From the forums

Seems to be an issue with the process hanging on acquiring the IP address

Just got my tessel today (6/12/2014). Went through a couple of tutorials, and tried wifi.

First time around, the blinking orange light turned solid, and the console did the Acquiring IP address. with the dots. That eventually timed out, and it started "Retrying..."

and there it sits. This time it is not printing the little progress dots.

The orange light is solid, but it never gets out of the Acquiring IP address. on the command line.

I checked the router, and everything is fine there. I moved to within 15 feet of the router now, just to make sure.

I'm running from Windows 8.1

After a hard reboot of the tessel (removing usb cable, waiting a bit) the board comes up with a solid orange light.

tessel run wifi.js

Seems to run, returning 200 OK on the command prompt.

So, I guess it's just a hangup with acruiring the IP address. It must have acquired it, and then either the command itself hangs, or something else?

Issue for tracking progress on remote code pushing integration into firmware.

Power-saving during data collection with periodic batch upload.

https://forums.tessel.io/t/power-draw-will-a-battery-last-all-day/320/3

Originally posted in private Beta repo.

After updating to version 0.1.8 my http servers no longer sends data to the clients.

Sample code is:

var http = require( 'http' );
console.log("Creating server");
http.createServer( function (req, res) {
console.log("Got a request!");
res.writeHead( 200, {'Content-Type': 'text/plain'} );
res.end('Hello World!\n');
console.log("Finished writing request");
}).listen( 8080 );

Tessel logs:

INFO Deploying bundle (5.00 KB)...
INFO Running script...
Creating server
Got a request!
Finished writing request

Client output:

curl http://192.168.0.10:80
curl: (52) Empty reply from server

Originally posted on the private Beta repo.

GPIOs allow you to set interrupts on them and a time argument is returned in the callback.

Our documentation specifies that the time argument is in milliseconds but it's actually implemented as microseconds.

Which should we change? Microseconds seems useful from an embedded perspective but milliseconds is more in-line with how JS traditionally handles time (mostly thinking about setInterval and setTimeout).

Originally posted in private Beta repo.

var net = require('net'),
    firstServer = net.createServer(),
    conflicting = net.createServer();
firstServer.listen(0, function () {
  var firstPort = firstServer.address().port;
  console.log(firstPort, "assigned a port");
  conflicting.listen(firstPort, function () {
    console.log("this should not be called!");
  });
  conflicting.on('error', function (e) {
    console.log(e, "got error as expected");
  });
});

After a fresh power cycle, this outputs:

1024 assigned a port
this should not be called!

See also tessel/t1-runtime#331 (comment)

If WiFI fails to initialize, Tessel will just hang forever and can't receive USB commands.

We need to have a verification scheme for allowing remote pushes.

I don't have a repro test case yet, but sometimes when running tessel wifi -l, Tessel will output the names of the same networks in a loop, forever.

runtime/test/suite/net.js fails with this error:

…

server-basic

ok 107 ok
ok 108 ok
ok 109 ok
ok 110 ok
Error: ENOENT Cannot connect to 127.0.0.1 Got: err-1
stack traceback:
[T]:src/colony/modules/net.js:205: in function <[T]:src/colony/modules/net.js:209>

@kevinmehall @tcr Every two hours or so, the clock on the tessel will jump back 72 minutes.

Originally posted in private Beta repo.

See this forum post.

