simple search api (was Re: mimetype standardisation by testsets)
Jos van den Oever
jvdoever at gmail.com
Sun Nov 26 16:02:34 EET 2006
2006/11/26, Thiago Macieira <thiago at kde.org>:
> If they don't, they should probably be fixed to support it.
> For instance, seeing an .ogg file and assuming it's Vorbis is wrong (it
> could be FLAC). Just as it would be wrong to not inspect and know what it
> is internally.
> If a file type is a container type, the indexer has to be smart enough to
> look inside it. A good example are zip files, which can be OpenDocument
This is a good point to repeat the original message of this thread.
We need a set of files as a testset. For each of these files we should
determine what the metadata of each file is. One of the most important
fields is 'mimetype'. With such a test set, every indexer can be unit
More information about the xdg