CircuitPython is a derivative of MicroPython designed to simplify experimentation and education on low-cost microcontrollers. It makes it easier than ever to get prototyping by requiring no upfront desktop software downloads. Simply copy and edit files on the CIRCUITPY drive to iterate.

Set Up CircuitPython

Follow the steps to get CircuitPython installed on your MagTag.

Option 1 - Load with UF2 Bootloader

This is by far the easiest way to load CircuitPython. However it requires your board has the UF2 bootloader installed. Some early boards do not (we hadn't written UF2 yet!) - in which case you can load using the built in ROM bootloader.

Still, try this first!

If you're using Windows and you get an error at the end of the file copy that says Error from the file copy, Error 0x800701B1: A device which does not exist was specified. You can ignore this error, the bootloader sometimes disconnects without telling Windows, the install completed just fine and you can continue. If its really annoying, you can also upgrade the bootloader (the latest version of the UF2 bootloader fixes this warning)

Option 2 - Use esptool to load BIN file

If you have an original MagTag with while soldermask on the front, we didn't have UF2 written for the ESP32S2 yet so it will not come with the UF2 bootloader.

You can upload with esptool to the ROM (hardware) bootloader instead!

Option 3 - Use Chrome Browser To Upload BIN file

If for some reason you cannot get esptool to run, you can always try using the Chrome-browser version of esptool we have written. This is handy if you don't have Python on your computer, or something is really weird with your setup that makes esptool not run (which happens sometimes and isn't worth debugging!) You can follow along on the Web Serial ESPTool page and either load the UF2 bootloader and then come back to Option 1 on this page, or you can download the CircuitPython BIN file directly using the tool in the same manner as the bootloader.

One of the great things about most Espressif microcontrollers are their built-in WiFi capabilities. This page covers the basics of getting connected using CircuitPython.

The first thing you need to do is update your code.py to the following (it will error until WiFi details are added). Click the Download Project Bundle button to download the necessary libraries and the code.py file in a zip file. Extract the contents of the zip file, and copy the entire lib folder and the code.py file to your CIRCUITPY drive.

# SPDX-FileCopyrightText: 2020 Brent Rubell for Adafruit Industries
#
# SPDX-License-Identifier: MIT

import os
import ipaddress
import ssl
import wifi
import socketpool
import adafruit_requests

# URLs to fetch from
TEXT_URL = "http://wifitest.adafruit.com/testwifi/index.html"
JSON_QUOTES_URL = "https://www.adafruit.com/api/quotes.php"
JSON_STARS_URL = "https://api.github.com/repos/adafruit/circuitpython"

print("ESP32-S2 WebClient Test")

print(f"My MAC address: {[hex(i) for i in wifi.radio.mac_address]}")

print("Available WiFi networks:")
for network in wifi.radio.start_scanning_networks():
    print("\t%s\t\tRSSI: %d\tChannel: %d" % (str(network.ssid, "utf-8"),
                                             network.rssi, network.channel))
wifi.radio.stop_scanning_networks()

print(f"Connecting to {os.getenv('CIRCUITPY_WIFI_SSID')}")
wifi.radio.connect(os.getenv("CIRCUITPY_WIFI_SSID"), os.getenv("CIRCUITPY_WIFI_PASSWORD"))
print(f"Connected to {os.getenv('CIRCUITPY_WIFI_SSID')}")
print(f"My IP address: {wifi.radio.ipv4_address}")

ping_ip = ipaddress.IPv4Address("8.8.8.8")
ping = wifi.radio.ping(ip=ping_ip)

# retry once if timed out
if ping is None:
    ping = wifi.radio.ping(ip=ping_ip)

if ping is None:
    print("Couldn't ping 'google.com' successfully")
else:
    # convert s to ms
    print(f"Pinging 'google.com' took: {ping * 1000} ms")

pool = socketpool.SocketPool(wifi.radio)
requests = adafruit_requests.Session(pool, ssl.create_default_context())

print(f"Fetching text from {TEXT_URL}")
response = requests.get(TEXT_URL)
print("-" * 40)
print(response.text)
print("-" * 40)

print(f"Fetching json from {JSON_QUOTES_URL}")
response = requests.get(JSON_QUOTES_URL)
print("-" * 40)
print(response.json())
print("-" * 40)

print()

print(f"Fetching and parsing json from {JSON_STARS_URL}")
response = requests.get(JSON_STARS_URL)
print("-" * 40)
print(f"CircuitPython GitHub Stars: {response.json()['stargazers_count']}")
print("-" * 40)

print("Done")

