On Sat, Jun 28, 2003 at 09:58:37PM +0200, Dirk Meyer wrote: > Like Thomas said, mmpython does caching. When I integrate the whole > set of features of mmpython we can fast access all metadata from > files. Not only mp3 id tags, also informations in videos (I was > suprised to see what avis and mov have about the file). mmpython can > do caching based on the directory (or cd/dvd). Accessing a cached > directory is very fast and you since the caching is done by mmpython, > you can cache all your directories without freevo (e.g. cron). What > you can't do right now is 'where is song xy on my harddisc'. But if > you need that, too, you can get this informations by writing a search > plugin over the cache files.
What I'm talking about involves more than caching; I'm looking for a way to relate data arbitrarily. By putting tag/filename data into a sqlite database, which is essentially a single file with no daemon or anything, you can generate smart playlists on the fly, like the examples I suggested. Doing that via cache files doesn't make sense, first, because it requires data that isn't stored in the MP3/video, but also because it needs to relate things. A list of files is not related, but "songs with year=2000" is related. I've already started playing around with sqlite and it's very easy to use. What might be a good idea would be coming up with a way for a piece of code to inheret mmpython's base caching class, and override the storage part. Aubin ------------------------------------------------------- This SF.Net email sponsored by: Free pre-built ASP.NET sites including Data Reports, E-commerce, Portals, and Forums are available now. Download today and enter to win an XBOX or Visual Studio .NET. http://aspnet.click-url.com/go/psa00100006ave/direct;at.asp_061203_01/01 _______________________________________________ Freevo-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/freevo-devel
