wms/nflg-qms-pdf-extract/src/main/java/service/PdfExtractionService.java

package service;

import extraction.DimensionIdentifier;
import extraction.PositionedTextStripper;
import extraction.TextGrouper;
import lombok.extern.slf4j.Slf4j;
import model.DimensionResult;
import model.TextElement;
import model.TextGroup;
import org.apache.pdfbox.Loader;
import org.apache.pdfbox.pdmodel.PDDocument;

import java.io.ByteArrayOutputStream;
import java.io.File;
import java.io.IOException;
import java.io.InputStream;
import java.io.StringWriter;
import java.net.MalformedURLException;
import java.net.URL;
import java.nio.file.Path;
import java.util.List;
import java.util.stream.Collectors;

@Slf4j
public class PdfExtractionService {

    private final TextGrouper textGrouper;
    private final DimensionIdentifier dimensionIdentifier;

    public PdfExtractionService(TextGrouper textGrouper,
                                DimensionIdentifier dimensionIdentifier) {
        this.textGrouper = textGrouper;
        this.dimensionIdentifier = dimensionIdentifier;
    }

    public ExtractionResult extractDimensions(Path pdfPath, String fileId) throws IOException {
        long start = System.currentTimeMillis();
        File file = pdfPath.toFile();

        try (PDDocument document = Loader.loadPDF(file)) {
            int totalPages = document.getNumberOfPages();

            PositionedTextStripper stripper = new PositionedTextStripper();
            stripper.setSortByPosition(true);
            StringWriter writer = new StringWriter();
            stripper.writeText(document, writer);
            List<TextElement> allElements = stripper.getElements();

            List<TextGroup> groups = textGrouper.groupTextElements(allElements);
            List<DimensionResult> dimensions = dimensionIdentifier.identifyDimensions(groups);

            long elapsed = System.currentTimeMillis() - start;
            log.info("Extracted {} dimensions from {} in {}ms", dimensions.size(), pdfPath.getFileName(), elapsed);

            return new ExtractionResult(dimensions, totalPages);
        }
    }

    /**
     * 区域提取：不限公差过滤，包含所有尺寸，不保存到数据库
     */
    public List<DimensionResult> extractAllDimensionsForRegion(String pdfUrl) throws IOException {
        byte[] pdfBytes;
        try (InputStream in = new URL(pdfUrl).openStream();
             ByteArrayOutputStream buffer = new ByteArrayOutputStream()) {
            byte[] chunk = new byte[8192];
            int len;
            while ((len = in.read(chunk)) != -1) {
                buffer.write(chunk, 0, len);
            }
            pdfBytes = buffer.toByteArray();
        }
        try (PDDocument document = Loader.loadPDF(pdfBytes)) {
            PositionedTextStripper stripper = new PositionedTextStripper();
            stripper.setSortByPosition(true);
            StringWriter writer = new StringWriter();
            stripper.writeText(document, writer);
            List<TextElement> allElements = stripper.getElements();
            List<TextGroup> groups = textGrouper.groupTextElements(allElements);
            return dimensionIdentifier.identifyDimensions(groups, false);
        }
    }

    public List<DimensionResult> extractDimensions(String pdfUrl) throws IOException {
        byte[] pdfBytes;
        try (InputStream in = new URL(pdfUrl).openStream();
             ByteArrayOutputStream buffer = new ByteArrayOutputStream()) {
            byte[] chunk = new byte[8192];
            int len;
            while ((len = in.read(chunk)) != -1) {
                buffer.write(chunk, 0, len);
            }
            pdfBytes = buffer.toByteArray();
        }
        try (PDDocument document = Loader.loadPDF(pdfBytes)) {
            int totalPages = document.getNumberOfPages();
            PositionedTextStripper stripper = new PositionedTextStripper();
            stripper.setSortByPosition(true);
            StringWriter writer = new StringWriter();
            stripper.writeText(document, writer);
            List<TextElement> allElements = stripper.getElements();

            List<TextGroup> groups = textGrouper.groupTextElements(allElements);
            return dimensionIdentifier.identifyDimensions(groups);
        }
    }

    public static class ExtractionResult {
        private final List<DimensionResult> dimensions;
        private final int totalPages;

        public ExtractionResult(List<DimensionResult> dimensions, int totalPages) {
            this.dimensions = dimensions;
            this.totalPages = totalPages;
        }

        public List<DimensionResult> getDimensions() { return dimensions; }
        public int getTotalPages() { return totalPages; }
    }
}