#circuitpython-dev

1 messages Β· Page 404 of 1

solar whale
#

nice

slender iron
#

all the other files are the same

manic glacierBOT
#
dhalbert
[adafruit/circuitpython] New comment on issue #5818: Chat using open source platform?

Discord is better -- the other moderators are there. Please try to resolve your discord registration issues on a desktop when you get a chance. If you fail then open a forum thread. I saw you join as @kj7rrv and have pinged you in the #help-with-community channel. Please don't use GitHub as a chat server. Dozens of people get an email every time there's a post in this issue.

solar whale
#

just comes up to a frozen REPL

#

let me try reflashing my SDCard

#

argg -- linux box also frozen....

slender iron
#

hrm

#

looks for any rogue while (true) {}

solar whale
#

rebooting -- do you have the full .zip handy -- I wll try a new SDCard

slender iron
#

sure

solar whale
#

so far so good -- boots and file manager is happy .

slender iron
#

🀞

solar whale
#

I may have corrupted the previous card

slender iron
#

ya, I had one where it was unhappy until I did an erase_filesystem

solar whale
#

I was able to eject it and then disconnect and replug in -- booted OK

#

should I try taht with the previous build or are you going to keep the changes

slender iron
#

I think I'll keep these changes. they make the pi4b work better for me

solar whale
#

OK -- great -- I'll let you know if/when I break it πŸ˜‰

#

It is a little weird that Raspberry Pi Imager does not clear the CIRCUITPY drive. I got very confused by that.

slender iron
#

I was hoping that'd be the case

#

since that'll keep all your python files there

#

not good when the filesystem can be corrupted easily

solar whale
#

That is good, It just really confused me when I reflashed an old card and there were files in CIRCUITPY that I forgot had been there.

slender iron
#

ugh, it is still unreliable for me

solar whale
#

😦

manic glacierBOT
slender iron
#

it might be an interrupt handler clobbering a neon register

#

it's the same interrupt handler on the 2w though too

solar whale
#

My 4B was ok when I tested it this morning. Minimal testing, but no problems

slender iron
#

ya, it's weirdly flaky for me

solar whale
#

like the 2w for me

slender iron
#

it worked once this morning

solar whale
#

all the other builds worked for me -- just the 2w was flaky

slender iron
#

I think I'll pull these delays out before a PR

solar whale
#

OK -- I am going to ry this mornings build on my 2w with the ne SDCard

slender iron
#

you are seeing something different from me.

#

I'm not seeing corruption. Well... maybe I was before adding the delays

#

right now I see startup errors with SD card init failed 2

solar whale
#

I went back to this mornings build and My 2W is working - CIRCUITPY comes up and I can use the file manager. My issues must have been with the old SDCard....

slender iron
#

kk, so one without the delays

solar whale
#

yes

manic glacierBOT
slender iron
#

☝️ is what I sent you minus a bunch of delays

solar whale
#

odd thing -- power cycled -- unplugged USB and replugged (after eject) and it booted, but no CIRCUITPY...then I unplugged and repluggedd and CIRCUITPY came up fine!!! this happend last time I tried it as well

slender iron
#

any debug output?

#

specifically the SD card init failed line

solar whale
#

noting unusual on UART

slender iron
#

hrm

solar whale
#

it is working fine after the second power cycle

slender iron
#

I was wondering if there is an SD card power pin that I'm resetting

solar whale
#

are you on 4b or 2w?

slender iron
#

4b

solar whale
#

ok -- I was on 2w

manic glacierBOT
#
ladyada
[adafruit/circuitpython] Pull request review submitted: #5819 Add Piunora and Zero\. Also, enable full build

lol at
`void board_init(void) {
videocore_framebuffer_obj_t *fb = &allocate_display_bus()->videocore;
fb->base.type = &videocore_framebuffer_type;
common_hal_videocore_framebuffer_construct(fb, 640, 480);

framebufferio_framebufferdisplay_obj_t *display = &displays[0].framebuffer_display;
display->base.type = &framebufferio_framebufferdisplay_type;
common_hal_framebufferio_framebufferdisplay_construct(
    display,
    MP_OBJ_FROM_PTR(fb),
    0...
solar whale
#

If you have a 4b .zip handy, I can try it -- or I'll try it when the artifacts are ready.

slender iron
solar whale
#

thanks

#

Seems happy -- I am booting from USB drive -- File system on SDCard

#

no debug output

slender iron
#

but no drive?

solar whale
#

yes CIRCUITPY OK

slender iron
#

that's good

solar whale
#

Note: don't use blue leds on the Pi adaptors... the forward voltage is 3V and they are vey dim.... I used red orange yellow an blue. can barely see the blue....

slender iron
#

πŸ™‚

#

I think I have two green and two blue

#

haven't used them much

solar whale
#

My quick test cycles them

slender iron
#

nice!

solar whale
#

so far my 4b is behaving -- anything you want me to try on it?

slender iron
#

nah, I think we need to let this latest version soak and see what people say

solar whale
#

Sounds good. Thanks for all you've done on this

slender iron
#

the never reset thing was obvious

solar whale
#

only to you πŸ˜‰

slender iron
#

this failure I see is so weird

solar whale
#

I've been doing assembly language coding for the past few weeks. I've very familiar with "weird" bugs....

slender iron
#

πŸ™‚

#

I suspect/blame poor sync between the cpu and the emmc

solar whale
#

Timing is so hard to debug.

manic glacierBOT
solar whale
#

Are you using "debug" builds. I used to find that the changes in optimization cause weird problems in critical timing.

slender iron
#

these have been opt builds

#

but prints can cause timing changes too and I ❀️ print debugging

solar whale
#

In the "old days" those prints at 9600 baud took 1ms per character.....

#

yawn,,,

slender iron
#

πŸ™‚ yikes

#

I'm amazed that the linux kernel has functions to print out register state

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

So I think the problem caused by resetting the SD card pins is fixed.

I'm still seeing occasional init problems. Here are the debug logs of the init. From my rpi_flash_dmb branch.

First try:

base 250000496 target 400000
new divisor 1252 reg 313
CONTROL1 000ee401
new clock ok
sending cmd slot=0 op=52 arg=80000c08 flags=1c00 data=0x0 blklen=0 datalen=0 timeout=1000
command timeout
cmd=52, sdmmc_req_run returned 0...
solar whale
#

I have to shovel some snow - got about 6 inches -- almost done -- I hope.

slender iron
#

good exercise! it's rainy here

#

I wonder if DMA would fix my problem

manic glacierBOT
#
tbjers
[adafruit/circuitpython] New comment on issue #5538: Sending Requests Failed on second network call in CircuitPython 7\.0\.0

My MagTag just arrived yesterday and I spent today walking through the Calendar tutorial. Glad I found this thread because I was hitting the same errors. I've downloaded the example above, but have a question about calendar_id.

In the original code from the tutorial, the calendar_id for my personal calendar was hardcoded into the code.py. Does the full working example mentioned above now have the calendar_id in the secret.py file? Or, should it go here:

now = self._iso_...
crimson ferry
#

I'm trying #5800 images on Zero W with SPI display and getting hangs and disconnects. Are there artifacts yet for the latest changes?

crimson ferry
#

oh, it does seem to continue to display correctly if I power it on the power connector and not the USB connector (for a while at least; consistently lasts longer on power-only than on USB)

manic glacierBOT
#
pmdevita
[adafruit/circuitpython] Issue opened: #5821 pulseio module missing from Adafruit Trinket M0

CircuitPython version

Adafruit CircuitPython 7.1.0 on 2021-12-28; Adafruit Trinket M0 with samd21e18
Board ID:trinket_m0

Code/REPL

import pulseio

Behavior

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

Description

Between versions 6 and 7, support for pulseio was removed from the Adafruit Trinket M0. You can also see it between the module support matrix pages for [version ...

manic glacierBOT
viscid pine
#

i have the esp32-s3-devkitc-1-n8r8 board but i'm hitting this:

E (864) psram: PSRAM ID read error: 0x00ffffff
E (864) cpu_start: Failed to init external RAM!```
#

r8 means ram, right?

#

espressif_esp32s3_devkitc_1_nopsram board seems to be working fine

manic glacierBOT
#
ladyada
[adafruit/circuitpython] Pull request review dismissed: #5819 Add Piunora and Zero\. Also, enable full build

lol at
`void board_init(void) {
videocore_framebuffer_obj_t *fb = &allocate_display_bus()->videocore;
fb->base.type = &videocore_framebuffer_type;
common_hal_videocore_framebuffer_construct(fb, 640, 480);

framebufferio_framebufferdisplay_obj_t *display = &displays[0].framebuffer_display;
display->base.type = &framebufferio_framebufferdisplay_type;
common_hal_framebufferio_framebufferdisplay_construct(
    display,
    MP_OBJ_FROM_PTR(fb),
    0...
viscid pine
#

still doesnt start cpy though

crimson ferry
#

@viscid pine CONFIG_SPIRAM_SIZE=2097152 is for the N8R2 it looks like? But jepler may have done some auto-sensing, I don't see RAM size in the Box.

viscid pine
crimson ferry
#

no decode()? ```py
Adafruit CircuitPython 7.2.0-alpha.1 on 2021-12-29; Adafruit QT Py ESP32S2 with ESP32S2

dir(b'abcdef')
['class', 'count', 'endswith', 'find', 'format', 'index', 'isalpha', 'isdigit', 'islower', 'isspace', 'isupper', 'join', 'lower', 'lstrip', 'replace', 'rfind', 'rindex', 'rsplit', 'rstrip', 'split', 'startswith', 'strip', 'upper', 'center', 'decode', 'partition', 'rpartition', 'splitlines']butpy
Adafruit CircuitPython 7.2.0-alpha.1-47-g3b083c547 on 2022-01-06; Raspberry Pi Zero W with bcm2835
dir(b'abcdef')
['count', 'endswith', 'find', 'format', 'index', 'isalpha', 'isdigit', 'islower', 'isspace', 'isupper', 'join', 'lower', 'lstrip', 'replace', 'rfind', 'rindex', 'rsplit', 'rstrip', 'split', 'startswith', 'strip', 'upper']```

viscid pine
#

got it working! had to copy some board.c stuff from the esp32 box

#

@atomic summit ^

#

like i mentioned, for octal it needed some pin stuff

onyx hinge
# viscid pine tried with the right number for 8M, still not working

yeah we decided that autodetecting SPI ram was not the thing to do. I'm glad you figured out the other pins that have to be "never reset", I had figured that out for 'box' but hadn't remembered it until I scrolled down and saw you'd discovered it too. I think the problem with WHICH pins become unusable with WHICH RAM modules is part of the reason

#

.. that we went away from autodetection

atomic summit
onyx hinge
#

please feel free to PR a board definition, so it doesn't bite anybody else.

