/usr/share/psychtoolbox-3/PsychHardware/Daq/DaqDeviceIndex.m is in psychtoolbox-3-common 3.0.11.20131230.dfsg1-1build1.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 | function daq = DaqDeviceIndex(DeviceName, IShouldWarn)
% daq = DaqDeviceIndex([DeviceName][, ShowInterfaceNumberWarning=1])
% Returns a list of all your USB -1208FS, -1408FS, or -1608FS daqs.
%
% CAUTION: This routine works well on GNU/Linux and MS-Windows, but is very
% unreliable on Apple OSX, especially with multiple DAQ devices connected.
% You may be better off guessing a proper daq index.
%
% Also implements experimental code for detection of the USB-1024LS.
% However that code has not been tested yet and may need some small amount
% of tweaking to make it really work. Search the Psychtoolbox forum for
% corresponding messages...
%
% TECHNICAL NOTE: When we call PsychHID('Devices'), each USB-1208FS/1408FS
% box presents itself as four HID "devices" sharing the same serial number. They
% are interfaces 0,1,2,3. They usually appear in reverse order in the
% device list. Nearly all the commands use only interface 0, so
% we will select that one to represent the whole. All our Daq routines
% expect to receive just the interface 0 device as a passed designator. The
% few routines that need to access the other interfaces do so
% automatically.
%
% ADDENDUM: The above statement is correct for the 1208FS and the 1408FS,
% not for the 1608FS. Number of Devices found by PsychHID for a
% 1608FS and Leopard varies from five to seven with little rhyme or reason. It
% appears that the correct number of interfaces is seven. As with the 1208FS,
% most communication is through interface 0. However, when acquiring data
% (e.g., using DaqAInScan), output is via interfaces 1 through 6. Because
% PsychHID's enumeration of the interfaces is flaky, you may need to run this
% function more than once (with calls to DaqReset and probably "clear PsychHID"
% in between successive calls) in order to get device enumeration completed
% correctly. In DaqTest, user will be warned if they try to test a 1608 and
% this function doesn't return the right number of interfaces, so I added the
% second optional argument as a flag to suppress the warning that would be
% generated here. -- mpr
%
% See also Daq, DaqFunctions, DaqPins, DaqTest, PsychHIDTest,
% DaqFind, DaqDIn, DaqDOut, DaqAIn, DaqAOut, DaqAInScan,DaqAOutScan.
% 4/15/05 dgp Wrote it.
% 8/26/05 dgp Added support for new "USB" name, as suggested by Steve Van Hooser <vanhooser@neuro.duke.edu>.
% 8/26/05 dgp Incorporated bug fixes (for compatibility with Mac OS X Tiger)
% suggested by Jochen Laubrock <laubrock@rz.uni-potsdam.de>
% and Maria Mckinley <parody@u.washington.edu>.
% The reported number of outputs of the USB-1208FS has changed in Tiger.
% http://groups.yahoo.com/group/psychtoolbox/message/3610
% http://groups.yahoo.com/group/psychtoolbox/message/3614
%
% 11-12/xx/07 mpr Added possibility of specified input (with defaults) and
% changed number of expected device outputs; tested with
% USB-1608FS and Leopard. Known device names: PMD-1208FS,
% USB-1208FS, USB-1408FS, USB-1608FS.
% 1/7/08 mpr Rewrote sections to take advantage of input, added checks to deal
% with problems I encountered with 1608 and PsychHID's device
% enumeration.
% 1/14/08 mpr added second argument
% 5/22/08 mk Add (untested!) support for detection of USB-1024LS box.
% 5/08/13 mk Fix enumeration of multiple connected devices (untested).
% Use numeric USB productID instead of USB product name string
% for matching of devices, as the latter isn't supported on
% Windows, at least not reliably, whereas productID's are. Also
% reject non-MCC devices based on the well-defined MCC
% vendorID.
