#circuitpython-dev

1 messages · Page 402 of 1

stuck elbow
#

that was fast

crimson ferry
#

the esp32-s2 devkits are $9, with PSRAM. and they have both USB connectors and are skinnier so fit on a breadboard with a row of pins to spare on both sides 🙂

#

just like the s3 devkits

gilded cradle
#

It's probably best to tag CircuitPython Librarians in case I'm not in like yesterday.

solar whale
#

It may only apply to the WROOM module -- does not appear to be an issue with the WROVER

Screen_Shot_2021-12-16_at_11.59.40_AM.png
manic glacierBOT
#
jerryneedell
[adafruit/circuitpython] New comment on issue #5258: esp32s2 \- imagecapture sometimes adds extra byte

FYI -- I updated CP and libraries and retried -- now only the first capture after reboot is valid -- all subsequent images are not able to be decoded by AIO

Press any key to enter the REPL. Use CTRL-D to reload.

Adafruit CircuitPython 7.2.0-alpha.0-146-g190aee551 on 2021-12-16; Saola 1 w/Wrover with ESP32S2
>>> 
soft reboot

Auto-reload is on. Simply save files over USB to run them or enter REPL to disable.
code.py output:
Connecting to WIFI
Connecting to Adafruit IO
Captur...
manic glacierBOT
solar whale
#

@slender iron Yay! Compute modules are back in stock -- ordered one with an I/O board

slender iron
#

nice! at adafruit?

solar whale
#

yup

slender iron
#

I haven't tried it with an emmc module yet

#

(I do have one with emmc already)

solar whale
slender iron
#

yup, I have one of those

solar whale
#

ok -- I got one of those as well -- only 1 left

slender iron
#

great!

#

emmc shouldn't be too hard. the sd card code should be able to handle it. I just haven't tested it

solar whale
#

What could go wrong 😉

manic glacierBOT
solar whale
#

I am having lots of issues with the 4b and zero2w now --- I put some very confusing results in the 4b issue

manic glacierBOT
slender iron
#

I haven't had a chance to look at it yet

#

will look in a bit and see if it triggers anything in my mind

manic glacierBOT
#
scath999
[adafruit/circuitpython] New review comment on pull request #5718: Nrf internal pullup support

Ok, so do away with the existing code beneath this comment?

// This appears to be a board-specific way of supporting internal pullups without 
    // going through all of the changes I've made - @dhalbert can you comment?
    // Not going to change anything here before discussion - will just throw internal_pullup away
    #if CIRCUITPY_I2C_ALLOW_INTERNAL_PULL_UP
    gpio_pullup_en(sda->number);
    gpio_pullup_en(scl->number);
    #endif
#
scath999
[adafruit/circuitpython] New review comment on pull request #5718: Nrf internal pullup support

If move this up to line 124 and remove the #if CIRCUITPY_REQUIRE_I2C_PULLUPS, won't the pin pulls be set down after I set them up by the code following the else you suggest?

I think you mean I should put all of the test code into the else:

  // Set pulls up if internal_pullup is true
    if (internal_pullup){
        nrf_gpio_cfg_input(scl->number, NRF_GPIO_PIN_PULLUP);
        nrf_gpio_cfg_input(sda->number, NRF_GPIO_PIN_PULLUP);
    }
    else (!nrf_gpio_pin_read(sda->nu...
manic glacierBOT
#
scath999
[adafruit/circuitpython] New comment on issue #5689: Support for nrf52833 internal pull ups and pull downs\.

@dhalbert Short answer: yes, and I get a voltage but no device addresses.

The steps required are as follows:

  1. Set the sensor Power Pin to 1: When I measure voltage on the declared I2C pins I get ~3.3V
  2. Create an I2C interface and look for device addresses using the following code
scl = microcontroller.pin.P0_21
sda = microcontroller.pin.P0_20
i2c = busio.I2C(scl, sda, internal_pullups = True)


while not i2c.try_lock():
    pass

try:
    while True:
        print...
manic glacierBOT
#
jftheoret
[adafruit/circuitpython] New comment on issue #5734: Please add collections deque to CircuitPython

I'm using the Raspberry Pi Pico.

I was actually looking in the code to see if I could activate it and
recompile?

Thanks :)

On Thu, Dec 16, 2021 at 3:14 PM Scott Shawcroft @.***>
wrote:

What board are you using? Some of them won't have space for it.

—
Reply to this email directly, view it on GitHub
https://github.com/adafruit/circuitpython/issues/5734#issuecomment-996162096,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAKBDH547IIX4GKCJAPT44DURJCC...

#
tannewt
[adafruit/circuitpython] New comment on issue #5706: broadcom port for raspberrypi4b often corrupts SDCard

start 0 count 0 is the print of the partition for the CIRCUITPY drive. Having it zero will mean CP doesn't have flash. I can add more prints to output the partition table info.

The command timeout on Zero 2 W is what I see too. Somehow, it isn't recovering from an error correctly.

The last Pi4b output is working I think. These are debug statements, not errors. My guess is that it's initializing the filesystem but is slow due to the debug prints. You can see that arg is increasing by o...

#
jftheoret
[adafruit/circuitpython] New comment on issue #5734: Please add collections deque to CircuitPython

Thanks!!

I will definitely try it. It's also going to be an opportunity to learn how
to build it :)

On Thu, Dec 16, 2021 at 3:25 PM Scott Shawcroft @.***>
wrote:

You can set MICROPY_PY_COLLECTIONS_DEQUE to (1) to enable it.

—
Reply to this email directly, view it on GitHub
https://github.com/adafruit/circuitpython/issues/5734#issuecomment-996169022,
or unsubscribe
https://github.com/notifications/unsubscribe-auth/AAKBDHZN2GR3V5E6T3FZQJ3URJDLNANCNFSM5KG5MDDA
.
Tr...

manic glacierBOT
#
PaintYourDragon
[adafruit/circuitpython] New comment on issue #5725: RGBW NeoPixel setter: W only works with 4\-element tuples, is ignored in packed integers

Is this a good place to object to the behavior of setting the white channel when R==G==B?

  • There are at least three different color temperatures of RGBW NeoPixels, and in none of them will R+G+B white accurately correspond to W white in either tint or brightness.
  • Altering the numbers in results in the getter returning a value different than was passed to the setter, and this is certain to result in misunderstandings. While not exactly the same thing, the way the Arduino library handles...
manic glacierBOT
onyx hinge
#

I just noticed this on github today.. It created a log like this for my python library: ```

What's Changed

Full Changelog: https://github.com/jepler/wwvbpy/compare/2.0.0...2.0.1``` -- looks good & super convenient!

unknown.png
manic glacierBOT
idle wharf
#

@slender iron heads up that you can't merge this PR without setting up the required secrets first.
@tulip sleet my guess is you all built adabot before gh cli existed. I was able to do PRs much more simply without using the web api.

https://github.com/adafruit/deep-dive-notes/pull/24

GitHub
Automate Show Notes PRs by askpatrickw · Pull Request #24 · adafrui...

Before merging this please add a secret YOUTUBE_API_KEY to this repo.

This replaces the original script which scraped the data from the YouTube webpage to a process where the latest episode is ret...

manic glacierBOT
meager adder
#

Are there unit tests to verify library changes for backwards compatibility without having a variety of physical hardware devices to test on?
For example, could changes to https://github.com/adafruit/Adafruit_CircuitPython_framebuf formats be tested on a local machine without having access to the different display hardware?

GitHub
GitHub - adafruit/Adafruit_CircuitPython_framebuf: CircuitPython fr...

CircuitPython framebuf module, based on the Python framebuf module - GitHub - adafruit/Adafruit_CircuitPython_framebuf: CircuitPython framebuf module, based on the Python framebuf module

lone axle
onyx hinge
#

hum, main is failing to build ... ulab/numpy/__init__.pyi:51: error: Cannot assign multiple types to name "_float" without an explicit "Type[...]" annotation ulab/numpy/__init__.pyi:54: error: Cannot assign multiple types to name "_bool" without an explicit "Type[...]" annotation I don't think anything was changed in that area lately. however, mypy was updated yesterday and I saw problems elsewhere due to it.

#

yeah verified it's the 0.920 update that caused this problem

manic glacierBOT
#
jepler
[adafruit/circuitpython] Pull request opened: #5737 fix mypy requirement at 0\.910, for compatibility

This should fix the doc build, which currently fails

ulab/numpy/__init__.pyi:51: error: Cannot assign multiple types to name "_float" without an explicit "Type[...]" annotation
ulab/numpy/__init__.pyi:54: error: Cannot assign multiple types to name "_bool" without an explicit "Type[...]" annotation

since mypy 0.920 was released.

After https://github.com/v923z/micropython-ulab/pull/461 or another fix is merged and our submodule ref is updated, we can un-pin mypy. However, as w...

orchid basinBOT
meager adder
lone axle
manic glacierBOT
manic glacierBOT
#
tyomitch
[adafruit/circuitpython] New comment on issue #5707: synthio\.MidiTrack produces single note/beep only

Pinging @tyomitch, who contributed synthio.

On my Meowbit (which has audiopwmio but not a DAC-based audioio) with 7.1.0-beta.3, the example from the docs (with board.BUZZ.play(melody)) works with no problem.

I'd be happy to help, but I don't have a pybadge to test on.

BTW, what sound should it make? Is tempo=640 too high?

http://atari.ruvds.com/level.mp3 (the start-level sound from QBasic Nibbles)

The tempo affects only the interpretation of the MIDI stream; as op...

manic glacierBOT
#
ErlendFj
[adafruit/circuitpython] Issue opened: #5739 Core code crashed hard

CircuitPython version

Adafruit CircuitPython 7.0.0 on 2021-09-20; Adafruit Feather M4 Express with samd51j19

Code/REPL

USB keept dropping out, plugged in battery and got error message Crash into the HardFault_Handler. Maybe this is a brownout related bug/behaviour?

Behavior

cannot run any code

Description

You are in safe mode because:
CircuitPython core code crashed hard. Whoops!
Crash into the HardFault_Handler.
Please file an issue wit...

manic glacierBOT
manic glacierBOT
onyx hinge
#

@tulip sleet which do you prefer -- upgrading ulab [there aren't many changes, just one other bugfix to any()/all() in their main branch] or pinning mypy?

tulip sleet
#

you can do it in 7.1.x and we can merge to main immediately

onyx hinge
#

@tulip sleet yeah I at least assume the build problem is affecting all our branches

tulip sleet
#

we'd have to pin mypy in both branches anyway, so we may as well fix the new issue

#

do you need to wait for v923z or can you proceed yourself

onyx hinge
#

do we care if there's a tag in ulab for the commit we use?

#

if not, I don't think I need to wait

tulip sleet
#

no, a commit is a commit, and when v923z does a release, we can move up if needed

manic glacierBOT
tulip sleet
#

@onyx hinge so does #5740 supersede #5737?

manic glacierBOT
#
jftheoret
[adafruit/circuitpython] New comment on issue #5734: Please add collections deque to CircuitPython

Another quick question for you - I have changed the code, recompiled and
re-flashed the firmware.

The import now works.

When I try to use it:

self.stored_chars = deque([], 4)

I get a ValueError. I am trying to create an empty queue with a maximum of
4 elements.

Can you please give me a hint as how I can investigate this further, as all
I have is a ValueError without any other trace information?