#

@atomic summit I assume that all the n8r8 would be octal

#

did anybody start to make headway on the wifi problem with s3 yet?

atomic summit
onyx hinge
#

huh I stand corrected!

atomic summit
manic glacierBOT
onyx hinge
#

except you can get espressif parts at the moment πŸ˜€

atomic summit
slender iron
#

@crimson ferry try with my PR version where I do FULL_BUILD

silent crane
#

@slender iron thank you for the suggestion on the waveshare epaper display. It is now initing but not yet drawing correctly (display looks like snow). I'll be doing more tinkering and I'll submit a pull request once I have verified it is working. Thanks again!

manic glacierBOT
main furnace
#

CircuitPython + Pi Zero W + ThinkInk epaper bonnet. Think it will work?

manic glacierBOT
crimson ferry
#

@main furnace have you gotten any SPI stuff to work? I've been having issues, stuff kinda works but not fully

main furnace
#

Haven't tried anything (yet). The HDMI display is just too alluring.

#

What would it take to get the mu talking to it over USB serial? Mu 1.0.3 on Windows 10.

crimson ferry
#

I don't know, I've been using tio and screen on macOS and RPi OS (+ Mini PiTFT)

viscid pine
main furnace
#

Well, it's working. Black boxes on a white background. I just installed the libraries using circup and entered the code in the 2.13-in-e-in-bonnet guide.

#

All the pins I needed were present in board. I'm using Scott's 'enable full build' #7800.

#

I'm running wsl tio /dev/ttyS30, but I don't get any text back from the pi.

#

It's a little weird typing on one computer and the display is on the second computer.

#

Hmm, seems to have hung during soft reboot after driving EPD.
It's late, have to try again tomorrow.

crimson ferry
#

I got SPI ethernet working on Zero W, but SPI displayio is still quirky.

onyx hinge
#

@viscid pine same vid/pid in CP, add a special exception in tools/ci_check_duplicate_usb_vid_pid.py

warm stump
#

most unreliable connectivity, i'm constantly hearing the windows 10 usb connect and disconnect sound right now
but there it is, LUFA Bootloader USB Device under Disk Drives with 0 bytes available. Hopefully soon to be CIRCUITPY with code.py, not sure when.
that's up to people currently making 16U2's which is Microchip. Also interested to see what happens when this is used to decouple the
USB drive part of CircuitPython with the Python part of CircuitPython. CircuitDrive and MicroPython?

unknown.png
manic glacierBOT
orchid basinBOT
manic glacierBOT
#
Neradoc
[adafruit/circuitpython] Issue opened: #5823 keypad reset does not get pressed keys

The description of keypad.KeyMatrix.reset(self) states:

Any key that is already pressed at the time of this call will therefore immediately cause a new key-pressed event to occur.

However it seems that's not the case, and instead you have to wait for the time interval to pass.
I ran this code on latest on a raspberry pico with 7.1.0 with a wire connecting the pins.

Adafruit CircuitPython 7.2.0-alpha.1-54-gfb6ab7239 on 2022-01-07; Raspberry Pi Pico with rp2040

import boar...
lone axle
#

What is the best way to test a change to a library to ensure that it won't cause the library to be too big for any core builds that it is frozen into?

#

In my core repo I've gone into frozen/Adafruit_CircuitPython_APDS9960/ and changed to the PR branch using gh pr checkout then I ran make fetch-submodules. Is that sufficient to then start making builds to see if they succeed?

lone sandalBOT
orchid basinBOT
lone sandalBOT
lone sandalBOT
manic glacierBOT
#
dglaude
[adafruit/circuitpython] New comment on issue #5792: Vectorio Ignores Palette colors \(pi zero 2W\)

Dear @rsbohn I know this ticket is closed, but I did keep a link to it as an example of graphics on Pi Zero 2W.
I am testing with adafruit-circuitpython-raspberrypi_zero2w-en_US-20220107-fb6ab72.disk.img (latest).

Your code is working in the REPL, but does nothing visible (it stay in "text mode" when I run that from code.py).
And I have print("something") for debug, and my (your) code is running.
I was wondering if you had any clue on what I am doing wrong.

Thanks in advance.

viscid pine
#

and in that case, should the regular one be renamed to be clearer

crimson ferry
#

On the ESP32-S2 TFT Feather, are these both active-high?c { MP_ROM_QSTR(MP_QSTR_NEOPIXEL_POWER), MP_ROM_PTR(&pin_GPIO34) }, { MP_ROM_QSTR(MP_QSTR_I2C_TFT_POWER), MP_ROM_PTR(&pin_GPIO21) },

onyx hinge
manic glacierBOT
manic glacierBOT
#
prplz
[adafruit/circuitpython] Issue opened: #5824 ESP32\-S3\-DevKitC\-1\-N8R8

The espressif_esp32s3_devkitc_1 board definition does not work with the N8R8 devkit because it uses octal psram. I was able to get it working by coping the sdkconfig and some never_reset pins from espressif_esp32s3_box board.

I opened an issue instead of a PR because there are decisions to be made about how to name and manage these different variants (more discussion at #5678).

rigid kestrel
#

I have a wESP (which runs a ESP32-WROOM-32 and 16 MiB flash) and it'd be nice to have the option of CircuitPython instead of just MicroPython. I see https://github.com/adafruit/circuitpython/issues/1629 contains

Using the ESP32 as a co-processor is advantageous in several ways: we don't need to port CircuitPython itself to the ESP32, which is a major task

but I'm curious about the scope of that task given that there's MicroPython support for it?

unknown.png
solar whale
rigid kestrel
#

Gotcha, okay. Thank you!

manic glacierBOT
manic glacierBOT
#
CrackXT
[adafruit/circuitpython] Issue opened: #5825 Please add degree \(Sign Β°\) in terminalio\.FONT

The standard font (terminalio.FONT) does not contain the degree (Β°) sign (?) And thus it is ignored. Interestingly, there is no error if you "define" it in the code as a symbol to be displayed.

Example line of code:
cpu_label.text = 'CPU Temp. [R]: {: .1f} Β° C'.format (microcontroller.cpu.temperature)

Is then output / shown on the display as follows:
CPU Temp. [R]: 38.7C

As far as ok except for the Β° is not displayed and the C slips right away.

If I replace the terminalio.Font w...

manic glacierBOT
#
deshipu
[adafruit/circuitpython] New comment on issue #5825: Please add degree \(Sign Β°\) in terminalio\.FONT

The terminalio.FONT uses the font that is built in into the CircuitPython firmware itself. For space limits reasons, that font only includes the characters that are used in the internal messages in CircuitPython itself, because its main use is to display those messages.

You can load the same font from a file, and then all the characters will be included. The font used by default is ter-u12n.bdf.

orchid basinBOT
lone sandalBOT
manic glacierBOT
tulip sleet
lone axle
manic glacierBOT
manic glacierBOT
#
TG-Techie
[adafruit/circuitpython] Issue opened: #5828 Error in str\.format for non number formatting

CircuitPython version

Adafruit CircuitPython 7.2.0-alpha.1-58-g211dc53d4-dirty on 2022-01-10; TG-Watch with nRF52840

Code/REPL

'{:09}'.format('Jan')

Behavior

The behavior circuitpython is:

>>> '{:09}'.format('Jan')
Traceback (most recent call last):
  File "", line 1, in 
ValueError: '=' alignment not allowed in string format specifier

Description

the behavior on python 3.10 is:

>>> '{:09}'.format('Jan')...
manic glacierBOT
manic glacierBOT
onyx hinge
idle owl
idle owl
tulip sleet
onyx hinge
#

I don't know, I've never had one more than a week old be of value to me in adabot reports

lone axle
# idle owl Did you get an answer to this?

I didn't, but I saw a PR this morning for the Thermsistor library that ultimately raises the same question for me about what is the best way to test whether a library change will make certain builds with it frozen in too big.

idle owl
#

That's how I do it anyway, though I'm usually changing the Circuit Playground library, so.... only one board to test.

tulip sleet
#

I'll keep the last two months. @idle owl, do you have any use for older ones?

idle owl
crimson ferry
idle owl
lone axle
#

Thank you. The languages factor in as well don't they? Is there a specific one that I could test that is known to be the largest?

idle owl
crimson ferry
#

cool, thanks ...I get mine tomorrow πŸ™‚

idle owl
#

And yes, languages matter.

tulip sleet
#

german or japanese

lone axle
#

Thank you. I'll test both of those for the Circuit Playground builds, I'm guessing those are likely the tightest ones that would include Thermsistor library.

idle owl
#

Most likely, yes.

orchid basinBOT
lone sandalBOT
orchid basinBOT
manic glacierBOT
daring wharf
#

Hi everyone, looking at the source code of micropython... how would one go about implementing a python core? The purpose of accomplishing something like this for me is to simply educate myself. The question is, how should I start going about it? what sorta concepts do I need to wrap my head around?

tulip sleet
lone axle
#

Is there a way to make a distinction between different pins for pin alarms? i.e. can I go to sleep with PinAlarms on multiple buttons and then when one is pressed it wakes up and knows which one specifically got pressed so that it can have different behavior for each.

tulip sleet
thorny jay
idle owl
orchid basinBOT
slender iron
onyx hinge
#

ahoy all <@&356864093652516868> --- we'll be meeting in about 1 hour in this text channel and in the circuitpython voice channel. Please take a few moments to add your hug reports, status updates, and in the weeds topiccs to the document, https://docs.google.com/document/d/1f-olt8jHqYvCX-7hbH1kZF7qILHwaqk1mgiOKxqVhV8/edit?usp=sharing

Google Docs
CircuitPython Weekly for 10 January 2022

CircuitPython Weekly for 10 January 2022 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 participate,...

orchid basinBOT
onyx hinge
#

Hello! We just finished up our internal meeting, so I'll be back in 2 or 3 minutes and we'll get the Discord meeting going! <@&356864093652516868>

serene warren
#

Okay!

#

Sounds good!

onyx hinge
manic glacierBOT
turbid radish
#

SOunds great

slender iron
#

1 in the weeds topic so far

lone axle
#

https://www.youtube.com/watch?v=V2ivH2PEoiA&t=2s

YouTube
TEC_IST
Circuitpython Raspberry Pi Pico USB HID Keyboard (the smallest keyb...

If you missed the build video, check that out here: https://www.youtube.com/watch?v=iWWTJKWFNok

In this video, we walk through the Circuitpython + Raspberry Pi Pico USB HID keyboard code (matrix decoding, sending keypresses, etc.)

Code, design files, notes, etc. are available on the hackaday.io project page: https://hackaday.io/project/178204-...

β–Ά Play video
cobalt grail
#

πŸ‡ΈπŸ‡ͺ πŸ‘‹

turbid radish
#

No problem

idle owl
#

I am trying to bring my CircuitPython clone up to date, and did a git pull followed by make fetch-submodules, and I am left with ports/espressif/esp-idf still reporting modified content in git status. Tried running the make again simply to see if that would work; it did not change anything.