% Use combined locationID and serialNumber as unique ID for
% each instance of multiple identical DAQ devices to
% disambiguate if user connects multiple devices of same model.
% Bits of cleanup and refactoring.
% 5/20/13 mk Verify correct enumeration of devices with a USB-1408FS device
% on 10.5 Leopard, 10.7 Lion, Linux, and Windows-7. Remove 1408FS
% "untested" warnings. It is tested now.
%
% 10/4/13 mk More futile attempts to fix OSX: Use a heuristic of always
% choosing the maximum device index of eligible entries as
% returned daq index, assuming that one is interface#0. This
% heuristic only used if only 1 DAQ device is detected and only
% on OSX, as the regular matching code works well on Linux and
% Windows.
if nargin < 2 || isempty(IShouldWarn)
IShouldWarn=1;
end
if ~nargin || isempty(DeviceName)
% DeviceName can be anything -- doesn't matter for further processing.
DeviceName = '';
AcceptAlternateNames = 1;
else
if ~ischar(DeviceName)
error('DaqDeviceIndex expects a character string as input.');
end
AcceptAlternateNames = 0;
% Validate given DeviceName against list of supported devices:
switch DeviceName
case {'PMD-1208FS','USB-1208FS','PMD-1408FS','USB-1408FS'}
case {'PMD-1608FS','USB-1608FS'}
case {'PMD-1024LS','USB-1024LS'}
otherwise
error('I did not recognize your specified device name.');
end % switch DeviceName
end % if ~nargin || isempty(DeviceName)
NumInterfaces = [];
% Retrieve preprocessed list of PsychHID('Devices'), with some filtering
% applied to supported DAQ devices:
devices = PsychHIDDAQS;
daq=[];
for k=1:length(devices)
% Skip all devices with a vendorId other than 2523 aka 0x09db, ie. all
% devices that are not from Measurement Computing:
if devices(k).vendorID ~= 2523
continue;
end
% It is a MCC device. The product name string is only well defined on
% OSX and Linux, but non-sensical on Windows. The USB productID is well
% defined and accessible on all systems, so we remap the productID to
% the name string: Database with id's available at
% ftp://ftp.measurementcomputing.com/Customer/JBR/ULHelp/Users_Guide/Appendix/BoardType_Codes.htm
% This remapping of product name string from productID is actually done
% in our helper function PsychHIDDAQS above.
% Set NumOutputs threshold value known to be consistent with
% interface 0 on OSX. On Window and Linux we don't need this hack.
%
% Use as flag to prevent processing of unsupported devices, therefore
% init to empty.
NumOutputs = [];
% 1024-LS type product ID?
if devices(k).productID == 118
% MK: This NumOutputs threshold needs to be tinkered with to find the
% correct number for the actual output interface of 1024LS:
NumOutputs = 0;
end
% 1608FS type product ID?
if devices(k).productID == 125
NumOutputs = 65;
end
% 1208FS type product ID?
if devices(k).productID == 130
NumOutputs = 69;
end
% 1408FS type product ID?
if devices(k).productID == 161
NumOutputs = 65;
end
% It is a MCC device.
if AcceptAlternateNames
if ~isempty(NumOutputs)
if isempty(NumInterfaces)
NumInterfaces = 1;
MatchedDeviceName = devices(k).product;
MatchedSerialNumbers = devices(k).serialNumber;
else
% Device already known - stored in our list?
matchedIt = 0;
for l=1:size(MatchedDeviceName,1)
if ~isempty(strfind(MatchedDeviceName(l,:),devices(k).product)) && ...
