#circuitpython-dev

1 messages Β· Page 71 of 1

stuck elbow
#

which is effectively the same as --break-system-packages

tulip sleet
#

I think I considered --user but it wasn't going to work? I could try a test run of that.

stuck elbow
#

this is the pep

#

This applies both to system-wide installs (sudo pip install) as well as user home directory installs (pip install --user), since packages in either location show up on the sys.path of /usr/bin/python3.

#

you are right about --user

tulip sleet
#

ok, now I re-remember, thanks!

stuck elbow
#

I hate it when project politics take precedence over user needs

lone axle
tulip sleet
#

I clean that up from time to time

lone axle
#

Okay thank you. Do you know how circuitpython-build-tools chooses which version to use when it runs in actions? Is there a specific point at which that will switch to using 9.x?

tulip sleet
#

I thought it was using 9.0.0-alpha.something?

#

it's in a .github file somewhere, I think

#

I would do a repo or org search for mpy-cross in Github

lone axle
#

maybe because it builds both 8.x and 9.x right now. However I don't see any other s3 links with a 9.x version in the URL in that actiosn log.

tulip sleet
#

the failure is actually bad python?

Traceback (most recent call last):
  File "/tmp/tmpv_f3cj59", line 127
SyntaxError: invalid syntax
lone axle
#

the mpy-cross used doesn't support f-string with repr format f"Unsupported image format {magic_number!r}" seemingly

tulip sleet
lone axle
#

(I have not tried a 9.x mpy-cross yet but I am about to)

tulip sleet
#

i think micropython does not support that syntax

#

or only started to do so recently

lone axle
#

Ahh, I see. Yeah it appears the 9.x mpy-cross does compile it successfully with the f-string and !r. I have not tested it specifically on hardware, but was just thinking maybe that it's not so much that mpy-cross doesn't / didn't support it, but rather micropython itself.

tulip sleet
#

you can use repr(...) instead, it's just an alternate syntax. Since we want the library to still work with 8.x, I think that's the right thing to do

lone axle
#

So we definitely will want to keep the 8.x compatibile version until whenever we make the final bundle build for 8.x

tulip sleet
#

yes, exactly

lone sandalBOT
tulip sleet
slender iron
#

I think it's simplest to say you fully light sleep only if you've never received a character from the usb to serial chip

#

we have a notion of pretend deep sleep for a similar purpose

cunning sundial
#

It's possible to make a UF2 file based on your current setup so that it can be flashed on other board with everything included, correct? (e.g. code, libraries, additional files)?

devout jolt
cunning sundial
#

Is what I have

nimble sparrow
#

@devout jolt fancy giving this vgm class a go for me?

devout jolt
# cunning sundial Thanks - is is possible on other devices as well (in theory at least)? Trying to...

On SAMD21, yes. If you boot into UF2 mode, then the "CURRENT.UF2" file will contain the entire flash contents.
All other architectures, no. I believe you only get the internal flash of the chip and not the SPI flash. So on an ESP32-S2 board, CURRENT.UF2 only contains the 128kB internal flash and not the 8MB (or whatever) SPI flash. The TinyUF2 bootloader could be written to read out the external flash (and I wish it would!), but that would make it more complex

cunning sundial
#

Ok, thank you!

devout jolt
nimble sparrow
#

Just spent a few minutes extricating the bits and pieces that make it depend on my wider code

vital reef
slender iron
tulip sleet
#

@devout jolt foamyguy is working on wwshell. It's been mentioned in discord

devout jolt
midnight ember
#

much better this week

#

sounding much louder and clearer πŸ‘

turbid radish
#

test

quiet scroll
#

hello devs

nimble sparrow
quiet scroll
#

need help wiht 2 missing drivers for cirpy

midnight ember
#

We're starting the circuit python meeting now. All text visible in this channel will be on youtube.

quiet scroll
#

driver for the si570 and for xpy2046 touch

#

ok

#

post link to channel pls

midnight ember
slender iron
midnight ember
#

you can listen in by joining the CircuitPython audio channel or wait until the video is released later.

lone axle
midnight ember
#

#CircuitPythonDay2024 (just to put it out there)

#

appreciate all the effort you put in every week @turbid radish always cool stuff to read in the newsletter.

#

text only please

slender iron
#

@gilded cradle mute please

midnight ember
#

very very very low paul, look into your mic settings.

gilded cradle
#

Thanks @slender iron

midnight ember
slender iron
#

tony started it

#

then I took it over

midnight ember
#

bill88t or a different bill?

slender iron
#

phil, pt

midnight ember
#

oh phil my bad

turbid radish
#

Congratulations!

midnight ember
#

🐱

#

asyncio with RFM? awesome!

#

i have a few rfm projects i can test with sure.

slender iron
#

I have a tab open for the changes

gilded cradle
midnight ember
#

recommended boards to use with it, any boards you'd like tested in particular? or doesn't matter?

gilded cradle
#

Any boards

solar whale
#

OOps -- taht was not meant for me.... sorry

midnight ember
#

well it was a good question for you too since i only have rfm featherwings i think.

dense glacier
#

I have RFM9x M0 which is too small unfortunately. Cannot even fit 9.1 plus RFM9x library

solar whale
#

An rfm feather wing on anything except a feather M0 should be fine.

midnight ember
#

didn't know m0 can't do asyncio so that's good info.

tulip sleet
#

just not enough room in the firmware

random junco
midnight ember
#

I definitely intend on using your cookie cutter web form. Just a lot of projects ahead of it to be done with imageload.

#

It seems like a much more friendly way to get a cookie cutter thing going. The current cookie process is absolutely daunting. The web form looks sooo much easier to use.

solar whale
midnight ember
#

The way some of the issue forms are formatted is very easy to use too. I'm ok with that too.

gilded cradle
#

You can use JSZip

stuck elbow
#

I wonder if cookiecutter could work in pyscript

midnight ember
#

I've held back on contributing libraries to cookie cutter because it's so daunting and i just don't want to deal with it. The complexity and inconvenience is a barrier.

short tendon
#

I have seen some repos run pre-commit as a github task and then committing the changes

midnight ember
#

I can do pre-commit, pylint, black, or ruff. Those are a bit easier to figure out than cookie cutter as it is now.

lone axle
midnight ember
#

oh so would be looking at an action eventually anyway.

gilded cradle
#

The advantage to using Jinja would be that we could keep all of the same source files.

midnight ember
#

As long as a repo is formatted to a normal standard that most other libraries use, I'd love to just point adabot or something at my library repo and it does it all for you.

#

Just in general need something easier to use.

#

Great meeting everyone. Thank you for hosting @tulip sleet blinka_heart

errant grail
#

Thanks all.

solar whale
#

πŸ‘‹

random junco
#

thanks everybody!

lone axle
#

Thanks for hosting Dan, Have a great week everyone.

midnight ember
#

oddly enough this is actually one of the libraries that i'll probably test out foamyguy's cookie cutter web form with.

quiet scroll
#

yes

#

the si570 is a single clock chip like the si5351

#

ok I will test this out as we are working on developing a vfo interface for a 2meter ssb radio and it will have a touch enu

#

when done

#

enu/menu

#

i have the base of a py driver I will find it and post fer further help

#

but I have to say learning circuitpython in 1 week has been fun .

manic glacierBOT
#
mMerlin
[adafruit/circuitpython] Issue opened: #9481 SocketPool requirements and documentation

CircuitPython version

Adafruit CircuitPython 9.0.5 on 2024-05-22; Adafruit-Qualia-S3-RGB666 with ESP32S3
Β«Using 9.0.5 instead of 9.1.1 due to display issues (jitter) on the newer version for that boardΒ»

Code/REPL

import os
import time
import socketpool
import wifi
print(f'{wifi.radio.connected =}')
pool1 = socketpool.SocketPool(wifi.radio)
addr_info = pool1.getaddrinfo('0.adafruit.pool.ntp.org', 123)
print(f'pool1 {addr_info =}')
pool2 = socketpool...
quiet scroll
tulip sleet
# quiet scroll the si570 is a single clock chip like the si5351

I found several Arduino and Python libraries that support Si570, so I think it will be a simple task to port the functionality of those to CircuitPython. I don't know the details of the differences between Si570 and Si5351 but you could probably use the SI5351 CircuitPython library as a base.

midnight ember
#

specifically the change to FourWire

quiet scroll
#

ok

midnight ember
#

my driver also allows for display rotation independent or aligned with the rotation of the touch x/y 0,0 point.

#

some minor updates that make it a little easier to use in landscape or portrait mode.

quiet scroll
#

ok

#

thanks

midnight ember
#

I coded it to use the ST7796S display. I've not used it with any other display so any feedback is appreciated.

quiet scroll
#

ok we are using the ili 2.8

#

with touch

midnight ember
#

similiar enough should be pretty easy to drop in the ili driver and just go... hopefully. 😬

quiet scroll
#

ok

#

back to code issues

acoustic sapphire
#

awesomeeee

#

lol didnt know this existed i mainly use it for programming my picos and rp2040 stuff

slender iron
#

@random junco just listened to the bootloader. My hope is that circuitmatter will run on all wifi boards that support ipv6. You don't need BLE for it (it is only used if the device isn't on the wifi already.)

tulip sleet
#

@Here is the notes document for next Monday’s CircuitPython Weekly Meeting. It is at the normal time of 11am Pacific / 2pm US Eastern here on Discord. Add your hug reports and status updates to the document before the meeting. If you are unable to attend but would still like to contribute, feel free to add your notes and we’ll read them off during the meeting. Hope to see you there! <@&356864093652516868> https://docs.google.com/document/d/1izl2ShgUGlfcqk4x2GsMl09FlTILnFV6XS57sEJ-mH8/edit?usp=sharing

tulip sleet
random junco
slender iron
manic glacierBOT
#
tannewt
[adafruit/circuitpython] New comment on issue #2676: audio: continuous DMA from AudioIn, real-time audio filtering

It would be helpful to be able to specify a sample rate by reference to another peripheral, i.e. "use whatever sample rate the ADC pacer is set to." One way this could be done is to define some high sample rates as special-purpose codes, e.g. a sample rate of 0xFFFFFF00 would be interpreted as "use the ADC DREQ", etc. Thoughts?

Would it be enough for us to coordinate internally so that the ADC and PWM output dma's use the same pacing timer when set to the same rate? I don't really like t...

#
tannewt
[adafruit/circuitpython] New comment on issue #9463: Enable WiFi/BLE automatic light sleep mode on ESP32

What about waiting a few minutes after boot for a serial connection before enabling sleep, and just accepting any new connections after that will lose the first char and give some kind of error, unless you have native USB?

You don't need to wait too long I think. I think the first character is known to be unreliable. I think it is common to do multiple ctrl-c.

I'm not sure how to handle the supervisor serial console through, and other similar background tasks.

