Posted at 2017-11-27 by PaddeK

Well it seems to me you are trying to parse a XML response with JSON.parse. You will need a XML parser to get what you want.

If you are just interested in the BinaryState Tag value you can get by with a regex if they are supported.

// Gives you the value of BinaryState
let binaryState = /<BinaryState>(\d)<\/BinaryState>/.exec(pdata)[1];

Posted at 2017-11-27 by @gfwilliams

Basic Regex are supported in Espruino latest builds now, but they may not be supported in 512k ESP8266 builds because of the lack of flash

Posted at 2017-11-27 by Vasily

Thank you both.

I thought response looked as JSON string so I was wondering what I m doing wrong.
I will give a try asap.

BTW, Gordon, another issue was reported already.... still not resolved.

onPageRequest(req, res)
value req.method=="POST" is OK
but inside res.method it is still "GET"

Txs.

Posted at 2017-11-27 by @gfwilliams

but inside res.method it is still "GET"

I'm pretty sure this was covered in another post quite recently. If you do url.parse("url") then Espruino has no idea whether to use GET or POST - so it defaults to GET. That's likely what you're seeing.

Posted at 2017-11-27 by Vasily

yep, understood.

Posted at 2017-11-27 by Vasily

I found a stand-alone package that could be very usefull for XMLtoJSON.
I will give a try. It has also xmlToJSON.min.js...interesting.

https://github.com/metatribal/xmlToJSON

Posted at 2017-11-27 by Wilberforce

Here is your string you want to parse.. that lib will be too big to use.

<?xml version="1.0" encoding="utf-8"?><s:Envelope xmlns:s="http://schemas.xmlsoap.org/soap/envelope/" s:encodingStyle="http://schemas.xmlsoap.org/soap/encoding/">
	<s:Body>
		<u:GetBinaryState xmlns:u="urn:Belkin:service:basicevent:1">
			<BinaryState>1</BinaryState>
		</u:GetBinaryState>
	</s:Body>
</s:Envelope>

You should be able to treat as a string and search on <BinaryState> and then get the character at the next location - which would be 0 or 1

Posted at 2017-11-27 by Vasily

Thanks.... I will try tonight both solutions regexp and xmltojson class

actually as there are 2 options only, the fastest way would be using 2 times
String.contains(1)
String.contains(0)

I am wondering about the memory size... device itself has quite a big memory (divided by half due to boot-1 and boot-2). I was writing a small script and get already out-of-memory for a script less than 2 pages. If I understood correctly in case there is a module (incorporated into the firmware) than there is no issue with memory usage.

Posted at 2017-11-27 by PaddeK

Its String.includes and Espruino doesn't support it i guess, at least it is not listed in the reference.
But a good alternative would be indexOf and check for a return value greater -1 .

Posted at 2017-11-28 by Vasily

Thanks for suggestions. works fine with string.indexof().

The script has been cleaned .... but big problem came - "out of memory" and even Minimization did not help.
Popup showing "Minimization 15000 to 7200", but while uploading I still see error
"ERROR: Too big to save to flash (14508 vs 12284 bytes)"

any hint how I can overcome it (apart from obvious code restructure)?

complete code attached (change your wifi id first)

Attachments:

• alexa_nok2workonprogress.js

Posted at 2017-11-28 by PaddeK

Try minify yourself.. at least some parts.. first place i would start is console.log. You have quite a lot of them and multiple times in a row.

In your setupId function for example:

  console.log("NAME: " + gFriendlyName);
  console.log("SERIAL: " + gSerialNumber);
  console.log("UUID: " + gUuid);

Try to combine this to a single console.log

  console.log("NAME: " + gFriendlyName, "SERIAL: " + gSerialNumber, "UUID: " + gUuid);

I am not aware of a minifier that does this for you without breaking everything.
Using arrow functions could scrape another few bytes. I could take a stab at minifying this
later today.

Edit

Offloading fixed strings etc. to a eeprom could help too.

Posted at 2017-11-28 by Wilberforce

Get rid of the comments. They are taking up a heap of space...

Posted at 2017-11-28 by Vasily

You will not believe me.... I tried this way and the code got LONGER, surprisingly :)
Anyway I will not come down 2500 chars with that :(

p.s. the thing is doing on-off works with Alexa, but after a while get nuts and give numbers due to memory leaks.

Posted at 2017-11-28 by Vasily

I use minification - comments should be removed.

I tried all possible minification options. .... suprized that minification tells me "reduced to 7200" but still uploading 14508 to the board.

Posted at 2017-11-28 by Wilberforce

As I said, remove the comments. You are assuming they get stripped and may not be.

You could also try minimising the code with an external tool and upload that.

Posted at 2017-12-19 by PaddeK

Well.. if found a thread regarding UDP multicast with AT firmware http://bbs.espressif.com/viewtopic.php?f=16&t=234&p=6382&hilit=multicast#p6382 .
The thread is a bit older but since i can't find anything thats stating the opposite it seems the AT firmware does not support this.