#

I can't either.

tulip sleet
gilded cradle
#

I think my microphone isn't working

idle owl
turbid radish
#

sudo melissa

idle owl
#

PT made it! I simply asked for it πŸ™‚

gilded cradle
#

I'll try again

turbid radish
#

sudo alsa -mic:melissa

idle owl
turbid radish
#

There are

#

Thanbks Kattni

thorny jay
#

πŸ‘Ά 🍼

turbid radish
#

Congratulations Scott - being a dad is great

manic glacierBOT
errant grail
#

Congratulations Scott! This also means we'll have development continuity for CircuitPython v2039 and beyond.

manic glacierBOT
cobalt grail
#

Gotta go. Have a nice week peeps!

gilded cradle
#

I need to head out

errant grail
#

From last week, here's the link to a resistive touchscreen calibration utility for built-in displays. The calculator project has 35 very small buttons and needed more precision than the usual calibration defaults. https://github.com/CedarGroveStudios/Touchscreen_Calibrator

GitHub
GitHub - CedarGroveStudios/Touchscreen_Calibrator: A simple Circuit...

A simple CircuitPython DisplayIO touchscreen calibrator helper for built-in displays. - GitHub - CedarGroveStudios/Touchscreen_Calibrator: A simple CircuitPython DisplayIO touchscreen calibrator he...

thorny jay
#

Bye

errant grail
#

Thanks!

lapis hemlock
#

See you!

manic glacierBOT
idle owl
#

Lagrange points are fascinating.

manic glacierBOT
onyx hinge
#

Here is the notes document for Tuesday's CircuitPython Weekly meeting. It is on TUESDAY the 18th, 24 hours later than usual, at 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/1adH6JMqjSxUkdThbXvb9EbfQHxXELwffE2QLgQ-NLmo/edit?usp=sharing

Google Docs
CircuitPython Weekly for Tuesday January 18, 2022

CircuitPython Weekly for Tuesday January 18, 2022 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 par...

onyx hinge
manic glacierBOT
manic glacierBOT
onyx hinge
idle owl
manic glacierBOT
#
CharString
[adafruit/circuitpython] New comment on pull request #5820: WIP: Waveshare RP2040\-Zero board definition

@tannewt Thanks, I found the guide and have been looking at the QT PY code and schematics (hurrah for open source hardware) as an example.
tl;dr I have Zero (no pun intended) experience with python on mcu's. So I still have planned to test the UART with an FTDI board I still have laying around from my Huzzah. And setup a pot as a voltage divider to test the ADC A0-3 pins. And try the I2C oled displays I have, following one of the Adafruit tutorials.

That said. It builds, installs and th...

lone axle
orchid basinBOT
orchid basinBOT
manic glacierBOT
#
tannewt
[adafruit/circuitpython] Issue opened: #5830 Support Pi4 second HDMI screen

From @TheMindVirus on https://github.com/tannewt/circuitpython/issues/14:

Hi Scott, not an issue as such but more of a feature suggestion.
Great work on the weekly streams for Adafruit, it shows people the kind of stuff we have to deal with as embedded developers.

I was looking at this file that you have for displayio in CircuitPython for Raspberry Pi SBC's:
https://github.com/tannewt/circuitpython/blob/rpi/ports/broadcom/bindings/videocore/Framebuffer.c
There might be mo...

manic glacierBOT
manic glacierBOT
strong rock
#

Anyone know if there are examples for the ESP32-S2 TFT Feather yet? I can't seem to find anything

crimson ferry
daring wharf
daring wharf
#

thanks a ton

tulip sleet
#

The core language part of MicroPython (and CircuitPython) is shared among all the different board ports

daring wharf
tulip sleet
#

The initial work was done by Damien George before it appeared in GitHub.

manic glacierBOT
manic glacierBOT
manic glacierBOT
#
CrackXT
[adafruit/circuitpython] New comment on issue #5825: Please add degree \(Sign Β°\) in terminalio\.FONT

Thanks @deshipu for your feedback. However, I cannot understand why I should include the already integrated font again using additional lines of code, just so that more characters are enabled (?) There. That takes away unnecessary storage space because it actually exists twice ... or I didn't understand it ;(.

With your reference to which font it is in CP, I looked "deeper" into the code and found it here
https://github.com/adafruit/circuitpython/blob/main/tools/fonts/ter-u12n.bdf

If y...

stuck elbow
manic glacierBOT
#
Thynix
[adafruit/circuitpython] Issue opened: #5833 ulab documentation mentions nonexistent \`ulab\.spectrum\`

At least as of Adafruit CircuitPython 7.1.0, places like ulab.numpy.fft documentation refer to a nonexistent "spectrum" function:

See also ~ulab.extras.spectrum, which computes the magnitude of the fft, rather than separately returning its real and imaginary parts.

I haven't been able to figure out whether it's supposed to exist, or its removal was intentional. 6.3 had the similarly-named [...

manic glacierBOT
#
arturo182
[adafruit/circuitpython] New branch created: rpi\-uart\-claim\-later
wanton topaz
#

(Apologies, I've posted this in #help-with-circuitpython but on reflection it might be better here.. I can delete either if someone wants me to)

Ahoy hoy... I was hoping someone here could help me out, I am trying to compile circuitpython for my feather_m4_can so I can add some modules to the mix. I followed the build instructions word for word, but I just keep getting the following error when I call make BOARD=feather_m4_can

I am doing this on an M1 MacBook Air - macOS Monterey - 16GB RAM

I have the source on a dmg volume set to APFS Case Sensitive

The error I get is

mkdir -p build-feather_m4_can/genhdr
GEN build-feather_m4_can/genhdr/moduledefs.h
[Errno 2] No such file or directory: 'arm-none-eabi-gcc'
[Errno 2] No such file or directory: 'arm-none-eabi-gcc'
[Errno 2] No such file or directory: 'arm-none-eabi-gcc'
<375 more lines of the same thing>
QSTR updated
/bin/sh: arm-none-eabi-gcc: command not found
ERROR: Empty preprocessor output - check for errors above
make: *** [build-feather_m4_can/genhdr/qstrdefs.enum.h] Error 1
make: *** Deleting file `build-feather_m4_can/genhdr/qstrdefs.enum.h'

the mpy-cross build seemed to work fine

blissful pollen
wanton topaz
onyx hinge
#

no, mpy-cross is a program that runs on the host computer

#

so it uses the host compiler

wanton topaz
onyx hinge
#

seems like PATH would do the trick, at least on Linux

#

(that's what I use)

#

also once you do resolve that, so that which arm-none-eabi-gcc prints a full path instead of nothing, you may need to make BOARD=boardname clean to delete incorrect files created while things weren't working right yet

wanton topaz
stuck elbow
#

you can always do echo $PATH to check your PATH variable

manic glacierBOT
manic glacierBOT
#
dhalbert
[adafruit/circuitpython] Issue opened: #5836 RP2040 busio\.SPI\(\) fails to note initial SPI frequency

CircuitPython version

7.1.0

Code/REPL

Adafruit CircuitPython 7.1.0 on 2021-12-28; Adafruit QT Py RP2040 with rp2040
>>> import busio,board
>>> spi = busio.SPI(board.SCK, board.MOSI, board.MISO)
>>> spi.frequency
0
>>> spi.configure(baudrate=50000)
Traceback (most recent call last):
  File "", line 1, in 
RuntimeError: Function requires lock
>>> spi.try_lock()
True
>>> spi.configure(baudrate=250000)
>>> spi.frequency
250000

Behavior

a...

lone sandalBOT
wanton topaz
stuck elbow
#

personally I would prefer my harvest to be blessed by Freyja, but I'm not going to complain

wanton topaz
wanton topaz
stuck elbow
#

it's a goddess of fertility after all

solar whale
manic glacierBOT
#
tannewt
[adafruit/circuitpython] New comment on pull request #5820: WIP: Waveshare RP2040\-Zero board definition

It seems that many of the pins can do that, and it's a matter of choice. And that the QT PY board definition is the way it is, because of the silk screen markings and qwiic/Stemma qt connector.

Correct, default buses are those marked by the silkscreen of the board. SDA and SCL typically for I2C, MOSI, MISO and SCK for SPI and TX and RX for UART. If those names aren't marked then there's no default bus.

I went with pin 0 and 1 for the UART because those are mentioned as default UART ...

manic glacierBOT
manic glacierBOT
idle owl
manic glacierBOT
proven garnet
#

How would we ask for a Learn guide to be update per changes in a library? Do I submit an issue somewhere or just mention it here?

idle owl
#

You can mention it here and tag me, and I can help you find the right person to get it updated.

#

(But submit it in the guide too. πŸ™‚ )

proven garnet
#

Okay, thanks!

manic glacierBOT
proven garnet
#

Well the update is for DRV2605 Learn guide, submitted throught "Feedback? Corrections?"!

#

@idle owl

idle owl
#

Thanks!

idle owl
lone axle
idle owl
#

@proven garnet I think this process will work out well. Submit through the guide, and tag me here with it. We'll get it sorted one way or another!

manic glacierBOT
#
dhalbert
[adafruit/circuitpython] Issue opened: #5838 Release stubs to pypi only when a release is done

Right now the circuitpython stubs are released to pypi on every merge build. This seems excessive, and generates a lot of pre-releases with strange version numbers in pypi: https://pypi.org/project/circuitpython-stubs/#history. I thinik it would make more sense to release the stubs to pypi only when we actually do a release, in create_website_pr.yml, which we might rename to release_tasks.yml or something like that).

Is there a good reason for doing the stub releases so frequently that...

proven garnet
#

Hooray, thanks for the guidance!

manic glacierBOT
#
wavesailor
[adafruit/circuitpython] Issue opened: #5839 New version of CircuitPython 7\.1\.0 and associated new libraries seems to have broken deep sleep on MagTag

CircuitPython version

Adafruit CircuitPython 7.1.0 on 2021-12-28; Adafruit MagTag with ESP32S2
Board ID:adafruit_magtag_2.9_grayscale

Code/REPL

import time
import terminalio
import displayio
import adafruit_imageload
from adafruit_display_text import label
from adafruit_magtag.magtag import MagTag
from secrets import secrets

# --| USER CONFIG |--------------------------
METRIC = False  # set to True for metric units
# ---------------------------...
#
atctwo
[adafruit/circuitpython] New comment on issue #5680: ESP32\-S2 latest S3 \(2021\-12\-07\) I2C crash

Hi! I've been having a similar problem, but on an ESP32-S3 (ESP32-S3-DevKitC-1 w/ 2MB PSRAM), running on tinyuf2 (although i had the same problem without tinyuf2).

I first had the problem trying to use a DS1337 (with the DS1307 library), and a PCF8523 (CircuitPython hard crashed when calling the libraries' constructors). I looked at the libraries in the issue history, and the SI7021, DS1307, PCF8523 libraries call write_then_readinto() in their constructors (which calls `writeto_then...

#
m-mcgowan
[adafruit/circuitpython] New comment on pull request #5837: Feature/larger program storage swan r5

Hi @jepler. These are good points! It had crossed my mind to target 1MB devices also, but since there were no other L4R5 ports as yet, I figured it wasn't urgently needed.