Can you run CP back...

manic glacierBOT
manic glacierBOT
orchid basinBOT
manic glacierBOT
manic glacierBOT
#
timchinowsky
[adafruit/circuitpython] New comment on issue #2676: audio: continuous DMA from AudioIn, real-time audio filtering

Re: the TAC5212, these things are great, up to 768 kHz sampling at up to 32 bits, and you can daisy-chain the devices to get up to 8 channels over one pin. To do them justice, planning to use rp2pio with whatever modifications are needed to support continuous streaming from input to output. That module is kind of the ultimate as far as commanding low-level hardware performance with high-level CP functions. There is a lot of variety in how codecs like to be spoken to, and it will be nice ...

manic glacierBOT
#
EternityForest
[adafruit/circuitpython] New comment on issue #9463: Enable WiFi/BLE automatic light sleep mode on ESP32

I haven't seen a lot of unreliability with the first char myself but it does seem like the sort of thing applications and users should be able to handle.

We could just limit the sleep time to 1s or 100ms at a time(I think usual WiFi sleep is 100ms but will be longer when target wake time is common), regardless of what WiFi is up to, and run a CP/supervisor tick in between each sleep?

tiny peak
#

foamyguy, tannewt, and danh: reviewing the recording of today's meeting, I heard your In The Weeds discussion on improving the cookiecutter-for-libraries situation.
I don't know if you're looking for additional help with developing proof of concept alternatives. But, if you are, I might have some ideas.

#

In particular, if you wanted to make an easily deployable local http-server based web-GUI configuration form thing with the capability to generate zip files, Go might be a good fit. Its standard library has great support for templating, working with zip files, and implementing http servers. Also, the go compiler can build static binaries that are easy to distribute without dependency installation hassles.

#

Doing the same local-http-server to web-GUI to zip-archive workflow in Python should work fine too. The one mild advantage of Go would be that, in theory (haven't tested this), it ought to be pretty easy to use a GitHub Action workflow to have the Go compiler build static binaries for macOS and Linux (perhaps also Windows?) as part of releases to the library-generator tool. Again, in theory, I think those static binaries should be able to include all the necessary string templates for files to start a new library.

#

Were you to try that approach, you could potentially have a cross-platform "make me the start of a new library" CLI binary tool that could be distributed as a single static binary file for each supported platform.

#

So, you could have a learn guide with "Download Library-Maker Tool" buttons that work kinda like project bundle downloads. People would still need to run the thing in a shell, then open the web-GUI link in their browser. But, the procedure could potentially be much simpler than the steps involved in setting up a Python environment.

tiny peak
#

If a go-based tool sounds like a thing you'd like to have, I could potentially help get a proof of concept going.

lone axle
#

If I am trying to make a build on the raspberrypi port but getting a lot of errors like this:

./mpconfigport.h:9:10: fatal error: src/rp2040/hardware_regs/include/hardware/platform_defs.h: No such file or directory
    9 | #include "src/rp2040/hardware_regs/include/hardware/platform_defs.h"
      |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.

What is the the likely cause? It feels to me like it's missing some required library like the rp2040 SDK or something? I ran make fetch-port-submodules already and it seems to complete successfully. I'm not sure if I missed some other step somewhere. I tried running make fetch-all-submodules to see if it would pick up some thing I might be missing, but it does not complete successfully for me, I think it's problem is unrelated but am unsure.

tulip sleet
#

You could try make-remove-all-submodules and the fetch them agian

#

the contents of the submodule might have been removed by accident, but the submodule is still there

#

also try a clean build

lone axle
#

Ports/raspberrypi/sdk/ is empty except for a .git file I think

#

Heading to bed soon and already shut down will follow up tomorrow

tulip sleet
#

that's wrong. go to sdk and do a git restore .

#

or git submodule update --init sdk maybe in the directory above

#

but there may be other missing things too

manic glacierBOT
lone axle
lone axle
# tiny peak So, you could have a learn guide with "Download Library-Maker Tool" buttons that...

I haven't got any experience with Go so I can't make much in the way of informed comment on that possibility. Off the top of my head I think I'd find an end solution that doesn't need to be compiled to be ideal. But I could perhaps be persuaded otherwise if the compilation is taken completely out of the end users hands.

The idea of using JSZip and some JS templating library together to try to re-implement the core concept of cookiecutter question results getting "inflated" into a template folder (or zip) as a front end static HTML/JS page does seem like a promising route to me because the resulting tool could be used locally, or hosted on the internet and won't require any compile step or even "installing" any libraries since JS can just load modules with script tags or other ES6 methods.

All of that being said, I wouldn't say that I'm opposed to any options at this point. I'm not positive if I fully understand the mechanism you propose and a PoC that to look at would help ensure I'm understanding the idea you've got. Realistically I am pretty much in favor of any solution that 1) simplifies the process for people who haven't done it before, or don't do it often, and 2) makes the process less tedious which I find it to be even though I'm not unfamiliar with it. Personally I can say that I'd be interested in at least checking out any attempt that anyone wants to make at achieving those goals. So if you've got some ideas and are interested / willing to try to build out a small PoC to demo I'd say go for it. If you do, feel free to ping me when you're ready to share I'd be happy to take a look and try it out.

tulip sleet
#

It looks like other people are already inventing this wheel.

lone axle
#

Nice!, thank you I'll take a look through some of the existing efforts

slender iron
lone axle
#

I noticed in the core some functions accept a typed self argument like this one: https://github.com/adafruit/circuitpython/blob/main/shared-bindings/vectorio/Rectangle.h#L15 vectorio_rectangle_t *self

But others take a void *obj such as: https://github.com/adafruit/circuitpython/blob/main/shared-bindings/vectorio/Rectangle.h#L24 and then inside of shared-module the first thing they do is "convert" that obj into the self instance like this: https://github.com/adafruit/circuitpython/blob/main/shared-module/vectorio/Rectangle.c#L52 vectorio_rectangle_t *self = obj;

Is there different behavior between these two ways of declaring and using that first argument? Or are both equivalent?

manic glacierBOT
#
tannewt
[adafruit/circuitpython] New comment on issue #2676: audio: continuous DMA from AudioIn, real-time audio filtering

To do them justice, planning to use rp2pio with whatever modifications are needed to support continuous streaming from input to output.

Note that this isn't portable. An I2SIn would be better because it could be implemented on other platforms. RP2 can use PIO internally.

That would be better, it's just more code. Using a magic number is a ~two-line change.

More code is fine. Magic numbers are too much magic imo.

#
tannewt
[adafruit/circuitpython] New comment on issue #9463: Enable WiFi/BLE automatic light sleep mode on ESP32

We could just limit the sleep time to 1s or 100ms at a time(I think usual WiFi sleep is 100ms but will be longer when target wake time is common), regardless of what WiFi is up to, and run a CP/supervisor tick in between each sleep?

I'd only do it if the tick timer is enabled. Will light sleep be woken by interrupts? Much of what background tasks are used for is queued by an interrupt.

manic glacierBOT
#
timchinowsky
[adafruit/circuitpython] New comment on issue #2676: audio: continuous DMA from AudioIn, real-time audio filtering

Note that this isn't portable. An I2SIn would be better because it could be implemented on other platforms. RP2 can use PIO internally.

Yeah, I was thinking to do both, like right now there's both https://github.com/adafruit/Adafruit_CircuitPython_PIOASM/blob/main/examples/pioasm_i2sout.py and the same pio code in https://github.com/adafruit/circuitpython/blob/main/ports/raspberrypi/common-hal/audiobusio/I2SOut.c.

spare jacinth
# lone axle I noticed in the core some functions accept a typed self argument like this one:...

vectorio_rectangle_t *self = obj;
that code doesnt really do any anything (at runtime) it just tells the compiler that we have a variable self (points to the same memory address as obj) of type vectorio_rectangle_t instead of void, so that compiler can do stuff like self->some_field

at the end of the day, the pointer is just the memory address where the object is stored, lets say 0x1234

the difference here is that one function is taking that as "there's a vectorio item at 0x1234" while the other is taking it as "there's something(void) on 0x1234" + then you add the type information with the extra type *new_var = var; + operate on new_var

orchid basinBOT
#
BrainBoardz
[adafruit/circuitpython-org] New comment on pull request #1444: Update brainboardz_neuron.md

Wonderful! Thanks!

Jeremy Littler

LEAD Emerging Technologies

The Creative School

Toronto Metropolitan University

416-979-5000 ext. 552222

RCC 353

On Tue, Aug 6, 2024 at 1:17β€―PM Melissa LeBlanc-Williams <
@.***> wrote:

Merged #1444 https://github.com/adafruit/circuitpython-org/pull/1444
into main.

β€”
Reply to this email directly, view it on GitHub
https://github.com/adafruit/circuitpython-org/pull/1444#event-13781791231,
...

spare jacinth
# spare jacinth > `vectorio_rectangle_t *self = obj;` that code doesnt really do any anything (...

that said, i dont know why one would have the argument as void and the other have it as a concrete type (not that i've even clicked the links, though πŸ™ƒ )

my guess would be that the one with void has to adhere to some typing rules, as in...```c
typedef void (generic_one_arg_func)(void);

void some_func(void *self) {
a_concrete_t *obj = self;
}

// then you have something like
typedef struct {
generic_one_arg_func func;
} foo_t;

// if function was defined as void some_func(a_concrete_t *self) you wouldnt be able to use it here
foo_t bar = {.generic_one_arg_func = some_func};

lone axle
#

Okay, thank you for the explanation. So if I understand correctly the one that takes void *obj is more generic in nature. Theoretically it could have pointers to different types of objects passed for it.

But in it's usage within vectorio in the linked functions as far as I can tell there is no situation where anything would be passing a pointer to anything except a vectorio_rectangle_t. So if I understand the function is able to be used more generally but in practice within this code at least it's not using any of that more general functionality. And theoretically those void *obj could be changed to vectorio_rectangle_t *self. I may give that a try locally to test whether I've actually got my head wrapped around it correctly.