*code.py output:Traceback (most recent call last): File "code.py", line
130, in <modu...

manic glacierBOT
onyx hinge
#

@tulip sleet frustratingly, the CI of #5740 skipped building docs due to path checks, so we won't really know if it's fixed-fixed. doc building passes in my local testing, though. The relevant regular expression looks like it intended to fire on changes in ulab but seemingly it didn't. ```python
doc_pattern = re.compile(
r"^(?:docs|(?:(?:extmod/ulab|ports/\w+/bindings|shared-bindings)\S+.c|conf.py|tools/extract_pyi.py|requirements-doc.txt)$)|(?:-stubs|.(?:md|MD|rst|RST))$"
)

tulip sleet
onyx hinge
#

I'm not sure why the forward slash characters are escaped either

#

I was literally just typing that 🙂

tulip sleet
#

lol

#

does the ?: bind to the whole parenthesized thing or just the ulab part

#

nesting is confusing:

unknown.png
onyx hinge
#

the ?: modifies the whole (...) group

tulip sleet
#

so what's the purpose of the first (?:

onyx hinge
# 
jepler@bert:~/src/circuitpython/tools$ PYTHONPATH=../docs:adabot CHANGED_FILES='["extmod/ulab"]' python3 ci_set_matrix.py
Adding docs/boards to build based on changed files
Building docs: False
::set-output name=build-doc::False
Building boards:
  datum_imu

good news: I think I can test locally

tulip sleet
#

i will not merge without your ok

manic glacierBOT
onyx hinge
#

I can't explain the whole design of the regular expression, but I think I know why it didn't do what was desired in this case... I'm pretty sure extmod/ulab has to be followed by \S+\.c i.e., a file with a C extension. But it won't be, because it's a submodule.

# 
 PYTHONPATH=../docs:adabot CHANGED_FILES='["extmod/ulab/blorp.c"]' python3 ci_set_matrix.py | head -2
Adding docs/boards to build based on changed files
Building docs: True

If a filename was received, it would have built the docs

tulip sleet
#

if you are going to fix it, I would say remove the escaped forward slashes, at least.

manic glacierBOT
tulip sleet
#

can be tested in an online re tester?

onyx hinge
#

do you have one you trust?

#

OK updated one more time with the escaping of forward slashes removed..

tulip sleet
onyx hinge
#

I did a couple of local invocations of the script and the results look plausible still

manic glacierBOT
lunar gull
#

Hey there. Is there any Adafruit code that uses exponential backoffs for CircuitPython? I was mainly looking at reusing, or coding something new, for devices like the MagTag when they get network errors to not loop and retry with time.sleep(), but with exponential backoffs and deep sleep.

#

I am guessing I can store the current backoff number in sleep_memory

manic glacierBOT
#
jftheoret
[adafruit/circuitpython] New comment on issue #5734: Please add collections deque to CircuitPython

Excellent, thank you!

On Fri, Dec 17, 2021 at 10:30 AM Dan Halbert @.***>
wrote:

Try deque((), 4). The MicroPython code does not support initialization,
and only allows an empty tuple, not an empty list.

—
Reply to this email directly, view it on GitHub
https://github.com/adafruit/circuitpython/issues/5734#issuecomment-996813270,
or unsubscribe
https://github.com/notifications/unsubscribe-auth/AAKBDH7LYOZSFDBQU5DRATDURNJTDANCNFSM5KG5MDDA
.
Triage notifications ...

orchid basinBOT
manic glacierBOT
manic glacierBOT
manic glacierBOT
tulip sleet
manic glacierBOT
#
scath999
[adafruit/circuitpython] New review comment on pull request #5718: Nrf internal pullup support

Not to be thick-headed, but confused about what you mean by "define that constant directly".

To clarify, I assume you mean something like

mp_obj_t common_hal_board_create_i2c(void) {
    // All callers have either already verified this or come so early that it can't be otherwise.
    assert(i2c_singleton == NULL || common_hal_busio_i2c_deinited(i2c_singleton));
    busio_i2c_obj_t *self = &i2c_obj;
    self->base.type = &busio_i2c_type;

    // ---> Check if board supports i...
#
dhalbert
[adafruit/circuitpython] New review comment on pull request #5718: Nrf internal pullup support

My question is how can I know at runtime if the board has support for internal pullups?

With the addition of the internal_pullups argument to busio.I2C(), any board can now support internal pullups. You just have to ask for them by setting internal_pullups to be true in the constructor.

The question now for a particular board is whether board.I2C() asks for internal pullups by default. So I'm sayin to remove the older compile time constants, and add a `CIRCUITPY_BOARD_I2C_...

turbid radish
#

?serverinfo

#

serverinfo

#

?serverinfo

digital shoreBOT
#
adafruit
Owner

adafruit#3230

Category Channels

8

Text Channels

61

Voice Channels

6

Members
Roles

36

ID: 327254708534116352 | Server Created ·
onyx hinge
#

maybe dyno's under the weather

manic glacierBOT
#
rsbohn
[adafruit/circuitpython] New comment on issue #5707: synthio\.MidiTrack produces single note/beep only

I expected this to play two notes, then quit.
Instead it plays a warbling sound that only ends
when I call dac.stop().

# synthio test
# Raspbery Pi Pico RP2040
# Adafruit CircuitPython 7.1.0-beta.3 on 2021-12-13;
import time
import audiocore
import audiopwmio
import board
import synthio

dac = audiopwmio.PWMAudioOut(board.GP26)

G2 = 43
C3 = 60
midi = bytearray([
    0, 0x90, C3, 0,
    25, 0x80, C3, 0,
    50, 0x90, G2, 0,
    75, 0x80, G2, 0
])
audio = s...
onyx hinge
#

@turbid radish 32,443 users via the 'server settings'

manic glacierBOT
#
scath999
[adafruit/circuitpython] New review comment on pull request #5718: Nrf internal pullup support

Ah, I think I understand where I was confused; you're trusting the developer to specify (and therefore, know) whether the board has support for internal pullups - I was going the other way, requiring CPY developers to define this in .mk files.

If I understand, you're saying "assume the board doesn't, but allow the developer to tell us it does". If that's true, then I did need to remember internal_pullups, and can get this from the self pointer:

busio_i2c_obj_t *self = &i2c_o...
manic glacierBOT
#
BlueBackbite
[adafruit/circuitpython] Issue opened: #5745 Adafruit ESP32\-S2 Feather hard crashes with wifi\.radio\.ping

CircuitPython version

Adafruit CircuitPython 7.1.0-beta.3 on 2021-12-13; Adafruit Feather ESP32S2 with ESP32S2
Board ID:adafruit_feather_esp32s2

Code/REPL

# SPDX-FileCopyrightText: 2019 ladyada for Adafruit Industries
# SPDX-FileContributor: Modified by Reppad
#
# SPDX-License-Identifier: MIT

import os
import board
import neopixel
import ipaddress
import wifi
import rtc
import time
import random
import socketpool

# Get wifi details and mor...
manic glacierBOT
onyx hinge
#

^^ I'm open to opinions about the choice of whether to enable rgbmatrix for s3/c3.

manic glacierBOT
#
dhalbert
[adafruit/circuitpython] New review comment on pull request #5718: Nrf internal pullup support

Did you post and then delete (retract) a comment there? To answer that, nearly all chips have always had the capability of supporting internal pullups, we just didn't expose that. Now we are. It's a choice of the user about whether they are going to connect some I2C devices without pullups. That could happen on any board.

And, there might be some boards, with board-mounted I2C devices, without external pullups. In that case, we might want board.I2C() to use internal pullups by default....

#
scath999
[adafruit/circuitpython] New review comment on pull request #5718: Nrf internal pullup support

Ok. I'm confused about some of this, mainly because I'm trying to relate now to set your suggested CIRCUITPY_BOARD_I2C_USE_INTERNAL_PULLUPS at runtime, but perhaps it's better to put a pin in this?

In the interim, I can simply pass true to that part of the method and move on to the bigger issue: i2c.scan() is not returning any devices, even after I set the pulls up and read a proper voltage (3.3V) on them at runtime.

#
dunkmann00
[adafruit/circuitpython] Pull request opened: #5746 Explicitly cast float to mp\_int\_t

Not sure why this is necessary, but it prevents an off-by-one error in some (rare?) circumstances.

Fixes the bug found in #5725:

>>> import board, neopixel
>>> pixels = neopixel.NeoPixel(board.D5, 8, brightness=1.0, auto_write=False, pixel_order=neopixel.GRBW)
>>> pixels[2] = 0x400000FF
>>> pixels[2]
(0, 0, 255, 0)

This doesn't close #5725, since that it is also about whether or not the behavior of integer values with RGBW neopixels is appropriate.

stuck elbow
#

Is anybody planning further work on the paralleldisplay module? Things like allowing arbitrary pins, or 16-bit mode?

manic glacierBOT
#
dhalbert
[adafruit/circuitpython] New review comment on pull request #5718: Nrf internal pullup support

Ok. I'm confused about some of this, mainly because I'm trying to relate now to set your suggested CIRCUITPY_BOARD_I2C_USE_INTERNAL_PULLUPS at runtime, but perhaps it's better to put a pin in this?

I'm suggesting that the compile-time constant need not exist at all. The user code will decide whether or not to use internal pullups.

In the interim, I can simply pass true to that part of the method and move on to the bigger issue: i2c.scan() is not returning any devices, even after I ...

tulip sleet
#

@daring pumice can we have an audio chat re the internal pullups PR? Would save some typing and we can get on the same page, and I can hear about your testing difficulties.

daring pumice
tulip sleet
#

do you have a mic? We can talk in the Amelia Earhart channel. Just click on it.

daring pumice
#

I do. Headed there now

manic glacierBOT
#
anecdata
[adafruit/circuitpython] New comment on issue #5745: Adafruit ESP32\-S2 Feather hard crashes with wifi\.radio\.ping

I've reproduced your result.

But this ran a few hundred loops without issue:

import supervisor
import time
import random
import wifi
import socketpool
import ipaddress
from secrets import secrets

SERVERS = ("1.1.1.1", "1.0.0.1",
           "1.1.1.2", "1.0.0.2",
           "8.8.8.8", "8.8.4.4",
           "9.9.9.9", "149.112.112.112",
           "208.67.220.220", "208.67.222.222",
           "208.67.220.123", "208.67.222.123")

print("Connecting...")
wifi.radio.conn...
quasi veldt
#

@tulip sleet I'm trying to select a microcontroller for scaling across several high schools and colleges that I work with. My curriculum is definitely not the typical approach to intro material in that I am trying to teach async and event triggering.
We currently have a SAMD21 board and a bunch of micro:bit v2 boards (we liked them because makecode also works and has libraries unlike with other ARM boards), but looking at the list it seems like some of the libraries are not enabled on the micro:bit v2. Could we quickly hop into an audio room to chat about what can be enabled or what other alternatives I should consider?

tulip sleet
#

micro:bit v2 is only experimentally supported by CPy, and doesn't have the normal USB workflow. SAMD21 has too little RAM and flash to support asyncio and numerous other things

#

Circuit Playground Bluefruit I think has experimental MakeCode support, but I'm not sure.

#

It has a good amount of RAM and flash

#

if MakeCode is a strong requirement

quasi veldt
tulip sleet
#

microbit:v2 (and v1) only have serial connections over USB. They cannot present a CIRCUITPY drive, like all the other CircuitPython boards.

quasi veldt
manic glacierBOT
tulip sleet
#

We are developing a BLE-based workflow to allow editing programs and talking the REPL over BLE instead of over USB. That would be the way to talk to the microbit:v2. However, it's early. See the PyLeap and Glider apps, which are still in beta.

#

Advantage of BLE-based workflow is that you don't need a USB host: you can use tablets and phones. Disadvantage is that there's more setup, less reliable connections, etc.

quasi veldt
tulip sleet
#

(yesterday!)

quasi veldt
#

the other MCUs I had initially shortlisted were rPi Pico (seems to be the cheapest RP2040 board but no wifi or BLE), ESP32 (I believe this WROOM version is what I have to test https://www.amazon.com/KeeYees-Development-Bluetooth-Microcontroller-ESP-WROOM-32/dp/B07QCP2451/ref=sr_1_4?keywords=esp32&qid=1639785020&sr=8-4), or Particle Argon (but seems way too expensive).

