Hi Eugen, On Monday 30 March 2009 23:24:15 Eugen Dedu wrote: > Laurent Pinchart wrote: > > Hi Eugen, > > > > On Monday 30 March 2009 17:26:49 Eugen Dedu wrote: > >> Hi, > >> > >> (This e-mail is because we have a bug about a camera name which seems > >> not to be in utf-8, > >> https://bugs.launchpad.net/ubuntu/+source/ekiga/+bug/345192?comments=all > >>) > >> > >> For example, I see at > >> > >> http://www.mail-archive.com/linux-uvc- [email protected]/msg01225.html: > >>> uvcvideo: Found UVC 1.00 device Microsoft? LifeCam NX-6000 (045e:00f8) > >> > >> The character ? is in fact trademark/registered. > >> > >> My questions: Is this string (camera name) found in the camera or is it > >> created by uvcvideo? Is it in utf-8 or another encoding? > > > > The string comes directly from the camera where it is encoded in UTF-16. > > The kernel USB subsystem transcodes it to ISO-8859-1, hence the '?'. > > Thanks, Laurent, for the fast reply.
To compensate for the fast reply, here's a late reply :-) Just kidding, I was on holidays last week away from my mailbox. > - Is it true that the name for all the camera come from the camera itself? It is, unless the camera doesn't provide a name, in which case the driver uses the VID:PID. > - Is it true that all the camera store their name in utf-16? That's right. > - Why the kernel usb subsystem transcode them to latin-1 and not to utf-8? You will have to ask the USB subsystem developers. > - Do the linux kernel transcodes to latin-1 or it depends on how it was > configured? Is it the same for linux, windows and macos? It's because > I want to retranscode it to utf-8, and I need to know its current encoding. The USB subsystem transcodes the UTF-16 strings to Latin-1 regardless of configuration options. I have no idea how Windows and MacOS handle that. Best regards, Laurent Pinchart _______________________________________________ Linux-uvc-devel mailing list [email protected] https://lists.berlios.de/mailman/listinfo/linux-uvc-devel