#

I do not necessarily follow the code example all the way, but that is mostly on me only sort-of knowing C.

manic glacierBOT
spare jacinth
# lone axle Okay, thank you for the explanation. So if I understand correctly the one that t...

is more generic
yeah, you could pass the address of anything into it, but if you wanted to operate on the contents of it, it would have to be casted into some type, otherwise you just have a memory location but no idea what it has inside

however, you would need some context to know what to cast into (eg you wouldnt want to operate the contents of a struct as if it was a boolean)

pointers to different objects passed
yeah, when you call function(&something) it is equivalent to function((void *)&something) -- aka: type information is removed, as expected because function is not type-aware

would be passing a pointer to anything except a vectorio_rectangle_t
as i said before, there might be other factors in play so that you must define the function as void *, and there is no harm on having it as void * (other than slight readability)... i would bother rewriting

lone axle
#

I don't intend to rewrite it permaently, just try it the other way to validate my understanding.

spare jacinth
#

writing a basic foo.c is a better way to mess some code up and learn from it, in my experience

#

rather than battling a huge existing codebase

#

for example, i would start with something like this and do some fiddling from it ```c
void func_void(void *ptr) {}
void func_type(int *ptr) {}

int main(int argc, char **argv) {
bool foo = true;
func_void(&foo);
func_type(&foo);
}

#

mostly, reading what the compiler yells me about

lone axle
#

I would agree and I'd probably be able to learn C more easily by working on something smaller. But in this case my goal really is to (attempt to) implement a specific new feature within vectorio objects more than to learn the language more broadly (though I do grant if I hunkered down and went through the process to learn the basics better it'd make it easier to make these sorts of changes). If I can get the feature I'm attempting to work even without a full understanding it'll be a win in my book.

low sentinel
#

The void * might be related to the draw protocol support I added (foggy memory). I remember wanting to make vectorio able to draw from arbitrary python objects that implement draw protocol. I haven't had another project to pull me back and add it yet though :-).

lone axle
low sentinel
#

I also wanted to add native floating point rotations and scales πŸ˜…. Lots of possibilities.

lone axle
#

I was initially trying thigs like self->x or self->draw_protocol_instance->x but those were not working (also probably showing my lack of actual knowledge of the basics.)

#

also I know it's likely not something you've looked at lately, totally understand if you are unsure, please don't go digging into it on my behalf!

low sentinel
#

Yeah, if you're in the abstract, in vectorshape / draw protocol or layer territory, you have to ask through the adapter. Do you know what a vtable is? If so, this is basically the guts of that.
Rectangle doesn't have an x though - that's a common property of all shapes

#

rectangle gets its common shape properties and implementations from vectorshape, along with circle and polygon

lone axle
#

Okay cool, thank you! I don't know what vtable is (by name atleast) I'll read up. I did eventually realize the rect didn't have the x property directly and figured out how to get it from that common_hal call using the draw_protocol but wanted to make sure I wasn't way over-complicating it somehow if there was an easier way to grab those values that come from the VectorShape rather than the subclasses.

low sentinel
#

I think of them as related by composition rather than inheritance πŸ˜‰ but yeah

#

I recall the protocol stuff being fairly interesting to wire

lone axle
#

I am trying to make some basic collision detection support for Circle and Rectangle, maybe eventually Polygon, but I'll need to find some geometry resources for that one, rectangle and circle I've got an idea of how to do already.

low sentinel
#

It's funny, I wanted to do that too. You can kind of cheat by asking the shape to draw an x/y and inspecting the pixel πŸ™‚

lone axle
#

I implemented it in python first, but I'm curious to see if / how much faster it is if it's in the core API for them. I'm working on a breakout clone that is using vectorio shapes.

low sentinel
#

It's really cool to see people using it. I had a lot of fun with it. Anyway, best of luck! And if you change behaviors remember to test a lot haha

lone axle
#

Thanks again for the insight!

manic glacierBOT
#
EternityForest
[adafruit/circuitpython] New comment on issue #9463: Enable WiFi/BLE automatic light sleep mode on ESP32

The tick timer doesn't seem to be compatible with sleep at all, since the IDF automatic light sleep doesn't trigger when there's stuff scheduled every millisecond, right now I'm just disabling it when entering sleep and re-enabling it on exit.

Interrupts work in light sleep unless the peripheral is just disabled entirely like most of them are, but I'm not sure if/how you can detect the automatic background wakeups in general, so the main thread is only going to get notified by stuff that sp...

tiny peak
manic glacierBOT
#
FoamyGuy
[adafruit/circuitpython] Pull request opened: #9483 Vectorio is_colliding()

This change is an initial working implementation of collision detection for vectorio. Specifically it supports Circle and Rectangle only right now, and currently the is_colliding() function has been added to Rectangle only but it accepts either another Rectangle or a Circle to test against. But my intention would be to also this function to the Circle as well so that the test can be done either way.

I wanted to open this up to get feedback before I went further adding it to Circle though...

manic glacierBOT
manic glacierBOT
manic glacierBOT
manic glacierBOT
manic glacierBOT
#
johnnohj
[adafruit/circuitpython] Pull request opened: #9484 Add Waveshare RP2040 One to Circuit Python

Submitting to include the Waveshare board RP2040 One with Circuit Python. It's essentially an RP2040 Zero clone from the same company, except for the larger flash and the Trinkey-like form factor.

In fact, it's so much of an RP2040 Zero clone that Waveshare doesn't offer a separate bootloader for it, which makes the Product ID something of an issue. When I compiled a .u2if for it, I had temporarily assigned 0x111F for the One , agains...

slender iron
manic glacierBOT
#
dhalbert
[adafruit/circuitpython] New comment on pull request #9484: Add Waveshare RP2040 One to Circuit Python

In fact, it's so much of an RP2040 Zero clone that Waveshare doesn't offer a separate bootloader for it,

The RP2040 bootloader is in ROM, and identifies the same for every RP2040 on any board. So you can't distinguish based on that. To get an official PID for this board, see https://github.com/raspberrypi/usb-pid (explained in https://learn.adafruit.com/how-to-add-a-new-board-to-circuitpython/frequently-asked-questions#faq-3130481)

#
FoamyGuy
[adafruit/circuitpython] New comment on pull request #9483: Vectorio intersects()

The addition of intersects() function on Circle class pushed a few other devices over the storage limit. The latest commit disables vectorio on all of the ones that failed during the last action run.

It appears the docs build failed too, but it may have been an intermittent issue with the network in the container, or the server that hosts the svg file that dogs build attempts to fetch. I found these warnings in the action log:

WARNING: Could not fetch remote image: https://github....
#
dhalbert
[adafruit/circuitpython] New comment on pull request #9483: Vectorio intersects()

The addition of intersects() function on Circle class pushed a few other devices over the storage limit. The latest commit disables vectorio on all of the ones that failed during the last action run.

These failures are not due to overflow, I believe, but instead are multiple GitHub actions network issues. See https://github.com/adafruit/circuitpython/actions/runs/10277146885. I think you should revert https://github.com/adafruit/circuitpython/pull/9483/commits/f11cd0c40507c986838743a6e...

manic glacierBOT
#
FoamyGuy
[adafruit/circuitpython] New comment on pull request #9483: Vectorio intersects()

I'm curious about how to tell the difference between those types of actions failures? I noticed this time it seems to have skipped building the ports so it didn't re-attempt any of the builds from last time that didn't succeed.

In the actions log I found a few of them with errors like this: https://github.com/adafruit/circuitpython/actions/runs/10277146885/job/28438883955#step:10:1361 and 20 lines down from here at line 1381 it shows the negative space readout:

1441808 bytes used, ...
#
dhalbert
[adafruit/circuitpython] New comment on pull request #9483: Vectorio intersects()

It is a mix of actual storage overflows and transient CI failures. The C3 failures look like actual overflow. The nordic failures are not. I was able to see the transient failures easily in the list at the link I gave above: https://github.com/adafruit/circuitpython/actions/runs/10277146885. Some of these even succeeded after retries, but the transient errors still show up in that log.

It's worth considering whether to turn something else off to make things fit on the C3 builds, especial...

manic glacierBOT
#
dhalbert
[adafruit/circuitpython] New comment on pull request #9484: Add Waveshare RP2040 One to Circuit Python

Thanks for the clarification. The README in https://github.com/raspberrypi/usb-pid links to a Google Form. I think you should get Waveshare to ask for a new PID via that form. This issue, https://github.com/raspberrypi/usb-pid/pull/14, was submitted by @EngineerWill, who appears to work for Waveshare, based on their profile. You could contact them and ask them to request a new PID. Maybe the ping of them here will also be helpful.

manic glacierBOT
manic glacierBOT
#
johnnohj
[adafruit/circuitpython] New comment on pull request #9484: Add Waveshare RP2040 One to Circuit Python

The Build CI / ports (raspberrypi) / board [...] test seems to be tripping over the entry for the flash chip. The product spec on the website lists it as "W25Q32JVSSIQ - 4MB NOR-Flash," which is twice the size of the Zero and it is by that number I refer to it in the .mk file. Again my inexperience is showing, but I knew I wished to avoid the controller being able to 'see' only 2MB.

orchid basinBOT
manic glacierBOT
lone sandalBOT
manic glacierBOT
manic glacierBOT
#
gbechtinger
[adafruit/circuitpython] Issue opened: #9485 Strange keypad behaviour with CP 9.1.0-beta.2

CircuitPython version

Adafruit CircuitPython 9.1.0-beta.2 on 2024-05-15; Waveshare RP2040-Zero with rp2040

Code/REPL

import usb_hid
import keypad
import storage
import board
import supervisor
import analogio
from time import sleep

def range_map(x, in_min, in_max, out_min, out_max):
    return (x - in_min) * (out_max - out_min) // (in_max - in_min) + out_min
    

# This is only one example of a gamepad descriptor, and may not suit your needs.
...
manic glacierBOT
#
tannewt
[adafruit/circuitpython] New comment on issue #9463: Enable WiFi/BLE automatic light sleep mode on ESP32

Does that conflict with some other process that might be enabling or disabling ticks?

Do you know what is enabling ticks? In theory it should only be on if used.

so the main thread is only going to get notified by stuff that specifically calls the port wake main thread function.

I think its safe to assume we should wake the main thread when needed. Doing so is required.

What about the keypad and the other activities that happen in the polled supervisor tick? Could we make it...

#
tyeth
[adafruit/circuitpython] Issue opened: #9486 FunHouse flashes Yellow lights then nothing, won't boot after installing circuitpython v9.1.1

CircuitPython version

circuitpython v9.1.1 (en_GB UF2) and adafruit-circuitpython-adafruit_funhouse-en_GB-20240806-main-PR9479-7f92099.uf2

Code/REPL

none

Behavior

it gives me Yellow flashing dotstars at circuitpython boot then nothing.
No circuitpython drive shows up and nothing on display.

Description

No response

Additional information

I've tried two funhouses, both erased first then bootloader 0.18.2 then UF2 of 9.1.1 en_GB
...

manic glacierBOT
orchid basinBOT
orchid basinBOT
orchid basinBOT
manic glacierBOT
orchid basinBOT
orchid basinBOT
lone sandalBOT
lone sandalBOT
manic glacierBOT
#
mMerlin
[adafruit/circuitpython] Issue opened: #9487 f-string comma format adds leading comma for 12 digits (and 15…)

CircuitPython version

Adafruit CircuitPython 9.0.5 on 2024-05-22; Adafruit-Qualia-S3-RGB666 with ESP32S3

Code/REPL

v = 1
for _n in range(30):
    l = f'{v:,}'[0:1]
    if l == ',':
        print(f'{v:,}')
    v *= 10

Behavior

,100,000,000,000
,100,000,000,000,000
,100,000,000,000,000,000
,100,000,000,000,000,000,000
,100,000,000,000,000,000,000,000
,100,000,000,000,000,000,000,000,000
,100,000,000,000,000,000,000,000,000,000

Descr...

manic glacierBOT
#
wyrdsec
[adafruit/circuitpython] Pull request opened: #9488 Driver for Pervasive Display's Aurora based E-ink display

A driver for Pervasive Displays Aurora CoG chipsets:
https://www.mouser.com/new/pervasive-displays/pervasive-displays-aurora-mb-displays/

My group used these displays for a Circuitpython badge we created, and this driver came from that effort.

