如何将从MySQL获取的Java Blob对象保存为本地图片文件？

Got it, let's break this down clearly. The key thing to note here is that your Blob object doesn't contain raw image binary data — instead, it stores a base64-encoded data URI string (like data:image/png;base64,iVBORw0KGgo...). So we need to process that string first before saving it as an image file.

Here's a step-by-step solution with code:

Step 1: Extract the base64 string from the Blob #

Since the Blob holds text content, we'll read it as a string instead of using binary streams directly:

Blob imageBlob;
String base64DataUri = null;
while (rs.next()) {
    imageBlob = rs.getBlob("face");
    // Read the Blob content as a string
    try (Reader reader = imageBlob.getCharacterStream()) {
        StringBuilder sb = new StringBuilder();
        char[] buffer = new char[1024];
        int bytesRead;
        while ((bytesRead = reader.read(buffer)) != -1) {
            sb.append(buffer, 0, bytesRead);
        }
        base64DataUri = sb.toString();
    }
}

Step 2: Strip the data URI prefix #

We need to remove the data:image/png;base64, part to get just the base64-encoded image data. If your images might use other formats (like JPEG), you can make this logic more flexible, but for your current case:

// Split the URI to isolate the base64 image content
String base64Image = base64DataUri.split(",")[1];

Step 3: Decode the base64 string to raw binary data #

Use Java's built-in Base64 class to convert the base64 string back into the original image byte array:

byte[] imageBytes = Base64.getDecoder().decode(base64Image);

Step 4: Write the byte array to a disk file #

Use FileOutputStream (with try-with-resources to auto-close the stream) to save the binary data as an image file:

// Replace with your desired file path and name
String outputPath = "/your/target/path/image.png";
try (FileOutputStream fos = new FileOutputStream(outputPath)) {
    fos.write(imageBytes);
    System.out.println("Image saved successfully to " + outputPath);
}

Full Complete Code (with error handling) #

Here's the combined code with proper exception handling and resource management:

import java.sql.Blob;
import java.sql.ResultSet;
import java.io.*;
import java.util.Base64;

// ... your existing database connection/query code ...

Blob imageBlob;
String base64DataUri = null;

try {
    while (rs.next()) {
        imageBlob = rs.getBlob("face");
        // Read Blob content as a string
        try (Reader reader = imageBlob.getCharacterStream()) {
            StringBuilder sb = new StringBuilder();
            char[] buffer = new char[1024];
            int bytesRead;
            while ((bytesRead = reader.read(buffer)) != -1) {
                sb.append(buffer, 0, bytesRead);
            }
            base64DataUri = sb.toString();
        }
    }

    if (base64DataUri != null && base64DataUri.contains(",")) {
        String base64Image = base64DataUri.split(",")[1];
        byte[] imageBytes = Base64.getDecoder().decode(base64Image);
        
        // Save to disk
        String outputFile = "output_image.png";
        try (FileOutputStream fos = new FileOutputStream(outputFile)) {
            fos.write(imageBytes);
            System.out.println("Image saved to: " + new File(outputFile).getAbsolutePath());
        } catch (IOException e) {
            System.err.println("Failed to write image file: " + e.getMessage());
            e.printStackTrace();
        }
    } else {
        System.err.println("Invalid base64 data URI format in Blob");
    }
} catch (Exception e) {
    System.err.println("Error processing Blob data: " + e.getMessage());
    e.printStackTrace();
}

Key Notes: #

• Why not use getBinaryStream()? Because your Blob contains a text-based data URI, not raw image bytes. Reading it as a binary stream would give you the bytes of the string itself, not the actual image data.
• Support for other image formats: If your database might store different image types (JPEG, GIF), you can parse the MIME type from the data URI (the part between data: and ;base64) to set the correct file extension dynamically.

内容的提问来源于stack exchange，提问作者Deyan koba