marcus/pdf-umbenenner
1
0
Fork 0
Code Issues 14 Releases 2 Activity

Fix #60: SHA-256-Fingerprint streaming statt Files.readAllBytes berechnen

Browse Source 
Files.readAllBytes laedt grosse PDFs vollstaendig in den Heap und
riskiert OutOfMemoryError. Die Berechnung nutzt jetzt einen
DigestInputStream mit 8-KB-Puffer in try-with-resources. Das
Hash-Ergebnis ist bitidentisch zur vorigen Implementation, die
Exception-Semantik bleibt unveraendert.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
Marcus van Elst
2026-04-29 06:21:25 +02:00
parent d10a572b50
commit 8aaa3331d7
1 changed files with 16 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files
pdf-umbenenner-adapter-out/src/main/java/de/gecheckt/pdf/umbenenner/adapter/out/fingerprint/Sha256FingerprintAdapter.java
+16 -5
View File
@@ -1,6 +1,8 @@
package de.gecheckt.pdf.umbenenner.adapter.out.fingerprint; package de.gecheckt.pdf.umbenenner.adapter.out.fingerprint;
import java.io.IOException; import java.io.IOException;
import java.io.InputStream;
import java.security.DigestInputStream;
import java.nio.file.Files; import java.nio.file.Files;
import java.nio.file.InvalidPathException; import java.nio.file.InvalidPathException;
import java.nio.file.Path; import java.nio.file.Path;
@@ -45,7 +47,7 @@ public class Sha256FingerprintAdapter implements FingerprintPort {
* The implementation: * The implementation:
* <ol> * <ol>
* <li>Resolves the opaque locator to a filesystem path</li> * <li>Resolves the opaque locator to a filesystem path</li>
* <li>Reads the entire file content</li> * <li>Reads the file content in chunks via a streaming digest</li>
* <li>Applies SHA-256 hashing</li> * <li>Applies SHA-256 hashing</li>
* <li>Returns the hex-encoded result wrapped in a {@link FingerprintSuccess}</li> * <li>Returns the hex-encoded result wrapped in a {@link FingerprintSuccess}</li>
* </ol> * </ol>
@@ -113,8 +115,9 @@ public class Sha256FingerprintAdapter implements FingerprintPort {
/** /**
* Computes the SHA-256 hash of the file content at the given path. * Computes the SHA-256 hash of the file content at the given path.
* <p> * <p>
* Reads the entire file content and applies SHA-256 hashing to produce * Liest die Datei blockweise über einen {@link DigestInputStream}, um den Heap-Bedarf
* a lowercase hexadecimal representation of the digest. * bei großen PDFs zu minimieren. Das erzeugte Hash-Ergebnis ist bitidentisch zur
* byteweisen Verarbeitung des gesamten Dateiinhalts.
* *
* @param filePath the path to the file to hash; must not be null * @param filePath the path to the file to hash; must not be null
* @return the lowercase hexadecimal representation of the SHA-256 digest (64 characters) * @return the lowercase hexadecimal representation of the SHA-256 digest (64 characters)
@@ -123,8 +126,16 @@ public class Sha256FingerprintAdapter implements FingerprintPort {
*/ */
private String computeSha256Hash(Path filePath) throws IOException, NoSuchAlgorithmException { private String computeSha256Hash(Path filePath) throws IOException, NoSuchAlgorithmException {
MessageDigest digest = MessageDigest.getInstance("SHA-256"); MessageDigest digest = MessageDigest.getInstance("SHA-256");
byte[] fileBytes = Files.readAllBytes(filePath); // Streaming-Verarbeitung: Die Datei wird in 8-KB-Blöcken gelesen, damit auch
byte[] hashBytes = digest.digest(fileBytes); // sehr große PDFs nicht vollständig in den Heap geladen werden müssen.
byte[] buf = new byte[8192];
try (InputStream is = Files.newInputStream(filePath);
DigestInputStream dis = new DigestInputStream(is, digest)) {
while (dis.read(buf) != -1) {
// DigestInputStream leitet jeden Block automatisch an den MessageDigest weiter
}
}
byte[] hashBytes = digest.digest();
return bytesToHex(hashBytes); return bytesToHex(hashBytes);
} }