The driver uses framebufferio.FramebufferDisplay (very similar to the sharpdisplay module) to interact with displayio. I has support for the 1.44", 2", 2.7", 1.9", and 2.6" sized Aurora EPD products, however I only have the 2" sized EPD so...

orchid basinBOT
#
bryanjhv
[adafruit/circuitpython-org] Pull request opened: #1447 Restore deleted Waveshare RP2040-Zero

Board Waveshare RP2040-Zero was deleted by:

https://github.com/adafruit/circuitpython-org/commit/6ec46b6a51823d7e134d964288428a9df6fa1f4e#diff-12e6a57bd2b4284f5f5f93d010e3d3ecd0aafe75a5c10991432549441c27f873

Where file _board/waveshare_rp2040_zero.md was moved to _board/waveshare_rp2040_one.md.

Also, the board Waveshare RP2040-PiZero is not to be confused with Waveshare RP2040-Zero.

This commit gets that fixed by:

  • Fixing Waveshare RP2040-PiZero title and name.
  • Restoring ...
orchid basinBOT
onyx hinge
#

just ducking in for a moment, forgive me for not looking into it in detail.

lone sandalBOT
lone sandalBOT
upper radish
onyx hinge
upper radish
onyx hinge
manic glacierBOT
orchid basinBOT
manic glacierBOT
gilded cradle
manic glacierBOT
manic glacierBOT
manic glacierBOT
#
EternityForest
[adafruit/circuitpython] New comment on issue #9463: Enable WiFi/BLE automatic light sleep mode on ESP32

I do remember seeing some scattered discussion somewhere about getting rid
of the OTA partition

On Thu, Aug 8, 2024, 3:04β€―PM Scott Shawcroft @.***>
wrote:

It seems like a lot of ESP32 boards don't have the alarm module at all
though, I wonder how many actually have space for it?

Is that due to the 4MB flash limit? I think we plan on changing the
partition layout with CP 10 so that those boards can have more features
like BLE.

β€”
Reply to this email direc...

manic glacierBOT
manic glacierBOT
manic glacierBOT
#
tannewt
[adafruit/circuitpython] New tag created: 9.2.0-alpha.2350
slender iron
#

Its been a while since I did a release

#

I've filled up the CI

blissful pollen
#

What an exciting release too! Does most of the 2040 code work on the 2350?

slender iron
#

Ya, most of it just worked

#

once the build system could handle a different chip

#

There are definitely bugs. I just realized there was an additional UART pinout

#

and the USB host code is crashing

blissful pollen
#

That's a bonus most of it works. Minus the host. Just scanning the PR and was wondering how much would just work while at work today.

slender iron
#

No better way to find bugs than to release an alpha πŸ™‚

blissful pollen
#

As soon as people can get a hold of the boards I do not think you will lack testers!

orchid basinBOT
orchid basinBOT
orchid basinBOT
orchid basinBOT
orchid basinBOT
orchid basinBOT
manic glacierBOT
tulip sleet
orchid basinBOT
manic glacierBOT
#
bill88t
[adafruit/circuitpython] New comment on pull request #9480: LILYGO T-Watch-S3

IMPORTANT!

This board does some pretty weird AXP2101 with the RTC, and claims differently in the docs.

To get all the hw to work, following configuration commands are run after the pmic is loaded:

pmic._aldo1_voltage_setpoint = 3300
pmic._bldo1_voltage_setpoint = 0
pmic._bldo2_voltage_setpoint = 3300
pmic._aldo1_voltage_setpoint = 3300
if pmic._read_register8(24) != 13:
    pmic._write_register8(24, 13) # Charge VBackup at 3v3

If the coin cell battery is not cha...

manic glacierBOT
#
wyrdsec
[adafruit/circuitpython] New comment on pull request #9488: Driver for Pervasive Display's Aurora based E-ink display

I'm sorry but I can't seem to wrangle sphinx to do what I need. For some reason the second line of the docs comment is being included in the title of the page.

//| class AuroraMemoryFramebuffer:
//|     """A framebuffer for Pervasive Displays Aurora E-paper displays.
//|
//|     These displays are 2 color only.

Do I need another line break or special character to stop it from doing this?

#
tyeth
[adafruit/circuitpython] Issue opened: #9498 Qualia / DotClockFrameBuffer unexpectedly renders a few lines of pixels at once in the most bizarre way

CircuitPython version

9.1.1

Code/REPL

Adapted version of Liz's countdown clock for CircuitPython Day 2024, bitmap background + bitmap label that scrolls

Behavior

Image renders about 5 lines at a time, but seemingly randomly out of order and not the whole line at once.
Probably made worse by my use of display.rotation=90

Description

No response

Additional information

No response

manic glacierBOT
manic glacierBOT
manic glacierBOT
#
FoamyGuy
[adafruit/circuitpython] New comment on pull request #9483: Vectorio intersects()

Maybe it would make sense to move this out of vectorio, or at least out of the Rectangle and Circle classes. If they were separated from those it would mean we could have a single something.cirlce_rectangle_intersects(rect_x, rect_y, rect_w, rect_h, circle_x, circle_y, circle_r) or similar. It wouldn't accept vectorio.Rectangle or Circle directly, but instead just the relevant coordinates and sizes. That way user code can use it with a variety of different objects rather than just Rectangl...

manic glacierBOT
manic glacierBOT
manic glacierBOT
slender iron
empty salmon
#

@slender iron - WOOHOO!!

slender iron
#

found the schematic for the pinout

manic glacierBOT
empty salmon
#

@slender iron - I may have time tomorrow. Today was teaching DCNextGen Python and CircuitPython 🀞🏼

manic glacierBOT
manic glacierBOT
spare jacinth
#

#help-with-git message after stumping into this a few hours ago, i've done a very minimal research and it seems like mypy already has a tool called stubtest that compares hints in .pyi(aka, circuitpython-stubs) with actual code in .py (aka, blinka)
i was thinking of doing a similar thing myself from scratch if it didnt exist yet

think it is worth investigating and perhaps add some testing with it, or at least a "we know these dont match" list

didnt feel like opening an issue or anything, as i dont use either project and dont want to (potentially) get a lot of notifications about the topic πŸ˜›

manic glacierBOT
manic glacierBOT
#
graeme-winter
[adafruit/circuitpython] New comment on issue #9471: Building precompiled module for itsybitsy esp32 gives incompatible .mpy arch

Update on progress: the system locks up at mp_call_function_0(module_fun); in do_execute_raw_code - looking at the address of the function:

function at: 3f8077f0

and the memory map from the ESP32 technical reference manual:

Screenshot 2024-08-10 at 16 58 43

I suspect that the issue is that the module is loaded into data memory not instruction => system throws exception when hits the f...

versed lark
#

there's an inconsistency in the "Get Setup to Add Your Board" page at https://learn.adafruit.com/how-to-add-a-new-board-to-circuitpython/get-setup-to-add-your-board in the tutorial for adding a new board -
The following files must be updated. The board_name folder will be your board. The first three must be customized for your board and your board must added to the last:
there are 3 files listed, but not a fourth, as the text implies.

How to Add a New Board to CircuitPython

Add your new CircuitPython compatible board to CircuitPython!

austere acorn
stuck elbow
#

you don't normally need to edit board.c, unless your board has a display that needs to be initialized or some other special hardware

austere acorn
#

I think the language is unclear there, since it's talking about setting up a new port, with the port folder and the 4 initial files

versed lark
#

i'll probably need to edit board.c, as this board (sparkfun thing plus esp32s3) has an SD slot on the bottom

austere acorn
#

it lists all the files needed

versed lark
#

seems like a case of either too much info for the 'getting started' page, or not enough

#

πŸ™‚

#

also having trouble with the initial setup - doing the 'make fetch-all-submodules' thing, and got this at the end:
Cloning into '/home/brian42/repos/circuitpython/tools/python-semver'... Cloning into '/home/brian42/repos/circuitpython/tools/uf2'... Submodule path 'data/nvm.toml': checked out '6b678f15e378edce820f2ffdef3286b3e55449e7' remote: Total 0 (delta 0), reused 0 (delta 0), pack-reused 0 fatal: transport 'file' not allowed fatal: Fetched in submodule path 'extmod/ulab', but it did not contain eacb0c9af47f85f5d4864b721c3b28661364e8e3. Direct fetching of that commit failed. fatal: ::endgroup:: Duration: 168.70768673699786 Traceback (most recent call last): File "/home/brian42/repos/circuitpython/tools/ci_fetch_deps.py", line 238, in <module> main(target) File "/home/brian42/repos/circuitpython/tools/ci_fetch_deps.py", line 220, in main fetch(submodules) File "/home/brian42/repos/circuitpython/tools/ci_fetch_deps.py", line 134, in fetch run( File "/home/brian42/repos/circuitpython/tools/ci_fetch_deps.py", line 110, in run subprocess.run(shlex.split(command), stderr=subprocess.STDOUT, check=True, cwd=cwd) File "/usr/lib/python3.10/subprocess.py", line 526, in run raise CalledProcessError(retcode, process.args, subprocess.CalledProcessError: Command '['git', 'submodule', 'update', '--init', '--depth', '1', '.', 'frozen']' returned non-zero exit status 128. make: *** [Makefile:343: fetch-all-submodules] Error 1

austere acorn
#

no all in there

versed lark
#

this is a fresh clone of my fork from adafruit/circuitpython:main, FWIW

versed lark
#

yes, it has the 'all'

versed lark
#

definitely something git doesn't like... running that commandline separately gives the same failure

manic glacierBOT
#
Axlfc
[adafruit/circuitpython] Issue opened: #9503 File Copy Error and Corruption in CircuitPython Device (multisparkline.py, polygon.py, rect.py, roundrect.py, sparkline.py, triangle.py)

Similar issue here.

Description:

When attempting to copy the files multisparkline.py, polygon.py, rect.py, roundrect.py, sparkline.py, triangle.py into my CircuitPython device (drive E: named CIRCUITPY), the file becomes corrupted with strange characters, and the following error message appears:

No puede copiar el archivo debido a un error inesperado. Si sigue recibiendo este error,...
#
dhalbert
[adafruit/circuitpython] New comment on issue #9503: File Copy Error and Corruption in CircuitPython Device (multisparkline.py, polygon.py, rect.py, roundrect.py, sparkline.py, triangle.py)

Transferred issue to https://github.com/adafruit/circuitpython, because this issue is not related to the particular files.

There are various reasons the copies might be corrupted.