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

# SPDX-FileCopyrightText: 2020 Brent Rubell for Adafruit Industries
#
# SPDX-License-Identifier: MIT

import os
import ipaddress
import ssl
import wifi
import socketpool
import adafruit_requests

# URLs to fetch from
TEXT_URL = "http://wifitest.adafruit.com/testwifi/index.html"
JSON_QUOTES_URL = "https://www.adafruit.com/api/quotes.php"
JSON_STARS_URL = "https://api.github.com/repos/adafruit/circuitpython"

print("ESP32-S2 WebClient Test")

print(f"My MAC address: {[hex(i) for i in wifi.radio.mac_address]}")

print("Available WiFi networks:")
for network in wifi.radio.start_scanning_networks():
    print("\t%s\t\tRSSI: %d\tChannel: %d" % (str(network.ssid, "utf-8"),
                                             network.rssi, network.channel))
wifi.radio.stop_scanning_networks()

print(f"Connecting to {os.getenv('CIRCUITPY_WIFI_SSID')}")
wifi.radio.connect(os.getenv("CIRCUITPY_WIFI_SSID"), os.getenv("CIRCUITPY_WIFI_PASSWORD"))
print(f"Connected to {os.getenv('CIRCUITPY_WIFI_SSID')}")
print(f"My IP address: {wifi.radio.ipv4_address}")

ping_ip = ipaddress.IPv4Address("8.8.8.8")
ping = wifi.radio.ping(ip=ping_ip)

# retry once if timed out
if ping is None:
    ping = wifi.radio.ping(ip=ping_ip)

if ping is None:
    print("Couldn't ping 'google.com' successfully")
else:
    # convert s to ms
    print(f"Pinging 'google.com' took: {ping * 1000} ms")

pool = socketpool.SocketPool(wifi.radio)
requests = adafruit_requests.Session(pool, ssl.create_default_context())

print(f"Fetching text from {TEXT_URL}")
response = requests.get(TEXT_URL)
print("-" * 40)
print(response.text)
print("-" * 40)

print(f"Fetching json from {JSON_QUOTES_URL}")
response = requests.get(JSON_QUOTES_URL)
print("-" * 40)
print(response.json())
print("-" * 40)

print()

print(f"Fetching and parsing json from {JSON_STARS_URL}")
response = requests.get(JSON_STARS_URL)
print("-" * 40)
print(f"CircuitPython GitHub Stars: {response.json()['stargazers_count']}")
print("-" * 40)

print("Done")

Your CIRCUITPY drive should resemble the following.

To get connected, the next thing you need to do is update the settings.toml file.

The settings.toml File

We expect people to share tons of projects as they build CircuitPython WiFi widgets. What we want to avoid is people accidentally sharing their passwords or secret tokens and API keys. So, we designed all our examples to use a settings.toml file, that is on your CIRCUITPY drive, to hold secret/private/custom data. That way you can share your main project without worrying about accidentally sharing private stuff.

If you have a fresh install of CircuitPython on your board, the initial settings.toml file on your CIRCUITPY drive is empty.

To get started, you can update the settings.toml on your CIRCUITPY drive to contain the following code.

# SPDX-FileCopyrightText: 2023 Adafruit Industries
#
# SPDX-License-Identifier: MIT

# This is where you store the credentials necessary for your code.
# The associated demo only requires WiFi, but you can include any
# credentials here, such as Adafruit IO username and key, etc.
CIRCUITPY_WIFI_SSID = "your-wifi-ssid"
CIRCUITPY_WIFI_PASSWORD = "your-wifi-password"

1
2
3
4
5
6
7
8
9

# SPDX-FileCopyrightText: 2023 Adafruit Industries
#
# SPDX-License-Identifier: MIT

# This is where you store the credentials necessary for your code.
# The associated demo only requires WiFi, but you can include any
# credentials here, such as Adafruit IO username and key, etc.
CIRCUITPY_WIFI_SSID = "your-wifi-ssid"
CIRCUITPY_WIFI_PASSWORD = "your-wifi-password"

This file should contain a series of Python variables, each assigned to a string. Each variable should describe what it represents (say wifi_ssid), followed by an = (equals sign), followed by the data in the form of a Python string (such as "my-wifi-password" including the quote marks).

At a minimum you'll need to add/update your WiFi SSID and WiFi password, so do that now!

As you make projects you may need more tokens and keys, just add them one line at a time. See for example other tokens such as one for accessing GitHub or the Hackaday API. Other non-secret data like your timezone can also go here.

