[TIKA-474] Tika parsing corrupt mp3

• Type: Improvement
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 0.7
• Fix Version/s: 0.8
• Component/s: cli

Environment

Linux Mandriva 2010 based OS (Linux Caixa Mágica 15)
java version "1.6.0_20"
Java(TM) SE Runtime Environment (build 1.6.0_20-b02)

• Assignee: Nick Burch
• Reporter: André Ricardo
• Created: 06/Aug/10 14:42
• Updated: 09/Aug/10 15:23
• Resolved: 09/Aug/10 14:49

Description

I was trying some mp3s in tika-app cli coming from Nutch 0.9/1.0 samples and with "A corrupt MP3 file that has been truncated half way through the ID3v2 frames" returned this:

$ java -jar tika-app-0.7.jar -v -m ~/nutch-0.9/src/plugin/parse-mp3/sample/test.mp3
Exception in thread "main" org.apache.tika.exception.TikaException: TIKA-198: Illegal IOException from org.apache.tika.parser.mp3.Mp3Parser@1bf3d87
at org.apache.tika.parser.CompositeParser.parse(CompositeParser.java:138)
at org.apache.tika.parser.AutoDetectParser.parse(AutoDetectParser.java:99)
at org.apache.tika.cli.TikaCLI.process(TikaCLI.java:169)
at org.apache.tika.cli.TikaCLI.main(TikaCLI.java:62)
Caused by: java.io.IOException: Tried to read 259186 bytes, but only 65526 bytes present
at org.apache.tika.parser.mp3.ID3v2Frame.readFully(ID3v2Frame.java:160)
at org.apache.tika.parser.mp3.ID3v2Frame.<init>(ID3v2Frame.java:110)
at org.apache.tika.parser.mp3.ID3v2Frame.createFrameIfPresent(ID3v2Frame.java:81)
at org.apache.tika.parser.mp3.Mp3Parser.getAllTagHandlers(Mp3Parser.java:128)
at org.apache.tika.parser.mp3.Mp3Parser.parse(Mp3Parser.java:64)
at org.apache.tika.parser.CompositeParser.parse(CompositeParser.java:132)
... 3 more

Also tried with the latest trunk from github reproducing the problem:

$ java -jar tika-app-0.8-SNAPSHOT.jar -v -m ~/nutch-0.9/src/plugin/parse-mp3/sample/test.mp3
Exception in thread "main" org.apache.tika.exception.TikaException: TIKA-198: Illegal IOException from org.apache.tika.parser.mp3.Mp3Parser@e79839
at org.apache.tika.parser.CompositeParser.parse(CompositeParser.java:169)
at org.apache.tika.parser.AutoDetectParser.parse(AutoDetectParser.java:110)
at org.apache.tika.cli.TikaCLI.process(TikaCLI.java:193)
at org.apache.tika.cli.TikaCLI.main(TikaCLI.java:72)
Caused by: java.io.IOException: Tried to read 259186 bytes, but only 65526 bytes present
at org.apache.tika.parser.mp3.ID3v2Frame.readFully(ID3v2Frame.java:160)
at org.apache.tika.parser.mp3.ID3v2Frame.<init>(ID3v2Frame.java:110)
at org.apache.tika.parser.mp3.ID3v2Frame.createFrameIfPresent(ID3v2Frame.java:81)
at org.apache.tika.parser.mp3.Mp3Parser.getAllTagHandlers(Mp3Parser.java:133)
at org.apache.tika.parser.mp3.Mp3Parser.parse(Mp3Parser.java:64)
at org.apache.tika.parser.CompositeParser.parse(CompositeParser.java:163)
... 3 more

The mp3 is here: http://github.com/apache/nutch/raw/tags/release-1.0/src/plugin/parse-mp3/sample/test.mp3

All the other mp3 samples were parsed well by Tika.