We're just testing the large user space out for now, but when it comes to submitting a PR, I'll be sure to make the linker file more explicitly named (include 2M) and include more defines that identify if the flash size is 1 or 2 MB on the STM32L4R5 in code so that it's easy to set up the board definition for 1MB devices...

manic glacierBOT
manic glacierBOT
viscid pine
#

how do i edit the board sdkconfig files? I found the menuconfig makefile task but that doesn't seem to work correctly right now because there's a lot of configs missing from esp-idf-config/sdkconfig.defaults

slender iron
#

@viscid pine I just saw that too. I think the defaults file needs to be update. I think for now just copy the unique bit into the board file

#

@atomic summit do you know of a way to enable the APP CPU instruction cache?

#

the "Application Startup Flow" page says "The duty of enabling cache for APP CPU is passed on to the application." but doesn't say how

manic glacierBOT
atomic summit
slender iron
#

k, np

atomic summit
#

Maybe something Jimmo or Damien has dealt with?

slender iron
#

I'll bug igrr tomorrow about it

manic glacierBOT
#
RetiredWizard
[adafruit/circuitpython] New comment on issue #5628: Zero 2W hangs sporadically when using USB mass storage

This seems to be happening faster on the latest build

      Adafruit CircuitPython 7.2.0-alpha.1-74-g9c6e34df0-dirty on 2022-01-11; Raspberry Pi Zero 2W with bcm2837

I've tried 6 or 7 boots now and when connecting via the serial USB connector I haven't gone more than 10 minutes without freezing and most times the serial interface hung in less than 5 minutes.

I tried a short run connected via the UART pins and that seemed to be working fine but I didn't let that test run more than...

rapid slate
#

Good evening from California... I'm here to report
"Adafruit CircuitPython 7.2.0-alpha.1-74-g9c6e34df0 on 2022-01-11; Raspberry Pi Zero W with bcm2835" seems to launch fine on the original Raspberry Pi Zero (no wifi, no camera)

crimson ferry
#

is there an SPI equivalent to i2cperipheral?

viscid pine
#

what's the difference between that and busio's i2c?

#

oh it's peripheral side instead of host

stuck elbow
manic glacierBOT
manic glacierBOT
manic glacierBOT
#
CrackXT
[adafruit/circuitpython] New comment on issue #5825: Please add degree \(Sign Β°\) in terminalio\.FONT

Thanks @tannewt for your feedback. Ok, if I understand that correctly then the code line in its link reduces the font to "ASCII-128" and then decodes to UTF-8.
visible_ascii = bytes (range (0x20, 0x7F)). decode ("utf-8")

It's kind of confused (and for me inconsistent) that the REPL then apparently gets its characters for display from somewhere else ... and is therefore actually not compliant with the code.

Example line of code:
print ('CPU Temperature [R]: {: .1f} Β° C'.format (microco...

crimson ferry
#

@stuck elbow thanks, looking now

stuck elbow
#

I'm not entirely sure it can read data with a clock

#

and it will be inefficient not using the SPI peripheral

manic glacierBOT
orchid basinBOT
manic glacierBOT
crimson ferry
#

@stuck elbow the other possibility is that SPI may be symmetrical enough with just one peripheral, one clock, fudge the CS

stuck elbow
#

no, you need the clock line

crimson ferry
#

but just one of them could generate the clock

stuck elbow
#

but the other one needs to get it

crimson ferry
#

I guess it could get out of sync

stuck elbow
#

especially with circuitpython which tends to send data in bursts

#

because it's executing python code in between

crimson ferry
#

you're right, the SPI main isn't set up to receive clock

stuck elbow
#

this is the only difference -- the source of the clock

#

the rest is identical

#

well, and handling of cs

lone axle
#

@proven garnet if you end up having a moment sometime can you take a quick look at the DRV2605 Learn Guide and let me know if you find any remaining references in there that need to be updated. I looked that over last night when I merged your PR to the repo and changed everything that I found, but would appreciate an extra set of eyes to make sure I didn't miss anything.

idle owl
#

@tidal kiln You around for a shallow dive into a schematic? Working on the TFT Feather, and I think something I copied from the OG ESP Feather isn't quite right.

tidal kiln
#

yah. sure.

idle owl
#

OK, the thing I copied is that SCL/SDA (the pins) have 5K pullups. I don't think they have any pullups, but the STEMMA QT connector has 10k pullups.

#

But good chance I'm missing something.

#

Or if the pullups are anywhere, do they apply to the whole thing?

#

I feel like there's stuff between the pullups and the rest of the SCL/SDA setup.

tidal kiln
#

rev c, right?

idle owl
#

ESP32-S2 TFT Feather Rev C is the brd/sch file name.

#

Yeah.

tidal kiln
#

one sec

idle owl
#

Thanks

tidal kiln
#

looks like 10k pullups on SCL and SDA

idle owl
#

OK, near the STEMMA connector, right?

#

Or at least on the schematic anyway

tidal kiln
#

on schematic, yah

idle owl
#

Ok

#

So if the pullups are anywhere in-line, they apply to the whole thing?

tidal kiln
idle owl
#

Right yeah

tidal kiln
#

"whole thing" little confusing. but, yah, anywhere else you see SDA or SCL, those resistors are attached

idle owl
#

Ok got it

#

Wonder if the original pinouts page is wrong, or if there were 5k on that one for some reason. Eh.

#

Fixing this up. Thanks!

tidal kiln
#

which page?

idle owl
#

Feather ESP32-S2

#

Limor said it's the same thing, except + TFT

#

So I pasted most of it from there.

tidal kiln
#

non-TFT shows 5k?

idle owl
#

Now wondering if I got that one wrong to begin with.

#

The pinouts page does

#

I don't know whether that's correct

#

looking now

#

Nope. 10k.

tidal kiln
idle owl
#

So yeah that page is wrong.

#

Fixing that too now.

tidal kiln
#

schematic shows 5k?

idle owl
#

Oh did I misread it?

#

looking closer

#

That's what I have for "Basic ESP32-S2 Feather rev c"

#

wait did we ship rev B?

tidal kiln
idle owl
#

Ohhhhhh

#

Rev B is 5k, which is the one that has the panel file in MBAdafruitBoards

#

Rev C is 10k. Which I don't think we shipped yet.

#

I guess Limor changed her mind. She changed a couple of things, I did know that. Didn't know what though.

#

Ok, fixing it back to 5k, heh.

#

For now.

tidal kiln
#

not seeing a rev mark on the non-TFT

idle owl
#

There's only one panel file in the repo, and it's for rev B.

#

So there was never a panel generated for rev C. Which is what's shipped to the PCB house, as far as I understand it.

tidal kiln
#

yah, was just seeing if i could check actual feather - looking for rev mark on the one i have here

idle owl
#

Ah. Fair enough.

#

That's odd

#

The silk in Eagle has B on it for rev B

#

in the center of the back

#

Almost dead center.

tidal kiln
#

there?

unknown.png
idle owl
#

Oof. Fair enough.

#

Β―_(ツ)_/Β―

tidal kiln
#

the two i have here are first run boards. product photos prolly same. there could've been updates since then?

idle owl
#

Indeed.

tidal kiln
#

just measured mine and they are 5k

#

but is weird that rev mark didn't show up

unknown.png
idle owl
#

Right?

tidal kiln
#

yah. if C is showing 10k. then could've been a decision change on the values? and that's same decision used for TFT version.

idle owl
#

I'm assuming she did a rev C, and then spun the TFT off of that.

tidal kiln
#

seems like

idle owl
#

Or maybe rev B of both have 5k.

#

Dunno. Didn't look.

#

Either way, it changed. πŸ™‚

stuck elbow
#

if there is no mark, it must mean it's the newest revision, right? it will magically appear when a newer revision is released ;-)

manic glacierBOT
#
dhalbert
[adafruit/circuitpython] Pull request opened: #5842 Fixes SAMD PDMIn

SAMD PDMIn has not worked right for a while.

  • The main problem was that the needed IRQ was never actually enabled.

  • Made a flag variable volatile, as it should have been.

  • Called to pdmin_reset() was commented out, maybe left over from some debugging.

  • Fixes #5797 (both on SAMx5x and SAMD21). Tested with a sound level program on CPX and with a simpler program on Metro M4.

I will probably backport this to 7.1.x, but there is maybe one other regression that should be checked ...

tulip sleet
#

I think those circled letters don't necessarily correspond to the revisions at all.

manic glacierBOT
solar whale
#

my espressif build for teh esp32s2-tft is failing ```Loading defaults file /home/jerryneedell/projects/circuitpython/ports/espressif/esp-idf-config/sdkconfig-4MB.defaults...
Loading defaults file /home/jerryneedell/projects/circuitpython/ports/espressif/boards/adafruit_feather_esp32s2_tft/sdkconfig...
CMake Error at esp-idf/tools/cmake/component.cmake:301 (message):
Include directory
'/home/jerryneedell/projects/circuitpython/ports/espressif/esp-idf/components/mbedtls/mbedtls/include'
is not a directory.
Call Stack (most recent call first):
esp-idf/tools/cmake/component.cmake:472 (__component_add_include_dirs)
esp-idf/components/mbedtls/CMakeLists.txt:10 (idf_component_register)

-- Configuring incomplete, errors occurred!
See also "/home/jerryneedell/projects/circuitpython/ports/espressif/build-adafruit_feather_esp32s2_tft/esp-idf/CMakeFiles/CMakeOutput.log".
make: *** [Makefile:302: build-adafruit_feather_esp32s2_tft/esp-idf/config/sdkconfig.h] Error 1
make: *** Deleting file 'build-adafruit_feather_esp32s2_tft/esp-idf/config/sdkconfig.h'

#

id did do the make fetch-submodules after my git pull tehn i re-installed the esp-idf

proven garnet
manic glacierBOT
#
dhalbert
[adafruit/circuitpython] New comment on pull request #5842: Fixes SAMD PDMIn

didnt test but.... is there any linter that will warn if irq code is accessing a non-volatile variable? i feel like its such a killer bug that keeps popping up in microcontroller code :/

Totally agree. I looked at the ones in https://github.com/caramelomartins/awesome-linters and also did some general searching; I could not find any checkers, but it would not be so hard as long as IRQ routines were labeled as such.

idle owl
#

@proven garnet We're going to run a fairly simple Adabot patch soon, and I wanted to make sure you're still interested in helping with any potential cleanup following the patch. Are you still up for doing that?

proven garnet
#

Yes!

idle owl
#

Excellent! I'll keep you posted on it. Thank you!

proven garnet
#

@lone axle here's what I found:

In the CircuitPython Installation of DRV2605 Library on Python & CircuitPython page, the line about what to do when not using express boards tries to give an example of one but that example is missing.

And actually that's it for what I could find!

proven garnet
crimson ferry
#

@idle owl on the TFT S2, what's board.R2? A pun because it's also board.D2?

idle owl
#

Ok, trying to build CircuitPython on an M1 Mac for the first time, the user data was copied from the previous machine, so I have my previous setup going. Failing to build now. Also, I'm not sure how many cores I have in terms of the -jXX at the end of the make BOARD. Because I was running -j12 on my previous machine, and system overview says 10 cores on this one. Which seems odd that I was using more on the previous machine?

idle owl
#

Getting a CMake error.