For the correct time zone string, look at http://worldtimeapi.org/timezones and remember that if your city is not listed, look for a city in the same time zone, for example Boston, New York, Philadelphia, Washington DC, and Miami are all on the same time as New York.

Of course, don't share your settings.toml - keep that out of GitHub, Discord or other project-sharing sites.

If you connect to the serial console, you should see something like the following:

In order, the example code...

Checks the ESP32's MAC address.

print(f"My MAC address: {[hex(i) for i in wifi.radio.mac_address]}")

1

print(f"My MAC address: {[hex(i) for i in wifi.radio.mac_address]}")

Performs a scan of all access points and prints out the access point's name (SSID), signal strength (RSSI), and channel.

print("Available WiFi networks:")
for network in wifi.radio.start_scanning_networks():
    print("\t%s\t\tRSSI: %d\tChannel: %d" % (str(network.ssid, "utf-8"),
                                             network.rssi, network.channel))
wifi.radio.stop_scanning_networks()

1
2
3
4
5

print("Available WiFi networks:")
for network in wifi.radio.start_scanning_networks():
    print("\t%s\t\tRSSI: %d\tChannel: %d" % (str(network.ssid, "utf-8"),
                                             network.rssi, network.channel))
wifi.radio.stop_scanning_networks()

Connects to the access point you defined in the settings.toml file, and prints out its local IP address.

print(f"Connecting to {os.getenv('WIFI_SSID')}")
wifi.radio.connect(os.getenv("WIFI_SSID"), os.getenv("WIFI_PASSWORD"))
print(f"Connected to {os.getenv('WIFI_SSID')}")
print(f"My IP address: {wifi.radio.ipv4_address}")

1
2
3
4

print(f"Connecting to {os.getenv('WIFI_SSID')}")
wifi.radio.connect(os.getenv("WIFI_SSID"), os.getenv("WIFI_PASSWORD"))
print(f"Connected to {os.getenv('WIFI_SSID')}")
print(f"My IP address: {wifi.radio.ipv4_address}")

Attempts to ping a Google DNS server to test connectivity. If a ping fails, it returns None. Initial pings can sometimes fail for various reasons. So, if the initial ping is successful (is not None), it will print the echo speed in ms. If the initial ping fails, it will try one more time to ping, and then print the returned value. If the second ping fails, it will result in "Ping google.com: None ms" being printed to the serial console. Failure to ping does not always indicate a lack of connectivity, so the code will continue to run.

ping_ip = ipaddress.IPv4Address("8.8.8.8")
ping = wifi.radio.ping(ip=ping_ip) * 1000
if ping is not None:
    print(f"Ping google.com: {ping} ms")
else:
    ping = wifi.radio.ping(ip=ping_ip)
    print(f"Ping google.com: {ping} ms")

1
2
3
4
5
6
7

ping_ip = ipaddress.IPv4Address("8.8.8.8")
ping = wifi.radio.ping(ip=ping_ip) * 1000
if ping is not None:
    print(f"Ping google.com: {ping} ms")
else:
    ping = wifi.radio.ping(ip=ping_ip)
    print(f"Ping google.com: {ping} ms")

The code creates a socketpool using the wifi radio's available sockets. This is performed so we don't need to re-use sockets. Then, it initializes a a new instance of the requests interface - which makes getting data from the internet really really easy.

pool = socketpool.SocketPool(wifi.radio)
requests = adafruit_requests.Session(pool, ssl.create_default_context())

1
2

pool = socketpool.SocketPool(wifi.radio)
requests = adafruit_requests.Session(pool, ssl.create_default_context())

To read in plain-text from a web URL, call requests.get - you may pass in either a http, or a https url for SSL connectivity. 

print(f"Fetching text from {TEXT_URL}")
response = requests.get(TEXT_URL)
print("-" * 40)
print(response.text)
print("-" * 40)

1
2
3
4
5

print(f"Fetching text from {TEXT_URL}")
response = requests.get(TEXT_URL)
print("-" * 40)
print(response.text)
print("-" * 40)

Requests can also display a JSON-formatted response from a web URL using a call to requests.get. 

print(f"Fetching json from {JSON_QUOTES_URL}")
response = requests.get(JSON_QUOTES_URL)
print("-" * 40)
print(response.json())
print("-" * 40)

1
2
3
4
5

print(f"Fetching json from {JSON_QUOTES_URL}")
response = requests.get(JSON_QUOTES_URL)
print("-" * 40)
print(response.json())
print("-" * 40)

