[MPlayer-dev-eng] [PATCH] subtitle reload, slave mode - 2nd try

[kiriuja]

>     Please, read my proposal in the other mail I sent. The problem I have with
> your approach is that the position a subtitle is loaded to is never communicated outside,
> so the user (or the program communicating through the slave mode) shouldn't know
> it. It is obvious that, if a single sub is loaded, it is at position 0, but
> the user shoudn't rely on this information.

I would expect the first subtitle file you give MPlayer to have the index 0,
the second one to have the index 1, and so on. Is that not the case? A host
program should have no problem keeping track of subtitles it passed to MPlayer.
Plus of course using an index would eliminate that for loop with string
comparison you have in your patch, so it would be more efficient, too.

>     Furthermore, I would prefere to implement a human-friendly command, so
> it's better to use names, rather than numbers.

I am pretty sure that the slave mode is only supposed to be used by frontends
rather than end users. But in either case, if a command called sub_reload adds
a subtitle rather than reloads it, that would look like a rather ugly hack.
But I may be wrong.

Another thing: how are spaces in file/directory names handled? Especially with
two path arguments it would be a pretty bad mess, you'd pretty much have to
URL encode them. With one index and one path it would be possible to have a
non-encoded path, but for that the index would have to be the first argument.

--
kiriuja