  1. How are you copying the files? Drag and drop?
  2. The CIRCUITPY drive may be corrupted due to previously unplugging or resetting the device before a write was complete. Try this in the REPL serial connection. It will erase and reformat CIRCUITPY. Then copy the files agan.
import storage
storage.erase...
#
graeme-winter
[adafruit/circuitpython] Issue opened: #9504 analogbufio.BufferedIn fails on ItsyBitsy esp32

CircuitPython version

Adafruit CircuitPython 9.1.1 on 2024-07-22; Adafruit ItsyBitsy ESP32 with ESP32

Code/REPL

import board
import analogbufio
import array

N = 1000
rate = 1000

buffer = array.array("H", [0x0000] * N)
adc = analogbufio.BufferedIn(board.A3, sample_rate=rate)
adc.readinto(buffer)
adc.deinit()
for i in range(N):
    print(i, buffer[i])

Behavior

>>> %Run -c $EDITOR_CONTENT
Traceback (most recent call last):
  Fil...
versed lark
#

got past the above. now, during build:
`brian42@CarasGaladhon:~/repos/circuitpython/ports/espressif$ make BOARD=adafruit_feather_esp32s3_4mbflash_2mbpsram

  • Verbosity options: any combination of "steps commands rules", as make V=... or env var BUILD_VERBOSE
    Traceback (most recent call last):
    File "/home/brian42/repos/circuitpython/ports/espressif/../../tools/gen_web_workflow_static.py", line 7, in <module>
    import minify_html
    ModuleNotFoundError: No module named 'minify_html'
    make: *** [../../supervisor/supervisor.mk:210: build-adafruit_feather_esp32s3_4mbflash_2mbpsram/autogen_web_workflow_static.c] Error 1`
#

where is minifiy_html supposed to be? there isn't a file by that name anywhere in the circuitpython tree

sleek relic
#

Pages
https://docs.circuitpython.org/en/9.1.x/shared-bindings/usb_cdc/index.html
https://docs.circuitpython.org/en/8.2.x/shared-bindings/usb_cdc/index.html
https://docs.circuitpython.org/en/latest/shared-bindings/usb_cdc/index.html
under class usb_cdc.Serial Include

You cannot create an instance of usb_cdc.Serial. The available instances are in the usb_cdc.serials tuple.
However, with
Adafruit CircuitPython 9.0.5 on 2024-05-22; Adafruit-Qualia-S3-RGB666 with ESP32S3
usb_cdc.serials does not exist:
>>> import usb_cdc
>>> dir(usb_cdc)
['__class__', '__name__', 'Serial', '__dict__', 'console', 'data', 'disable', 'enable']
>>> usb_cdc.serials
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
AttributeError: 'module' object has no attribute 'serials'
Is that specific to the build for the qualia board?

stuck elbow
#

I think they are usb_cdc.console and usb_cdc.data

#

looks like some leftover from an earlier version

manic glacierBOT
#
Axlfc
[adafruit/circuitpython] New comment on issue #9503: File Copy Error and Corruption in CircuitPython Device (multisparkline.py, polygon.py, rect.py, roundrect.py, sparkline.py, triangle.py)

Hello and so many thanks for the quick response.

1. How are you copying the files? Drag and drop?

Yes, I download manually the source code and copy/paste to the board's folder, I went file by file because I noticed when copying the whole folder the behavior happens too.

2. The CIRCUITPY drive may be corrupted due to previously unplugging or resetting the device before a write was complete. Try this in the REPL serial connection. It will erase and reformat CIRCUITPY. Then ...
manic glacierBOT
#
Axlfc
[adafruit/circuitpython] New comment on issue #9503: File Copy Error and Corruption in CircuitPython Device (multisparkline.py, polygon.py, rect.py, roundrect.py, sparkline.py, triangle.py)

Are you using a hub or a direct connection?

Direct connection to the pc.

Please try a different cable as well.

I have another cable and it still happening...

Could you try CircuitPython 9.0.5 and see if it makes any difference?

Sure I will try again, I'm resetting using the command you gave me earlier and try the whole process again and write it here in few minutes.

import storage
storage.erase_filesystem()
#
Axlfc
[adafruit/circuitpython] New comment on issue #9503: File Copy Error and Corruption in CircuitPython Device (multisparkline.py, polygon.py, rect.py, roundrect.py, sparkline.py, triangle.py)

Sorry I do not have a Mac nor RapsberryPi, but I can try tomorrow to replicate whole file copying and file editing process in my dual boot ubuntu GNU/Linux (it's the same pc I run Windows 10) and see what happens.

It blocked again like this
imagen
trying to edit the code.py file...

😰

It's only happening randomly at some times... But fortunately the last changes I made it worked.

But for ...

lone axle
# versed lark where is minifiy_html supposed to be? there isn't a file by that name anywhere i...

It's a python module. It's listed as a requirement inside of the requirements-dev.txt file https://github.com/adafruit/circuitpython/blob/5f7c42e277a20aef1863aaeadd5464a3e3c2ef67/requirements-dev.txt#L33 It's mentioned, albeit only briefly in this section of the building circuitpytyhon guide: https://learn.adafruit.com/building-circuitpython/build-circuitpython#install-required-python-packages-3086211

versed lark
# lone axle It's a python module. It's listed as a requirement inside of the `requirements-d...

apparently i'd missed one of the pip3 installs. now, esp-idf doesn't like what it just installed for itself:

`brian42@CarasGaladhon:~/repos/circuitpython/ports/espressif$ source esp-idf/export.sh
Detecting the Python interpreter
Checking "python3" ...
Python 3.10.12
"python3" has been detected
Checking Python compatibility
Checking other ESP-IDF version.
Adding ESP-IDF tools to PATH...
Checking if Python packages are up to date...
Constraint file: /home/brian42/.espressif/espidf.constraints.v5.2.txt
Requirement files:

  • /home/brian42/repos/circuitpython/ports/espressif/esp-idf/tools/requirements/requirements.core.txt
    Python being checked: /home/brian42/.espressif/python_env/idf5.2_py3.10_env/bin/python
    The following Python requirements are not satisfied:
    Requirement 'setuptools<71.0.1,>=21' was not met. Installed version: 72.1.0
    To install the missing packages, please run "install.sh"
    Diagnostic information:
    IDF_PYTHON_ENV_PATH: /home/brian42/.espressif/python_env/idf5.2_py3.10_env
    Python interpreter used: /home/brian42/.espressif/python_env/idf5.2_py3.10_env/bin/python`
#

hrm. something in the pip3 installs tools 72.1... install.sh removes it an puts 71 back on. go figure.

#

yay... finally... a build. πŸ™‚

thorny jay
#

I have not been actively writing CP since 6+(?) months, but I still listen to the weekly meeting (not LIVE).

I figured that I could restart working on a library (IS31FL3730) and maybe finish for CP Day (15 August is a day off).
So I try cookiecutter again to have fresh start... this is where I start to understand the last "in the weed" discussion.
So I have to answer 14 questions right, the first time, and if ever I make a mistake I have to start all over again???
I never succeed first time (typo, question I don't know the answer and need to search, ...).
I don't remember the value I used the last time I used cookiecutter for that project (because I did not save the answer).
Apparently the last time I worked on https://github.com/dglaude/CircuitPython_IS31FL3730 is a year before 2024 CP Day. !!!

UPDATE: I found .cookiecutter_replay/cookiecutter-adafruit-circuitpython.json and that seems to contain my answer from previous run (and luckily I only worked on one library on that Pi400 where I started the development).

There is also this that could be of interest: https://stackoverflow.com/questions/60458434/cookie-cutter-whats-the-easiest-way-to-specify-variables-for-the-prompts with ~/.cookiecutterrc file where you can put your frequent answer.
Having a way to save a "per library" file, in the repo of that library, so that re-running cookiecutter 6-12 months later become painless... would be great.

(ping @lone axle)

manic glacierBOT
#
FoamyGuy
[adafruit/circuitpython] Issue opened: #9505 Adding math.dist()?

I wondering if there is interest in having an math.dist() (CPython ref: https://docs.python.org/3/library/math.html#math.dist) in the core?

I had a try at implementing it in this branch: https://github.com/FoamyGuy/circuitpython/tree/math_dist. It only supports 2D, whereas the CPython one seems to allow higher dimensions, and I have done only some minimal "happy path" testing, but it does seem to work correctly and mostly match CPython results with the points I've tested.

I started w...

manic glacierBOT
#
FoamyGuy
[adafruit/circuitpython] New comment on issue #9505: Adding math.dist()?

I have learned about an alternative way to calculate euclidean distance from here: https://stackoverflow.com/a/1401828/507810 This one uses numpy.linalg.norm which it turns out that we already have access to by way of ulab

The syntax to use it is a bit different, needing you to create two numpy arrays, so maybe the math.dist() version could be slightly more convenient to use. But I'm not sure if "slightly more convenient" really makes it worth using up the extra space.

The `nump...

thorny jay
# thorny jay I have not been actively writing CP since 6+(?) months, but I still listen to th...

I copied stuff I found in .cookiecutter_replay/cookiecutter-adafruit-circuitpython.json and edited to be in the right format for ~/.cookiecutterrc... nothing automated, but I could keep one such file per library.
Here is how it look and I just have to press Enter to approuve the new default value:

default_context:
    target_bundle: Community
    github_user: dglaude
    author_name: David Glaude
(env) dglaude@pi400-64:~/CC $ cookiecutter gh:adafruit/cookiecutter-adafruit-circuitpython
You've downloaded /home/dglaude/.cookiecutters/cookiecutter-adafruit-circuitpython before. Is it 
okay to delete and re-download it? [y/n] (y): y
  [1/14] Select target_bundle
    1 - Community
    2 - Adafruit
    3 - CircuitPython Org
    4 - test
    Choose from [1/2/3/4] (1): 
  [2/14] github_user (dglaude): 
  [3/14] author_name (David Glaude):
manic glacierBOT
manic glacierBOT
versed lark
#

Anyone know if it's possible to get the USB VID/PID from a board somehow? Or do I need to ask Sparkfun for their IDs? (working on board definition for the Thing Plus ESP32S3).

edit: i see the IDs for the jtag device that shows up in Windows for it. i did read that UF2 and CP should have their own PIDs. Should I just put the jtag ones in mpconfigboard.mk?

orchid basinBOT
manic glacierBOT
brazen hatch
#

Uhh, I just tried building current master with the t-watch PR.. It doesn't even boot now.. Did I miss a big PR?