Finally, you can fetch and parse a JSON URL using requests.get. This code snippet obtains the stargazers_count field from a call to the GitHub API.

print(f"Fetching and parsing json from {JSON_STARS_URL}")
response = requests.get(JSON_STARS_URL)
print("-" * 40)
print(f"CircuitPython GitHub Stars: {response.json()['stargazers_count']}")
print("-" * 40)

1
2
3
4
5

print(f"Fetching and parsing json from {JSON_STARS_URL}")
response = requests.get(JSON_STARS_URL)
print("-" * 40)
print(f"CircuitPython GitHub Stars: {response.json()['stargazers_count']}")
print("-" * 40)

OK you now have your ESP32 board set up with a proper settings.toml file and can connect over the Internet. If not, check that your settings.toml file has the right SSID and password and retrace your steps until you get the Internet connectivity working!

IPv6 Networking

Starting in CircuitPython 9.2, IPv6 networking is available on most Espressif wifi boards. Socket-using libraries like adafruit_requests and adafruit_ntp will need to be updated to use the new APIs and for now can only connect to services on IPv4.

IPv6 connectivity & privacy

IPv6 addresses are divided into many special kinds, and many of those kinds (like those starting with FC, FD, FE) are private or local; Addresses starting with other prefixes like 2002: and 2001: are globally routable. In 2024, far from all ISPs and home networks support IPv6 internet connectivity. For more info consult resources like Wikipedia. If you're interested in global IPv6 connectivity you can use services like Hurricane Electric to create an "IPv6 tunnel" (free as of 2024, but requires expertise and a compatible router or host computer to set up)

It's also important to be aware that, as currently implemented by Espressif, there are privacy concerns especially when these devices operate on the global IPv6 network: The device's unique identifier (its EUI-64 or MAC address) is used by default as part of its IPv6 address. This means that the device identity can be tracked across multiple networks by any service it connects to.

Enable IPv6 networking

Due to the privacy consideration, IPv6 networking is not automatically enabled. Instead, it must be explicitly enabled by a call to start_dhcp_client with the ipv6=True argument specified:

wifi.start_dhcp_client(ipv6=True)

1

wifi.start_dhcp_client(ipv6=True)

Check IP addresses

The read-only addresses property of the wifi.radio object holds all addresses, including IPv4 and IPv6 addresses:

>>> wifi.radio.addresses
('FE80::7EDF:A1FF:FE00:518C', 'FD5F:3F5C:FE50:0:7EDF:A1FF:FE00:518C', '10.0.3.96')

1
2

>>> wifi.radio.addresses
('FE80::7EDF:A1FF:FE00:518C', 'FD5F:3F5C:FE50:0:7EDF:A1FF:FE00:518C', '10.0.3.96')

The wifi.radio.dns servers can be IPv4 or IPv6:

>>> wifi.radio.dns
('FD5F:3F5C:FE50::1',)
>>> wifi.radio.dns = ("1.1.1.1",)
>>> wifi.radio.dns
('1.1.1.1',)

1
2
3
4
5

>>> wifi.radio.dns
('FD5F:3F5C:FE50::1',)
>>> wifi.radio.dns = ("1.1.1.1",)
>>> wifi.radio.dns
('1.1.1.1',)

Ping v6 networks

wifi.radio.ping accepts v6 addresses and names:

>>> wifi.radio.ping("google.com")
0.043
>>> wifi.radio.ping("ipv6.google.com")
0.048

1
2
3
4

>>> wifi.radio.ping("google.com")
0.043
>>> wifi.radio.ping("ipv6.google.com")
0.048

Create & use IPv6 sockets

Use the address family socket.AF_INET6. After the socket is created, use methods like connect, send, recfrom_into, etc just like for IPv4 sockets. This code snippet shows communicating with a private-network NTP server; this IPv6 address will not work on your network:

>>> ntp_addr = ("fd5f:3f5c:fe50::20e", 123)
>>> PACKET_SIZE = 48
>>> 
>>> buf = bytearray(PACKET_SIZE)
>>> with socket.socket(socket.AF_INET6, socket.SOCK_DGRAM) as s:
...     s.settimeout(1)
...     buf[0] = 0b0010_0011
...     s.sendto(buf, ntp_addr)
...     print(s.recvfrom_into(buf))
...     print(buf)
... 
48
(48, ('fd5f:3f5c:fe50::20e', 123))
bytearray(b'$\x01\x03\xeb\x00\x00\x00\x00\x00\x00\x00GGPS\x00\xeaA0h\x07s;\xc0\x00\x00\x00\x00\x00\x00\x00\x00\xeaA0n\xeb4\x82-\xeaA0n\xebAU\xb1')