#

Was I pretending to have that many cores? Or is that number in the command supposed to be a multiple of the actual number of cores?

onyx hinge
idle owl
#

It's this three different times with three different compiler files or whatever ```CMake Error at esp-idf/tools/cmake/project.cmake:290 (__project):
The CMAKE_C_COMPILER:

xtensa-esp32s2-elf-gcc

is not a full path and was not found in the PATH.

Tell CMake where to find the compiler by setting either the environment
variable "CC" or the CMake cache entry CMAKE_C_COMPILER to the full path to
the compiler, or to the compiler name if it is in the PATH.
Call Stack (most recent call first):
CMakeLists.txt:12 (project)```

#

Do I need to do what it says? I don't know how to do that.

#

The other two: ```CMake Error at esp-idf/tools/cmake/project.cmake:290 (__project):
The CMAKE_CXX_COMPILER:

xtensa-esp32s2-elf-g++

is not a full path and was not found in the PATH.

Tell CMake where to find the compiler by setting either the environment
variable "CXX" or the CMake cache entry CMAKE_CXX_COMPILER to the full path
to the compiler, or to the compiler name if it is in the PATH.
Call Stack (most recent call first):
CMakeLists.txt:12 (project)

CMake Error at esp-idf/tools/cmake/project.cmake:290 (__project):
The CMAKE_ASM_COMPILER:

xtensa-esp32s2-elf-gcc

is not a full path and was not found in the PATH.

Tell CMake where to find the compiler by setting either the environment
variable "ASM" or the CMake cache entry CMAKE_ASM_COMPILER to the full path
to the compiler, or to the compiler name if it is in the PATH.
Call Stack (most recent call first):
CMakeLists.txt:12 (project)```

onyx hinge
#

Two causes for errors like that in my experience, let's find out if they're applicable .. 1. didn't do esp-idf/install.sh in a fresh terminal window (it doesn't work right if you already did step 2) 2. didn't do . esp-idf/export.sh in this terminal window

idle owl
#

I absolutely did neither of those.

onyx hinge
#

oh!

idle owl
#

Never built for ESP.

onyx hinge
#

okay

#

so in the circuitpython/ports/espressif directory, start by doing esp-idf/install.sh which SHOULD download & install some stuff. This needs to be done "just once" (or sometimes when there's a big change in the esp-idf version)

idle owl
#

Running.

onyx hinge
#

ok, if anything looks suspicious then we can circle back. otherwise, when it's done, the next command (which you'll need to run once in each new terminal window where you want to build for espressif) is . esp-idf/export.sh

idle owl
#

Oof. I ran it from a venv, and it gave me an error. Installing Python environment and packages fatal: No tags can describe '214d62b9ad859318520e38870c828278c2caee2a'. Try --always, or create some tags. WARNING: Git describe was unsuccessful: b'' ERROR: This script was called from a virtual environment, can not create a virtual environment again

onyx hinge
#

yeah it is picky about some things

idle owl
#

Run it outside the venv?

onyx hinge
#

so close that terminal, get a fresh one without a venv, and do the install.sh again

idle owl
#

ok

jaunty juniper
#

yeah it creates its own a venv basically

idle owl
#

Running again.

#

OK, it's happy.

#

What is a .fish file?

#

(My tab completion ran by .fish files)

jaunty juniper
#

it's for the fish shell

onyx hinge
#

if you use fish as your shell you may need to use export.fish instead of export.sh, I am not familiar with that.

idle owl
#

Does anyone have an answer on the cores in the make BOARD command?

#

I don't use fish.

crimson ferry
#

anyone else get safe mode on Adafruit CircuitPython 7.2.0-alpha.1 on 2021-12-29; Adafruit Feather ESP32-S2 TFT with ESP32S2 with:```py
import time
import board
from adafruit_lc709203f import LC709203F

time.sleep(1)
print("LC709203F Battery Fuel Gauge")
i2c = board.I2C()
print("LC709203F init...")
sensor = LC709203F(i2c)
print(f"IC version: {hex(sensor.ic_version)}")

onyx hinge
#

oh do -j4 for starters and it'll be good enough

idle owl
onyx hinge
idle owl
#

I'm pretty sure I was lying to it with 12 before. Not sure how I ended up there.

crimson ferry
#

oh, thanks, how did I miss that

onyx hinge
#

it's something to do with the updated esp-idf, I haven't dug into it, was hoping someone else would πŸ™‰

idle owl
#

New fail.

#

wait..

#

let me try a Thing.

onyx hinge
# 
$ python3 -c 'import multiprocessing; print(multiprocessing.cpu_count())'
4
#

you can try whatever number this prints next time

idle owl
#

Sigh. My circuitpython build stuff is in a venv, and the esp stuff is not.

#

So it's not working because dependencies aren't there in the venv.

#

But I can't get dependencies there because it won't run in the venv.

#

Does ESP not need the standard build stuff?

onyx hinge
#

I ... dunno? it's been OK for me

idle owl
#

Or do I have to install it globally. blech.

jaunty juniper
#

you should be able to pip install requirements-dev.txt in the IDF venv maybe ?

onyx hinge
#

what commands are failing for you? can you pip install them after . esp-idf/install.sh?

idle owl
#

It works without the venv.

onyx hinge
#

since you are in a venv (it just doesn't show in your prompt like most venvs do)

idle owl
#

So I guess it installed what it needed.

#

Or at least it's further than it was

#

This is chatty.

#

Fail.

#

Sigh.

#

Exception occurred. During handling of the above exception, another exception occurred... twice.

#

It's big, I can't paste all of it. Any idea what bits might be most useful?

# 
FAILED: esp-idf/esp_system/ld/sections.ld /Volumes/circuitpython.git/circuitpython/ports/espressif/build-adafruit_feather_esp32s2_tft/esp-idf/esp-idf/esp_system/ld/sections.ld```
slender iron
#

esp expects to be in it's own venv I think

idle owl
#

Then a bunch of stuff, then a traceback: Traceback (most recent call last): File "/Users/kattni/.espressif/python_env/idf4.4_py3.9_env/lib/python3.9/site-packages/pyparsing.py", line 1548, in _parseNoCache loc,tokens = self.parseImpl( instring, preloc, doActions ) File "/Users/kattni/.espressif/python_env/idf4.4_py3.9_env/lib/python3.9/site-packages/pyparsing.py", line 2622, in parseImpl if (instring[loc] == self.firstMatchChar and IndexError: string index out of range

idle owl
#

Then two more tracebacks following that one.

slender iron
#

do -j1

#

and clean and then build again

#

sometimes concurrency can mess up the build

idle owl
#

running.

#

Wow, this is the first time I've seen the difference between building with concurrency and without. Oof.

#

That difference is massive.

slender iron
#

πŸ™‚

idle owl
#

If I didn't know better, I'd think it was hanging. (It's not.)

onyx hinge
#

this isn't exactly a circuitpython question but .. anybody know how to build a .uf2'able project directly with esp-idf?

idle owl
onyx hinge
#

notices there's an idf.py uf2 target now

idle owl
#

shakes fist at concurrency fail

slender iron
#

generally you can do concurrency later. I think it's often the first things that it does that don't work well with concurrency

idle owl
#

My update worked! ESP32-S2 TFT Feather BOOT button is now an input

onyx hinge
#

wooo!

manic glacierBOT
onyx hinge
#

the uf2 files generated by idf.py uf2 don't seem to be liked by tinyuf2

slender iron
#

maybe the wrong board id?

onyx hinge
#

yes, something like that .. python3 /home/jepler/src/circuitpython/tools/uf2/utils/uf2conv.py -f 0xbfdd4eee -b 0x0000 -c -o app.uf2 build/sntp.bin this made a working uf2 file

#

not sure how to make idf.py uf2 do anything different than what it does right now πŸ˜•

slender iron
#

Β―_(ツ)_/Β―

#

@tulip sleet are you deleting all of the old pypi releases?

tulip sleet
#

I wanted to see how much work it was. It's quite tedious. I'm thinking of changing the email address temporarily so y'all won't get spammed.

manic glacierBOT
slender iron
#

I think you should just leave them

manic glacierBOT
tulip sleet
#

I will do a PR about moving the building of the stubs

slender iron
#

ya, I'm ok changing the release cadence. it's not worth changing the history though

tulip sleet
#

if it were easy I would delete them, but there's no API that I see. Getting rid of the yanked one was good because it was always at the top.

manic glacierBOT
tulip sleet
#

not to mention that poetry problem

slender iron
#

yup, agreed

manic glacierBOT
orchid basinBOT
manic glacierBOT
#
tannewt
[adafruit/circuitpython] New comment on issue #5825: Please add degree \(Sign Β°\) in terminalio\.FONT

It's kind of confused (and for me inconsistent) that the REPL then apparently gets its characters for display from somewhere else ... and is therefore actually not compliant with the code.

What do you mean? I'm not sure what you are pointing out.

I suspect that "ASCII-256" aka Extended ASCII is not a future option or is not implemented for other reasons (space)?

We could bring in more characters into builds that have a lot of flash space. However, we tend to run out of room on mo...

solar whale
manic glacierBOT
idle owl
solar whale
#

Mine was some missing folder -- not sure why my usual process did not work...but it was an easy fix

manic glacierBOT
solar whale
manic glacierBOT
manic glacierBOT
#
dhalbert
[adafruit/circuitpython] Pull request opened: #5845 Push stubs to pypi only on a release
  • Fixes #5838.

Previously, circuitpython-stubs was pushed to pypi on every merge. This generates a lot of releases in https://pypi.org/project/circuitpython-stubs/, most with strange version numbers. Instead, push stubs to pypi on when a release is done.

Stubs are still built and pushed to https://adafruit-circuit-python.s3.amazonaws.com/index.html?prefix=bin/stubs/ on every merge. I just cleaned this up to include only releases, but it will repopulate quickly enough. I will add it to...

solar whale
#

@idle owl Does the magic thing of loading an arduino sketch to the esp32s2-tft get you a free UF2 bootlaoder?

idle owl
#

I loaded Arduino on the TFT Feather, but my UF2 bootloader was ok to begin with.

solar whale
#

do they ship with UF2 -- I probalby blew mine away because I assumed it did not and I erased the Flash

idle owl
#

Ahhh. Completely recoverable. Give Arduino a shot.

solar whale
#

I will - thanks

#

@idle owl worked perfectly!

idle owl
#

Great!

manic glacierBOT
#
oliwetti
[adafruit/circuitpython] Issue opened: #5847 rotaryio pin behaviour

CircuitPython version

Adafruit CircuitPython 7.2.0-alpha.1-74-g9c6e34df0 on 2022-01-11; Raspberry Pi Pico with rp2040

Code/REPL

# uses circuitpython
from time import sleep
import board
import rotaryio

enc = rotaryio.IncrementalEncoder(board.GP10, board.GP11, divisor=1)
last_position = enc.position
while True:
    if(enc.position != last_position):
        print(enc.position)
        last_position = enc.position
        sleep(0.01)

Beha...

crimson ferry
#

Did something change with the TFT/displayio between (Feather ESP32-S2 TFT) 7.1.0 and 7.2.0-alpha.1?

solar whale
#

@onyx hinge Thanks for the quick fix to the ESP32S2-tft SPI

crimson ferry
#

does that fix the 7.1.0 issue where doing this sequence:```spi = board.SPI()
display = board.DISPLAY

#

looks like it would

onyx hinge
#

@solar whale it was quick because Ladyada ran into it the other day and I had made the fix but not PR'd it yet πŸ˜•

#

but you're welcome πŸ™‚

crimson ferry
#

will that fix get applied to 7.1.x as well as main (unless 7.2.0 is released first)?

