Diederick C. Niehorster (12021-06-08): > Done. Only problem is that now a device of mine that should be called > "Microphone Array (Intel® Smart Sound Technology (Intel® SST))" is > instead printed as "Intel┬« Smart Sound Technology (Intel┬« SST)". > Since devices are selected by name for some avdevices (dshow), thats > an issue. > > Googling tells me that printing UTF-8 (which is whats returned by > dshow's get_device_list) with printf is complicated to say the least > on Windows. It all works when logging because log.c has implemented > win_console_puts(), which looks like it would be easy to make > available more broadly. > > Do you see a solution that would allow printing the device name correctly?
What matters is not what you see in the console but what data is really written on the stream. Programs that read from the ffmpeg process and use the output to build a command line, all in a binary-clean way, should succeed. You can test with Perl (you can probably achieve the same with any language, but I know Perl, and I know it will not automagically mess things up): my $out = `ffmpeg -devices`; my $dev = $out =~ /(Microphone.*)/; print "Using \"$dev\"\n"; system "ffmpeg", "-f", "dshow", "-i", $dev; If the “Using "..."” message is mangled but the command line succeeds, then the issue is not ffmpeg's problem. Another test you can run: take a Matroska file with accents in the metadata, make sure it standard-compliant, and see what ffprobe -show_stream prints about it. Regards, -- Nicolas George
signature.asc
Description: PGP signature
_______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
