个工具类，当然，在使用这个工具类之前，要先引用aspose.pdf.dll。

官方的dll文件只能处理前4页内容，而且还带版权信息，我用的是破解版的，没有限制，但是个别文件处理效果不如官方正式的。破解版在csdn上有下载，需要的可以自己去找。 

 

在使用我这个工具类转换为html之前，必须对filepath和folder进行定义，切记切记，否则，样式表、字体库、图片什么的就对不上号了。

生成编辑图什么的，定义不定义filepath和folder就无关紧要了

 

关于使用aspose.word，另有帖子做了详细说明，在此标记留用

http://www.cnblogs.com/wuhuacong/archive/2012/08/30/2662961.html

 

class PdfTools
{
    /// <summary>
    /// 源文件
    /// </summary>
    public string filepath { get; set; }

    /// <summary>
    /// 
    /// </summary>
    private string filename {
        get {
            string _filename = Path.GetFileName(filepath);
            _filename = _filename.Substring(0, _filename.IndexOf('.'));
            return _filename;
        }
    }

    /// <summary>
    /// 最终文件目标夹
    /// </summary>
    public string folder { get; set; }

    public PdfTools() { }
    public PdfTools(string _filepath, string _folder)
    {
        filepath = _filepath;
        folder = _folder;

        folder = folder.Replace("/",@"\");
        if (folder.EndsWith(@"\") == false)
            folder = folder + @"\";
    }

    /// <summary>
    /// 将PDF文件转换成一个完整的HTML
    /// filepath为PDF源文件，必须将文件路径填写完整
    /// folder为最终保存目录，生成文件均在此目录下
    /// </summary>
    /// <returns></returns>
    public bool Pdf2Html_NoSplit()
    {
        //判断PDF文件是否存在
        if (File.Exists(filepath) == false)
            return false;

        folder = folder.Replace("/", @"\");
        if (folder.EndsWith(@"\") == false)
            folder = folder + @"\";

        if (Directory.Exists(folder) == true)
            Directory.Delete(folder,true);
        Directory.CreateDirectory(folder);

        Document doc = new Document(filepath);

        HtmlSaveOptions hso = new HtmlSaveOptions();
        hso.SplitIntoPages = false;
        hso.FixedLayout = true;

        hso.CompressSvgGraphicsIfAny = true;

        //我偏向于所有资源存放在一个文件夹内，因此，需要重写下面三个函数
        hso.CustomResourceSavingStrategy = new HtmlSaveOptions.ResourceSavingStrategy(ResourceStrategy);
        hso.CustomCssSavingStrategy = new HtmlSaveOptions.CssSavingStrategy(CssStrategy);
        hso.CustomStrategyOfCssUrlCreation = new HtmlSaveOptions.CssUrlMakingStrategy(CssUrlStrategy);
        hso.FontSavingMode = HtmlSaveOptions.FontSavingModes.AlwaysSaveAsTTF;

        string outpath = folder + filename + ".html";
        try
        {
            doc.Save(outpath, hso);
            return true;
        }
        catch
        {
            Directory.Delete(folder,true);
            return false;
        }
    }
/// <summary>
    /// 根据PDF文件生成首页缩略图
    /// </summary>
    /// <param name="_input">PDF源文件完整路径</param>
    /// <param name="_output">图片保存路径</param>
    public void CreateThumb(string _input,string _output) {
        if (File.Exists(_output) == true)
            File.Delete(_output);

        using (Document doc = new Document(_input))
        using (FileStream imgstream = new FileStream(_output, FileMode.Create))
        {
            //图片显示质量
            Aspose.Pdf.Devices.Resolution resolution = new Aspose.Pdf.Devices.Resolution(100);
            Aspose.Pdf.Devices.PngDevice device = new Aspose.Pdf.Devices.PngDevice(resolution);
            device.Process(doc.Pages[1], imgstream);
            imgstream.Close();
        }
    }

    /// <summary>
    /// 从PDF文件中提取正文
    /// </summary>
    /// <param name="_input">PDF文件目录</param>
    /// <returns></returns>
    public string GetTxtFromPDF(string _input)
    {
        Document doc = new Document(_input);
        StringBuilder sb = new StringBuilder();
        for (int i = 1; i <= doc.Pages.Count; i++)
        {
            Aspose.Pdf.Text.TextAbsorber ab = new Aspose.Pdf.Text.TextAbsorber();
            ab.TextSearchOptions.LimitToPageBounds = true;
            try
            {
                doc.Pages[i].Accept(ab);
                sb.Append(ab.Text);
            }
            catch {
                return null;
            }
        }

        sb = sb.Replace("\r\n", " ").Replace("　"," "); 

        string output = sb.ToString();
        while (output.IndexOf("  ") >= 0)
            output = output.Replace("  "," ");
        return output;
    }
/// <summary>
    /// 对字体、图片进行重命名，如果不用该函数，页面可能出现异常
    /// </summary>
    /// <param name="resource"></param>
    /// <returns></returns>
    private string ResourceStrategy(SaveOptions.ResourceSavingInfo resource)
    {
        string resourcename = "";
        if (resource.ResourceType == SaveOptions.NodeLevelResourceType.Image)
            resourcename = Guid.NewGuid().ToString() + Path.GetExtension(resource.SupposedFileName);
        else
            resourcename = resource.SupposedFileName;

        if (!Directory.Exists(folder))
            Directory.CreateDirectory(folder);

        string outfile = folder + resourcename;
        if (File.Exists(resourcename) == true)
            return resourcename;

        System.IO.BinaryReader reader = new BinaryReader(resource.ContentStream);
        System.IO.File.WriteAllBytes(outfile, reader.ReadBytes((int)resource.ContentStream.Length));
        return resourcename;
    }

/// <summary>
    /// 生成页面引用的css文件
    /// </summary>
    /// <param name="resource"></param>
    private void CssStrategy(HtmlSaveOptions.CssSavingInfo resource)
    {
        string path = folder+ filename + @".css";
        BinaryReader reader = new BinaryReader(resource.ContentStream);
        File.WriteAllBytes(path, reader.ReadBytes((int)resource.ContentStream.Length));
    }

    /// <summary>
    /// 对页面引用的css地址进行重写
    /// </summary>
    /// <param name="resource"></param>
    /// <returns></returns>
    private string CssUrlStrategy(HtmlSaveOptions.CssUrlRequestInfo resource)
    {
        return filename + ".css";
    }
}

 

转载于:https://www.cnblogs.com/theluther/p/3920941.html