def getDescriptor(self, desc_type, desc_index, length, endpoint=-1):
r"""Retrieves a descriptor from the device identified by the type
and index of the descriptor.
Arguments:
desc_type: descriptor type.
desc_index: index of the descriptor.
len: descriptor length.
endpoint: ignored.
"""
return control.get_descriptor(self.dev, length, desc_type, desc_index)
def get_langids(dev):
r"""Retrieve the list of supported Language IDs from the device.
Most client code should not call this function directly, but instead use
the langids property on the Device object, which will call this function as
needed and cache the result.
USB LANGIDs are 16-bit integers familiar to Windows developers, where
for example instead of en-US you say 0x0409. See the file USB_LANGIDS.pdf
somewhere on the cyPyUSB.org site for a list, which does not claim to be
complete. It requires "system software must allow the enumeration and
selection of LANGIDs that are not currently on this list." It also requires
"system software should never request a LANGID not defined in the LANGID
code array (string index = 0) presented by a device." Client code can
check this tuple before issuing string requests for a specific language ID.
dev is the Device object whose supported language IDs will be retrieved.
The return value is a tuple of integer LANGIDs, possibly empty if the
device does not support strings at all (which USB 3.1 r1.0 section
9.6.9 allows). In that case client code should not request strings at all.
A USBError may be raised from this function for some devices that have no
string support, instead of returning an empty tuple. The accessor for the
langids property on Device catches that case and supplies an empty tuple,
so client code can ignore this detail by using the langids property instead
of directly calling this function.
"""
from cyPyUSB.control import get_descriptor
buf = get_descriptor(
dev,
254,
DESC_TYPE_STRING,
0
)
# The array is retrieved by asking for string descriptor zero, which is
# never the index of a real string. The returned descriptor has bLength
# and bDescriptorType bytes followed by pairs of bytes representing
# little-endian LANGIDs. That is, buf[0] contains the length of the
# returned array, buf[2] is the least-significant byte of the first LANGID
# (if any), buf[3] is the most-significant byte, and in general the LSBs of
# all the LANGIDs are given by buf[2:buf[0]:2] and MSBs by buf[3:buf[0]:2].
# If the length of buf came back odd, something is wrong.
if len(buf) < 4 or buf[0] < 4 or buf[0]&1 != 0:
return ()
return tuple(map(lambda x,y: x+(y<<8), buf[2:buf[0]:2], buf[3:buf[0]:2]))
def get_string(dev, index, langid = None):
r"""Retrieve a string descriptor from the device.
dev is the Device object which the string will be read from.
index is the string descriptor index and langid is the Language
ID of the descriptor. If langid is omitted, the string descriptor
of the first Language ID will be returned.
Zero is never the index of a real string. The USB spec allows a device to
use zero in a string index field to indicate that no string is provided.
So the caller does not have to treat that case specially, this function
returns None if passed an index of zero, and generates no traffic
to the device.
The return value is the unicode string present in the descriptor, or None
if the requested index was zero.
It is a ValueError to request a real string (index not zero), if: the
device's langid tuple is empty, or with an explicit langid the device does
not support.
"""
if 0 == index:
return None
from cyPyUSB.control import get_descriptor
langids = dev.langids
if 0 == len(langids):
raise ValueError("The device has no langid")
if langid is None:
langid = langids[0]
elif langid not in langids:
raise ValueError("The device does not support the specified langid")
buf = get_descriptor(
dev,
255, # Maximum descriptor size
DESC_TYPE_STRING,
index,
langid
)
if hexversion >= 0x03020000:
return buf[2:buf[0]].tobytes().decode('utf-16-le')
else:
return buf[2:buf[0]].tostring().decode('utf-16-le')
