Enable PTP support in libmtp's libusb1 glue code. We don't use the libusb0.1 or libopenusb versions.

Enable PTP support in libmtp's libusb1 glue code. We don't use the libusb0.1 or libopenusb versions.

BUG=none
TEST=none
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=149281

[Lei Zhang, 2012-07-26 05:26:43]

Some of this code was ifdef'd / commented out and no longer compiles as is.

[vandebo (ex-Chrome), 2012-07-26 16:59:35]

https://chromiumcodereview.appspot.com/10831026/diff/1/src/libusb1-glue.c
File src/libusb1-glue.c (right):

https://chromiumcodereview.appspot.com/10831026/diff/1/src/libusb1-glue.c#new...
src/libusb1-glue.c:300: return 1;
Should this return 0?  Above seems to indicate we've hit a PTP device.

https://chromiumcodereview.appspot.com/10831026/diff/1/src/libusb1-glue.c#new...
src/libusb1-glue.c:524: if (desc.bDeviceClass == LIBUSB_CLASS_PTP) {
does libusb resolve altsetting interface classes for us?

[Lei Zhang, 2012-07-26 23:17:02]

https://chromiumcodereview.appspot.com/10831026/diff/1/src/libusb1-glue.c
File src/libusb1-glue.c (right):

https://chromiumcodereview.appspot.com/10831026/diff/1/src/libusb1-glue.c#new...
src/libusb1-glue.c:300: return 1;
On 2012/07/26 16:59:35, vandebo wrote:
> Should this return 0?  Above seems to indicate we've hit a PTP device.

No, we want to return 1. I've updated the comment for the function to indicate
what it should do.

https://chromiumcodereview.appspot.com/10831026/diff/1/src/libusb1-glue.c#new...
src/libusb1-glue.c:524: if (desc.bDeviceClass == LIBUSB_CLASS_PTP) {
On 2012/07/26 16:59:35, vandebo wrote:
> does libusb resolve altsetting interface classes for us?

The above code does. I can't remember why I needed this change down here. So
I'll revert it for now.

[vandebo (ex-Chrome), 2012-07-31 21:28:18]

LGTM