#

(7.1.0 has spi issue, 7.2.0 has i2c issue)

proven garnet
idle owl
#

Oof. Should CP PRs still be queued for CI 3 hours after submission? I know there's a lot going on, but I want to make sure something isn't wonky.

#

Of course I ask, and it starts. πŸ€¦πŸ»β€β™€οΈ

idle owl
tulip sleet
idle owl
#

Ah fair enough. Didn't think to check there.

#

Will in the future!

#

Back into the queue. πŸ˜„

tulip sleet
#

saves time

#

maybe you know that already

idle owl
tulip sleet
#

I constantly realize there's a bug in my PR and push a change (sometimes git commit --amend and force push)

#

or there's a translation weblate PR which keeps resubmitting itself

idle owl
#

I think I do that more on libs than I do in the core. I'm always more careful in the core, I think. Feels like far more things could go wrong than with a lib.

#

Or I'm simply less comfortable with the core, so I tread more lightly.

manic glacierBOT
jaunty juniper
#

(in the cd commands and text)

manic glacierBOT
tulip sleet
manic glacierBOT
thorny jay
#

Hi, I just want to report that I tested the Pi Zero W firmware with all the hardware version I own: (1) the PiZero v1.2 that does have the camera connector (2) the PiZero v1.3 with that connector (3) the PiZero W. I did not use it, but the drive was visible on my host, the HDMI was working and displaying the REPL. So now I have two SD card, one for that family and one for the PiZero 2W. My plan is to replicate my test on both SD card to confirm the same things are working.

slender iron
#

great!

manic glacierBOT
manic glacierBOT
manic glacierBOT
#
dkulinski
[adafruit/circuitpython] Issue opened: #5850 EPaperDisplay sequences are limited to less than 128 bytes due to DELAY byte implementation

CircuitPython version

Adafruit CircuitPython 7.1.0 on 2021-12-28; Adafruit ItsyBitsy RP2040 with rp2040
Board ID:adafruit_itsybitsy_rp2040

Code/REPL

"""
`dkulinsk_ws2in9v2`
================================================================================
CircuitPython `displayio` driver for Waveshare 2.9in V2 Display
* Author(s): Daniel Kulinski
Implementation Notes
--------------------
**Hardware:**
* `WaveShare 2.9" Black and White eInk Display `_...
timber mango
#

I am curious whether there is a HIL-test framework for circuitpython available?

Here is the quick and dirty python unittest based approach that I've tested recently:

import unittest


def setUpModule():
    # unittest.TestCase.spi = Mock_SPI()
    unittest.TestCase.spi = Serial_SPI('/dev/ttyUSB0')


def tearDownModule():
    del unittest.TestCase.spi


class Test_SPI(unittest.TestCase):
    def setUp(self):
        unittest.TestCase.setUp(self)

    def tearDown(self):
        unittest.TestCase.tearDown(self)

    def test_SPI(self):
        spi = unittest.TestCase.spi
        frequency = spi.frequency()
        default_frequency = 100000
        self.assertEqual(frequency, default_frequency, f'Default frequency mismatch: {frequency} != {default_frequency}')

    def test_frequency(self):
        spi = unittest.TestCase.spi
        frequency = spi.frequency()
        default_frequency = 250000
        self.assertEqual(frequency, default_frequency, f'Default frequency mismatch: {frequency} != {default_frequency}')
#

Here is the definition of Serial_SPI class:

import serial
import time

class ISPI:
    def __init__(self):
        pass

    def configure(self, baudrate = 100000, polarity = 0, phase = 0, bits = 8):
        raise NotImplementedError('ISPI::configure() method is not implemented')
        pass

    def frequency(self):
        raise NotImplementedError('ISPI::get_frequency() method is not implemented')
        pass

class Serial_SPI(ISPI):
    def __init__(self, port, baudrate=115200, bytesize=8, parity='N', stopbits=1, timeout=None):
        self.__ser = serial.Serial(port, baudrate, bytesize, parity, stopbits, timeout)
        self.__ser.write(str('import busio, board\r\n').encode())
        time.sleep(0.2)
        print(f'[DEBUG]: {self.__ser.read_all()}')

        self.__ser.write(str('spi = busio.SPI(board.P1_20, board.P1_24, board.P1_23)\r\n').encode())
        time.sleep(0.5)
        print(f'[DEBUG]: {self.__ser.read_all()}')

    def __del__(self):
        self.__ser.write(str('spi.deinit()\r\n').encode())
        time.sleep(0.5)
        print(f'[DEBUG]: {self.__ser.read_all()}')

        self.__ser.write(str('del spi\r\n').encode())
        time.sleep(0.2)
        print(f'[DEBUG]: {self.__ser.read_all()}')

        self.__ser.write(str('del busio, board\r\n').encode())
        time.sleep(0.2)
        print(f'[DEBUG]: {self.__ser.read_all()}')
        self.__ser.close()

    def frequency(self):
        self.__ser.reset_input_buffer()

        self.__ser.write(str('spi.frequency\r\n').encode())
        time.sleep(0.2)

        reply = self.__ser.read_all()
        print(f'[DEBUG]: {reply}')
        frequency = int(reply.split(b'\r\n')[1], 10)
        return frequency
#

And here is the test console output:
`
nosetests --with-xunit --xunit-file=report.xml -s tests/circuitpython-regression/test_busio.py
[DEBUG]: b'import busio, board\r\n>>> '
[DEBUG]: b'spi = busio.SPI(board.P1_20, board.P1_24, board.P1_23)\r\n>>> '
[DEBUG]: b'spi.frequency\r\n250000\r\n>>> '
F[DEBUG]: b'spi.frequency\r\n250000\r\n>>> '
.[DEBUG]: b'spi.deinit()\r\n>>> '
[DEBUG]: b'del spi\r\n>>> '
[DEBUG]: b'del busio, board\r\n>>> '

======================================================================
FAIL: test_SPI (test_busio.Test_SPI)

Traceback (most recent call last):
File "/work/repos/external/circuitpython-bruker/tests/circuitpython-regression/test_busio.py", line 95, in test_SPI
self.assertEqual(frequency, default_frequency, f'Default frequency mismatch: {frequency} != {default_frequency}')
AssertionError: 250000 != 100000 : Default frequency mismatch: 250000 != 100000

Ran 2 tests in 2.015s
`

slender iron
#

I'm not sure what you mean by HIL-test

#

but generally we don't test as well as we should

#

on hardware would be ideal but is hard to do

timber mango
#

Hardware In the Loop

slender iron
#

ah

timber mango
slender iron
#

in the first year or so of CP I had a setup here that ran the micropython tests on hardware

#

it wasn't reliable

#

sommersoft has done some work on it since but has been busy with other stuff the last six months or so

timber mango
slender iron
#

np, I appreciate you working on CP

timber mango
#

I am glad to hear it from you!
We gained already a lot in our company by using CPy for hardware testing. This made my life much easier!
You have done a really good job!

slender iron
#

Great! I'm glad it's useful for you

#

If you have ideas on how to test CP better I'd love to hear them

timber mango
#

@slender iron BTW: Are you working night hours? Aren't you located in the US?

slender iron
#

I'm in Seattle. I'm not working now but have discord open

#

Just about to πŸ’€

timber mango
#

Fair enough. Have a good night then!

slender iron
#

good night πŸ™‚

manic glacierBOT
warm stump
#

Trying out Baremetal CircuitPython on a Raspberry Pi Zero 1 W (retroactive name change)
with a USB Hub HAT which is having some issues connecting to the repl. It got this far
because the USB Hub HAT is also a USB->UART with a CP210x and micro usb on it, a nice way to try it out.
The occasional lock up because as I understand it's new firmware, but great to see that over HDMI nonetheless.

unknown.png
warm stump
#

also a little bit of Pepper’s Ghost to go with that since it’s here set up at the same time

IMG_5163.jpg
manic glacierBOT
#
dhalbert
[adafruit/circuitpython] New comment on issue #5839: New version of CircuitPython 7\.1\.0 and associated new libraries seems to have broken deep sleep on MagTag

I also tested the program above, using 7.0.0 and 7.1.0, and got about 230uA during sleeping. I then also tested using an extremely simple deep sleep example, from https://learn.adafruit.com/deep-sleep-with-circuitpython/alarms-and-sleep#timealarm-deep-sleep-3078700-7. Again, the sleep current was about 230uA. This value is typical, and is what we first saw when deep sleep was first introduced in CircuitPython, in 6.2.0.

I am testing with a Nordic PPK2 power monitor, which simulates a batte...

slender iron
warm stump
slender iron
#

πŸ™‚ if you are ever able to reliably produce it please do file an issue

#

more info will make it easier to debug

warm stump
#

-_(\ I may give it a build and tweak at some point but correct information on broadcom stuff has been understandably scarce

slender iron
#

feel free to ask me if you have questions

crimson ferry
#

what's the DBLTAP circuit on Adafruit ESP32-S2 boards for? It doesn't seem to get used in hardware. RFU, or some software stores a charge there?

manic glacierBOT
manic glacierBOT
pallid echo
#

Hi, I saw MicroPython has support for ESP8266. I happen to have a Huzzah. Any reason why there's no CircuitPython support for it?

slender iron
#

it doesn't have native USB

pallid echo
#

I see.. Makes sense, given the target audience.

slender iron
#

(we did until version 3 and many folks expected the CIRCUITPY drive to show)

orchid basinBOT
idle owl
orchid basinBOT
manic glacierBOT
manic glacierBOT
manic glacierBOT
#
jepler
[adafruit/circuitpython] Pull request opened: #5852 Add floppyio

Initially enabled for samd51, this enables reading raw flux data as well as DOS/MFM formatted media.