➜  Twitter-Sumobot git:(master) ✗ tessel run index.js
TESSEL! Connected to TM-00-04-f000da30-005c4f48-422c2586.
INFO Bundling directory /Users/frijol/Twitter-Sumobot
INFO Deploying bundle (442.50 KB)...
INFO Running script...
Connecting...
TypeError: Uncaught, unspecified "error" event.
    at js_new (src/colony/lua/colony-init.lua:494)
    at emit (src/colony/lua/colony-node.lua:703)
    at _f (builtin/wifi-cc3000.js:136)
    at <anonymous> (builtin/wifi-cc3000.js:40)
    at connect (builtin/wifi-cc3000.js:40)
    at checkConnection (/app/index.js:23)
    at res (/app/index.js:10)
    at run (src/colony/lua/colony-node.lua:1082)
    at require (src/colony/lua/colony-node.lua:1077)
    at res (/_start.js:4)

Code:

// Get Tessel connected to Wifi
var tessel = require('tessel');
var wifi = require('wifi-cc3000');
var wifiSettings = {
  ssid: 'Gaumenkitzel',
  password: 'Gaumenkitzel',
  timeout: 40
};

checkConnection();

wifi.on('disconnect', function () {
  console.log('Disconnected.');
  checkConnection();
});

function checkConnection () {
  if (wifi.isConnected()) {
    console.log('Connected.');
    main();
  } else {
    console.log('Connecting...');
    wifi.connect(wifiSettings, function (err, res) {
      if(err) {
        console.log('Error connecting:', err);
      }
      checkConnection();
    });
  }
}

function main () {
  // Set up Twitter
  console.log('Setting up Twitter...');
  var twitter = require('twitter');
  var util = require('util');
  var twit = new twitter({
    consumer_key: 'O7oc0pvsZn4xjgcuHuYdX4FaC',
    consumer_secret: 'iJYuHFz2sD46Nvk3mcwzX8uih14aEAMgVWdWoR59nx8v6Zl7ZX',
    access_token_key: '2529232909-luARGU89K4CKFMvfzBjCgG6ubefzDkdDWkSB85i',
    access_token_secret: 'GXQfuzvGdjLEs3t1HEYfhQ9x9bdBcSBVXjBkbRgwYlOE0'
  });
  tessel.led[0].output(1);
  console.log('Twitter is all set up.');

  // Set up servos
  var servoLib = require('servo-pca9685');
  var servo = servoLib.use(tessel.port['A']);

  // When servo module is ready...
  servo.on('ready', function () {
    console.log('Servo module ready. Listening for tweets...');
    tessel.led[1].output(1);
    // Stream tweets that mention the keyword
    twit.stream('statuses/filter', {track:'#Tesselbot'}, function(stream) {
      stream.on('data', function(data) {
        var tweet = data.text;
        console.log(tweet);
        if (tweet.indexOf('forward') > 0) {
          forward();
        } else if ( tweet.indexOf('back') > 0 ) {
          back();
        } else if (tweet.indexOf('right') > 0 ) {
          right();
        } else if (tweet.indexOf('left') > 0) {
          left();
        } else {
          // Anything not an explicit command makes it stop
          stop();
        }
      });
      stream.on('error', function (err) {
        console.log('Twitter error:', err);
      });
    });
  });
}

// Motion functions
function forward() {
  console.log('Moving: forward');
  servo.move(1, 1);
  servo.move(2, 0);
}

function back() {
  console.log('Moving: back');
  servo.move(1, 0);
  servo.move(2, 1);
}

function right() {
  console.log('Moving: right');
  servo.move(1, 1);
  servo.move(2, 1);
}

function left() {
  console.log('Moving: left');
  servo.move(1, 0);
  servo.move(2, 0);
}

function stop() {
  console.log('Stopping.');
  servo.move(1, 0.5);
  servo.move(2, 0.5);
}

We need an analog to this function: http://arduino.cc/en/Reference/pulseIn
Currently blocking something I want to build.

