Copilot commented on code in PR #3130: URL: https://github.com/apache/tika/pull/3130#discussion_r3934558208
########## tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-image-module/src/test/java/org/apache/tika/detect/image/RawTiffDetectorFuzzTest.java: ########## @@ -0,0 +1,136 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ +package org.apache.tika.detect.image; + +import static org.junit.jupiter.api.Assertions.fail; + +import java.util.Locale; +import java.util.Random; + +import org.junit.jupiter.api.Test; + +/** + * Randomized boundary test for {@link RawTiffDetector}'s directory walk. + * <p> + * The detector runs ahead of every parser, and {@code Detector.detect} declares + * only {@link java.io.IOException}: anything else it throws aborts detection for + * the document and the remaining detectors never run. So the invariant is simply + * that no throwable escapes, whatever the directories say. + * <p> + * Inputs are well-formed TIFF and BigTIFF headers whose offsets, counts and + * entry values are drawn from the arithmetic boundaries -- 0, the 32 and 64 bit + * maxima, the prefix limit, and their neighbours -- since that is where the + * offset handling goes wrong rather than in random bytes. The seed is random per + * run and reported on failure. + */ +public class RawTiffDetectorFuzzTest { + + private static final int TRIALS = 20000; Review Comment: `TRIALS` is set to 20,000, which can noticeably increase CI/unit-test runtime (it’s an order of magnitude higher than other fuzz-style tests in this repo, e.g. `TikaEvalTokenizerFuzzTest`). Consider making this configurable via a system property (keeping the current default) so CI can dial it down if needed while still allowing deeper runs locally. ########## tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-image-module/src/test/java/org/apache/tika/detect/image/RawTiffDetectorTest.java: ########## @@ -224,6 +225,49 @@ public void testTruncatedPrefixIsHarmless() { } } + /** + * A BigTIFF directory offset near {@code Long.MAX_VALUE}: adding the entry + * count to it wraps negative, and a negative end must not read as "already + * in the prefix". The three pointer sources (this header field, a SubIFDs + * array, the follower below) all reach the same bounds check. + */ + @ParameterizedTest + @ValueSource(longs = {Long.MAX_VALUE, Long.MAX_VALUE - 7, Long.MAX_VALUE - 8, + 0x100000000L, 1024L * 1024L + 1}) + public void testDirectoryOffsetBeyondTheFileIsRejected(long firstIfd) throws Exception { + byte[] tiff = bigTiffHeader(firstIfd); + assertEquals(MediaType.OCTET_STREAM, RawTiffDetector.detect(tiff, tiff.length)); + try (TikaInputStream tis = TikaInputStream.get(tiff)) { + assertEquals(MediaType.OCTET_STREAM, + new RawTiffDetector().detect(tis, new Metadata(), new ParseContext())); + } + } + + /** + * The same offset as the follower of an otherwise good directory: the + * vendor the directory names still decides the type. + */ Review Comment: Javadoc grammar is off (“vendor the directory names…”). Rewording will make the intent clearer when reading the test. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