KeeYees Development Board 2.4 GHz Dual Core WLAN WiFi + Bluetooth 2...

KeeYees Development Board 2.4 GHz Dual Core WLAN WiFi + Bluetooth 2-in-1 Microcontroller ESP-WROOM-32 Chip CP2102 for ESP32 for Arduino (2PCS)

tulip sleet
#

Pi Pico is cheap, could use a reset button; there are many other RP2040 boards availble. ESP32 (plain ESP32) does not have CircuitPython support, again because no native full USB. We support ESP32 only as a wifi or BLE coprocessor. ESP32-S2 we support with a number of boards. We will be supporting ESP32-S3 (very new) and probably ESP32-C3 (only with BLE; no native USB)

quasi veldt
#

My original idea was get a cheap but reliable MCU + rPi 0W for every student.
microbit v2 came into the list because of MakeCode and onboard debugger

tulip sleet
#

The Circuit Playground Express is our workhorse educational board, with the Circuit Playground Bluefruit being very similar but more powerful

#

The Feather boards work well with rechargeable batteries, but if you are doing elementary or junior high, you may not want to have bare LiPo batteries around (too much of a hazard if punctured, etc.

#

But you can use them with USB also. Many add-ons for the Feather boards (FeatherWings)

quasi veldt
#

lowest is grade 7, highest is early university/community college. typical is grades 9-12

tulip sleet
#

Advantage of the CIrcuit Playground boards is onboard NeoPixels, sensors, mic, speaker (so no wiring needed) and easy connection via alligator clips for external stuff

quasi veldt
manic glacierBOT
quasi veldt
quasi veldt
tulip sleet
#

But the "latest" is very similar.

#

Under $15:

quasi veldt
tulip sleet
#

ItsyBitsy M4 ($15), ItsyBitsy RP2040 ($10), QT Py RP2040 ($10), Feather RP2040 ($12)

#

I am not including the SAMD21 boards

tulip sleet
#

Feather ESP32-S2 is $17.50 (< $15 in qty), there are other boards similarly priced, I think

crimson ferry
manic glacierBOT
#
BlueBackbite
[adafruit/circuitpython] New comment on issue #5745: Adafruit ESP32\-S2 Feather hard crashes with wifi\.radio\.ping

It hard faults with this code very fast

import ipaddress
import wifi

# Get wifi details and more from a secrets.py file
try:
    from secrets import secrets
except ImportError:
    print("WiFi secrets are kept in secrets.py, please add them there!")
    raise

modemIP = ipaddress.ip_address("192.168.100.1")
# count the number of pings
ping_counter = 0

print("Connect wifi")
wifi.radio.connect(secrets['ssid'],secrets['password'])

def check_connectivity():
    ...
quasi veldt
manic glacierBOT
#
jepler
[adafruit/circuitpython] Pull request review submitted: #5746 Explicitly cast float to mp\_int\_t

Thank you for finding this! It is "obvious in retrospect", for at least some definitions of obvious, but I had looked directly at this code just recently and still not seen it for myself.

I'm approving this PR, but with the caveat that we may wish to have the bugfix in the 7.1.x branch. @dhalbert please advise on that point.

Here's a rough expression based on my understanding of C: When using the conditional operator such as cond ? x1 : x2, the type of the whole expression depend...

onyx hinge
# stuck elbow Is anybody planning further work on the `paralleldisplay` module? Things like al...

I had done a tiny bit of investigating of adding 16 bit mode but ended up deciding it needed more time than I wanted to devote to it. My interest was in the screen of the esp32-s2 "hmi" devkit... where I identified trouble was, sometimes data needs to go out in 16 bit mode [display data] and other times it needs to go out in 8 bit data even if there are many bytes [register setting, mostly] and I wasn't sure how to accomplish it. So I set it aside.

lone axle
manic glacierBOT
manic glacierBOT
manic glacierBOT
meager adder
#

Is it possible to run circuitpython libraries directly on a local machine cpython? For example, something like adafruit_framebuf doesn't seem to have any hardware dependencies

#

Some libraries have test suites that I assume gets run on a build machine so I think it's probably possible to run some things locally?

solar whale
stuck elbow
#

even on a desktop, with a gpio adapter

manic glacierBOT
spiral elk
manic glacierBOT
lone axle
spiral elk
#

I have an assortment of various spi and i2c flash here. Will check on a Xicor X2402 later but that one might be iffy.

proven garnet
lone axle
#

The driver is currently I2C only. The one it's based on did contain SPI support, but I removed it because I didn't find any SPI based EEPROM breakouts searching adafruit.

spiral elk
#

x2402 is i2c, but its really old and I never really got them to work with any generic libraries before.

proven garnet
#

I'll give this a try now

proven garnet
#

Thanks I love it ❤️

#

Works great with Feather M0 Express and 24LC32 Breakout board at 7.1.0-beta.2

#

Let me know if you want any help like typing or testing. Also, it looks like you're using the __getitem__ and __setitem__ from the FRAM library so it should already support slices now 😄

#

Actually, I think it's slightly bugged and grabbing too large of a slice, let me fix that in the other library, I'll tag you so that change can carry over panic

#

I can also look into the issue you posted there

#

I just submitted a PR to your repo to fix it, I'll do the same for the other haha

lone axle
meager adder
# solar whale What system are you trying to run it on? There is an interface called "Bilinka" ...

On a windows desktop machine. I'm getting this on blinka: NotImplementedError: Adafruit-PlatformDetect version 3.18.0 was unable to identify the board and/or microcontroller running the Windows platform.

I'm currently only running libraries like adafruit_framebuf which works entirely within a bytearray() so there's no need to access pins or any other devices.
There is an agnonstic folder in blinka but I'm not sure how to use it. Would I have to go through the platformdetect guide to create a custom windows chip.py, board.py, etc. to create a fake placeholder windows board, or is there an already existing empty board I can reuse?

meager adder
# stuck elbow even on a desktop, with a gpio adapter

Thank you! I just learned that boards like the MPC2221 (https://www.adafruit.com/product/4471) exist for adding GPIO to desktop. This means it would be possible to start using the debugger during development instead of relying only on print() for tracing issues

Adafruit MCP2221A Breakout - General Purpose USB to GPIO ADC I2C

Wouldn't it be cool to drive a tiny OLED display, read a color sensor, or even just flash some LEDs directly from your computer?  Sure you can program an Arduino or ...

proven garnet
daring pumice
#

Anyone have any idea why i2c.scan() would cause a device to hang?

blissful pollen
meager adder
proven garnet
meager adder
#

Oh ok, no worries

proven garnet
manic glacierBOT
ornate breach
#

Okay, dumb question here..

#

I’m trying to build a new Espressif ESP32-S2 board and I’m getting the following error:

/bin/sh: xtensa-esp32s2-elf-gcc: command not found```
#

Is there something that changed that would cause this?

onyx hinge
#

@ornate breach you need to run the install.sh script within esp-idf first, which will download a bunch of stuff from the internet and install it in the secret meadows and valleys of your computer .. then, each time you open a terminal to build for espressif you need to . esp-idf/export.sh. Omitting either of those will get you an error like the above..

#

top two causes I can guess at from here. just because it's a new board shouldn't have much to do with it

ornate breach
#

ah... so even if I had it on my computer already i would need to do it in the esp-idf folder within circuitpython

onyx hinge
#

yep, that is how I have done it at any rate

ornate breach
#

odd.. okay, i have to clone down the lib again because macos made copies of things for some reason

#

what was the new script for pulling in the modules?

#

so i don't have to download all 20GB again

edgy drum
#

curious if anyone has any insight into this circup issue or if there's a way to have it give me more verbose error information

unknown.png
#

So it appears this was already fixed in a newer version of circup, the issue was a bad url but circup 1.0.1 works without issue:

There was a problem downloading the bundle. Please try again in a moment.
12/18/2021 19:18:03 ERROR: 404 Client Error: Not Found for url: https://github.com/adafruit/Adafruit_CircuitPython_Bundle/releases/download/20211217/adafruit-circuitpython-bundle-6.x-mpy-20211217.zip
Traceback (most recent call last):
  File "D:\Disasters in the Making\circuitpy\venv\lib\site-packages\circup\__init__.py", line 470, in ensure_latest_bundle
    get_bundle(bundle, tag)
  File "D:\Disasters in the Making\circuitpy\venv\lib\site-packages\circup\__init__.py", line 685, in get_bundle
    r.raise_for_status()
  File "D:\Disasters in the Making\circuitpy\venv\lib\site-packages\requests\models.py", line 953, in raise_for_status
    raise HTTPError(http_error_msg, response=self)
requests.exceptions.HTTPError: 404 Client Error: Not Found for url: https://github.com/adafruit/Adafruit_CircuitPython_Bundle/releases/download/20211217/adafruit-circuitpython-bundle-6.x-mpy-20211217.zip
onyx hinge
#

@edgy drum yes, recently we stopped publishing the "6.x" bundle. we don't have a way for circup to know which bundle versions are actively published, so for now we have to manually change circup when this happens (which is rarely)... hopefully someday we'll publish a json file or something that will let circup learn automatically about bundle versions. -- https://github.com/adafruit/Adafruit_CircuitPython_Bundle/issues/360

GitHub
Provide an official way for tools to enumerate supported bundle ver...

circup and possibly the vs code extension both encountered problems when we dropped the 6.x bundles. circup in particular hard coded the list of versions. if we provide an official way to enumerate...

edgy drum
onyx hinge
# 
PLATFORMS = {"py": "py", "6mpy": "6.x-mpy", "7mpy": "7.x-mpy"}
PLATFORMS = {"py": "py", "7mpy": "7.x-mpy"}```
edgy drum
#

I mean the commit that stopped the publishing of 6.x to the bundle repo releases

#

that appears to be the circup patch?

ornate breach
#

urgh... i'm really tired of macos not cloning down everything in a repo

onyx hinge
#

@edgy drum oh I misunderstood the question, apologies & glad you found the right commit

edgy drum
#

💪 I commented on your issue, might take a look at this issue next week to see if I can PR this sort of functionality in place without adding any new release steps

spiral elk
#

It had a date code of 1989

#

Does seem to work on most I2C EE devices, so might want to change the constructor to take a size as a keyword argument and not hardcode it in.

ornate breach
#

I may have found an ESP32-S2 bug with running neopixel animation libraries on an ESP32-S2 on battery.

#

it might not be battery related though

#

when running an animation loop of one animation, it gets a memory allocation error:

Traceback (most recent call last):
  File "code.py", line 44, in <module>
  File "adafruit_led_animation/sequence.py", line 112, in __init__
  File "adafruit_led_animation/animation/__init__.py", line 210, in add_cycle_complete_receiver
MemoryError: memory allocation failed, allocating 32768 bytes```
#

it typically runs few a few seconds, maybe 10-15 before erroring.

# SPDX-FileCopyrightText: 2021 Kattni Rembor for Adafruit Industries
# SPDX-License-Identifier: MIT

"""
This example uses AnimationsSequence to display multiple animations in sequence, at a five second
interval.