When specifying a bitValue of 'msb' (which is the default) the tessel throws an error at me.

this.spi = new port.SPI({
  clockSpeed: 4*1E6,
  cpol: 0,
  cpha: 0,
  bitOrder: 'msb',
  chipSelect: this.CS,
  chipSelectActive: 'low'
});

Error: Invalid property value
    at _new ([string "colony-init.lua"]:1)
    at propertySetWithDefault ([string "tessel.js"]:49)
    at f ([string "tessel.js"]:1183)
    at _new ([string "colony-init.lua"]:1)
    at f ([string "tessel.js"]:1425)
    at _new ([string "colony-init.lua"]:1)
    at f (/app//index.js:34)
    at _new ([string "colony-init.lua"]:1)
    at use (/app//index.js:78)
    at res (/app/test.js:5)

Removing just bitOrder: 'msb', fixes it...

a@UBUNTU02:~$ tessel update

/home/a/.nvm/v0.10.33/lib/node_modules/tessel/node_modules/usb/node_modules/bindings-shyp/bindings.js:107
throw err
^
Error: Could not locate the bindings file. Tried:
→ /home/a/.nvm/v0.10.33/lib/node_modules/tessel/node_modules/usb/build/usb_bindings.node
→ /home/a/.nvm/v0.10.33/lib/node_modules/tessel/node_modules/usb/build/Debug/usb_bindings.node
→ /home/a/.nvm/v0.10.33/lib/node_modules/tessel/node_modules/usb/build/Release/usb_bindings.node
→ /home/a/.nvm/v0.10.33/lib/node_modules/tessel/node_modules/usb/out/Debug/usb_bindings.node
→ /home/a/.nvm/v0.10.33/lib/node_modules/tessel/node_modules/usb/Debug/usb_bindings.node
→ /home/a/.nvm/v0.10.33/lib/node_modules/tessel/node_modules/usb/out/Release/usb_bindings.node
→ /home/a/.nvm/v0.10.33/lib/node_modules/tessel/node_modules/usb/Release/usb_bindings.node
→ /home/a/.nvm/v0.10.33/lib/node_modules/tessel/node_modules/usb/build/default/usb_bindings.node
→ /home/a/.nvm/v0.10.33/lib/node_modules/tessel/node_modules/usb/compiled/0.10.33/linux/x64/usb_bindings.node
→ /home/a/.nvm/v0.10.33/lib/node_modules/tessel/node_modules/usb/node_modules/usb-shyp-linux-x64/node-v11/usb_bindings.node
→ /home/a/.nvm/v0.10.33/lib/node_modules/tessel/node_modules/usb/node_modules/usb-shyp-linux-x64/usb_bindings.node
at bindings (/home/a/.nvm/v0.10.33/lib/node_modules/tessel/node_modules/usb/node_modules/bindings-shyp/bindings.js:104:9)
at Object. (/home/a/.nvm/v0.10.33/lib/node_modules/tessel/node_modules/usb/usb.js:1:124)
at Module._compile (module.js:456:26)
at Object.Module._extensions..js (module.js:474:10)
at Module.load (module.js:356:32)
at Function.Module._load (module.js:312:12)
at Module.require (module.js:364:17)
at require (module.js:380:17)
at Object. (/home/a/.nvm/v0.10.33/lib/node_modules/tessel/src/tessel_usb.js:10:44)
at Module._compile (module.js:456:26)

This one is a little weird, I haven't gotten to the bottom of it. If a script uses the net/tls modules, and you try to tessel run it before the WiFi connects, the Tessel just silently executes whatever was the last tessel push script instead.

So for whatever reason you can't even write a script that includes wifi.on('connect', function () { … }) since it never even gets that far.

Originally from private Beta repo.

this works

var tessel = require('tessel'); // import tessel
var gpio = tessel.port['GPIO']; // select the GPIO port
var myPin = gpio.analog[0]; // can be gpio.digital[0-5] or gpio.pin['G3'] through G6
// Turn on the pin
myPin.write(1); // setting ‘true’ has the same effect
// Read the pin
console.log('Reading pin:', myPin.read());
// Turn off the pin
myPin.write(0); // setting ‘false’ has the same effect
// Read the pin
console.log('Reading pin:', myPin.read());

but this doesn't

var tessel = require('tessel'); // import tessel
var gpio = tessel.port['GPIO']; // select the GPIO port
var myPin = gpio.analog[0]; // can be gpio.digital[0-5] or gpio.pin['G3'] through G6
// Turn on the pin
myPin.output(1); // setting ‘true’ has the same effect
// Read the pin
console.log('Reading pin:', myPin.read());
// Turn off the pin
myPin.write(0); // setting ‘false’ has the same effect
// Read the pin
console.log('Reading pin:', myPin.read());

cited error:
/app/scratch.js:5: attempt to call method 'output' (a nil value)

Are there docs on how to get arduino connected?