#

All I am getting on a debug build is:

[20:40:55.972] Connected to /dev/ttyACM0
I (624) esp_image: segment 0: paddr=00010020 vaddr=3c1500atency  : 0x02 (10 cycles@Fixed)
I (1004) octal_psram: DriveStrength: 0x00 (1/1)
I (1010) MSPI Timing: PSRAM timing tuning index: 5
I (1015) esp_psram: Found 8MB PSRAM device
I (1019) esp_psram: Speed: 80MHz
I (1446) esp_psram: SPI SRAM memory test OK
I (1455) cpu_start: Pro cpu start user code
I (1455) cpu_start: cpu freq: 240000000 Hz
I (1456) cpu_start: Application information:
I (1458) cpu_start: Project name:     circuitpython
I (1464) cpu_start: App version:      9.2.0-alpha.2350-19-ge0d326edfc
I (1471) cpu_start: Compile time:     Aug 11 2024 20:36:16
I (1477) cpu_start: ELF file SHA256:  8f06be8f5...
I (1483) cpu_start: ESP-IDF:          v5.2.1-548-gd58f9ce0b4
I (1489) cpu_start: Min chip rev:     v0.0
I (1494) cpu_start: Max chip rev:     v0.99 
I (1499) cpu_start: Chip rev:         v0.2
I (1504) heap_init: Initializing. RAM available for dynamic allocation:
I (1511) heap_init: At 3FCB6B10 len 00032C00 (203 KiB): RAM
I (1517) heap_init: At 3FCE9710 len 00005724 (21 KiB): RAM
I (1523) heap_init: At 3FCF0000 len 00008000 (32 KiB): DRAM
I (1529) heap_init: At 600FF064 len 00000F84 (3 KiB): RTCRAM

[20:41:05.334] Disconnected
manic glacierBOT
#
microdev1
[adafruit/circuitpython] New comment on issue #9468: ci_set_matrix: Builds no boards in specific circumstance

Hey, diffs between merge commits can get quite complex and the current implementation doesn't handle them very well. Here, the runs should have been scheduled according to the changes introduced in the 9.1.x branch since it diverged. Hopefully, the following sheds some light on this:

env_var commit ref
BASE_SHA https://github.com/adafruit/circuitpython/commit/901dd228cbdb2069e66fae8c2108d44466ab1e7f 9.1.x
GITHUB_SHA~ https://github.com/adafruit/circuitpython/com...
brazen hatch
#

Uhh, every S3 board is broken on main rn

#

Onto bisecting

brazen hatch
#

So yea, TinyUSB is broken.

#

Artifacts main-PR9479-7f92099 or earlier work fine.

#

The alpha release is of course broken too.

nimble sparrow
#

Oops

orchid basinBOT
#
marcobrianza
[adafruit/circuitpython-org] New comment on issue #1450: "lolin S3 mini" and "lolin S3 pro" web installer links to uf2 should be updated

Hi, consider I'm just starting to use circuitpython that is why I was counting very much on the web installer.
anyway I flashed the combined.bin using https://adafruit.github.io/Adafruit_WebSerial_ESPTool/
then I copied to the pseudo-usb-drive adafruit-circuitpython-lolin_s3_mini-en_GB-9.1.1.uf2

circuitpython runs and I just made the onboard neopixel blink.

board.out is the follwing:

Adafruit CircuitPython 9.1.1 on 2024-07-23; LOLIN S3 MINI 4MB Flash 2MB PSRAM with ESP32S3
Board I...

nimble sparrow
blissful pollen
slender iron
manic glacierBOT
slender iron
#

Thanks!

brazen hatch
#

Np, I was kinda confused, I should have done it alr.

slender iron
#

that's why it is alpha

brazen hatch
#

I was confused how nobody caught it before me.

versed lark
manic glacierBOT
#
bill88t
[adafruit/circuitpython] New comment on issue #9496: Calling micrcontroller.reset on the Qualia (with 3.2" bar 820x320) causes reset into TinyUF2 Bootloader mode

Oh I know this. It's the double reset setting in tinyuf2.

// GPIO that implement 1-bit memory with RC components which hold the
// pin value long enough for double reset detection.
#define PIN_DOUBLE_RESET_RC   4

Furthermore, it seems the latest commit there changed that from IO5 to IO4.
https://github.com/adafruit/tinyuf2/commit/da3eaf9949ef33e484106115b16b849034559664

Defining this without the proper circuit leads to this exact behavior.
Suggestion: Transfer this issu...

versed lark
#

@brazen hatch ya, that was it for me too. checked out 9.1.1, and my new board def built & is running AFAICT. At least I get the right name in the REPL, so that's a good sign. Now to figure out how to get the UF2 bootloader going... but that's something for another day.

manic glacierBOT
orchid basinBOT
manic glacierBOT
manic glacierBOT
manic glacierBOT
#
bill88t
[adafruit/circuitpython] New comment on issue #9496: Calling micrcontroller.reset on the Qualia (with 3.2" bar 820x320) causes reset into TinyUF2 Bootloader mode

Curiously I'm pretty sure I've been running firmware since that change, but to be fair I've no idea the actual last version I ran before 9.x

@tyeth The CircuitPython version has nothing to do with this. It's just the bug doesn't happen every time (it's a 80% from what I can tell).

CircuitPython resets correctly, but then the TinyUF2 bootloader performs the double reset check and (in most cases) succeeds, sending you to TinyUF2 instead of CircuitPython.

The TinyUF2 bootloader is als...

manic glacierBOT
#
bill88t
[adafruit/circuitpython] New comment on pull request #9480: LILYGO T-Watch-S3

The PR is kinda ready, I'll update the branch and fix the merge conflicts in a moment.

Hardware init guide:
<details><summary>PMIC (AXP2101)</summary>

import board
from axp2101 import AXP2101
pmic = AXP2101(board.I2C())
pmic._aldo1_voltage_setpoint = 3300
pmic._bldo1_voltage_setpoint = 0
pmic._bldo2_voltage_setpoint = 3300
pmic._aldo1_voltage_setpoint = 3300
if pmic._read_register8(24) != 15:
    pmic._write_register8(24, 15) # Charge VBackup at 3v3
if pmic._read_regist...
orchid basinBOT
manic glacierBOT
manic glacierBOT
#
tyeth
[adafruit/circuitpython] New comment on issue #9496: Calling micrcontroller.reset on the Qualia (with 3.2" bar 820x320) causes reset into TinyUF2 Bootloader mode

Ahhh yeah good points, morning brain wasn't thinking bootloaders.
Not sure which versions I'm running on each, but probably 0.18.2 (definitely on this problematic one), and half aware that the bootloaders on circuitpython.org/downloads are often older than latest tinyuf2 releases (most are on 0.18.2).

My INFO_UF2.txt:

TinyUF2 Bootloader 0.18.2 - tinyusb (0.15.0-566-gf1e006d09)
Model: Adafruit Qualia ESP32-S3 RGB666
Board-ID: ESP32S3-Qualia-revC
Date: Mar 22 2024
Flash Size: 0x00...
manic glacierBOT
lone axle
#

<@&356864093652516868> The weekly meeting will occur here on discord in about 1 hour, at the normally scheduled time of 2pm Eastern / 11am Pacific. We look forward to hearing from everyone!

manic glacierBOT
manic glacierBOT
#
FoamyGuy
[adafruit/circuitpython] New comment on issue #9505: Adding math.dist()?

I am trying to make intersection detection functionality for arcade / physics style games. (https://github.com/adafruit/circuitpython/pull/9483 is the beginning of this effort).

Some of the intersection functions rely on euclidean distance calculations so I looked into refactoring it into it's own function but went looking and found that CPython already has it.

I do intend for for these intersection detection functions to help me do something faster, specifically being able to check fo...

manic glacierBOT
solar whale
#

Cat safely ejected 😼

onyx hinge
#

Still traveling! Have a good meeting today. Group hug.

slender iron
#

the report is from 8/7

#

I must need to poke it

#

no newer ones

#

we hit the 60 day limit

brazen hatch
#

Breakout on the T-Watch-S3 would be lots of fun, looking forward to it!

thorny jay
#

Thanks @lone axle

slender iron
#

thanks for hosting!

gilded cradle
#

Thanks foir hosting @lone axle

brazen hatch
#

Thanks for hosting!

lone axle
#

Thanks you, hope everyone has a great week see on on Friday for CircuitPython day!

gilded cradle
#

@lone axle I didn't want to put you on the spot during the meeting, but was wondering if you could do some testing on code-beta.circuitpython.org focusing mostly on USB workflow?

solar whale
#

@lone axle It looks like the release of the RFM library worked OK

slender iron
#

@brazen hatch what board were you testing with?

brazen hatch
lone axle
gilded cradle
lone axle
brazen hatch
slender iron
#

yup. I tried a debug build on an esp devkit with 16mb flash and no external ram and it worked

#

trying an S2 now

brazen hatch
#

huh, it could be psram, but unlikely

#

I'll try an s2 with psram too then

slender iron
#

ya, that was my thinking too

#

could also be that I'm doing a debug build

brazen hatch
#

nah

#

I did a debug build too

slender iron
#

did you get a crash backtrace?

brazen hatch
#

nope, just bootloops

#

it doesn't hard crash and stop

slender iron
#

do you have the idf logging output from it?

brazen hatch
#

normal terminal logs, I sent them a bit before those msgs, one moment

brazen hatch
#

S2 w/psram broken too

slender iron
#

k, I'm on an N4R2 atm

#

gotta make the debug build work though

brazen hatch
#

without a backtrace things are very rough

#

curiously, this S2 board blinks the led once

slender iron
#

it could be resetting via reset_into_safe_mode() repeatedly

#

the idf logging will tell you why it reset

brazen hatch
#

true, but i'm not getting a usb connection at all

brazen hatch
slender iron
#

idf logging is over uart so you use a serial to usb converter

#

the esp devkits have this built in

#

that's why they have two usb ports

brazen hatch
#

Oh I have one of those boards, lemme load it, it is n16r8

slender iron
#

πŸ‘

#

I'm trying with a n32r8 now

#

(s3)

lone axle
#

Here is the notes document for next Monday’s CircuitPython Weekly Meeting. It is at the normal time of 11am Pacific / 2pm US Eastern here on Discord. Add your hug reports and status updates to the document before the meeting. If you are unable to attend but would still like to contribute, feel free to add your notes and we’ll read them off during the meeting. Hope to see you there! <@&356864093652516868> https://docs.google.com/document/d/13Q0vqnzMAVGwjdVG3-lXMxt62ouymSQ134-o-ROgXBM/edit?usp=sharing

Google Docs
CircuitPython Weekly Meeting for August 19, 2024

CircuitPython Weekly Meeting for August 19, 2024 Welcome to the CircuitPython Weekly meeting notes! Feel free to add your Hug Reports and Status Updates before the meeting, alphabetically by your username. During the meeting, we go through them in order. If you can’t make the meeting and woul...

lone axle
brazen hatch
#

SBC compiling go brrrr

slender iron
#

ok, I think n32r8 shows the issue

#

ok, my tlsf change is the issue

#

not sure why it causes the hang

brazen hatch
#

Yea watchdog timeout reset...
rst:0x10 (RTCWDT_RTC_RST),boot:0x8 (SPI_FAST_FLASH_BOOT)

slender iron
#

I'm going to get lunch. will push a fix later. I suspect it is writing off the end of psram

brazen hatch
#

Enjoy your break, i'll go back to watch app making, if you need me to test anything else later, ping.

slender iron
#

thanks!

brazen hatch
#

I also have implemented power states.

