在 Java 中使用 GroupDocs.Comparison 设置文档元数据

<repositories>
   <repository>
      <id>repository.groupdocs.com</id>
      <name>GroupDocs Repository</name>
      <url>https://releases.groupdocs.com/comparison/java/</url>
   </repository>
</repositories>
<dependencies>
   <dependency>
      <groupId>com.groupdocs</groupId>
      <artifactId>groupdocs-comparison</artifactId>
      <version>25.2</version>
   </dependency>
</dependencies>

import com.groupdocs.comparison.Comparer;

public class DocumentComparison {
    public static void main(String[] args) {
        // Initialize the comparer with source document path
        try (Comparer comparer = new Comparer("sourceFilePath")) {
            // Your comparison logic goes here
        }
    }
}

import com.groupdocs.comparison.examples.SampleFiles;

String outputFileName = SampleFiles.getOutputDirectoryPath("SetDocumentMetadataTarget");

try (Comparer comparer = new Comparer(SampleFiles.SOURCE_WORD)) {
    comparer.add(SampleFiles.TARGET1_WORD);

final Path resultPath = comparer.compare(outputFileName, new SaveOptions.Builder()
        .setCloneMetadataType(MetadataType.TARGET)
        .build());

public void processBatchComparison(List<String> documentPairs) {
    for (String[] pair : documentPairs) {
        try (Comparer comparer = new Comparer(pair[0])) {
            comparer.add(pair[1]);
            comparer.compare(generateOutputPath(pair), new SaveOptions.Builder()
                .setCloneMetadataType(determineMetadataStrategy(pair))
                .build());
        }
    }
}

private MetadataType selectMetadataType(Document source, Document target) {
    // Custom logic based on document age, author, or other criteria
    if (target.getLastModified().after(source.getLastModified())) {
        return MetadataType.TARGET;
    }
    return MetadataType.SOURCE;
}

// Verify your MetadataType setting
SaveOptions options = new SaveOptions.Builder()
    .setCloneMetadataType(MetadataType.SOURCE) // Make sure this matches your intention
    .build();

// Check if the output format supports your metadata
// Some formats (like plain text) don't support rich metadata

// Proper resource management
try (Comparer comparer = new Comparer(sourceDoc)) {
    // Your comparison logic
} // Automatic resource cleanup happens here

// For large documents, consider streaming approaches
// and increase JVM memory: -Xmx4g

// Robust path handling
Path outputDir = Paths.get(System.getProperty("user.dir"), "output");
if (!Files.exists(outputDir)) {
    Files.createDirectories(outputDir);
}

public class LegalDocumentProcessor {
    public Path compareContracts(String originalContract, String revisedContract) {
        try (Comparer comparer = new Comparer(originalContract)) {
            comparer.add(revisedContract);
            
            // Always preserve original metadata for legal compliance
            return comparer.compare(generateLegalOutputPath(), new SaveOptions.Builder()
                .setCloneMetadataType(MetadataType.SOURCE)
                .build());
        }
    }
}

public class VersionControlIntegration {
    public DocumentVersion createNewVersion(String baseDocument, String updatedDocument) {
        // Logic to determine if this is a major or minor update
        MetadataType metadataStrategy = isMajorUpdate(baseDocument, updatedDocument) 
            ? MetadataType.TARGET 
            : MetadataType.SOURCE;
            
        try (Comparer comparer = new Comparer(baseDocument)) {
            comparer.add(updatedDocument);
            Path result = comparer.compare(generateVersionPath(), new SaveOptions.Builder()
                .setCloneMetadataType(metadataStrategy)
                .build());
                
            return new DocumentVersion(result, extractMetadata(result));
        }
    }
}

// Good - automatic resource management
try (Comparer comparer = new Comparer(sourceDoc)) {
    // Your logic here
} // Resources automatically closed

// Avoid - manual resource management
Comparer comparer = new Comparer(sourceDoc);
// ... logic ...
comparer.close(); // Easy to forget!

public void processBatch(List<DocumentPair> documents) {
    // Process in smaller chunks to manage memory
    int batchSize = 10;
    for (int i = 0; i < documents.size(); i += batchSize) {
        List<DocumentPair> batch = documents.subList(i, 
            Math.min(i + batchSize, documents.size()));
        processBatchChunk(batch);
        
        // Optional: force garbage collection between batches
        System.gc();
    }
}

// More efficient
Path outputPath = Paths.get("output", "comparison-result.docx");

// Less efficient
String outputPath = "output" + File.separator + "comparison-result.docx";

documents.parallelStream()
    .forEach(this::processDocument);

// Simple performance monitoring
long startTime = System.currentTimeMillis();
Path result = comparer.compare(outputPath, options);
long processingTime = System.currentTimeMillis() - startTime;
logger.info("Document comparison took {} ms", processingTime);

SaveOptions advancedOptions = new SaveOptions.Builder()
    .setCloneMetadataType(MetadataType.SOURCE)
    .setGenerateSummaryPage(true) // Adds a summary of changes
    .setMarkChangedContent(true)  // Highlights modifications
    .build();

public class CustomMetadataProcessor {
    public Path compareWithCustomMetadata(String source, String target, 
                                        CustomMetadataStrategy strategy) {
        try (Comparer comparer = new Comparer(source)) {
            comparer.add(target);
            
            Path tempResult = comparer.compare("temp_output.docx", 
                new SaveOptions.Builder()
                    .setCloneMetadataType(MetadataType.SOURCE)
                    .build());
            
            // Apply custom metadata logic
            return applyCustomMetadata(tempResult, strategy);
        }
    }
    
    private Path applyCustomMetadata(Path document, CustomMetadataStrategy strategy) {
        // Your custom metadata processing logic here
        // This could involve reading metadata from both source and target,
        // then applying business rules to determine final metadata
        return document;
    }
}