For those of you who already have stacks and stacks of Arduino shields, don’t despair—you can still
use them! If we’re funded, we’ll make interfacing with Arduino a first-class experience. With just one 
cable, send and receive data directly from your Arduino stack using your existing C libraries. Or, use 
our firmware to control the Arduino board directly using JavaScript!

Apparently, it's possible:
http://processors.wiki.ti.com/index.php/CC3000_Host_Programming_Guide#MAC_Address_update_process.

https://github.com/tessel/firmware/blob/abcd7a8ea5ce5d0e6460d9c615039c084318de9b/src/hw/hw_digital.c#L26 adjusts the mode in the structure which is doubly wrong -- it doesn't change the state of the pin until another function happens to apply it, and it mutates the "default" setting that stays around after the script ends, in an array which should have remained const.

TI's SmartConfig allows you to use a mobile app to set the WiFI credentials of the CC3000 WiFi chip. We put the config button on the board specifically to tie in to this service but haven't exposed it yet. This could be useful for folks who have finished up a project, embedded it in something, and want to move it around.

It happend really often for me (during nodeconf) that my tessel script chrashed and I got a

Error: ENOENT Cannot connect to 10.11.83.177
stack traceback:
    [T]:src/colony/modules/net.js:109: in function <[T]:src/colony/modules/net.js:121>

I used http to send a get request like this:

http
          .get('http://10.11.83.177:8082/ws/remote?text= New Weather Report has arrived it is' + weather, function (res) {
            console.log("new weatherReport sent");
          })
          .on('error', function(e) {
            console.log("Got error: " + e.message);
          });

I guess it should hit the error handler and not just crash. Or did I do something wrong?

Our [documentation] mentions being able to switch the GPIO bank pin A2 and A4 between digital and analog signals. That is currently not implemented in the firmware.

First step here would be writing a quick Request For Comment (RFC) and posting it on the forums to discuss how to expose that API (pin.type('analog') or whatever).

Then, add that API to tessel.js. You'll want to use the underlying binding to the set_mode function.

In the UART initialization logic, all of the state variables are set, and then the actual initialization is put inside a setImmediate call. Because BLE opens a UART port, then toggles the reset line, it expects to receive bootup data immediately. Unfortunately, it misses the data because the port isn't initialized until after the data is sent from the BLE module to Tessel.

The initialization is currently within setImmediate so that you can create a UART port, then add listeners before the uart object emits any data. I believe the idea was that you would get that data once the event loop got to the setImmediate but that's not how it works.

I propose removing the setImmediate call and just initializing and registering the data listener right away.

I suspect this is trouble on my end, but filing in case anyone else hits this:

postel:firmware natevw$ make arm
AR=arm-none-eabi-ar AR_host=arm-none-eabi-ar AR_target=arm-none-eabi-ar CC=arm-none-eabi-gcc gyp  firmware.gyp --depth=. -f ninja-arm -D builtin_section=.text -D COLONY_STATE_CACHE=0 -D COLONY_PRELOAD_ON_INIT=0 -D enable_ssl=1 && ninja -C out/Release
ninja: Entering directory `out/Release'
[4/324] CC runtime/deps/c-ares/c-ares.ares__close_sockets.o
FAILED: arm-none-eabi-gcc -MMD -MF runtime/deps/c-ares/c-ares.ares__close_sockets.o.d -DCOLONY_EMBED -DCONFIG_PLATFORM_EMBED -DTM_FS_vfs -DHAVE_CLOSESOCKET -DENABLE_TLS -DHAVE_CONFIG_H -I../../../runtime/deps/axtls/config -I../../../runtime/deps/c-ares -mcpu=cortex-m3 -mthumb -mtune=cortex-m3 -march=armv7-m -mlong-calls -mfix-cortex-m3-ldrd -mapcs-frame -msoft-float -mno-sched-prolog -ffunction-sections -fdata-sections -std=c99 -Wall -Werror -Wno-error=unused-parameter -ggdb -Wno-unused-result -Wno-unused-value -Wno-unused-function -Wno-unused-variable -std=gnu99 -O3   -c ../../../runtime/deps/c-ares/ares__close_sockets.c -o runtime/deps/c-ares/c-ares.ares__close_sockets.o
dyld: Library not loaded: /usr/local/lib/libmpc.3.dylib
  Referenced from: /Users/natevw/Development/Pebble/arm-cs-tools/bin/../libexec/gcc/arm-none-eabi/4.6.3/cc1
  Reason: image not found
arm-none-eabi-gcc: internal compiler error: Trace/BPT trap: 5 (program cc1)
Please submit a full bug report,
with preprocessed source if appropriate.
See <https://github.com/jsnyder/arm-eabi-toolchain> for instructions.
[4/324] CC runtime/deps/c-ares/c-ares.ares__get_hostent.o
FAILED: arm-none-eabi-gcc -MMD -MF runtime/deps/c-ares/c-ares.ares__get_hostent.o.d -DCOLONY_EMBED -DCONFIG_PLATFORM_EMBED -DTM_FS_vfs -DHAVE_CLOSESOCKET -DENABLE_TLS -DHAVE_CONFIG_H -I../../../runtime/deps/axtls/config -I../../../runtime/deps/c-ares -mcpu=cortex-m3 -mthumb -mtune=cortex-m3 -march=armv7-m -mlong-calls -mfix-cortex-m3-ldrd -mapcs-frame -msoft-float -mno-sched-prolog -ffunction-sections -fdata-sections -std=c99 -Wall -Werror -Wno-error=unused-parameter -ggdb -Wno-unused-result -Wno-unused-value -Wno-unused-function -Wno-unused-variable -std=gnu99 -O3   -c ../../../runtime/deps/c-ares/ares__get_hostent.c -o runtime/deps/c-ares/c-ares.ares__get_hostent.o
dyld: Library not loaded: /usr/local/lib/libmpc.3.dylib
  Referenced from: /Users/natevw/Development/Pebble/arm-cs-tools/bin/../libexec/gcc/arm-none-eabi/4.6.3/cc1
  Reason: image not found
arm-none-eabi-gcc: internal compiler error: Trace/BPT trap: 5 (program cc1)
Please submit a full bug report,
with preprocessed source if appropriate.
See <https://github.com/jsnyder/arm-eabi-toolchain> for instructions.
ninja: build stopped: subcommand failed.
make: *** [firmware] Error 1
postel:firmware natevw$ 

x = 12.3456;
console.log(x.toString());

node:
12.3456

tessel:
12

Run this somewhere the Tessel can talk to:

require('net').createServer(function (sock) {
  sock.on('data', sock.write);
}).listen(5004, function () {
  console.log("listening", this.address().port);
});

Run this with tessel run -l all -s script.js:

require('net').connect(5004, "192.168.4.114", function () {
  console.log("connected, initiating writes");
  for (var i = 0; i < 25; i += 1) this.write(".");
  this.end();
});

Gets stuck in a loop like the following, and then needs a power cycle before doing any other WiFi stuff:

…
Sending 1 bytes <Buffer 2e>
tSLInformation.usNumberOfFreeBuffers 1
HCI Event error on command: 1008
HCI Event error on command: 1008
HCI Event error on command: 1008
HCI Event error on command: 1008
kicking out of buffer wait
HostFlowControlConsumeBuff is bad -5
Sending 1 bytes <Buffer 2e>
tSLInformation.usNumberOfFreeBuffers 1
HCI Event error on command: 1008
…

Full log at https://gist.github.com/natevw/6c39419e1e3bca9bf17c

Test Code:

var tessel = require('tessel');
var port = tessel.port['A'];
var uart = new port.UART({
  baudrate: 57600, dataBits: 8
});

console.log('set!');

Run this once, close the script (ctrl-c), then try running it again. You will only see 'set!' once.