For NeoPixel FeatherWing. Update pixel_pin and pixel_num to match your wiring if using
a different form of NeoPixels.
"""
import board
import neopixel
from digitalio import DigitalInOut, Direction
from adafruit_led_animation.animation.blink import Blink
from adafruit_led_animation.animation.comet import Comet
from adafruit_led_animation.animation.chase import Chase
from adafruit_led_animation.sequence import AnimationSequence
from adafruit_led_animation.color import PURPLE, AMBER, JADE


# Update to match the pin connected to your NeoPixels
pixel_pin = board.NP_STRIP
# Update to match the number of NeoPixels you have connected
pixel_num = 60

pixels = neopixel.NeoPixel(pixel_pin, pixel_num, brightness=0.1, auto_write=False)

blink = Blink(pixels, speed=0.5, color=JADE)
comet = Comet(pixels, speed=0.01, color=PURPLE, tail_length=10, bounce=True)
chase = Chase(pixels, speed=0.1, size=3, spacing=6, color=AMBER)

while True:
    animations = AnimationSequence(chase, advance_interval=1, auto_clear=True)
    animations.animate()```
#

i guess it's happening because the while loop keeps building the animation object

lone axle
ornate breach
#

i was trying to use this to setup where i could cycle through animations on button presses

#

which is another issue i'm having

#

QSTR defined buttons always show up as in use.

lone axle
#

Ah, I'm not sure about the QSTR button issue. For cycling through animations though you can create multiple Animations or AnimationSequences before the loop and then inside the loop check the button states and call animate() on whichever one makes sense based on what has been pressed.

ornate breach
#

seems reasonable

onyx hinge
ornate breach
onyx hinge
#

what's the full code? Why do you think "QSTR" is relevant to what you're seeing?

ornate breach
#

Well, I can’t use buttons I’ve defined in the QSTR table

#

I tried a modified funhouse build since that one is supposed to work

#

Even if I try to call from microcontroller.pin.GPIO31 it still shows in use

onyx hinge
#

modules/wroom.c: common_hal_never_reset_pin(&pin_GPIO31); a bunch of boards mark this pin as reserved. The associated comment indicates it's used by the SPI flash.

ornate breach
#

Ah

#

And pin GPIO26?

onyx hinge
#

wrover excludes that one

ornate breach
#

That’s terribly odd because it’s broken out on the module

#

IO26 is used on a lot of boards though

onyx hinge
unique beacon
#

Hey! I am not sure if I am supposed to ask this question here or in the 'circuitpython-help' thread. But since it is more related towards building and extending I chose this one. If I am wrong about that just say so and I post it again in the other channel:

I am currently trying to link a C library while building CircuitPython (The BSEC). While doing so there appear a number of errors about undefined math functions (Basic stuff like fabs). This shouldn't be a issue if I am not mistaken. I tried using (Or rather linking) the BSEC with gcc-arm-none-eabi-10-2020-q4-major outside of CircuitPython with the math library linked (-lm) and it didn't complain. After some digging it seems like CircuitPython uses its own libm version to safe some space. So I guessed that it is probably connected to that.

I tried disabling it by editing the INTERNAL_LIBM = 0. After trying to build a port (Without changing/linking/adding anything, a clean clone from Github where I didn't change anything besides the Value of INTERNAL_LIBM) for the Pico, I got a lot of new errors though - now certain functions like nearbyintf() aren't defined anymore. Other math functions seem to work as intended though, so I guess it is linking some kind of math library. Is this intended? Am I doing something wrong?

This just leaves me at a point where I am quite confused. Is there a way to link the toolchain math library while building CircuitPython?

I'd love any kind of help or maybe a pointer to what I could do next.

ornate breach
#

Guess I’ll change the pin usage on the next iteration

onyx hinge
#

@unique beacon I doubt anybody is testing/building circuitpython with this flag, so it's no big surprise if it's broken. I see at least one 'obvious' problem that relates to trying to set INTERNAL_LIBM=0...

unique beacon
#

Ahh, that is what I assumed. It makes sense since I guess it isn't relevant for most people.

onyx hinge
#

but now I'm 3 levels deep and there's still weird problems. It may not be simple to make it work out.

unique beacon
#

Oh, and thank you very much for the information! 🙂

onyx hinge
#

problem 1: ifndef INTERNAL_LIBM is the incorrect check. problem 2: $(LIBS) was not actually used. problem 3: libm depends on some symbols from libc (errno & one other?) 4: something about copysign that I "corrected" by removing a bit about it from PICO_LDFLAGS

#

the first 3 changes would probably be fine to take into CircuitPython, but not sure about the 'wrap' one.

stuck elbow
#

@ornate breach are you creating your digitalinout objects inside the loop as well?

ornate breach
#

No

onyx hinge
unique beacon
#

@onyx hinge Thank you for taking your time to try it out. I'll check both patches out.

onyx hinge
#

@unique beacon you're welcome. I don't have any code to actually test the double precision working so that's up to you. If you do get it working and there's some portion of it we should incorporate to make your project work better, open up a PR once the code is tested!

unique beacon
#

I'll make sure to do so!

subtle trail
#

<@&681912791711350805> I'm getting a "crashed hard" error when trying to add a library to my ESP32 board. Anyone up for helping me debug?

"CircuitPython core code crashed hard. Whoops!
Crash into the HardFault_Handler."

crimson ferry
daring pumice
#

If anyone else has any ideas, I'm still trying to figure out why i2c.scan() would cause a device to hang. If anyone has experienced this before, would appreciate any guidance 🙂

#

@blissful pollen To answer your question, I can acquire a lock - I've checked to make sure that happens.

blissful pollen
#

can you connect to the device without the scan?

daring pumice
#

How do you mean? If I try to do anything with the i2c interface before acquiring the lock, I get an error asking me to get a lock 😉

blissful pollen
#

I guess to take a step back what device are you trying to connect to?

daring pumice
#

I have two devices using an i2c interface - trying to talk to either of them (one at a time of course). I haven't tried to talk to both in the same script, only ever one at a time.

#

The first is a TI capacitance sensor, the other is a custom-built sensor

#

When I try to create an interface for the TI and run a scan, I get back no device addresses - i2c.scan() doesn't hang, but it returns an empty array.

#

When I try to talk to the custom sensor, the scan hangs.

#

Honestly, being able to talk to either would be a victory.

#

Put a different way, where is the code for i2c.scan CPY? Happy to trace through myself, but it's difficult for me to find out where to start looking.

blissful pollen
#

in the core the code starts in shared-bindings/busio/I2C.c

daring pumice
#

Thanks @blissful pollen. I've seen this before - these appear to be lower level constituent functions for initialization, locking, reading and writing, but not for scanning.

I realize these methods are likely used within i2c.scan(), but I have no idea how and in what order/combination.

Is there somewhere I can find that code?

blissful pollen
daring pumice
daring pumice
#

Is this a valid strategy to output some debugging info for that method?

 for (int addr = 0x08; addr < 0x78; ++addr) {
        bool success = common_hal_busio_i2c_probe(self, addr);
        if (success) {
            mp_obj_list_append(list,  
                               MP_OBJ_NEW_SMALL_INT(addr));
        }
        else {
            mp_printf(print, "probe failed for address :%d\n", 
                      addr)
        }
    }

?

#

ie. will I get that msg in the REPL?

blissful pollen
#

I usually use mp_printf(&mp_plat_print, "stuff");

daring pumice
#

ok, can I pass values in? like addr using standard printf syntax?

blissful pollen
#

yeah you can

daring pumice
#

Awesome. Cheers 🙂

daring pumice
#

Ok. So running i2c.scan seems to disconnect the device.

When I dig in, there is an empty while loop within common_hal_busio_i2c_probe that runs forever (ports/nrf/common-hal/busio/I2C.c)

# 
;
}```
#

Explains why this loops without ever returning.

#

So NRF_TWIM_EVENT_STOPPED is never getting generated.

#

Or is being generated incorrectly (e.g. with an incorrect offset)

manic glacierBOT
#
FoamyGuy
[adafruit/circuitpython] Pull request opened: #5749 Tilegrid size properties

This change introduces properties to access height, width, as well as tile_height, and tile_width for TileGrid objects. Can be accessed like this:

print("width: {}".format(tile_grid.width))
print("height: {}".format(tile_grid.height))
print("tile_height: {}".format(tile_grid.tile_height))
print("tile_width: {}".format(tile_grid.tile_width))

These values can make it easier to place a TileGrid in some relative position i.e. center it in the center of the display.

#
FoamyGuy
[adafruit/circuitpython] New comment on pull request #5749: Tilegrid size properties

This passed pre-commit locally:

circuitpython_adafruit on  tilegrid_size_props [$?] via 🐍 v3.8.10 
❯ pre-commit run --all-files
Check Yaml...............................................................Passed
Fix End of Files.........................................................Passed
Trim Trailing Whitespace.................................................Passed
Translations.............................................................Passed
Formatting.............................
#
FoamyGuy
[adafruit/circuitpython] New comment on pull request #5749: Tilegrid size properties

This also seems to have failed several of the builds in CI. But a local build of one of the same boards suceeded:

circuitpython_adafruit/ports/raspberrypi on  tilegrid_size_props [$?] via 🐍 v3.8.10 
❯ make BOARD=adafruit_itsybitsy_rp2040
Use make V=1, make V=2 or set BUILD_VERBOSE similarly in your environment to increase build verbosity.
mkdir -p build-adafruit_itsybitsy_rp2040/genhdr
GEN build-adafruit_itsybitsy_rp2040/genhdr/moduledefs.h
QSTR updated
   text    data     bss   ...
manic glacierBOT
orchid basinBOT
tulip sleet
onyx hinge
#

@tulip sleet hum I don't see an actions workflow to make a release from cascadetoml....?

tulip sleet
#

good point, the one on pypi is very old

onyx hinge
#

do you think scott manually runs commands to do it? that's not ideal.

tulip sleet
#

i agree

onyx hinge
#

I hope scott is in today

tulip sleet
#

btw re python versions, I am wondering if we should specify "3" instead of "3.whatever". Btw, should also be in quotes, because otherwise 3.10 == 3.1

#

yes; I think he is off until tomw

#

I have upgraded to 3.10 on a couple of things without issue

#

bundle and asyncio

onyx hinge
#

nice! I picked 3.9 since it matches what's on my debian bullseye machines, which means it's not crazy new (but probably not on ubuntu lts yet)

#

OK I popped in a couple of issues about releasing cascadetoml. in the meantime, do we need to do something to get circuitpython building again? we could put 'tomlkit==0.7.2 cascadetoml==0.3.1' in our requirements_dev.txt, for example.

#

or do we want to potentially just not be able to build all day until cascadetoml can get released

#

one of my personal projects has a matrix that covers each 3.x from 3.6 to 3.10 but I'm not sure doing that actually has any big advantage. well, it keeps me from introducing incompatibilities with old versions though I'm unsure why I care.

tulip sleet
#

can we just manually release cascadetoml, or add a release.yml now?

onyx hinge
#

I made a cascadetoml release in the github sense

tulip sleet
#

i have the credentials for a pypi release

#

i think

#

oh, I see, it's only owned by scott

onyx hinge
#

only "tannewt" is maintainer for cascadetoml. Looks like "adafruit-travis" would need to be added.

manic glacierBOT
onyx hinge
#

then we'll need to merge that up into main, there's one other pending change as well it looks like

#

oh that was a backport so it doesn't make a difference

tulip sleet
#

GitHub makes it really easy to do the merge up into main. I will click on that as soon as the pinning PR can be merged.

onyx hinge
#

yup, appreciate it!

tulip sleet
#

@slender iron I see you as Idle, will you be around today? See above

slender iron
#

@tulip sleet ya, I'll be around after I feed cats and myself until 3pm when I have a dentist appt and errands

#

ok, I added adafruit-travis as an owner

onyx hinge
#

@lone axle so if you are local to your git repo and you pre-commit run --files path/to/one/of/those/files.c it doesn't complain?

lone axle
#

I have always done pre-commit run -a so far. I will check that way pointing at the files

#

calling it with the --files flag does fail for the formatting (and make the required changes on disk). Experimenting a bit I rolled back the changes to try again. There is definitely difference for me with --all-files vs. --files [path]:

circuitpython_adafruit on  tilegrid_size_props [$!?] via 🐍 v3.8.10 
❯ pre-commit run --all-files
Check Yaml...............................................................Passed
Fix End of Files.........................................................Passed
Trim Trailing Whitespace.................................................Passed
Translations.............................................................Passed
Formatting...............................................................Passed

circuitpython_adafruit on  tilegrid_size_props [$?] via 🐍 v3.8.10 took 2s 
❯ pre-commit run --files shared-bindings/displayio/TileGrid.c 
Check Yaml...........................................(no files to check)Skipped
Fix End of Files.........................................................Passed
Trim Trailing Whitespace.................................................Passed
Translations.............................................................Passed
Formatting...............................................................Failed
- hook id: formatting
- files were modified by this hook

I wonder if something else locally could be causing --all-files not to be able to find everything?