1
2
3
4
5
6
7
8
9
10
11
12
13
14

>>> ntp_addr = ("fd5f:3f5c:fe50::20e", 123)
>>> PACKET_SIZE = 48
>>> 
>>> buf = bytearray(PACKET_SIZE)
>>> with socket.socket(socket.AF_INET6, socket.SOCK_DGRAM) as s:
...     s.settimeout(1)
...     buf[0] = 0b0010_0011
...     s.sendto(buf, ntp_addr)
...     print(s.recvfrom_into(buf))
...     print(buf)
... 
48
(48, ('fd5f:3f5c:fe50::20e', 123))
bytearray(b'$\x01\x03\xeb\x00\x00\x00\x00\x00\x00\x00GGPS\x00\xeaA0h\x07s;\xc0\x00\x00\x00\x00\x00\x00\x00\x00\xeaA0n\xeb4\x82-\xeaA0n\xebAU\xb1')

We will use a web service provided by the National Oceanic and Atmospheric Administration (NOAA) to get our tide information. One benefit of this is that no API key is needed, since it's your tax dollars at work. However, since this is a part of the US government, the service only covers regions of the United States.

If you live outside the United States, you may have to find a service which sends out tide data as a JSON feed and adapt the code to your new data source. The framework laid out here should provide you a great starting point.

NOAA CO-OPS API

Wow, acronyms! We already covered NOAA above. CO-OPS stands for Center for Operational Oceanographic Products and Services. Why is there a dash? Don't know, it's just what the government does. But also, it doesn't really matter. If you want to know more, read about it here.

API stands for Application Programming Interface, a general term used a lot in programming. The CO-OPS API is one of many web services that NOAA provides. There's a complete list here.

The main documentation for the web service we will use is here:

There are a lot of options and types of data that can be returned. We've worked out the magic invocation needed for the PyPortal and you can find it in the code later. The only item you will need to worry about is finding your closest tide monitoring station ID. This is how you will set your location in the code.

So let's see how you can figure that out next.

Find Your Tide Station ID

To find you station ID, start by going to this webpage:

Scroll down a bit and you'll see a map. You can simply click on the state of interest. Note that some states are not clickable. That's because they don't have tides :(

Once you've clicked on a state you'll get a familiar map like interface that you can drag and zoom around. Use that to zoom in and find the station marker that seems like it will work best for your location.

Click on the marker and it will bring up information about that marker.

The station ID will be the number shown near the top. This is what you will enter into your code.

Basic Tide Time Info

Here's the URL that gets the high and low times for the current day at the station location. The station ID is the very last thing in the URL, so you can change it for your location if you want.

https://tidesandcurrents.noaa.gov/api/datagetter?date=today&product=predictions&datum=mllw&interval=hilo&format=json&units=metric&time_zone=lst_ldt&station=9447130

1

https://tidesandcurrents.noaa.gov/api/datagetter?date=today&product=predictions&datum=mllw&interval=hilo&format=json&units=metric&time_zone=lst_ldt&station=9447130

If you put that address in your web browser you'll get something that looks like this:

Not a very pretty web page. That's because it's just JSON data. The key predictions is the root for all the data. Then there are several entries that look like:

{"t":"2019-04-15 02:55", "v":"3.401", "type":"H"}

1

{"t":"2019-04-15 02:55", "v":"3.401", "type":"H"}

The t entry gives us the date and time, the v entry gives us the tide level, and the type is H for high tide or L for low tide.

So we have what we need - the time for each of the high and low tides for the given day. You'll see how this is parsed out later in the code.

Tide Levels Throughout the Day

With a slight modification of the URL, we can get predicted water levels in 6 minute increments over the period of the current day. Here's the URL for that:

https://tidesandcurrents.noaa.gov/api/datagetter?date=today&product=predictions&datum=mllw&format=json&units=metric&time_zone=lst_ldt&station=9447130

1

https://tidesandcurrents.noaa.gov/api/datagetter?date=today&product=predictions&datum=mllw&format=json&units=metric&time_zone=lst_ldt&station=9447130

More data like before, but now much more of it!

With data every 6 minutes, that's 10 per hour, or 240 for the entire day. And now each entry contains simply t and v. So for the given time t, we get the predicted water level v. We can use that to make a neat little plot of water level vs. time for the day. This will give us a graphical representation of the tidal activity.