Given this test script:

var http = require('http');

setInterval(function() {
  var req = http.request(
    { hostname: 'httpstat.us',
      path: '/200',
      method: 'GET'}
    , function(res) {
      res.on('data', function(chunk) { console.log("BODY: " + chunk) })
    });

  req.on('error', function(e) { console.log("Problem with request: %s", e.message)});

  req.end();
},
500);

Only 4 sockets can be open simultaneously on the CC3000 at once. By the end of the log, 5 are in use, the last traces I get from the CC are:

 ← read  HCI_EVNT_SOCKET               <- socket #5 opened.
 → write HCI_CMND_SETSOCKOPT           24 bytes: <Buffer 05 00 00 00 ff ff 00 00 00 40 00 00 08 00 00 00 04 ...>
 ← read  HCI_EVNT_SETSOCKOPT           4 bytes: <Buffer ff ff ff ff>
 → write HCI_CMND_CONNECT              -> connecting socket #5... (8.8.8.8)
 ← read  HCI_EVNT_CONNECT              4 bytes: <Buffer 00 00 00 00>
 → write          DATA                 46 'bytes:' <Buffer 05 00 00 00 0c 00 00 00 1d 00 00 00 00 00 00 00 b6 ...>
 ← read  HCI_EVNT_SEND                 8 bytes: <Buffer 05 00 00 00 1d 00 00 00>
 → write HCI_CMND_SELECT               → r .... w .... e .... [non-blocking, timeout: 3125ms]

That timeout is incredibly long, especially since our sockopts are all 5ms. Maybe a bad SPI trace, maybe an error signal?

The full logs are here. There should have been 10+ requests that have gone through the CC3k, however only 5 made it up through the event loop.

WARN Warning. Deploying just this file.
INFO Bundling directory /Users/jialiya/projects/technical/test/wifi
INFO Deploying bundle (5.00 KB)...
INFO Running script...
BODY: 200 OK
BODY: 200 OK
BODY: 200 OK
BODY: 200 OK
BODY: 200 OK

Backtracing gets me to CC_BLOCKS. My guess is that we miss an IRQ somewhere and then get thrown into an inconsistent state.

(gdb) bt
#0  CC_BLOCKS () at ../../src/cc3000/host_spi.c:37
#1  0x1401b152 in hci_event_handler (from=0x0, fromlen=0x0,
    pRetParams=<optimized out>) at ../../src/cc3000/utility/evnt_handler.c:250
#2  SimpleLinkWaitEvent (usOpcode=<optimized out>, pRetParams=<optimized out>)
    at ../../src/cc3000/utility/evnt_handler.c:834
#3  0x1401be10 in select (nfds=<optimized out>, readsds=0x10089c90,
    writesds=0x10089c94, exceptsds=0x10089c98, timeout=0x10089c9c)
    at ../../src/cc3000/utility/socket.c:645
#4  0x140218ac in wait_ares (channel=0x281368f0)
    at ../../deps/runtime/src/colony/lua_cares.c:83
#5  tm__sync_gethostbyname (domain=<optimized out>)

Currently my guess is that if we enforce the 4 socket limit this shouldn't be resolved. Worth investigating further to see if:

1. Are the other 5+ GET responses somewhere in the event queue and we're going to handle them, but get caught up in CC_BLOCKS before we can?
2. Do the first responses I get back correspond to the first sockets getting data on the CC3k?
3. Is there a way to recover once we get into an inconsistent CC state?

Doing the same thing with POST requests gives me this log. The 5th socket request never completes.