#

This answer posted by the author of pre-commit suggests it may be smart enough to "only sending the changed files to the underlying tools". Maybe it's not finding it in this case because I made the commit with these actual changes a while ago so technically right now there are no changes in my local copy (or no uncommitted changes at least). https://stackoverflow.com/a/64772737/507810. I committed it previously without running pre-commit just to get it saved somewhere while I went to a different branch to test something unrelated.

#

Though the --help text makes no reference to only targeting changed files with the --all-files flag:

--all-files, -a       Run on all the files in the repo.
onyx hinge
#

that's interesting/weird!

#

I didn't suggest --files because I thought it would be different, but because I thought it would be quicker!

#

@lone axle some time ago you made a lib for storing structured data in nvm .. does it work with deep sleep memory? would it work with i2c fram/eeprom? [prompted by looking at your in the weeds topic just now]

lone axle
#

Now its failing in the CI due to changes in tests/extmod/utimeq1.py and tools/pydfu.py but these files were not actually touched as part of the functionality in my PR. I noticed that the formatter modified them when I did originally run pre-commit run -a when I made the PR (at the same time that it didn't fail for the files that I did actually change). It seems like it didn't format them in a way that CI likes. I will try to revert those.

lone axle
#

it is essentially using the first few bytes to store a length of the total data and then after those first few bytes writing the results of msgpack.pack() on the input object.

onyx hinge
#

neat!

#

puzzled by what you're seeing with pre-commit 😕

daring pumice
#

Afternoon everyone 🙂

slender iron
stuck elbow
#

thanks, I will give it a try

lone sandalBOT
daring pumice
#

How is the nrfx library integrated into CPY? Is it just dropped into the project, or has it been heavily edited to adapt it for CPY's use?

slender iron
#

I think it's essentially as-is.

#

If we want to do something differently, we just won't use it.

daring pumice
solar whale
#

Missing meeting today. Have a great week!

slender iron
#

you too @solar whale

onyx hinge
#

<@&356864093652516868> Hello! We'll be holding the weekly meeting starting in just a few minutes. Today's notes document is: https://docs.google.com/document/d/1L-I0fOE6yrnv7OV_jQZZiFGtgWmc86GiGP6_ecpnGQs/edit# -- if you will speak, or have notes to read, please drop them in there. otherwise, if you're just listening, we're happy to have you.

Google Docs
CircuitPython Weekly for December 20th, 2021

CircuitPython Weekly for December 20th, 2021 Welcome to the CircuitPython Weekly meeting notes! Feel free to add your Hug Reports and Status Updates early. During the meeting, we go through them as a round robin sorted by username. If you can’t make the meeting and would still like to particip...

slender iron
#

@daring pumice you'll need to update the submodule

tulip sleet
manic glacierBOT
#
vrtisworks
[adafruit/circuitpython] New comment on issue #4373: Intermittent "Connection Error: Unknown Failure 205" error on Adafruit MagTag when connecting to WiFi

Thought I would pass along my experience in case it provides a clue.. I have a Magtag running

Adafruit CircuitPython 7.0.0 on 2021-09-20; Adafruit MagTag with ESP32S2
Board ID:adafruit_magtag_2.9_grayscale

It was running the Weather code from the Adafruit examples fine for a few weeks. And then one day it started with the Unknown Failure 205 error. During the course of testing, I deleted everything and reloaded all the libraries and code. Still have the problem.

I have tried bo...

slender iron
turbid radish
#

It is

slender iron
turbid radish
#

Next week: Guest Editor Emeritus Phil Torrone

slender iron
#

@idle owl would you mind requesting a circuitpython2022 graphic from bruce? Here is the one we used last year: https://blog.adafruit.com/2021/02/16/circuitpython2021-round-up/

Adafruit Industries - Makers, hackers, artists, designers and engineers!
#CircuitPython2021 round up

Happy 2021 everyone, we’ve gotten started with a bang with the release of the Raspberry Pi Pico and RP2040. I just got a bit of time to post the “final” list of #CircuitPython2021…

idle owl
#

Absolutely!

slender iron
#

thank you!

idle owl
#

Thank you!!

blissful pollen
#

sorry forgot to add its some enhancements to the IS31 work

turbid radish
#

I';m going to wear my LED glasses to a New Year's gathering

blissful pollen
#

Those LED glasses are amazing, a picture of me made a news article while wearing them

gilded cradle
#

That's a great idea @turbid radish

idle owl
#

Let's wait so Foamyguy can be involved.

tulip sleet
daring pumice
tulip sleet
lone axle
#

@idle owl I don't think I have the permission needed to transfer directly to adafruit. When I look at create a new repo page I see it listed with (insufficient permission).

I did find the transfer button though and it looks like it can attempt the transfer and then get approved on the receiving side. Let me know if I should do that, or if you think it's easier to transfer to your account first?

unknown.png unknown.png
idle owl
daring pumice
lone axle
idle owl
lone axle
#

Drat. It seems like I do currently have permission to transfer it to circuitpython org. Perhaps that org then would have permission to transfer to adafruit?

low sentinel
idle owl
lone axle
idle owl
#

I got it!

#

Accepted. Waiting on it to finish.

idle owl
lone axle
#

Nice. It even brings over open issues.

idle owl
#

Yep!

manic glacierBOT
lone axle
#

I will PR references to FRAM to change them this week. Probably best to get that done before release / add to bundle. Thank you!

idle owl
manic glacierBOT
onyx hinge
#

Here is the notes document for Monday’s CircuitPython Weekly meeting. It is at the normal time of 11am Pacific / 2pm Eastern here on Discord. Everyone is encouraged to attend! Please add your hug reports and status updates even if you’ll be attending the meeting - it’s super helpful! If you are unable to attend but would still like to include updates, feel free to include them in the notes and we’ll read them off during the meeting. Hope to see you there! <@&356864093652516868>
https://docs.google.com/document/d/1NBK5WWMLkzOm3BYvKe3zdyLp1Z5S5DQfyDGdEudic5w/edit?usp=sharing

Google Docs
CircuitPython Weekly for December 27, 2021

CircuitPython Weekly for December 27, 2021 Welcome to the CircuitPython Weekly meeting notes! Feel free to add your Hug Reports and Status Updates early. During the meeting, we go through them as a round robin sorted by username. If you can’t make the meeting and would still like to participa...

onyx hinge
manic glacierBOT
#
tannewt
[adafruit/circuitpython] New comment on issue #5725: RGBW NeoPixel setter: W only works with 4\-element tuples, is ignored in packed integers

Is this a good place to object to the behavior of setting the white channel when R==G==B?

This is a good place to discuss it.

  • There are at least three different color temperatures of RGBW NeoPixels, and in none of them will R+G+B white accurately correspond to W white in either tint or brightness.

Ya, I get that. This was done as a "just works" option.

  • Altering the numbers in results in the getter returning a value different than was passed to the setter, and this is c...
daring pumice
#

A bit confused: Trying to grab a member of a busio_i2c_obj_t pointer (*self) which has been passed as a param:

self->scl_pin gives me error: 'busio_i2c_obj_t' has no member named 'scl_pin' - but it does.

Am I just doing the pointer notation incorrectly here?

#

1 moment - may be wrong name because of board

#

sry

#

appears so - sorry for the bother

solar whale
#

@slender iron If you have time, can you let me know how you have your cm4io board configured. I have mine now an it works fine with a "pristine" SDcard. It boots and I get a CIRCUITPY drive, but the access to the drive only lasts until I power cycle then It no longer mounts and os.listdir() is empty. Did you do any special configuration or do you have any jumpers installed? I'm hoping to get it booting from the USB drive. This is absolutely not a crisis an I thought you were out this week so it can certainly wait until the New Year. But if you have any quick pointers, that would be great.

idle owl
#

@slender iron Not that you're using a Mac anymore, but iShowU doesn't work on Big Sur. They recommend some other app, but I remembered BlackHole which didn't work last time I tried it. I installed it during the meeting and it worked on the first try - iShowU didn't even do that half the time. Anyway, FYI if you upgraded your Mac and you end up using it for the meeting at some point.

slender iron
#

@solar whale I'm around until thursday ish. I don't think I have anything special setup. I am doing network boot

solar whale
#

OK -- thanks -- how did you configure it for network boot?

slender iron
#

thanks @idle owl. I haven't been using it regularly

#

I have a modified version of fb_tftp that hosts the files for me

daring pumice
#

Has anyone tried to use nrfx_twi_bus_recover? Getting a linker error despite it being in the nrfx submodule, unsure how to resolve this.

solar whale
#

Thanks -- I'll review that!

#

I think that will also help me get it to boot from USB -- I'll try that tomorrow.

slender iron
#

👍

tulip sleet
daring pumice
manic glacierBOT
daring pumice
#

Am I alright to remove some modules from the build by editing the board-specific mpconfigport.mk file? I'm out of flash space, but there are a bunch of modules I don't need that are getting included.

onyx hinge
#

damien has been making micropython smaller for so long, he's deleted hundreds of thousands more lines than he's added. 🤣

unknown.png
tulip sleet
daring pumice
#

🙂 Thanks!

manic glacierBOT
daring pumice
#

Wondering how to properly print the value of an mp_obt_t and busio_i2c_obj_t using printf (wrapped in mp_printf)?

#

e.g. if self is a busio_i2c_obj_t ->

 NRF_TWIM_Type *reg = self->twim_peripheral->twim.p_twim;
    if (reg == NULL)
        mp_printf(&mp_plat_print, "reg is NULL\n");
    else
        mp_printf(&mp_plat_print, "reg is %s\n", reg);
blissful pollen
#

You just want to print one of the members of the busio_i2c_obj_t? Like:

mp_printf(&mp_plat_print, "Value is %d\n", self->somevalue);
daring pumice
#

Trying to see if self-twim_peripheral->twim.p_twim actually has any value - right now it looks empty, but may be because I'm trying to parse it as a string, rather than as the appropriate type.

#

The base definition I can find is for NRF_TWIM_Type - not joy, there. Not really helpful, but knowing what it contains might help me find out what's wrong.

manic glacierBOT
proven garnet
#

Catching up today on the meeting, Group Hug to @lone axle and dglaude for work on the EEPROM library!

manic glacierBOT
proven garnet
#

Also I hear word of repository patching for Python version fixing 👀 happy to help with the patch clean up whenever that is!

manic glacierBOT
manic glacierBOT
daring pumice
#

is there any way to prevent the REPL window from closing after device disconnect? Hard to see what's going wrong.