#

It clocks down to 40mhz while disconnected from wifi, 80mhz if connected.

#

Dimming based on accelerometer and touch.

#

It gets about 10 hours only though, so I still have more work to do.
I might need to actually consider deep sleep since we have interrupt pins.

#

I'd advise to at least steal code.

slender iron
#

nice! I'll keep an eye on it

manic glacierBOT
orchid basinBOT
lone axle
slender iron
#

@lone axle I think that may end up in the zip files

#

I think that VERSIONS.txt is autogenerated

lone axle
#

Ahhhhh, thank you. I do see it in the zipped bundle. Circup has spoiled me, I haven't actually downloaded and unzipped a bundle manually in a while πŸ˜…

orchid basinBOT
manic glacierBOT
lone axle
slender iron
lone axle
#

If I then go into pycharm and make a "nothing" change like add then remove a space and then save the file it will run successfully, but modifying in the web editor and then saving is always leading to that error. I'll try a different device as well.

gilded cradle
manic glacierBOT
manic glacierBOT
manic glacierBOT
manic glacierBOT
slender iron
#

@devout jolt want me to make a board def for the sparkfun board and then you can test it?

quiet scroll
#

hey guys who can i talk to about the si5351 lib

devout jolt
slender iron
#

uh oh! does it have swd pads?

devout jolt
slender iron
#

k, file an issue and I can debug with my feather

devout jolt
slender iron
#

the psram may not be configured correctly

slender iron
devout jolt
slender iron
devout jolt
slender iron
#

bryan from SF got right back to me with a PID too!

#

try doing b = bytearray(1000000) and see if it crashes

quiet scroll
#

where would I send a conversion of the si5351 setup that works for circuitpython vs a c ino ?

devout jolt
#

does not crash

Screenshot_2024-08-13_at_9.29.36_AM.png
slender iron
#

nice!

devout jolt
slender iron
#

I fully expect crashes when we try to DMA from psram

manic glacierBOT
devout jolt
#

perhaps that's why audio/synthio is crashing

slender iron
#

Β―_(ツ)_/Β―

devout jolt
# slender iron Β―\_(ツ)_/Β―

I've also been getting a few crashes like this (both with this firmware and the Pico2 firmware):

You are in safe mode because:
CircuitPython core code crashed hard. Whoops!
Hard fault: memory access or instruction error.
Please file an issue with your program at github.com/adafruit/circuitpython/issues.
Press reset to exit safe mode.

Is this known behavior?

slender iron
#

I haven't seen it

#

so if you can repro then please file an issue

devout jolt
#

yep

manic glacierBOT
#
todbot
[adafruit/circuitpython] Issue opened: #9517 RP2350 Audio crashes

CircuitPython version

Adafruit CircuitPython 9.2.0-alpha.2350-6-gf064354648-dirty on 2024-08-13; SparkFun Pro Micro RP2350 with rp2350a

Code/REPL

import board, audiocore, audiopwmio
audio = audiopwmio.PWMAudioOut(board.D0)  # or any GPIO
wave = audiocore.WaveFile("laser2.wav")
audio.play(wave)
while True:
  pass

Behavior

As soon as audio.play() is run, board resets.

Description

Playing a RawSample from RAM works, e.g. this does not...

slender iron
# devout jolt yep

thanks! I totally expect bugs like this as folks get hardware and start using it

manic glacierBOT
slender iron
#

@devout jolt you are saying what it was already set to

devout jolt
manic glacierBOT
slender iron
#

@devout jolt are you debugging the audio crash?

devout jolt
slender iron
#

I'll take a peek now before lunch with a friend

manic glacierBOT
#
tannewt
[adafruit/circuitpython] New comment on pull request #9315: ADC Multi-sampling

Honestly, I'm not sure I want to add sample number configuration as an API. I'd be ok if the ESP did multiple samples internally but adding an API impacts all ports (as you see) and all ADC drivers. It is backwards compatible but any code that starts using it won't work elsewhere that hasn't added it.

Multiple samples can be taken from Python too if you really want to too.

manic glacierBOT
#
bill88t
[adafruit/circuitpython] New comment on pull request #9315: ADC Multi-sampling

Sure, it's new api, but something that arguably should have been there from the start.
If you need precision you kinda have to take multiple samples.
Taking them from python is fine, but python is notoriously slow on for loops.
This is something not so complicated, that passes on the burden to the underlying core.

but any code that starts using it won't work elsewhere that hasn't added it.

Yea, but how often is an application developed and it's expected for it to work on earlier ...

#
tannewt
[adafruit/circuitpython] New comment on pull request #9315: ADC Multi-sampling

Sure, it's new api, but something that arguably should have been there from the start.
If you need precision you kinda have to take multiple samples.

analogbufio can be used to take multiple samples at once.

Taking them from python is fine, but python is notoriously slow on for loops.
This is something not so complicated, that passes on the burden to the underlying core.

It is simple but likely to not be used. (We've gone 7 years without it.) Furthermore, it has the cod...

slender iron
#

@devout jolt looks like the DMA interrupt isn't working correctly. I'm still debugging why

manic glacierBOT
#
BlitzCityDIY
[adafruit/circuitpython] New branch created: vindie_s2
manic glacierBOT
#
todbot
[adafruit/circuitpython] Issue opened: #9519 Metro RP2350 I2C pins swapped

CircuitPython version

n/a, don't have the board yet

Code/REPL

import board; i2c = board.I2C()

Behavior

What I expect one would see:

>>> import board; i2c = board.I2C()
Traceback (most recent call last):
  File "", line 1, in 
ValueError: Invalid pins

Description

In https://github.com/adafruit/circuitpython/blob/main/ports/raspberrypi/boards/adafruit_metro_rp2350/mpconfigboard.h, it has:

#define DEFAULT_I2C_BUS_SCL (&pi...
slender iron
#

@devout jolt got time to test?

devout jolt
#

Yep in about 40 minutes

slender iron
#

πŸ‘

manic glacierBOT
slender iron
devout jolt
manic glacierBOT
tulip sleet
manic glacierBOT
#
Sola85
[adafruit/circuitpython] Issue opened: #9521 Third sleep option: RAM-preserving "deep" sleep

CircuitPython currently supports two sleep modes: Light sleep and deep sleep. Given the specification of the two existing sleep modes, and the capabilities of many microcontrollers supported by CP, I think there is room for one more sleep mode, which would have a clear benefit. Here is my rationale:

  • Light sleep promises to keep background tasks such as WIFI, BLE, audio playback, etc alive. These promises make it hard to actually save any power with light sleep. This is even stated in the...
manic glacierBOT
#
bablokb
[adafruit/circuitpython] New comment on issue #9521: Third sleep option: RAM-preserving "deep" sleep

Have you tested anything of this? For a POC, this could be tested without an API change just to see if it is worth it. If you have any code to test, I would be happy to give it a try.

As a side note: there are currently a few issues open regarding sleep-modes (e.g. #9464 or #8977) so IMHO it would make more sense to fix the existing issues first with the existing sleep-modes before comparing results with a new third one.

spare jacinth
#

The circuitpython-stubs are based on doc comments (//| ... ) and dont get checked in any way against the actual logic exposed to the VM, right?

#

Mostly, im "worried" about using them to check whether blinka adheres to the APIs naming/typing
For example, i currently dont know if the readonly properties in here are blinka protecting them "for no reason" or the stubs not marking them as such πŸ˜…

manic glacierBOT
#
dhalbert
[adafruit/circuitpython] New comment on issue #9521: Third sleep option: RAM-preserving "deep" sleep

The original light-sleep/deep-sleep distinction was based on a misinterpretation of what was preserved in Espressif light sleep. In retrospect, we might have just dropped light sleep. As you mentioned, "Light sleep promises to keep background tasks such as WIFI, BLE, audio playback, etc alive.", which is not necessarily all that useful. So redefining light sleep as RAM-preserving, but not activity-preserving, might be worthwhile. I am not sure that many people even use light sleep now, as...

manic glacierBOT
slender iron
slender iron
#

gonna try and update to idf 5.3 today

manic glacierBOT
manic glacierBOT
manic glacierBOT
#
ladyada
[adafruit/circuitpython] New comment on pull request #9475: Feather ESP32-C6 Rev B pin changes

oki thats fine then

On Wed, Aug 14, 2024 at 1:50β€―PM Scott Shawcroft @.***>
wrote:

@.**** commented on this pull request.

In ports/espressif/boards/adafruit_feather_esp32c6_4mbflash_nopsram/pins.c
https://github.com/adafruit/circuitpython/pull/9475#discussion_r1717341366
:

@@ -66,18 +68,15 @@ static const mp_rom_map_elem_t board_module_globals_table[] = {
{ MP_ROM_QSTR(MP_QSTR_IO7), MP_ROM_PTR(&pin_GPIO7) },
...

manic glacierBOT
spare jacinth
lone sandalBOT
manic glacierBOT
manic glacierBOT
manic glacierBOT
#
tschundler
[adafruit/circuitpython] New comment on issue #673: DMX (lighting protocol)

I independently came up with something very similar to https://github.com/adafruit/circuitpython/issues/673#issuecomment-616831705 from @Tasm-Devil switching the transmit pin between digital out (for Break + Mark-after-Break) and uart (for start condition + data)

That works on the ESP32s3 I use.

Notable difference is that time.sleep has a minimum increment of 1ms. to asking it to sleep 8uS for the Mark-after-break ends up being really 1ms. But initializing the uart takes more than the...

lone sandalBOT
manic glacierBOT
manic glacierBOT
#
konstantint
[adafruit/circuitpython] New comment on commit `090f330`

How can one use the keyboard from the CircuitPython program? E.g. if I just try to manually create a DemuxKeyMatrix object, CircuitPython justifiably complains that "KB_A_0 is in use". Indeed, the keyboard has been configured for scanning at board initialization time in this commit.

Is there a way to get a hold of the event queue the keyboard events are being pushed by the pre-initialized object? Alternatively, is there a way to deinitialize the keyboard scanner (and free the pins) so that...

manic glacierBOT
#
joshua-beck-0908
[adafruit/circuitpython] New comment on commit `090f330`

How can one use the keyboard from the CircuitPython program? E.g. if I just try to manually create a DemuxKeyMatrix object, CircuitPython justifiably complains that "KB_A_0 is in use". Indeed, the keyboard has been configured for scanning at board initialization time in this commit.