This is only the low-level code for reading & decoding flux pulses from a floppy drive. high level details will live in a Python library.

adafruit-circuitpython-floppy will take care of details like stepping from track to track, etc.

misty garnet
#

how difficult would it be to modify PWMOut or PulseIO to send a specific # of pulses at a specific frequency?

slender iron
#

@misty garnet I think pulseout can do that now at the cost of ram

#

how many pulses?

misty garnet
#

more than 60ms worth

#

use case is driving a stepper motor

#

I tried chaining blocks of PulseIO together but it gets kinda bumpy

slender iron
#

ya, one big buffer would probably work best

#

some platforms use dma to do it iirc

manic glacierBOT
misty garnet
#

is there a way to gracefully stop the pulses from a big buffer, in case you hit a limit switch for exampole

slender iron
#

right now CP will block until the transmit is done

stuck elbow
#

a dedicated stepperio module might make sense at some point

misty garnet
#

yeah, that's what I remembered, which is why I used shorter pulse windows and checked the limit switch in between

#

so I use pwm, which is more async

stuck elbow
#

if it's an rp2040, you might be able to use the pio for this

slender iron
#

you might be able to do it with audiopwmout

misty garnet
#

yes, I thought about the pico

slender iron
#

the audio stuff will be non-blocking

misty garnet
#

yeah, that has potential as long as I have enough memory to stuff the pulses

stuck elbow
#

and you can loop it

manic glacierBOT
manic glacierBOT
manic glacierBOT
misty garnet
#

yeah, as long as looping doesn't stutter too badly might work

manic glacierBOT
manic glacierBOT
lone axle
manic glacierBOT
onyx hinge
#

@lone axle I'd recommend verifying with @idle owl but my understanding is you can do the steps inside Read The Docs, but then add the 'adabot' user as a co-maintainer which I don't see in the guide...

unknown.png
lone axle
#

Thank you. That looks right to me, I'll wait for confirmation though. I poked around a little last night trying to figure out the right name of the account but didn't think to try adabot.

manic glacierBOT
manic glacierBOT
#
dhalbert
[adafruit/circuitpython] New tag created: 7\.1\.1
solar whale
#

@tulip sleet not meaning to be nit-picking... in the release status CircuitPython has a number of "ports" that are the core implementations for different microcontroller families. Stability varies on a per-port basis. As of this release, atmel-samd, cxd56 (Spresense), esp32s2, nrf,raspberrypi, stm for the F4 family are stable. stm for other STM chip families is being actively improved but may be missing functionality and have bugs. broadcom (Raspberry Pi), litex and mimxrt10xx are in an alpha state and will have bugs and missing functionality. should there be a mention of esp32s3? Is it also considered alpha at this time?

tulip sleet
solar whale
#

Thanks -- I happened to think of it because I just tried it out. Still crashes on import wifi

orchid basinBOT
tulip sleet
solar whale
#

Sorry -- I was not really paying attention to the release being 7.1.1 -- I was testing 7.2.0-alpha

#

Was s3 even in 7.1.1?

proven garnet
#

I'll upload verification to the issue for reference

marsh stone
#

I started posting over in the #help-with-circuitpython page but I was looking for info on how to start with a new ESP32-C3 board I designed.

patent peak
solar whale
tulip sleet
#

right, maybe we should clean that part out, though fixing that minor thing might be ok

solar whale
#

Where are the ReadTheDocs files stored? -- Is it something anyone can submit a PR to in order to edit?

tulip sleet
#

There is a link in the upper right of the docs page that points to the doc source

idle owl
#

Checking the guide now.

#

@lone axle The guide covers adding adabot. It also covers adding it as a subproject to the CircuitPython project, for which you will need access to the CP project, if you do not already have it. Let me know.

solar whale
solar whale
tulip sleet
solar whale
#

I'll try it

idle owl
#

@proven garnet Wanted to give you an update. A series of patches are going to be run incrementally to try to catch as many libraries as possible (I can explain what this means if you're interested, but for now, I'll keep it simple), but once that's done, I'll have Eva put together a list of what's left to be done once the patches are completed and we can go over it then.

manic glacierBOT
#
jerryneedell
[adafruit/circuitpython] New branch created: jerryneedell\-patch\-1
solar whale
#

Ugh -- I guess it triggers a full CI build ... sorry about that -- argh -- and I even made a typo ...sigh

tulip sleet
#

how about we just close that PR and I'll cancel the actions?

#

you can do a proper one from your fork

solar whale
#

yes -- go ahead -- sorry about the churn

idle owl
#

I can make the change if that would help.

#

My CP environment is still open from my last change

solar whale
#

@idle owl are you talking to me?

tulip sleet
#

@marsh stone we don't have a writeup yet, because it's so early

idle owl
solar whale
#

Yes -- please -- go ahead and make that change if you don't mind.

idle owl
#

I don't mind at all.

#

Still have everything up from adding BUTTON pins to ESP32-S2 boards.

marsh stone
#

@tulip sleet I guess that explains why I couldn't find information on how to start.

manic glacierBOT
idle owl
#

@solar whale ^^

manic glacierBOT
slender iron
marsh stone
#

@slender iron Thanks, I saw that page but I don't see what to flash to a blank chip/board

slender iron
#

I think you can just flash CP

marsh stone
#

I know there's this, https://circuitpython.org/board/ai_thinker_esp32-c3s/ so do I just use esptool and flash? If so do I need any memory starting address?

CircuitPython - ESP-C3-32S Download

This is an entry-level development board based on Espressif ESP32-C3 SoC, which is equipped with a RISC-V 32-bit single-core processor, operating frequency up to 160 MHz, supports secondary development without using other microcontrollers or processors. The ESP32-C3 is an highly integrated low po...

#

It also sounds like the native USB port on the C3 isn't used for much in CP other than perhaps peripherals attached to USB, at least according to this statement:

A USB to UART convertor can be used for connecting to ESP32-C3 to get serial console, REPL interface and flashing CircuitPython

jaunty juniper
#

flash it at address 0 I asssume, no harm in trying

slender iron
#

and the c3 doesn't have native full usb

#

so you won't get a CIRCUITPY drive

#

or HID or MIDI

marsh stone
#

Hmm, I would have thought there's something that says what the first steps are

slender iron
#

it's not supported yet

marsh stone
#

Seems strange that the REPL doesn't come through on the USB though

slender iron
#

ya, I'd expect it to

#

@analog bridge is the one who added it

marsh stone
#

it's used in Arduino if I remember correctly

#

I thought I'd mess around with it and try things out but it seems like perhaps it would be on the bleeding edge and I'm not sure I've got enough spare blood to give!

manic glacierBOT
orchid basinBOT
manic glacierBOT
tulip sleet
#

https://blog.adafruit.com/2022/01/14/circuitpython-7-1-1-released/

Fixes a couple of specific bugs; otherwise unchanged. No need to upgrade in general unless you want those fixes.

Adafruit Industries - Makers, hackers, artists, designers and engineers!
CircuitPython 7.1.1 Released! @circuitpython

This is CircuitPython 7.1.1, the latest bugfix revision of CircuitPython, and is a new stable release. It has fixes for RGBMatrix on RP2040 and PDMIn (microphone input) on SAMx boards, such as Circ…

manic glacierBOT
orchid basinBOT
gilded cradle
idle owl
manic glacierBOT
idle owl
#

For those doing CircuitPython development on MacOS, what size is your development disk image? I feel like Broadcom has added a significant amount of data to the repo, and I'm wondering if I need to generate a new disk for it.

slender iron
#

did you do make fetch-submodules? broadcom will be 10G+ otherwise

idle owl
#

I did do that.

#

But it still takes ages to fetch the broadcom port.

#

It's not filled up, but I'd rather not decide to make a new disk when it fills up, since that's probably going to happen when I need to be doing other things, heh. Wanted to stay ahead of it, if it's needed.

slender iron
#

you could just deinit the broadcom thing

#

until you want to build broadcom

idle owl
#

My disk image is 10GB.

#

Hmm, fair enough. Does that not cause issues with updating? I run into problems updating the repo as it is.

slender iron
#

make fetch-submodules may re-init for you. I'm not sure

idle owl
#

Yeah, I'd rather avoid that. It's fine.

slender iron
#

my ports/broadcom folder is 1.4G

idle owl
#

I have the space. Simply wondered if I should make a new image for it.

slender iron
#

my whole CP folder is 8.9

idle owl
#

And you're doing broadcom dev, so 10 is probably ok for now for me.

slender iron
#

my image ended up getting hosed on my mac

idle owl
#

Oof

slender iron
#

not fun...

#

so only have cp in the image

#

I had all my repos...

idle owl
#

That's all I ever had in it

#

Oi...

#

The rest of my repos are local.

#

Never occurred to me to put them all in an image. I thought it was a CircuitPython-specific issue requiring the image.

tulip sleet
#

i thought it was for case-sensitivity during the build

idle owl
#

That sounds like what I remember.

gilded cradle
#

It looks like my circuitpython folder is almost 22GB.

idle owl
#

And it only has the circuitpython repo in it?

gilded cradle
#

yes

idle owl
#

Huh.

#

That seems odd.

#

But I obviously don't know πŸ˜„

#

(Or I wouldn't be asking.)

gilded cradle
#

I think that has all submodules in it

tulip sleet
#

you may have a lot of builds in it

slender iron
#

that could be if you got all of broadcom

gilded cradle
#

I probably did

idle owl
#

Ah so if I ever want to build broadcom, it could get that beefy?

#

Or can you build broadcom without all of it

slender iron
#

no, we used to fetch all the history you don't need

idle owl
#

Right ok

#

I vaguely remember reading that conversation.

gilded cradle
#

I mean I still have around 800GB of free disk space, so I'm not too concerned

idle owl
#

Yeah I have probably twice that. Which is why I figured make a bigger image for CP if necessary. But it sounds like it might not be necessary for now.

#

Thanks all!

gilded cradle
#

πŸ‘

orchid basinBOT
tulip sleet
#

@slender iron @idle owl @gilded cradle FYI, a fresh clone of circuitpython with no submodules is 179MB. With a full clone of submodules(not make fetch-submodules), it is 16GB. broadcom fetches 13.71GiB from the server, and that's before it sets anything up.

#

and that's not counting fetching the espressif dependencies, etc.

gilded cradle
#

I could try blowing it away and redownloading

tulip sleet
#

no reason to, if you're not bumping into something

gilded cradle
#

ok

solar whale
#

If I recall correctly, you only need the case-sensitive file system to build the espressif sdk -- not to build CP.

idle owl
idle owl
solar whale
#

It began with esp8266 support

idle owl
#

Ahh.

solar whale
#

but I stopped using one a long time ago -- I'm recloning and trying it now

#

I just cloned and updated -- 1.8 G on my M1 Mac -- trying espressif build now -- standard files system - not case-sensitive

idle wharf
#

@idle owl I meant to put this into my 2022 thoughts and forgot.
I was wondering if there should be a circuitpython-libraries channel specifically for library maintainers\supporters, design discussions, cookiecutter, etc.
Maybe which also has Repos linked (that might be too noisy).

idle owl
# idle wharf <@!330227457296957440> I meant to put this into my 2022 thoughts and forgot. I ...

That would be entirely too noisy πŸ˜„ As for a new channel, it's not a bad idea, but I worry about diluting thing too much, and/or creating another vector for support/dev folks to have to monitor. But I'm not completely against it. I'll think about it. And keep an eye on this channel to see if there's enough chatter to justify breaking it out. I also worry about folks thinking that's where you ask questions about using the libraries, meaning all of the support folks would have to keep a close eye on it to provide help.

idle wharf
#

"I also worry about folks thinking that's where you ask questions about using the libraries" that's a good point

idle owl
#

Since there's already confusion about the dev channel vs the help-with channel.

