java 获取pdf关键字坐标

要获取 PDF 关键字的坐标，可以使用 Apache PDFBox 库。以下是获取 PDF 关键字坐标的示例代码：

import java.io.File;
import java.io.IOException;
import java.util.List;

import org.apache.pdfbox.cos.COSArray;
import org.apache.pdfbox.cos.COSBase;
import org.apache.pdfbox.cos.COSDictionary;
import org.apache.pdfbox.cos.COSName;
import org.apache.pdfbox.pdmodel.PDDocument;
import org.apache.pdfbox.pdmodel.PDPage;
import org.apache.pdfbox.pdmodel.interactive.annotation.PDAnnotation;
import org.apache.pdfbox.pdmodel.interactive.annotation.PDAnnotationTextMarkup;
import org.apache.pdfbox.pdmodel.interactive.annotation.PDAnnotationTextMarkup.MarkupTypeEnum;
import org.apache.pdfbox.pdmodel.interactive.annotation.PDAnnotationTextMarkup.QuadPoints;

public class PDFKeywordCoordinates {
    public static void main(String[] args) throws IOException {
        String filePath = "example.pdf";
        String keyword = "example";
        PDDocument document = PDDocument.load(new File(filePath));
        List<PDPage> pages = document.getDocumentCatalog().getAllPages();
        for (PDPage page : pages) {
            List<PDAnnotation> annotations = page.getAnnotations();
            for (PDAnnotation annotation : annotations) {
                if (annotation instanceof PDAnnotationTextMarkup) {
                    PDAnnotationTextMarkup markup = (PDAnnotationTextMarkup) annotation;
                    if (markup.getMarkupType() == MarkupTypeEnum.HIGHLIGHT) {
                        COSDictionary dict = markup.getCOSObject();
                        COSArray quadPointsArray = (COSArray) dict.getDictionaryObject(COSName.getPDFName("QuadPoints"));
                        for (int i = 0; i < quadPointsArray.size(); i += 8) {
                            float x1 = quadPointsArray.getFloat(i);
                            float y1 = quadPointsArray.getFloat(i + 1);
                            float x2 = quadPointsArray.getFloat(i + 2);
                            float y2 = quadPointsArray.getFloat(i + 3);
                            float x3 = quadPointsArray.getFloat(i + 4);
                            float y3 = quadPointsArray.getFloat(i + 5);
                            float x4 = quadPointsArray.getFloat(i + 6);
                            float y4 = quadPointsArray.getFloat(i + 7);
                            if (containsKeyword(page, keyword, x1, y1, x2, y2, x3, y3, x4, y4)) {
                                System.out.println("Keyword '" + keyword + "' found on page " + (pages.indexOf(page) + 1) +
                                        " at (" + x1 + "," + y1 + ") (" + x2 + "," + y2 + ") (" + x3 + "," + y3 + ") (" + x4 + "," + y4 + ")");
                            }
                        }
                    }
                }
            }
        }
        document.close();
    }

    private static boolean containsKeyword(PDPage page, String keyword, float x1, float y1, float x2, float y2, float x3, float y3, float x4, float y4) throws IOException {
        StringBuilder sb = new StringBuilder();
        sb.append("q\n");
        sb.append(x1).append(' ').append(y1).append(" m\n");
        sb.append(x2).append(' ').append(y2).append(" l\n");
        sb.append(x3).append(' ').append(y3).append(" l\n");
        sb.append(x4).append(' ').append(y4).append(" l\n");
        sb.append("h\n");
        sb.append("W* n\n");
        sb.append("BT\n");
        sb.append("/Helv 12 Tf\n");
        sb.append("0 g\n");
        sb.append("1 0 0 1 ").append(x1).append(' ').append(y1).append(" Tm\n");
        sb.append("(").append(keyword).append(") Tj\n");
        sb.append("ET\n");
        sb.append("Q\n");
        return page.getContents().stream().anyMatch(content -> content.getString().contains(sb.toString()));
    }
}

在此示例代码中，我们首先加载 PDF 文件并获取所有页面。然后，我们遍历每个页面的所有注释，并查找类型为“高亮”的注释。对于每个高亮注释，我们获取该注释的四个顶点坐标，并检查关键字是否包含在其中。如果找到匹配项，则打印关键字的坐标。