#

in the Mu editor*

spiral elk
#

(And actually the larger ones too but you need multiple instances)

blissful pollen
daring pumice
blissful pollen
daring pumice
#

1.1.0.beta.6

blissful pollen
#

I have 1.0.2

#

What OS are you on too?

daring pumice
#

right now, windows

blissful pollen
#

I'm not sure why the latest Mu closes but you could try using PuTTY, its great and just have to find the close on exit/disconnect option and make sure it is off

daring pumice
#

I have putty 🙂 just use a serial connection?

spiral elk
#

I like tio, because it will reconnect after unplug/replug automatically

blissful pollen
#

Yup I just use the serial connection and put in the COM port I have.

daring pumice
#

(Y)

#

er

#

Thumbsup 🙂

robust quartz
#

HI I follow the instruction add the board in the Circuitpython, and I has generate the firmware.uf2, but when it draw it to the board, my board doesn't have any reaction, can someone help thanks

idle wharf
manic glacierBOT
robust quartz
#

Hi I want to add the new board in the Circuitpython, could anyone help thank, I has success build the uf2 file and work on my board.

tulip sleet
manic glacierBOT
lone axle
manic glacierBOT
idle owl
manic glacierBOT
idle owl
#

@tulip sleet I thought that it was going to be 3, not 3.x. Did you find out it needed to have the .x after the fact?

#

In your cookiecutter PR.

tulip sleet
#

it might work with 3

#

but 3.x is better

idle owl
#

Nah, that's fine, simply wanted to know the reasoning.

#

Thanks.

tulip sleet
#

I think we should try this on one library repo and do a release and check

idle owl
#

Ok.

tulip sleet
#

that all the steps work

idle owl
#

I'll run LED Animations through it.

tulip sleet
#

or if there is another library that needs a release anyway

idle owl
#

Well we have to do all the updates.

#

Cookiecutter doesn't update the libs.

#

So once I do all the updates, it'll need a release anyway.

orchid basinBOT
tulip sleet
#

good point, I was being parsimonious

idle owl
#

All the libs will eventually need this. 🙂

#

Once I get my mmpmhing IDE updated to work on this machine.

#

New laptop, who dis.

#

Wow that's fast now.

tulip sleet
#

i thought it would transfer all your stuff automatically if you wanted.

idle owl
#

So I have to get all the native apps downloaded. Finding them as I go.

orchid basinBOT
idle owl
#

@tulip sleet Release is done.

#

CI should begin running.

manic glacierBOT
idle owl
#

@tulip sleet 3.10.1 used. 🎉

tulip sleet
#

okee dokee!

idle owl
#

Thanks for the update! I'll let Eva know.

tulip sleet
#

thanks for good test ideas

idle owl
#

@tidal kiln Are you up for a schematic dive with me? I'm pretty sure it's entirely shallow end, but I've not seen this before on a board and wonder how to describe it for pinouts.

tidal kiln
#

@idle owl sure

idle owl
#

Do you want pngs or board files?

tidal kiln
#

which board?

idle owl
#

QT Py ESP32-S2

tidal kiln
#

rev A?

idle owl
#

B

tidal kiln
#

lemme fetch

idle owl
#

It has these two pads on the back labeled BAT and GND. And I traced them to the bit on the schematic labeled 3.3V LDO. And I looked up what an LDO is, so I know that it's providing the indicated voltage.... so are those pads simply a way to get 3.3V out of the board? I feel like it's something more than that.

tidal kiln
#

LDO = low drop out

idle owl
#

Yeah.

#

Voltage regulator, right?

tidal kiln
#

yep

idle owl
#

Right ok, so wait....

#

is it for connecting a battery?

#

Because I'm looking at the schematic and that arrow is going the opposite way it should if it's providing voltage... right?

tidal kiln
idle owl
#

Yeah.

tidal kiln
#

so two options for power: VBAT or +5V

#

both go into vreg

idle owl
#

And come out 3.3V.

tidal kiln
#

yep

idle owl
#

Right?

#

ok

#

So the pads on the back.... are they for a battery then?

tidal kiln
#

pretty sure. the M0 qt has some similar pads.

idle owl
#

And the 3.3V pin is for getting 3.3V out?

tidal kiln
#

yes

idle owl
#

Not putting it in?

#

ok

#

OK, thanks Carter. I was pretty sure, but wanted to make sure I understood it

tidal kiln
#

i think worth pinging limor to be sure. but my guess is those pads are general purpose. you'd just solder wires direct. but they may also be a footprint for something specific? like a JST?

#

but in eagle, they are just generic 1x3mm test points

idle owl
tidal kiln
#

yep

onyx hinge
#

The way it's presented as 2 pads, not one part, makes me think it's for soldering to

#

here's how the battery input looks on the CPX for example

unknown.png
idle owl
#

They're big pads too

#

Not test points on the actual board.

#

chonkpads.

onyx hinge
#

and here I assumed nothing on the qtpy was ever chonky

idle owl
#

Right?

#

She ninja'd them in.

tidal kiln
#

the qt m0 is simpler

idle owl
#

Ah yeah.

tidal kiln
#

no BAT, just 5V

idle owl
#

Right.

crimson ferry
#

in one of the livestreams... (maybe it was even last night), ladyada breezily showed the pads on the reverse and said it was for adding a battery, a nice little extra on a tiny board

idle owl
#

Saves me a step of bothering her. 😄

manic glacierBOT
#
bludin
[adafruit/circuitpython] Issue opened: #5757 watchdog stalling code execution on samd51

CircuitPython version

Adafruit CircuitPython 7.1.0-beta.1-27-g013e688c9-dirty on 2021-12-01; lis_43271a_onyx with samd51j19

Code/REPL

WATCHDOG_ENABLED = True
...
from watchdog import WatchDogMode
from microcontroller import watchdog
...
if WATCHDOG_ENABLED:
    watchdog.timeout = 16
    watchdog.mode = WatchDogMode.RESET
    ...
while True:
    if WATCHDOG_ENABLED:
        watchdog.feed()
    ...

Behavior

This is a preliminary report ...

crimson ferry
idle owl
#

Here's what might be a dumb question - is the 4MB of flash built into the module?

#

I don't see a flash chip. or I'm missing it.

#

Yep, it's a dumb question. Yes it's built in.

crimson ferry
#

yeah, presumably the S2FN4R2 with 4MB flash and 2MB PSRAM under the plastic (I think the Great Search a week or two ago was for that one) https://youtu.be/rMUil5q4i_8

#

(plus the 320k of regular RAM)

idle owl
#

Ugh. Discord stopped working. But it seems to be an issue with the canary build, which is the only build that runs natively on M1, so I guess I'm dealing with super slow Discord for a while.

#

Hmm they have a middle ground with a public test build. Which is native, and working.

#

Fingers crossed it continues.

idle owl
#

@slender iron I requested the CP2022 graphic from Bruce, I put a due date on it of 31 December - I can make it sooner if needed. Let me know.

idle owl
#

@slender iron Scratch that, it's already done! I'll get them to you. Ping me when you're around.

whole solstice
#

the feather m4 supposedly supports rs485 hardware wise, not sure about the rp2040. circuitpython has a couple lines of code in the busio library that are rs485 specific but I haven't seen anyone get circuitpython to do rs485. does anyone have experience with rs485 protocol in circuitpython?

tulip sleet
slender iron
#

@idle owl I see it in my basecamp overview email. Thanks!

idle owl
slender iron
#

sure 🙂

idle owl
#

I didn't want to keep pinging you while you were away.

slender iron
#

thanks!

idle owl
#

I really like them this year

slender iron
#

now I can draft the kick-off post

idle owl
#

That's what I figured

slender iron
#

ya, very cool

meager fog
ornate breach
#

😮

eternal lantern
#

is there some method to uninstall and reinstall circup to work around the fact that 6.x is breaking updates even if not trying to use 6.x libs? getting the "There was a problem downloading the bundle. Please try again in a moment." problem for about a week now.

onyx hinge
#

@eternal lantern assuming you originally installed circup with pip, you can upgrade it with pip install --upgrade circup. It should end with a line like Successfully installed circup-1.0.1.

eternal lantern
#

@onyx hinge. AWESOME SAUCE, that's the magic I needed thanks.

manic glacierBOT
#
tannewt
[adafruit/circuitpython] New branch created: revert\-5750\-pin\-cascadetoml
#
RetiredWizard
[adafruit/circuitpython] New comment on issue #5706: broadcom port for raspberrypi4b often corrupts SDCard

I've been working on this for a couple days with a RPIZero2W and so far haven't corrupted the SD card, however with both a clean build from the main github source and from the debug build from Scott's artifact page the flash is never mounted or accessible.

Strangely, if I download the prebuilt image from circuitpython.org, the flash mounts 90% of the time without a problem.

anyway, in case it's of any help, here's the output from the debug image:

sending cmd slot=0 op=52 arg=80000c08...

manic glacierBOT
proven garnet
#

So I found a file that could use some updates to the type hints, and wanted to ask: So the "font" parameter says it should be any class that has get_bounding_box() and get_glyph() methods, which from my understanding is a good case for using Protocol (duck typing). But realistically I think it's only like a few class classes so Union[fontio.BuiltInFont, adafruit_bitmap_font.BDF, adafruit_bitmap_font.PCF] is also valid. Is there a preferred method, being explicit vs. duck typing?

onyx hinge
#

@proven garnet that's a very valid question. Using that Union[] for now sounds like a good approach even if it's not fully correct.

tulip sleet
onyx hinge
#

we could put the font interface in the fontio pyi file, I think

#

needs some docstrings, but I think this would work. then a later revision of typing of the modules like label could use fontio.FontInterface [name is up for discussion!]

unknown.png
proven garnet
#

Got it, I'll use the Union. Trying to get some last minute PRs out before the holidays!

proven garnet
#

I think an Interface would be great! Would definitely be helpful for typing 🙂

#

I assume fontio.BuiltinFont would be a subclass then? Unless that interface already exists and just isn't public facing

#

I don't do much on the display side of things, some day I will haha

onyx hinge
proven garnet
#

Just learned about them!

onyx hinge
#

If I understand it, FontProtocol just needs to be seen by mypy, and nothing needs to derive from it

#

Resource is a subtype of the SupportsClose protocol since it defines a compatible close method. Regular file objects returned by open() are similarly compatible with the protocol, as they support close().

proven garnet
#

Yeah, my understanding is the methods need to be put in the Protocol, and it saves you having to use inheritance

#

I don't know how difference the font classes are all from each other and whether inheritance makes sense or not

onyx hinge
proven garnet
#

This looks great! And it's a subclass of typing.Protocol?

onyx hinge
#

nope, missed that 😅

proven garnet
#

Thanks @onyx hinge, this is awesome, should I add an issue to this library to eventually swap typing over the new protocol?

onyx hinge
#

let's see what happens with this PR first, I'll file it in circuitpython soon

#

I'll @ you on it (same name on github right?)

proven garnet
#

Yup! Please do so I don't lose it!

#

The post-holiday notification checking is the worst, so it'll help haha kgsPanic

onyx hinge
#

oh, drat. import _typing is widely used in the libs already 😦

#

I wanted to rename it, because I don't think the name _typing is a good one to squat on

manic glacierBOT
#
jepler
[adafruit/circuitpython] Pull request opened: #5759 Add a font protocol

@tekktrik noticed that there was no convenient way to specify that a function takes "any kind of Font", including the built in one and the ones from the adafruit_bitmap_font package. By defining this protocol in our stubs, it's hoped that fontio.FontProtocol can be used instead of an ad-hoc Union[] of several known font types.

image

onyx hinge
#

oh maybe I'm mistaken about that ... it is some other _typing. and not actually a bundle file, but in a virtualenv I have left over for some reason. good news? ```libraries/drivers/ili9341/_env/lib/python3.9/site-packages/pkg_resources/_vendor/packaging/_compat.py
8:from ._typing import TYPE_CHECKING

#

yeah yay it's okay 😹

manic glacierBOT
manic glacierBOT
#
RetiredWizard
[adafruit/circuitpython] New comment on issue #5706: broadcom port for raspberrypi4b often corrupts SDCard

I had not been erasing my SD cards when trying new images and it occurred to me that the problem I'm having is on the part of the card not occupied by the image, so I did a full erase (overwrite) of the SD card and reloaded Scott's debug image. This time the Flash file system was created and seemed to work on the first load, but after power cycling it failed once again. I guess this is exactly what the original issue reported.

