[
https://issues.apache.org/jira/browse/TIKA-1039?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13570390#comment-13570390
]
Nick Burch commented on TIKA-1039:
----------------------------------
I'm not sure there's much that we can do about this, as the starting "magic"
for MP3 files are a little generic. It's possible we could try to add a second
check for another frame header of a similar type (VBR means the first two frame
headers may differ) within a frame's size distance, but the mime magic checking
for that would be pretty icky given the current structure.
http://www.mars.org/pipermail/mad-dev/2002-January/000425.html indicates how
many bytes an audio frame could be (ID3 frames can be much much larger, and one
of those could be the first)
> Raw image file detected as audio/mpeg
> -------------------------------------
>
> Key: TIKA-1039
> URL: https://issues.apache.org/jira/browse/TIKA-1039
> Project: Tika
> Issue Type: Bug
> Components: mime
> Affects Versions: 1.2
> Reporter: Oliver Boldt
> Attachments: SimpleTestFile.raw
>
>
> A raw image file that starts with a long sequence of FFFF.... is recognised
> as audio/mpeg.
> The problem is that the raw file does not have a magic number itself and the
> FF...-pixeldata is wrongly interpreted as an mpeg file. The bug seems to be a
> general problem, because other image data could be misinterpreted as other
> magic numbers.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
