Have the TikaAuto content transformer skip .tar and .tar.* files, since these are handled by the new Archive transformer which gives control over recursion

(Will fix the build failure caused by adding quick.tar earlier today)


git-svn-id: https://svn.alfresco.com/repos/alfresco-enterprise/alfresco/HEAD/root@22861 c4b6b30b-aa2e-2d43-bbcb-ca4b014f7261
This commit is contained in:
Nick Burch
2010-10-04 22:18:10 +00:00
parent 2d690eb109
commit 40e6c1e1d9

View File

@@ -61,16 +61,20 @@ public class TikaAutoContentTransformer extends TikaPoweredContentTransformer
if(mt.getType().equals("image") ||
mt.getType().equals("audio") ||
mt.getType().equals("video") ||
mt.toString().equals("application/zip") ||
mt.toString().equals("application/tar"))
mt.getType().equals("video"))
{
// Skip these, as Tika mostly just does
// metadata rather than content
}
else if(mt.toString().equals("application/zip") ||
mt.toString().equals("application/tar") ||
mt.toString().equals("application/x-tar"))
{
// Skip these, as we handle container formats in a different
// transformer to give the user control over recursion
}
else
{
// Tika can probably do some useful text
SUPPORTED_MIMETYPES.add( mt.toString() );
}