~isempty(strfind(MatchedSerialNumbers(l,:),devices(k).serialNumber))
% Yes: Increment its interface count:
NumInterfaces(l) = NumInterfaces(l) + 1;
matchedIt = 1;
end
end % for l=1:size(MatchedDeviceName,1)
% Found at least 1 match in list of already known devices?
if ~matchedIt
% Nope: Add this devices name/id and serialNumber to array,
% init to at least one interface we found by definition of
% reaching this point:
MatchedDeviceName = char(MatchedDeviceName,devices(k).product);
MatchedSerialNumbers = char(MatchedSerialNumbers,devices(k).serialNumber);
NumInterfaces(end+1) = 1;
end
end % if isempty(NumInterfaces); else
% We want interface #0. We get it by checking for it on Linux and
% Windows, and indirectly detecting it by the number of outputs on OS/X:
if (devices(k).outputs > NumOutputs) || (devices(k).interfaceID == 0)
daq(end+1) = k;
end
end % if ~isempty(NumOutputs)
else % if ~AcceptAlternateNames
if streq(devices(k).product,DeviceName)
if isempty(NumInterfaces)
NumInterfaces = 1;
MatchedSerialNumbers = devices(k).serialNumber;
else
% Device already known - stored in our list?
matchedIt = 0;
for l=1:size(MatchedSerialNumbers,1)
if ~isempty(strfind(MatchedSerialNumbers(l,:), devices(k).serialNumber))
% Yes: Increment its interface count:
NumInterfaces(l) = NumInterfaces(l) + 1;
matchedIt = 1;
end
end
% Found at least 1 match in list of already known devices?
if ~matchedIt
% Nope: Add this devices serialNumber to array, init to at
% least one interface we found by definition of reaching
% this point:
MatchedSerialNumbers = char(MatchedSerialNumbers,devices(k).serialNumber);
NumInterfaces(end+1) = 1;
end
end % if isempty(NumInterfaces); else branch.
% We want interface #0. We get it by checking for it on Linux and
% Windows, and indirectly detecting it by the number of outputs on OS/X:
if (devices(k).outputs > NumOutputs) || (devices(k).interfaceID == 0)
daq(end+1) = k; %#ok<*AGROW>
end
end % if streq(devices(k).product,DeviceName)
end % if AcceptAlternateNames; else
end % for k=1:length(devices)
if IShouldWarn
for k=1:length(daq)
if ~isempty(strfind(devices(daq(k)).product,'1608'))
if NumInterfaces(k) < 7
ConfirmInfo(sprintf(['Only found %d interfaces for DeviceIndex %d. Execute ' ...
'"help DaqReset" for suggestions.'], ...
NumInterfaces(k),daq(k)));
end
end
end
end
if (length(daq) == 1) && IsOSX
% This is OSX, the most f$%@#$d up system in existence, and exactly 1
% DAQ device was detected. This is the common case. We know that the
% daq index corresponding to interface #0 seems to be always the one
% with the maximal index in the devices list, and the NumOutputs
% heuristic is badly broken due to the mostly random and varying number
% of outputs reported by OSX across OS versions.
% So let's try to fix this up for the common case of exactly 1 device
% on OSX with a new heuristic. Simply select the last device in the
% list, ie., the one with maximal index:
maxidx = -1;
fprintf('\nOnly one DAQ device detected and this is the horribly broken OSX OS.\n');
fprintf('Will return daq device index as the one with the highest index number,\n');
fprintf('assuming it to be interface #0, as our interface matching is too\n');
fprintf('unreliable on OSX, thanks to Apple''s broken operating systems. Good luck!\n');
for k=1:length(devices)
% Skip all devices with a vendorId other than 2523 aka 0x09db, ie. all
% devices that are not from Measurement Computing:
if devices(k).vendorID ~= 2523
continue;
end
% Skip all devices which don't match the proper product name if
% name filtering is requested by user:
if ~AcceptAlternateNames && ~streq(devices(k).product,DeviceName)
continue;
end
if devices(k).outputs == 0
continue;
end
% Ok, index k is a MCC devices which matches. Keep track of the
% device with maximum index:
maxidx = max(maxidx, k);
end
if maxidx >= 0
% Assign maximum index as proper daq index for interface #0 of only
% detected DAQ device on OSX:
daq = maxidx;
end
end
return;
|