#
jerryneedell
[adafruit/circuitpython] New comment on issue #5706: broadcom port for raspberrypi4b often corrupts SDCard

I had not been erasing my SD cards when trying new images and it occurred to me that the problem I'm having is on the part of the card not occupied by the image, so I did a full erase (overwrite) of the SD card and reloaded Scott's debug image. This time the Flash file system was created and seemed to work on the first load, but after power cycling it failed once again. I guess this is exactly what the original issue reported.

Yup -- same here.

meager adder
#

@lone axle @idle owl Is this issue still open for the taking? https://github.com/adafruit/Adafruit_CircuitPython_framebuf/issues/43
I've been working on the framebuf library to add a new format for my display so I'm already somewhat familiar with the code. I'm done with that so I can take on this annotation task if it's still available.

GitHub
Missing Type Annotations · Issue #43 · adafruit/Adafruit_CircuitPyt...

There are missing type annotations for some functions in this library. The typing module does not exist on CircuitPython devices so the import needs to be wrapped in try/except to catch the error f...

lone axle
manic glacierBOT
manic glacierBOT
idle owl
#

@slender iron Thank you so much for catching that Servokit issue.... I would have headed immediately into updating code.

slender iron
#

np 🙂

#

I had already seen it once earlier today

idle owl
#

Makes sense.

slender iron
#

it looks like arduino code to me 🙂

idle owl
#

Bonus! Not my problem 😄

manic glacierBOT
manic glacierBOT
#
RetiredWizard
[adafruit/circuitpython] New comment on issue #5706: broadcom port for raspberrypi4b often corrupts SDCard

With the image I built from github I get the following results over UART on the initial boot. On subsequent power cycles the flash file system isn't accessible.

            Adafruit CircuitPython 7.1.0-beta.3-166-gb60a80696 on 2021-12-20; Raspberry Pi Zero 2W with bcm2837
            Board ID:raspberrypi_zero2w
            write sectors result 3
            write sectors result 3

            Auto-reload` is on. Simply save files over USB to run them or enter REPL ...
manic glacierBOT
manic glacierBOT
#
dhalbert
[adafruit/circuitpython] Pull request opened: #5762 Fix recording of wake\_alarm
  • Fixes #5343.

alarm.wake_alarm was not being set properly after a real or fake deep sleep. I moved the responsibility for keeping track of whether the board had just hard-restarted to main.c, instead of having it in common-hal. The most recent wake alarm is now recorded in alarm.wake_alarm either after a hard-reset, or after a simulated (fake) deep sleep.

Light sleep was not setting alarm.wake_alarm. I did this in a port-independent way.

Also did a minor doc fix in alarm...

main furnace
#

If I run 7.1.0-beta.3-174-gbea052695 on 2021-12-22 on the pi zero 2W with USB I get read sectors result 3 and it tends to hang.
If I run the same but use the serial console the I still get the SD card errors but I can use the repl.
With the older 12-08 build USB doesn't work, but I can access the repl via serial (uart) and I can read and write the SD card.

manic glacierBOT
main furnace
#

No SD card means I have to type in anything I want to run, which is very retro indeed.

manic glacierBOT
manic glacierBOT
#
jpconstantineau
[adafruit/circuitpython] New comment on issue #5498: nRF52833 DK \(PCA10100\) CircuitPython UF2 does not work on Adafruit\_nRF52\_Bootloader

Good news everyone. I just tested changes on the bootloader side to make a build that includes the 7.2 Softdevice and there is no need to do anything special to load Circuitpython. Drag/drop of the CP UF2 as per other boards works perfectly fine.

I'll close this issue once all the bootloader changes are merged in and tested with merged binaries.

manic glacierBOT
manic glacierBOT
manic glacierBOT
onyx hinge
manic glacierBOT
#
dannystaple
[adafruit/circuitpython] Issue opened: #5763 Read the docs link for CircuitPython don't work

CircuitPython version

NA

Code/REPL

NA

Behavior

NA

Description

I get a 404 if I click the Edit On Github links in the Documentation. This may be a readthedocs issue, or a configuration issue in the doc generation.

Reproducing this:

Result:

-...

onyx hinge
#

we're still building the core with Collecting Sphinx<4 for some reason

manic glacierBOT
manic glacierBOT
#
tannewt
[adafruit/circuitpython] New branch created: revert\-5755\-fix\-cascadetoml\-more
slender iron
#

@tulip sleet did you know you can make revert PRs from github?

manic glacierBOT
manic glacierBOT
tidal kiln
manic glacierBOT
tulip sleet
tulip sleet
tidal kiln
#

thanks @tulip sleet

tulip sleet
tidal kiln
#

it's a pink feather rp2040 btw

manic glacierBOT
solar whale
#

@slender iron I'm trying to set up the tftp server for my cm4io - followed the link you gave me a few -- I copied everything from the /boot of my SDCard image made by Raspberry pi imager to my tftp server folder -- and configured the cm4io for netwrok boot, It seems to be trying, but it it is not getting anything from the server... Did copy anything else to the tftp server? Anything else I need to do?

slender iron
#

the folder should mirror the sdcard afaik

#

I've got a bit weird setup because I hacked my tftp version to look in two folders

solar whale
#

hmmm -- ok - thanks -- so far, I cannot get it to boot from USB or network -- only SDCard -- I'll keep trying...

slender iron
#

hrm

#

did you see the comment about pin reset causing the issue?

solar whale
#

yes -- sounds promising...

slender iron
#

I'm going to look into that after lunch

solar whale
#

ah -- it helps to enable the network on the server!!! It boots!

#

so -- now that it boots, where is the file system on the CM4io? How do you get to CIRCUITPY?

manic glacierBOT
dusky halo
#

@slender iron just spotted the request to test sideset opt - do I want the 7.1.0-beta 3, or the latest S3 build?

daring pumice
#

What does it mean if I get back a whack of i2c addresses from i2c.scan()?

tulip sleet
blissful pollen
#

I'm assuming you don't have a whack of i2c devices hooked up?

daring pumice
#

I have two. When I pass the sda and scl for the first to scan, I get one valid address. When I pass the sda and scl of the other, I get back every address.

tulip sleet
#

you have two different buses, right? The scanning is sending an address and waiting for an acknowledgment, which is a specific kind of response. I think the device on that bus may be acting up in some way.

#

is that the device with no pullups on board?

daring pumice
tulip sleet
#

did you get some resistors and a breadboard to test it with external pullups?

daring pumice
daring pumice
tulip sleet
#

i would vet it with that.

daring pumice
#

The issue is in common_hal_busio_i2c_probe

solar whale
#

@slender iron When you have a minute can you please confirm the configurations of your cm4io for me. On the cm4io. I am booting from the network. If I have no SDCard inserted, then I get no file system (os.listdir()) is empty and I see nothing mount on my host computer. I guess taht maes sense since the cm4-lite (no eMMC) has no place to put the FS. If I insert an SDCard made by Raspberry pi Imager (with the same image as on the network boot) it boots (and I think it is booting from the network) and I see my CIRCUITPY Drive and os.listdir() shows my files. But if I put in a blank SDCard, then I still get no file system. I'm just cuirious how you have it set up. Do you have an SD Card inserted and if so , wha it is on it and do you get a CIRCUITPY Drive?

daring pumice
#

When nrf_twim_task_trigger(reg, NRF_TWIM_TASK_STOP); is run, the code then sits in a while loop waiting for a NRF_TWIM_EVENT_STOPPED event. This never fires, and the device eventually hangs.

#

However

#

When I check for an error inside that loop using (nrf_twim_event_check(reg, NRF_TWIM_EVENT_ERROR)), an INTERNAL_ERROR event is found.

#

I can then recover the bus using nrfx_twi_bus_recover

#

This succeeds, and I get a single valid address: 0x08

tulip sleet
#

the sensor is acting up, I think. It would be good to test this in two other ways: 1. with external pullups. 2. on a non-NRF board, such as SAMD51

daring pumice
#

I should mention the aforementioned while loop properly exits after recovery, and the probe proceeds normally.

daring pumice
#

I'll wire something up over the holidays and see what happens 🙂

tulip sleet
#

We have not seen this behavior before. A few devices do not respond well to that kind of probing, but it's rare.

daring pumice
#

Interesting.

tulip sleet
#

The Feather Sense is still nRF. I would suggest ordering a Feather M4 or Metro M4 too. And did you get breaboard and jumpers, etc.?

daring pumice
#

Those have been ordered, and am expecting delivery before the new year.

tulip sleet
#

excellent. Is the acting-up device proprietary, or can you say?

daring pumice
#

The one returning 0x08 is proprietary - I'm wondering if it's all working correctly, now.

#

The one returning all addresses is 3rd party.

tulip sleet
#

i would websearch the all address one to see if there are any other reports of issues with it

daring pumice
#

Great idea 🙂 Thanks!

slender iron
#

@dusky halo you can snag a build from the PR if it's done or I can give you one

#

@solar whale CP expects the sd card to have a partition table already

solar whale
#

Ah… ok .. that makes sense. I’ll try that.

slender iron
#

and then CP fakes the partition table to USB like it normally does

solar whale
#

Step by step….

slender iron
#

🙂

manic glacierBOT
dusky halo
slender iron
#

👍

manic glacierBOT
manic glacierBOT
manic glacierBOT
#
RetiredWizard
[adafruit/circuitpython] Issue opened: #5768 Broadcom port doesn't include io module

CircuitPython version

Adafruit CircuitPython 7.1.0-beta.3 on 2021-12-13; Raspberry Pi Zero 2W with rp3a0
Adafruit CircuitPython 7.1.0-beta.3-174-gbea052695-dirty on 2021-12-22; Raspberry Pi Zero 2W with bcm2837

Code/REPL

import io

Behavior

Traceback (most recent call last):
File "", line 1, in
ImportError: no module named 'io'

Description

No response

Additional information

No response

manic glacierBOT
manic glacierBOT
manic glacierBOT
manic glacierBOT
proven garnet
#

Should adafruit_circuitpython_ble have blinka in the requirements.txt? Looking at things from the contributing page bu didn't know how Bleutooth is handled regarding libraries plus Blinka.

tulip sleet
manic glacierBOT
#
bigbearishappy
[adafruit/circuitpython] Issue opened: #5770 How can I make the circuitpython work with the bootloader\(\*nosd\.hex\) which don't include the softdevice?

I am working with nrf52840 board now. And I can make the circuitpython work fine with the adafruit bootloader which included softdevice.
However, when I change to the bootloader which don't included the softdevice.The circuitpython is not working.

I tried to modify the ports/nrf/bluetooth/bluetooth_common.mk to remove the "CFLAGS += -DSOFTDEVICE_PRESENT". I also tried to
change the ld file to delete the flash region of softdevice.
But I still can't make it work.

Is there any sug...

proven garnet
atomic summit
#

Has anyone been able to successfully build circuitpython or any ESP32-S3 board?

manic glacierBOT
solar whale
manic glacierBOT
#
bludin
[adafruit/circuitpython] New comment on issue #5757: watchdog stalling code execution on samd51

that's odd, ticks_ms and monotonic_ns both use ...

The link to ticks_ms() isn't corroborated. I only had 5 or 6 stalls in total and for the first two, I didn't even try a keyboard interrupt. I know for the last two that the line contained ticks_ms and I seem to remember this was true for the ones before, too, but I'm not sure. No stalls since I replaced ticks_ms with monotonic_ns, but that might well be luck. And ticks_ms is used in waiting loops, so these lines get called fairly oft...

manic glacierBOT
manic glacierBOT
#
dhalbert
[adafruit/circuitpython] New comment on issue #5770: How can I make the circuitpython work with the bootloader\(\*nosd\.hex\) which don't include the softdevice?