Thats it then.. or has someone another idea?

Posted at 2017-12-20 by @gfwilliams

I assume that if you know the IP address of the Alexa device there's no need to multicast? You can just send everything directly to it?

Posted at 2017-12-20 by PaddeK

I would have to try that.. but its a good idea. Other possibilities are maybe tcp broadcast in combination with a own skill thats using tcp broadcast for discovery or a solution with some kind of gateway and mqtt for example. But this is all far more work then necessary.. if espressif would just have included this feature in their AT firmware.. its in the SDK.. why not exposing it in the AT firmware grml

I did read its possible to modify the AT firmware and add features to it.. this would be another solution.. but maybe the hardest route to go.

Posted at 2017-12-26 by PaddeK

So i tried a few things now.. and sadly i am can't get alexa to discover. I tried starting the scan and then manualy responding to the UPNP search from alexa with multicast du 239.255.255.250:1900 but alexa does not care about it. I sniffed the traffic with tcpdump to make sure i send after the search from alexa and that my espruino wifi is sending the right response. While looking at the traffic running in with tcpdump i saw some devices answer to alexa directly and port of 50000 but still UDP. I tried that too.. but it it does not even show up in the tcpdump log. I guess with so much else i want to do i leave it as it is. Maybe i use a ESP8266 or ESP32 myself and round up the module but for espruino boards i see no chance without complete UDP support.

Posted at 2018-01-04 by opichals

It may be that this just isn't possible with the ESP8266 AT firmware. As far as I can tell the AT firmware provides no information on where a UDP packet came from (address/port), so it might not be possible to send a response back to the right address (unless you know for sure where Alexa is).

The CIPSTATUS (implementation source link) seems to be the command returning the remote_port/ip information. Perhaps that could be send for every +IPD received?

Posted at 2018-01-04 by @gfwilliams

Wow, thanks - good find! Yes, that sounds like it'd be perfect. It's never going to be super-fast so I don't think the extra overhead would be such a problem.

Posted at 2018-01-04 by opichals

There seems to be even better way which would be to use AT+CIPDINFO according to the docs PDF.

On github I could only find the CIPDINFO in the docs for ESP32 though and the implementation seems to be in the libat_core.a which looks like a closed source thing probably shared across the ESPs.

Posted at 2018-01-04 by PaddeK

Even in a scenario where i do know the address of alexa it did not work. And if i understood the AT documentation right there is a mode (mode 2) in CIPSTART where you can answer to the message origin. And yes.. this is all just a lack of the AT firmware.

Posted at 2018-01-04 by @gfwilliams

Interesting about CIPDINFO - I'll have to give that a try. I guess it may not be in the 0.40 firmware that the Espruino WiFis ship with, but it's worth a try :)

Posted at 2018-01-04 by PaddeK

Hm.. maybe in newer versions of the AT firmware UDP multicast is properly supported too.
Is it possible to update the firmware without risking to kill the espruino?

Posted at 2018-01-04 by opichals

According to this the 1.5.0 version might be required http://bbs.espressif.com/viewtopic.php?f=66&t=1677&p=5517&hilit=CIPDINFO#p5517. So perhaps the CIPSTATUS workaround (could be problematic for multiple subsequent incoming +IPDs of different origins) and when the CIPDINFO works use rather that.

But that gets complicated. Does EspruinoWifi support updating the ESP firmware? With that it could get integrated into the EspruinoWebIDE flasher which would be sufficient IMO.

Posted at 2018-01-05 by @gfwilliams

Theoretically you can update the ESP8266 firmware via the EspruinoWiFi, but it's not an easy process. I think I'd try and stick with the existing firmware if at all possible - I don't think many people would bother updating it if I added the option to do so... And actually I just tested and CIPDINFO=1 works fine, so it looks like there's no need.

It's just as well, because AT+CIUPDATE would have been perfect, but it seems to fail :(

@PaddeK I'm not sure multicast is supported in any firmwares? I could be wrong though

Posted at 2018-01-06 by PaddeK

I think at the moment not.. but its really not easy to find solid information about the AT firmware and its capabilities. But i did find a statement that older versions are open source and therefore are extensible.. no idea how feasible that is though.
My only hope at this point is that you come up with some magic to make UDP multicast happen or espressif gifts us with a new AT firmware :)

Posted at 2018-01-31 by @gfwilliams

Espruino WiFi now has UDP support: http://forum.espruino.com/conversations/316287/#comment14072238

There's no broadcast as far as I can tell but thanks to @opichals you can now get the UDP sender's IP and port, which I imagine will help a lot

Posted at 2018-01-31 by PaddeK

As soon as i got some spare time i will pick up where i left and see if this brings the hoped solution.