Originally Posted by wmcbrine
What's happening here is, pyTivo saves time in building the list by not fully parsing the metadata for a program until more details are requested. (And when I say "saves time", I mean you really wouldn't want it not to work this way.)
What do you use? Lately I've just been doing it with FFmpeg, after manually noting the cut points in VLC.
How much slower would it be? Since I have each show in its own folder (ooh, except I guess when it initially shows me the non-foldered list on the tivo?), I would hope it would be O(n) for the # of files..
I used to use MPEG Streamclip.. But it has gotten flaky on later versions of OS X.. (yes I know it uses the old MPEG 2 plugin) Flaky as in many recordings don't have video but just audio -- though strangely I could often still use it to edit in this state.
Though conversely, at least if I only wanted ONE segment, I _can_ use Quicktime Player often to just Trim that segment. But sometimes I want(ed) multiple segments (e.g. 2 musical performances from SNL).
Though wait -- can't it look at the metadata for EACH file as it "presents" that to the Tivo? Paging up/down sure seems to reload info.. so it seems to me like it could "do the right thing".. But if it took MINUTES, yeah, that would be bad..
(Though I guess that would mean it'd be cool to write a little tool to set the mod date of each file to the info in its corresponding metadata file... as a workaround)