#

Though adding -dev helped that a lot.

solar whale
#

@idle owl just built qtpy esp32s2 on my M1 Mac in standard file-ssytem -- so I "think" it is ok in general.

idle owl
idle owl
solar whale
#

If I recall correctly it was a holdover -- you need it to build some espressif tools locally but we don't do that duing a CP build.

manic glacierBOT
#
jepler
[adafruit/circuitpython] New comment on issue #5680: ESP32\-S2 latest S3 \(2021\-12\-07\) I2C crash

Investigating this further I found https://github.com/espressif/esp-idf/issues/7405

Per the comments there, our sequence of API calls in common_hal_busio_write is not acceptable in esp-idf, because we omit the stop call for write_then_readinto according to the transmit_stop_bit flag:

    if (transmit_stop_bit) {
        i2c_master_stop(cmd);
    }                                                                           
    esp_err_t result = i2c_master_cmd_begin(self->i2c_num,...
idle owl
#

Huh. Fair enough.

#

I mean, it's fine, and I'm used to it being in an image at this point. But still good to know I guess.

#

Would mean not having to guess at an image size or waste space on an image though.

solar whale
#

It "does no harm" to use a case-sensitive FS, but I don't think itis necessary. and it is a pain to maintain.

#

also built esp32s3-n8r2 and loaded onto a board to verify it actually works πŸ˜‰ All OK

idle owl
#

Hah good plan. Thanks!

solar whale
onyx hinge
#

how/when is shared-module/busio/I2C.c used? It looks like it 'adapts' busio to use bitbangio... but I couldn't find what enables/disables it.

manic glacierBOT
slender iron
#

@onyx hinge it might be a holdover

lone axle
idle owl
#

@lone axle You should have GitHub access, give it a try. I added you to the CircuitPython project, you should be able to add it there too.

lone axle
idle owl
#

Can you DM me a screenshot of what you're seeing on GitHub?

lone axle
#

Possibly the settings tab is admin permission above write access. Yep. will do.

idle owl
#

Ok, and you're for sure logged in?

#

That's odd.

#

I can add it.

lone axle
#

and yep. definitely logged in.

#

Thank you. I'll add it as a subproject in CircuitPython RTD project.

manic glacierBOT
#
ThomasAtBBTF
[adafruit/circuitpython] New comment on issue #4639: Unable to send HIDService keyboard events to specific connections

I am "suffering" from the same problem as described initially by ATMakesBill.
We are trying to make a BrailleKeyboard.
While investigating the problem and trying to find a solution I think there are two very different ways to achieve what we want:

  1. Checking the connections and disconnecting the "undesired" centrals. This would require a API which is able to identify the central at the "other side" of the connection.
  2. Changing the MAC Address of the peripheral to different values and ...
idle owl
lone axle
#

I think it's building now. I left the default branch blank originally and it seemed to try the wrong one first. Just set it to main though and it seems to be going now. Thanks again!

idle owl
#

Set manually.

#

RTD isn't quite caught up with main yet.

lone axle
#

Not directly related. But I noticed the change for some URLs to use docs.circuitpython.org/projects...

Do you have interest in changing the links in the existing repos to point there?

idle owl
#

Whatever's left, Tekktrik wanted to help with, so the final cleanup will be on his plate.

lone axle
#

Nice. Sounds good.

idle owl
#

Difficult to patch things that have repo-specific text on the same line, which much of the docs links do. So that's a sticking point for this update.

manic glacierBOT
lone axle
idle owl
#

Um..

#

checking.

manic glacierBOT
idle owl
#

I think you're seeing one of the alternate docs links.

#

As long as the right one works, the alternates are irrelevant.

lone axle
#

Oh I do see it now on the right link. It was ASCII art hallway not found thing when I looked before.

idle owl
#

Takes a few minutes to propagate sometimes.

manic glacierBOT
idle owl
#

For future reference, the alias should be lowercase though.

#

Lib name minus Adafruit_CircuitPython_ and all lowercase.

manic glacierBOT
lone axle
# idle owl Lib name minus `Adafruit_CircuitPython_` and all lowercase.

I think I may have done the wrong name initially too. Without "Adafruit" in it. I believe I had "CircuitPython 24lc32" originally and it seems that it created the project slug from that and set it to "circuitpython-24lc32" instead of "adafruit-circuitpython-24lc32" which causes the badge image in the readme file to come back with the unknown badge instead of passing. It looks like it may not be easy to change the slug. You think it's okay to PR the readme to use the current slug? or better to re-create or try to send a request to them to change the slug? Apologies for all the trouble, I did not read the guide text closely enough and followed the screenshots a little to blindly.

manic glacierBOT
#
rsbohn
[adafruit/circuitpython] Issue opened: #5861 Bluetooth workflow: Can't connect to nRF52840

CircuitPython version

Adafruit CircuitPython 7.1.0 on 2021-12-28; Adafruit Feather Bluefruit Sense with nRF52840

Code/REPL

n/a

Behavior

When I try to "Request Bluetooth Device" from Chrome (code.circuitpython.org)
I pick the 'CIRC' device and click 'Pair', then I get a Bluetooth busy spinner that never goes away.

Description

Chrome 96.0.4664.110 (Official Build) (64-bit)
--enable-experimental-web-platform-features
--enable-features=Web...

idle owl
manic glacierBOT
idle owl
#

@lone axle Are you alright with doing that?

#

Let me know so I can delete and redo the webhook when it's set. Otherwise, I can redo the project.

lone axle
manic glacierBOT
#
v923z
[adafruit/circuitpython] New comment on pull request #5859: update to ulab 4\.0\.0

Does this change any APIs?

@v923z can you take a moment to answer?

I know it optionally introduces new APIs, but the big thing (complex numbers) remains disabled in CP.

If complex is disabled, then there isn't any difference. There are two new functions, but you can disable them in the header file.

I don't know if you want to resolve this https://github.com/v923z/micropython-ulab/issues/435 before merging into CP.

lone axle
idle owl
lone axle
idle owl
#

Should be good to go!

lone axle
#

Thank you for helping me get it straightened out and set up!

idle owl
#

You're welcome! Thanks for doing it!

proven garnet
manic glacierBOT
#
wavesailor
[adafruit/circuitpython] New comment on issue #5839: New version of CircuitPython 7\.1\.0 and associated new libraries seems to have broken deep sleep on MagTag

This is strange now - I went back and reloaded the code from the Weather App in code.py and reset it.
It seems to run fine now with the original code from the Weather App.

I have my own version where I added and changed a few things which is where I first discovered the problem. So I'm loading my code into code.py and testing it out. I'll give feedback over the next few days.

PS. What do you use to measure the current draw?

manic glacierBOT
manic glacierBOT
manic glacierBOT
viscid pine
#

^ working on my board πŸ‘

manic glacierBOT
orchid basinBOT
silent crane
#

@slender iron Thank you for the help, this is now working!

dgwQNxcno2OircWM5vuBwNu6GQC0T-MTWeeR91zrOY1fMMErc_Nl7bXYqSbN7NDzfhi7x6J-VTdjquBMzGbWGpK5ypfdQcWD4mVGMq8gRuUaw3WR8SkgmhjaItkBWNKA75gfYRE-4AM051vXm_lovwE6cGL-ZjBdSjOP14DhjvArEUrTCkrhCfLPXgznE923QK162XTy-A_AYW-m31udKoHHK8Yyo2DK_lZb2kpRHinxJXWOA8UvkItsvdk3wnTuUGhj3RxERWokvd9Wb2jAAxFdebSIO5ON4QDxmkqL3hi0ur1Qq_Z1kMLi7y8Car4xAvqOUavqpdIZ2Jz_7dTAkN7pQody3YSKs6N2WsS_ymkkUT3sLR1Y3rDmaxjtZoi48JvYCF9J3kvdrVmz76XwHOR12mSaz2sH8tDlrXo_aj50ZmkrO3mW2yHWeA54A5KeV0FdSGnzyszQG0vCQ5iEZEJoCHHpc49GrZCmOSEMWE0KDky3yld8XNumKjcM..png
warm stump
#

broadcom port is downloading overnight, I want to try applying "step-over" behaviour for the interrupts to see if it helps with the lock-ups on Pi Zero W.
It's something I've already done before that helped with instructions that don't work when the MMU is switched off,
documented here: https://github.com/themindvirus/pix-ies/tree/step-over for anyone interested or who has already done the same
There will be a different Interrupt Vector Table between the original Zero and the Pi 4 so there is also that

manic glacierBOT
slender iron
solar whale
#

Yay! first successful wifi connection in esp32s3! Thank you all for getting it working!

warm stump
manic glacierBOT
#
TheMindVirus
[adafruit/circuitpython] Issue opened: #5864 Hug Report for CircuitPython BareMetal on Raspberry Pi Zero W and Interrupt Step\-Over

CircuitPython version

Adafruit CircuitPython 7.2.0-alpha.1-95-g936c9b205n 2022-01-13;
Raspberry Pi Zero W with bcm2835;
BoardID:raspberrypi_zero_w

Code/REPL

print("Hello World!")
input(">>>")

Behavior

REPL Lock-up with no errors or traceback, potentially ARM1176 Interrupt being called and returning to the same place in a loop.
The stack trace and other information matches that for the Pi Zero 2W: [Deep Dive w/Scott: Pi Zero 2W + CircuitPython...

lone sandalBOT
orchid basinBOT
proven garnet
#

@idle owl I was looking at things that still use gamepadshift and found some projects/examples that do. Since it's being depreciated, I submitted a PR for one example that uses keypad instead. If that's something that's wanted for other examples let me know but I also understand if the idea is to let that code live on CircuitPython pre-8

manic glacierBOT
#
TheMindVirus
[adafruit/circuitpython] New comment on issue #5864: Hug Report for CircuitPython BareMetal on Raspberry Pi Zero W and Interrupt Step\-Over

Additional required steps for Debian WSL (Windows Subsystem for Linux) include but are not limited to the following:

sudo apt update # Update Debian repositories list
apt-cache search # Find missing packages (information also available at Debian's websites)

sudo apt-get install build-essential git make # Base tools for development
# Additional cross-compilers such as arm-none-eabi-gcc may also be required

sudo apt-get install python3-pip # For installing python huffman module
s...
manic glacierBOT
manic glacierBOT
manic glacierBOT
#
TheMindVirus
[adafruit/circuitpython] New comment on issue #5864: Hug Report for CircuitPython BareMetal on Raspberry Pi Zero W and Interrupt Step\-Over

I'm closing this issue for now because it seems that the most recent main branch is working as expected.
However, there still remains the issue of it being updated on Amazon S3 which the Learn Guide is currently pointing to.

ItsAlive

Both this and the CircuitPython build for the Adafruit Macropad RP2040 will help in creating an AVR8 port
for the ATMEGA16U2 on the Arduino Uno and US...

manic glacierBOT
manic glacierBOT