Search results for:

The command-line still evaluates everything in order, as you type it... The if statement won't evaluate again unless you tell it to.

You could do something like:

setInterval(function() {
 if (a != b) {
    var b = a;
    var c = 1;
    var d = 1;
    var e = 1;
    var f = b+c+d+e;
 }
},100);

And now, that code will keep being called every 100ms so if you write a=20, everything will update as you expect.

I just tried this, and it worked. End result was:

>dump()
var a = 10;
var b = 10;
var c = 1;
var d = 1;
var e = 1;
var f = 13;

Which version of Espruino were you using? the one from the website, espruino-nightly, or your own build?

Thanks! Take a look at build_docs.py or build_jswrapper.py - common.py has code to read in all the json data, then you should just need to call something like: print(json.dumps(mydata)) and it'll dump it all out.

Just for anyone trying today, the nightly builds are back up: https://espruino-nightly.noda.se/2013-12-13/

Actually, Ganehag just fixed the nightly builds. You could try this one:

https://espruino-nightly.noda.se/2013-12-13/

Yes, sadly it's a huge regression in that version of Espruino that causes it to leave off the last few characters of the function.

It should have been fixed in last night's nightly build, but because of the filename change, that hasn't appeared!

For now, type:

var onInit = function () {
     SPI1.setup({sck:A5,miso:A6,mosi:A7});
     SPI1.send([0x90],A4);
};123

If you add a few characters after every function definition then you'll be fine.

Sorry about that - everything is in flux at the moment. I'm trying really hard to get everything ready for KickStarter and I don't have time to make sure everything works on cutting edge git all the time.

Great! Could you post up some details about it? I'd love to get a list of 'stuff you can do' going :)

Publishing to npm would be great - it'd be great to be able to give people easier steps to get it!

As far as getting that list of builtins, it might be easier to just copy the JSON directly from source files and let you do what you need with it. For instance:

https://github.com/espruino/Espruino/blob/master/src/jswrap_string.c#L18

There's some documentation on the format here: https://github.com/espruino/Espruino/blob/master/scripts/common.py#L32

It's actually been added now - sorry. That's the problem, you can't keep everyone happy.

In the case of I2C/SPI/OneWire it's fine as they are ignored (and the relevant SPI setup command is added) - it is only when you've set a pin as a simple output or have added pullup/pulldowns to an input.

For beginners, there is something nice about saying LED1.set();save(); though

Sorry about that - the espruino-nightly is actually something that someone else is running.

I changed the filenames recently (so they included the version number) and it seems to have broken his builds :( I've just asked him so hopefully they'll come back online soon.

Also I don't think you're using the latest version of the Web IDE either.

15 minutes before you wrote that post I merged some of Jurgen's changes which it turns out have fixed that issue (if you don't use modules, they don't call 'Modules.removeAllCached()').

You're using an old version of the Espruino firmware with a new version of the Espruino Web IDE.

Try https://espruino-nightly.noda.se/

Hi Andrey,

That's you problem I'm afraid. We haven't uploaded the latest version to the Chrome Web Store yet. However if you install the version from GitHub it should work:

https://github.com/espruino/EspruinoWebIDE

You will probably need to install an espruino-nightly image as well

Did you have any more information? An example?

Are you using the latest web IDE from git, or the one on the chrome web store.

At the moment there isn't a direct replacement. You can do:

digitalWrite(pin, 1);
setTimeout(function() {
  digitalWrite(pin, 0);
  setTimeout(function() {
   ...
  }, 0.35);
},0.35);

or you can do:

digitalPulse(pin,1,0.35); 

Which will be substantially more accurate. This actually starts executing subsequent commands (apart from other digitalPulses) before the pulse has ended, so for something more accurate, you could try:

digitalPulse(pin,1,0.35); 
digitalPulse(pin,0,0.35); 

I'm trying to discourage people from writing code that delays (because then they start to get annoyed that Espruino isn't multithreaded), however I think delays do need to be added for this kind of thing - I've put a bug in for it, so hopefully I'll get around to it at some point.

Hi Gamaiel,

That looks really cool - it doesn't seem to handle multi-line editing yet though?

There's actually some simple autocomplete code in the Espruino Web IDE at the moment - as it's also JS I wonder if it can be shared... https://github.com/espruino/EspruinoWebIDE/blob/master/js/codemirror/addon/hint/espruino-hint.js

I intend to improve that quite a lot at some point (I'll export some data from build_docs.py in Espruino) so it'd be cool if your project could take advantage of that (or could help the Web IDE as well ;)

As it looks like you're caching the whole JS command before sending it to Espruino, syntax highlighting would be awesome :)

Great! There's actually twice that amount of memory available on there too, but at the moment I'm having issues keeping the code size down - so any more memory and you can't save :(

On the video, I just soldered headers on. They're cheap to buy, and it makes hacking stuff up easier as you can just plug and unplug wires. Also, at the bottom I'd soldered on a Bluetooth module (HC-05) so it could be controlled (or could send data wirelessly).

Just wanted to add that you should check this out: http://forum.espruino.com/conversations/340/

The 3.2" board had an issue where there was at least 4x less memory than there should be

If you have the 3.2, try this: http://www.espruino.com/files/espruino_hystm32_32_vc_1v43_beta.zip

Hi - which screen size board do you have?

I've just noticed something a bit odd with the 3.2" binaries - it does seem like they're using far less of their memory than expected for some reason. If that is the case then I'll have a look at recompiling a newer version for you with that fixed.

This is expected behaviour. If you want to force LED1 on at startup, do:

function onInit() {
  LED1.write(1);
}
save();

I'd been considering saving pin state, but I'm not sure if it's such a good idea as often connected devices will actually need more complex initialisation done to them than just setting pin state back exactly as it was. I think very often it's actually better to force people to explicitly set the pin state.

I'd be interested to hear what everyone thinks though.

Great news! It's worrying that it doesn't work on your new Mac - I guess it could be a driver problem... I think some of the boards tend to use a cheap PL2303 USB-serial convertor that might be a knock-off of the 'real' Prolific one.

Can you try adding -b 9600 to the command-line? It'll force a much slower baud rate, which might help?

I can't say I've ever come across this problem before. Perhaps you could try another computer?

Hi Mudderman,

It looks like you're doing the right thing...

When the flasher says Chip replied with a NACK, have you tried just running the flasher again - without resetting the board? I've had that happen sometimes.

It's not a big deal if flashing fails and you need to do it again - the bootloader is built in to the chip, so you can never kill it.

Just to say, if you have any luck getting it going , I'd love to know.

Given how soon I want to get the Espruino boards out, if I don't get this fixed in the next few days I will have to order the Adafruit modules.