We have never tested without the SoftDevice, so there may be errors in the logic, and also the linking map may be wrong, since it assumes the firmware is placed after the SoftDevice in flash.

The attraction of using CircuitPython on the nRF boards is that it has BLE support, so we have always assumed the SoftDevice would be present. There are other chip families without BLE that we would recommend otherwise.

deft jay
#

Sorry for crossposting with main help, this is probably more of a Development question as it's related to the releases and not function: for the Seeeduino XIAO, there is a "keyboard optimized" version that states it does NOT include pulseIO module, but the regular version also does not include the pulseIO module. Is there any version of the XIAO firmware that includes the pulseIO module? This seems to be an erroneous omission on the main firmware due to the language of it's specific exclusion on the keyboard optimized firmware. Please explain what's up with this :D!

lone axle
# deft jay Sorry for crossposting with main help, this is probably more of a Development qu...

many modules are excluded on SAMD21 devices because it cannot fit all of them. I would agree the way it's listed on the page makes it seem like pulsio might be present in the normal build. But I think it may not actually be included in the normal build either. You could make a custom build that enables it if you want perhaps though. If there are other modules you don't need you could turn them off to make room.

deft jay
#

@lone axle is there a way to use the build system to generate a version with pulseio enabled and maybe disable something else to make room for it? Or am I going to have to make a build environment on my own PC to do a custom build?

(Is there a 4 dummies guide that I can follow?)

manic glacierBOT
#
jerryneedell
[adafruit/circuitpython] New comment on issue #5706: broadcom port for raspberrypi4b often corrupts SDCard

So this obviously isn't the solution, I'm sure it breaks all sorts of things, but hopefully it will help ferret out the issue.

I commented out the reset_all_pins(); statement on line 109 of ports/broadcom/supervisor/port.c and my build is now working with the flash.

Cool! I tried this as well on my Pi 4B -- -- It now boots from the USB Drive and I can access CIRCUITPY on the SDCard.
Thanks!

lone axle
tulip sleet
# 
class Blink:
    def __init__(self, on, off, pin):
        self.prev_time = -1
        self.on = on
        self.off = off
        self.pin = pin
#

and then refer to to the fields as led.pin instead of led["PIN"] etc.

lone axle
#

Would it make sense to call it BlinkTask? If "Task" is used as an actually concept somewhere It's probably best to not use it here since they aren't "real" Tasks. But as a general term I think it fits what they do.

tulip sleet
#

tasks are a concept (and a class) in asyncio

#

BlinkState? Blinker?

#

you could add methods to the class as well to do the blinking or checking, not sure if that makes sense for this example

#

a task is a wrapped coroutine

manic glacierBOT
onyx hinge
lone axle
#

I see. Probably best to avoid using that term then. I do like either BlinkState or Blinker. I've got it noted down on my todo list to rewrite the examples. The ones on the Servo pages could be updated similarly as well.

tulip sleet
#

ok! thanks for thinking about this

manic glacierBOT
#
RetiredWizard
[adafruit/circuitpython] New comment on issue #5743: Time methods report "No long integer support" on BCM

I'm not sure if this is the correct approach, but I stole the following code from the raspberrypi port and placed it at the top of ports/broadcom/mpconfigport.mk which seems to have enabled long integer support.

     # Define an equivalent for MICROPY_LONGINT_IMPL, to pass to $(MPY-TOOL) in py/mkrules.mk
     # $(MPY-TOOL) needs to know what kind of longint to use (if any) to freeze long integers.
     # This should correspond to the MICROPY_LONGINT_IMPL definition in mpconfi...
deft jay
manic glacierBOT
tulip sleet
deft jay
#

@tulip sleet is the pulseio module unsupported for the seeeduino xiao?

#

it appears to be missing from the matrix, but no explanation for why or why not

#

the xiao rp2040 module is supported, but this is some other kind of cpu

#

on the straight regular mainline xiao chip

deft jay
#

THANK Y'ALL I GOT IT WORKING 😄

SPOILER_unknown.png
#

totally defeating the purpose of moving from arduino to python by compiling my own firmware but also loving how easy that was. Thank you all and thank you to whoever wrote that beautiful guide on how to DIY compile circuitpython. that was easy as pie. hopefully it actually works for what I'm trying to build 😄 (Or I can figure out that coding! lol)

Happy holidays

slender iron
manic glacierBOT
stoic rain
#

This is probably a dumb question, but are the ssd1306 drivers expected to work well an M0-based board like the Metro M0 Express? I’m asking because I seem to be running into memory allocation issues using either the framebuff or displayio based drivers.

solar whale
solar whale
slender iron
#

@solar whale I tried whittling down which pin was causing issues on reset but didn't nail it down

stoic rain
solar whale
slender iron
#

I think the zero2w definitely had reset issues

#

but the 4b has secret sd card pins

solar whale
stoic rain
#

Yes, I am

slender iron
#

are you getting the memory issue on import?

#

@solar whale the issue I'm seeing on 4b isn't pin dependent from what I can tell

manic glacierBOT
#
dhalbert
[adafruit/circuitpython] New tag created: 7\.1\.0\-rc\.0
tulip sleet
solar whale
stoic rain
slender iron
#

@solar whale I'm seeing an error that data is corrupt after increasing the SD card clock. It's sporadic and caused by in incorrect load before the clock rate change

#

@stoic rain terminalio has a tilegrid for it's use

solar whale
#

Is this on the 2w?

slender iron
#

no, the 4b. on the 2w I actually get init failure when doing to high speed mode

#

(I wasn't checking that the card init actually finished ok)

manic glacierBOT
deft jay
#

I followed the guide on how to configure modules and removed midi

manic glacierBOT
tulip sleet
manic glacierBOT
manic glacierBOT
solar whale
#

@slender iron I know you are heading off for the holidays. Feel free to let me know if there is something in particular you want me to try to look at. I think I finally have my setups all useable. I had a bit of a soldering mess on one of the pi adapters and I had to put the SWD header on it instead of the JTAG. Now I need to figure out the proper configuration for using it.... But I have two other with JTAG so I can use them until I do. Now that I have some working systems, I'll be happy to try to help. Still a steep learning curve to catch up, but any pointes would be welcome -- enjoy your break -- it'll all be here when you get back 😉

slender iron
#

thanks @solar whale . I'm spending these last few hours looking. something has got to be off with how I'm doing data transactions

manic glacierBOT
solar whale
#

Good luck, I have so much to learn -- and relearn about debugging.... it's been too long...

slender iron
#

the broadcom chips are definitely the deep end

#

sdcards would be easier if we could sniff the data lines easily

#

getting this right will hopefully speed up sd card access

solar whale
#

Have you ever noticed that it takes a long time to get contol-c to interrupt a running gloop?

slender iron
#

no, I haven't looked at that specifically

atomic summit
solar whale
slender iron
#

over usb or uart?

solar whale
#

uart

#

I have not tried USB

slender iron
#

kk, I didn't think uart worked at all 😄

solar whale
#

Ah -- it's immediate over USB -- never-mind

main furnace
#

Ah, long press ctrl-C. I'll give it a try.

solar whale
#

over uart -- very long press

#

then it fires a bunch of times....

main furnace
#

Could I do a Broadcom build with no USB support? Sure it breaks the workflow, but that isn't really working just now.

manic glacierBOT
manic glacierBOT
solar whale
#

@slender iron FYI -- the SWD connector seems to work with the JTAG configuration -- I guess I thought some changes were needed, but I guess not -- at least at first glance. Nice! Now I have 3 working adapters.

slender iron
#

👍

#

I've gone in more circles with this SD stuff so it's lunch and pie crust time

solar whale
#

time to work on dinner here -- have a great holiday!

slender iron
#

you too! thanks!

solar whale
#

I have way too many toys to play with !!

orchid basinBOT
atomic summit
#

I just blew away my fork, and I'm starting from scratch again 😦

orchid basinBOT
manic glacierBOT
tulip sleet
atomic summit
#

I've been out of the CPY loop for ages, sorry, but well done to everyone on hitting 7.1 rc ! the new stuff looks awesome, and it's great to see prelim asyncio support!

tulip sleet
#

i tested it with the new qtpy esp32 (which, true to its name, is quite cute), talks to my wifi just fine,

manic glacierBOT
atomic summit
#

Ok, I got it compiling finally - after a clean fork and clone and submodule update... but either my tinyuf2 isn't right, or my CPY uf2 isn't right. because I can boot into uf2, but if I copy my CPY firmware.uf2 over, it just copies. No flashing orange on the LED... just stays green, and the file just appears on the drive 😦

#

Possibly something to do with the UF2 family ID changing for S3? Not really sure what that is.

viscid pine
atomic summit
#

Hmm, seems since yesterday the tinyuf2 S3 stuff has been merged and the add-s3 branch is gone. I'm going to go back to master and grab latest and remake my uf2 bootloader

atomic summit
#

I'm not able to build tinyuf2 - this is such a pain.

# 
 {
 ^
/Users/seon/tinyuf2/lib/esp-idf/components/hal/adc_hal.c:64:17: warning: 's_previous_init_code' defined but not used [-Wunused-variable]
 static uint32_t s_previous_init_code[SOC_ADC_PERIPH_NUM] = {-1, -1};
                 ^~~~~~~~~~~~~~~~~~~~
[290/1065] Building C object esp-idf/hal/CMakeFiles/__idf_hal.dir/spi_slave_hd_hal.c.obj
ninja: build stopped: subcommand failed.
ninja failed with exit code 1
make: *** [all] Error 2```
#

I wish CPY and TinyUF2 could at least be on the same IDF version.

lunar gull
#

I was on adafruit-circuitpython-adafruit_magtag_2.9_grayscale-en_US-20211223-605753c.uf2 and for whatever reason it wasn't working with I2C, but, rc0 seems to have fixed it!

atomic summit
#

just killed off my tinyuf2 fork as well and starting again from scratch. I just don't understand how this all keeps getting messed up. yesterday I could build from the add-s3 branch. today going back to master and making sure I pulled and my submodules were all good... I can no longer build.

viscid pine
#

submodules are a pain aren't they

atomic summit
# viscid pine submodules are a pain aren't they

Yup, especially when the IDF submodule has its own submodules. It would be way better if we could provide our own IDF and instead of hard coded file paths it used the IDF env paths. Just like MP used to do before cmake.

#

ok, finally got it..

Board ID:unexpectedmaker_tinys3
#

though it seems to be flashing the RGB LED green for a split second every 5 seconds... hmm, is that new?

viscid pine
#

it's signifying "Code done running"

atomic summit
viscid pine
#

unsure

#

might be since cpy 7

manic glacierBOT
#
bigbearishappy
[adafruit/circuitpython] New comment on issue #5770: How can I make the circuitpython work with the bootloader\(\*nosd\.hex\) which don't include the softdevice?

We have never tested without the SoftDevice, so there may be errors in the logic, and also the linking map may be wrong, since it assumes the firmware is placed after the SoftDevice in flash.

The attraction of using CircuitPython on the nRF boards is that it has BLE support, so we have always assumed the SoftDevice would be present. There are other chip families without BLE that we would recommend otherwise.

Thank you for your reply.
It seems that we need to do more work to make ...

#
dhalbert
[adafruit/circuitpython] Pull request review submitted: #5776 enable I2C by default

This will enabled I2C power after a hard reset, but will not change the pin state after a soft reload. For that we would add it to reset_board().

So if you disabled power in boot.py by setting the pin high, it will stay high. Maybe this is what we want. If we moved pin setting to code.py, it would get changed every time code.py started.

I think that makes sense. It is similar to the CPB, which also turns off stuff on hard reset, but doesn't adjust it on each soft reload.

This fix...