Hi Alan, > It's possible that this is more than just a private convention. In > principle we could try to support it: If the descriptor type and length > are both 0 then request 255 bytes and assume it's a null-terminated > string. > > I'm concerned because this is so contrary to the specification and subject > to errors. (If the descriptor type is 0, how can you be sure you're > looking at the right sort of descriptor?) > > If there's any evidence that more than one device or vendor uses this > convention, then we ought to adopt it. Otherwise I say to forget about > it.
I propose the following strategy: (1) Add a flag "support bullshit string descriptors", which tries to extract a null terminated string when the type and length are zero. (2) Turn the flag on only for this device. (3) Rummage around in windows, and try to work out if they have general support for this kind of string descriptor, or just for this particular device (this may take some time). If they support it for general devices, then we can drop the flag. Ciao, Duncan. ------------------------------------------------------- This SF.Net email is sponsored by BEA Weblogic Workshop FREE Java Enterprise J2EE developer tools! Get your free copy of BEA WebLogic Workshop 8.1 today. http://ads.osdn.com/?ad_id=4721&alloc_id=10040&op=click _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