I'm having trouble outputting PWM on my TM-00-002 board:

var port = require('tessel').port.GPIO,
    pin = port.pwm[0];
port.pwmFrequency(50);
console.log(pin);
pin.pwmDutyCycle(0.6);

Outputs:

{ pin: 37, isPWM: true, interrupts: {} }
Error: PWM is not suported on this pin

Originally posted in private Beta repo.

Seems potentially related to #60. My Tessel is on LAN (TomatoUSB router) and has IP 192.168.1.5. Laptop has IP 192.168.1.6. I can ping from laptop (.6) to Tessel (.5) just fine.

But when I start nc -kl 8080 on laptop and run this script:

//require('http').get("http://192.168.1.6:8080", function(res) {
require('http').get("http://postel:8080", function(res) {
//require('http').get("http://ipcalf.com:8080", function(res) {
  console.log("Got response: " + res.statusCode);
  res.on('data', function (d) { console.log(d.toString()); });
}).on('error', function(e) {
  console.log("Got error: " + e.message);
});

I get the following error:

Got error: ENOENT Cannot connect to 192.168.1.6 Got: err-5

So Tessel has DNS resolved from "postel" to "192.168.1.6" via my router, but now can't connect. (I don't see any TCP handshake attempt in Wireshark.) I'm certain this is not just an interface binding issue, I can connect to the nc listener on my laptop just fine from any other machine on my network.

If, however, I set up a remote tunnel to my laptop via a remote server (third commented line), Tessel can connect via that to my laptop just fine O_o

Originally reported on the forums.

var tessel = require('tessel');
var port = tessel.port['A'];
var blueToothSerial = new port.UART();

// Throws Error
blueToothSerial.setDataBits(8);

// Throws Error
blueToothSerial.setParity("none");

Error that is thrown:

> blueToothSerial.setParity("none");
Error: Invalid parity value
    at redo (/usr/local/lib/node_modules/tessel/node_modules/structured-clone/serialize.js:136:16)
    at rez (/usr/local/lib/node_modules/tessel/node_modules/structured-clone/serialize.js:168:25)
    at redo (/usr/local/lib/node_modules/tessel/node_modules/structured-clone/serialize.js:147:17)
    at rerez (/usr/local/lib/node_modules/tessel/node_modules/structured-clone/serialize.js:174:7)
    at Function.exports.deserialize (/usr/local/lib/node_modules/tessel/node_modules/structured-clone/serialize.js:222:10)
    at Tessel.<anonymous> (/usr/local/lib/node_modules/tessel/src/commands.js:112:32)
    at Tessel.emit (events.js:95:17)
    at InEndpoint.<anonymous> (/usr/local/lib/node_modules/tessel/src/tessel_usb.js:217:14)
    at InEndpoint.emit (events.js:95:17)
    at Transfer.transferDone (/usr/local/lib/node_modules/tessel/node_modules/usb/usb.js:273:9)

The parameters are not validated properly in tessel.js.

See http://processors.wiki.ti.com/index.php/CC3000_Host_Programming_Guide#MAC_Address_update_process for more information.

Hi,

I was very pleased to note that the Tessel built-in classes, like Pin, support chaining:

var pin = gpio.digital[0].input();

However, this seems to have been applied quite randomly to the codebase. Most functions do return this but others, such as .toggle() and .pulse(), do not. Some, like .write(), explicitly do not return the instance.

Is this intentional or just an oversight?

Originally posted in private Beta repo.

I propose we add each pin's name as one of its properties. The reason being that it's currently useless to call something like pin.analog as it returns:

[ { pinmode: 'input', pin: 28 }, { pinmode: 'input', pin: 29 }, { pinmode: 'input', pin: 30 }, { pinmode: 'input', pin: 31 }, { pinmode: 'input', pin: 32 }, { pinmode: 'input', pin: 33 } ]
That doesn't tell me which pins I can use for analog reads.

Originally posted in private Beta repo.