Itext7解析pdf获取pdf文件具体详情并存储

由于原来使用的是itext5 导致有些PDF的格式类型不支持 解析失败

所以想升级到7，目前我的测试版本是7.0.3 发现有些实体类都没有了，心累 比对了新老版本 不同的实体类和实现逻辑 终于找到一点点相关的信息

非常感谢 使用iText 7读取PDF文件中的文本和图片 - 佛西亚 - 博客园  的回答，让我有点领悟到了真谛。

博文里面提到的是获取整个page的String文本，但是我这里需求是按照行读，所以尝试在原来的基础上小改动， 具体的逻辑如下 省略了一些不关键的代码

官网地址如下： iText 7 Community | iText PDF

main(){

imageTool.readEnPdfData("D:/33.pdf");

}

public Map readEnPdfData(String filePath) throws Exception {

PDDocument document = null;
List<String> contentList = CollUtil.newArrayList();
String errorMsg = "系统错误";
try {
    boolean sort = true;// 是否排序
    int startPage = 1;// 开始提取页数
    int endPage = Integer.MAX_VALUE;// 结束提取页数

    URL url = URLUtil.url(filePath);
    InputStream inputStream = URLUtil.getStream(url);
    document =PDDocument.load(inputStream);
    PDFTextStripper pts = new PDFTextStripper();
    pts.setSortByPosition(sort);
    endPage = document.getNumberOfPages();
    // System.out.println("Total Page: " + endPage);
    pts.setStartPage(startPage);
    pts.setEndPage(endPage);

    String content = pts.getText(document);
    String[] contents = StrUtil.split(content,"\n");
    contentList = Convert.toList(String.class, contents);

    PDFTextStripperByArea stripper = new PDFTextStripperByArea();
    stripper.setSortByPosition(true);

    List<Map<String, Rectangle2D.Float>> xyAllList = new PdfCoordinateKit().readXY(filePath);

    Rectangle2D.Float ingredientsXY = null, inciXY = null,byWeightXY = null,functionXY = null,specificationsXY = null,heightWidthXY = null ;

    for (Map<String, Rectangle2D.Float> m : xyAllList) {
        if (ObjectUtil.isNotNull(m.get("a")) && ObjectUtil.isNull(ingredientsXY)) {
            ingredientsXY = m.get("a");

// 省略若干行解析获得数据并存储

                }        

        }

if (specificationsXY.y >= ingredientsXY.y) {
    specificationsXY.setRect(0, 0, 0, 0); // 当Specifications跨行变成0
}

// System.out.println("============获得数据结果集=================");
List<String> dataList = this.getTextByRectangle(stripper,firstPage,ingredientsXY,heightWidthXY,
        Math.round(inciXY.x - ingredientsXY.x),
        Math.round(ingredientsXY.y - specificationsXY.y - ingredientsXY.height - ingredientsXY.height));

}

Class:

PdfCoordinateKit

  public List<Map<String, Rectangle2D.Float>> readXY(String filePath){
        List<Map<String,Rectangle2D.Float>> list_text_all=new ArrayList<Map<String,Rectangle2D.Float>> ();
        try{

            PdfReader reader = new PdfReader(filePath);
            PdfDocument redDocument = new PdfDocument(reader);


            PdfPage onePage = redDocument.getPage(1); //得到pdf整页
         Rectangle ps = onePage.getPageSize();
            float height = ps.getHeight();
            float width = ps.getWidth();
            Map<String,Rectangle2D.Float> mapWH = new HashMap<String,Rectangle2D.Float>();
            Rectangle2D.Float f = new Rectangle2D.Float();
            f.setFrame(0, 0, width, height);
            mapWH.put("heightWidth$", f);
            list_text_all.add(mapWH);

            //2、try two 采用bufferedReader

          PdfDocumentContentParser parser = new PdfDocumentContentParser(redDocument);

         for(int i = 1;i <= redDocument.getNumberOfPages();i++) {
//实现具体的listener
                TextRenderListener listener = new TextRenderListener();
      

                //获取文字的矩形边框
                List<Map<String,Rectangle2D.Float>> list_text = listener.rows_text_rect;
                for(int k = 0;k < list_text.size();k++){
                    list_text_all.add(list_text.get(k));
                }
         }
        }catch(Exception e){

        }
        return list_text_all;
    }

// itext5 和 itext7的解析类 都不太一样，7要实现自己的EventLister

public class TextRenderListener implements IEventListener {

    //用来存放文字的矩形
    List<Rectangle2D.Float> rectText = new ArrayList<Rectangle2D.Float>();
    //用来存放文字
    List<String> textList = new ArrayList<String>();
    //用来存放文字的y坐标
    List<Float> listY = new ArrayList<Float>();
    //用来存放每一行文字的坐标位置
    List<Map<String,Rectangle2D.Float>> rows_text_rect = new ArrayList<Map<String,Rectangle2D.Float>>();

    @Override
    public void eventOccurred(IEventData data, EventType type)
    {
        TextRenderInfo renderInfo;
        if (data instanceof TextRenderInfo)
        {
            renderInfo = (TextRenderInfo)data;
            try
            {
                String text = renderInfo.getText();
                if(text.length() > 0){
                    Rectangle rectBase = renderInfo.getBaseline().getBoundingRectangle();
                    //获取文字下面的矩形
                    Rectangle rectAscen = renderInfo.getAscentLine().getBoundingRectangle();
                    //计算出文字的边框矩形
                    float leftX = rectBase.getX();
                    float leftY =  rectBase.getY()-1;
                    float rightX = rectAscen.getX()+rectAscen.getWidth();
                    float rightY = rectAscen.getY()+rectAscen.getHeight()+1;

                    Rectangle2D.Float rect = new Rectangle2D.Float(leftX, leftY, rightX - leftX, rightY - leftY);

                    if(listY.contains(rect.y)){
                        int index = listY.indexOf(rect.y);
                        float tempx = rect.x > rectText.get(index).x ? rectText.get(index).x : rect.x;
                        rectText.set(index,new Rectangle2D.Float(tempx,rect.y,rect.width + rectText.get(index).width,rect.height));
                        textList.set(index,textList.get(index) + text);
                    }else{
                        rectText.add(rect);
                        textList.add(text);
                        listY.add(rect.y);
                    }

                    Map<String,Rectangle2D.Float> map = new HashMap<String,Rectangle2D.Float>();
                    map.put(text,rect);
                    rows_text_rect.add(map);
                }

            }
            catch(Exception e)
            {

            }
        }
    }

    @Override
    public Set<EventType> getSupportedEvents(){
        return null;
    }