Is there a way to get a hold of the event queue the keyboard events are being pushed by the pre-initialized object? Alternatively, is there a way to deinitialize the keyboard scanner (and free the pins) s...

plucky tulip
#

Hi, do we have any option to access translations that CircuitPython uses inside .py files?

blissful pollen
plucky tulip
#

I want to add strftime to adafruit_datetime and I am looking for a way to get translated months and days names. Getting them from CircuitPython installed on the board seems logical, as the uf2 files are already localized.

blissful pollen
manic glacierBOT
manic glacierBOT
#
bablokb
[adafruit/circuitpython] New comment on issue #9521: Third sleep option: RAM-preserving "deep" sleep

I am not sure that many people even use light sleep now, as it's not that useful.

@dhalbert : I am using it in all my battery-based projects, at least for RP2040-based boards replacing time.sleep() with light-sleep. It saves about 30%.

Light-sleep is currently not fully activity preserving. One thing I noticed is that driving a ST7789 with backlight set to 0.7 fails during light-sleep, presumably because of the PWM not working as expected.

manic glacierBOT
#
konstantint
[adafruit/circuitpython] New comment on commit `090f330`

I didn't realize the keys are accessible via stdin (just curious, where in this commit is this functionality hidden?)
I do not think this would be enough, though. I'm trying to use the keyboard as a sound controller of sorts and would like to get access to individual key up / key down events. Would that be possible?
If not now, could the board_keyboard object be somehow exposed as board.KEYBOARD to the Python code here?

manic glacierBOT
#
joshua-beck-0908
[adafruit/circuitpython] New comment on commit `090f330`

I didn't realize the keys are accessible via stdin (didn't guess by just looking at the code that board_serial_xxx does that).

I do not think this would be enough, though. I'm trying to use the keyboard as a sound controller of sorts and would like to get access to individual key up / key down events. Would that be possible? If not now, could the board_keyboard object be somehow exposed as board.KEYBOARD to the Python code here?

How is it set up

The keyboard is set up l...

#
Sola85
[adafruit/circuitpython] New comment on issue #5318: nRF: deep sleep increases power usage

This might be completely unrelated, but I was also debugging why my nrf draws around 0.5-2 mA in deep sleep. I had detachted the USB data connection, so I was measuring true deep sleep and not fake deep sleep. My mistake was that while I had disconnected USB data, USB power was still connected. The mcu appears to draw ~1mA from VDD in sleep mode if VBUS is at 5V.

If I detach VBUS, I get ~ 2Β΅A deep sleep with only time alarms and around 20 Β΅A with a pin alarm active.

manic glacierBOT
#
konstantint
[adafruit/circuitpython] New comment on commit `090f330`

I see, thanks for the detailed write-up.

Having access to individual key-up/key-down events is pretty important to be able to implement any sort of game-like interaction, however (which, for my personal purposes, is way more important than the built-in terminal. Neither do I want to be able to ctrl+C out of the uploaded program).

It would be nice then, if we could come up with a slighly looser integration here. Either of the following would work:

  • Ability to disable the built-in key...
manic glacierBOT
manic glacierBOT
#
tyeth
[adafruit/circuitpython] New branch created: update-c6-IO14-pin-name
manic glacierBOT
#
tyeth
[adafruit/circuitpython] Pull request opened: #9524 Update C6 IO14 pin name

There's a pin named IO14 for the Feather ESP32-C6, but it's marked 12 on the silkscreen so was sticking in a PR to rename IO12 matching the other silkscreen pin naming (IO pin names), but not quite sure how I verify the GPIO name is correct as there is no PCB repo / guide yet. It's set as GPIO14 currently:
https://github.com/adafruit/circuitpython/blob/main/ports/espressif/boards/adafruit_feather_esp32c6_4mbflash_nopsram/pins.c#L57-L58

This closes #9523

turbid radish
#

Hey CircuitPython Networking folks: We've created a new guide to help people with networking in CircuitPython: https://learn.adafruit.com/networking-in-circuitpython - We know that it isn't a baby steps guide or that it completely lets one know the differences between when you need to use requests vs connection_manager vs both but it's a start.

Networking in CircuitPython

Connect to the internet with microcontrollers

#

I'd like your help. While the guide is 13 pages, it likely needs to contain more. It needs to be unambiguous and be sure it explains things and not just show examples.

#

If you'd like to help, please email me anneb(at)adafruit . com with your constructive suggestions. If you're busy, it can wait a few days, as long as the community works to get this right. I know we can get to a guide that addresses the needs of beginners and seasoned folks. Thanks!

manic glacierBOT
slender iron
manic glacierBOT
slender iron
#

I gotta get back to merging translations first

#

The bundle action was suspended too. We should move the daily reports and bundle updates to the circuitpython repo

manic glacierBOT
slender iron
willow totem
#

1sec

#

nah same flashing yellows. Oooh wait, reset and got same, and now it's self reset and cleaned up filesystem and seems to have booted. Success!

devout jolt
willow totem
slender iron
#

great!

#

I think it may have been that the status pixels are dotstars

willow totem
#

oh classic

slender iron
#

and the deinit would hang because we lock them but don't release

willow totem
#

aha, thanks for the details!

devout jolt
willow totem
#

maybe a debug build. gotta drop so much

slender iron
#

I added a wait for a lock in deinit when I added the max support

#

ya, debug build

devout jolt
#

gotcha

slender iron
#

gotta make space for logging

manic glacierBOT
slender iron
#

@tulip sleet just pushed the IDF update PR. It'll likely overfill some boards but I did build one of each chip locally first

tulip sleet
slender iron
#

I checked that it runs on funhouse

#

yup, ideally we get it out with the next alpha

tulip sleet
#

maybe i'll test a tft feather

slender iron
#

gonna eat while the CI churns

manic glacierBOT
manic glacierBOT
#
ozzloy
[adafruit/circuitpython] Issue opened: #9527 seeed xiao ble circuitpyhton 9.1.1 bluetooth, core code crashed hard whoops!

CircuitPython version

adafruit-circuitpython-bundle-9.x-mpy-20240730
adafruit-circuitpython-seeed_xiao_esp32c6-en_US-9.1.1.bin
seeed xiao ble

Code/REPL

from adafruit_ble import BLERadio

radio = BLERadio()
print("scanning")
found = set()
for entry in radio.start_scan(timeout=60, minimum_rssi=-80):
    addr = entry.address
    if addr not in found:
        print(entry)
    found.add(addr)

print("scan done")

Behavior

Auto-reload ...
#
ozzloy
[adafruit/circuitpython] New comment on issue #9527: seeed xiao ble circuitpyhton 9.1.1 bluetooth, core code crashed hard whoops!

got some help in discord, tried this sample code
https://github.com/adafruit/Adafruit_CircuitPython_BLE/blob/main/examples/ble_simpletest.py
got this output

Auto-reload is on. Simply save files over USB to run them or enter REPL to disable.
code.py output:
scanning
ESP-ROM:esp32c6-20220919
Build:Sep 19 2022
rst:0xc (SW_CPU),boot:0x1f (SPI_FAST_FLASH_BOOT)
Saved PC:0x4001975a
SPIWP:0xee
mode:DIO, clock div:2
load:0x40875720,len:0x124
load:0x4086c410,len:0xbe0
load:0x4086e610...
manic glacierBOT
#
ozzloy
[adafruit/circuitpython] New comment on issue #9527: seeed xiao ble circuitpyhton 9.1.1 bluetooth, core code crashed hard whoops!

after hitting reset and reconnecting serial, running the same sample code,

i saw meaningful output:

Auto-reload is on. Simply save files over USB to run them or enter REPL to disable.
code.py output:
scanning
<Address 66:df:81:01:b7:62> <Advertisement flags=<AdvertisingFlags  > >
        Advertisement(data=b"\x0e\xff\x03\x37\x51\x10\x7a\x87\xdb\x36\xcd\xbc\x36\x54\x56\x02\x01\x18\x03\x0...
manic glacierBOT
manic glacierBOT
manic glacierBOT
manic glacierBOT
#
djairjr
[adafruit/circuitpython] New comment on issue #9384: Pi Pico-Filesystem (CIRCUITPY) mounted Read-Only for versions >= v9 on Debian 11.10

I'm working on a version of Circuitpython for the Xiao ESP32S3 Sense and I had the same problem. In my case, I have a machine with Dual Boot (Ubuntu 22.03 and Windows 11). I carried out all the customization of the source files in my fork on Github and cloned the Adafruit repository. Then, on Linux, I moved the custom files and compiled the firmware.
When I start the machine in Linux, I have the CIRCUITPY drive mounted and read and write access. But when I boot into Windows, I don't have acc...

manic glacierBOT
#
djairjr
[adafruit/circuitpython] Issue opened: #9528 Filesystem READ ONLY on Circuitpython 9.1.1 and 9.2.0 builds for Seeed Xiao ESP32S3_Sense

CircuitPython version

Adafruit CircuitPython 9.1.1 on 2024-08-16; Seeed Xiao ESP32-S3 Sense with ESP32S3

Code/REPL

>>> f = open('settings.toml', 'w')
f.write('CIRCUITPY_WIFI_SSID = "wifissid"\n')
f.write('CIRCUITPY_WIFI_PASSWORD = "wifipassword"\n')
f.write('CIRCUITPY_WEB_API_PASSWORD = "webpassword"\n')
f.close()
Traceback (most recent call last):
  File "", line 1, in 
OSError: [Errno 30] Read-only filesystem
>>>

Behavior

Can't save any...

#
konstantint
[adafruit/circuitpython] Pull request opened: #9529 M5Stack Cardputer keyboard handling improvements.
  • Exposed the DemuxKeyMatrix object as board.KEYBOARD.
  • Implemented the ability to not have the keyboard assigned as serial input (sys.stdin) by specifying M5STACK_CARDPUTER_KEYBOARD_SERIAL=0 in settings.toml. This enables custom handling of individual key up/key down events.

For context, see discussion in https://github.com/adafruit/circuitpython/commit/090f330ad5e30acf2030900b45386d784cbcc1da

manic glacierBOT
manic glacierBOT
#
dhalbert
[adafruit/circuitpython] New comment on issue #9528: Filesystem READ ONLY on Circuitpython 9.1.1 and 9.2.0 builds for Seeed Xiao ESP32S3_Sense

The read-only is by design. CIRCUITPY is read-only from your code unless you remount it as read/write in boot.py. It is read/write from the host computer that has it mounted as a drive. This is to prevent corruption of CIRCUITPY if both the host computer your code try to write to it at the same time.

This is not true for boards like C3, C6, and plain ESP32, which do not present a CIRCUITPY USB drive.

#
joshua-beck-0908
[adafruit/circuitpython] New comment on commit `090f330`

It's great that you've already written some code to flesh out how this will work.
I do like the idea of controlling the keyboard mode with a config variable. Have you tested this code?
However, I wonder if there's a method to toggle whether the serial driver takes events from the queue.
One that can be controlled by the program.
I'll look over this tomorrow when I have some time.

Are you on the Adafruit Discord?
You'd be able to get support there, and you'd be able to send me a DM i...

manic glacierBOT
#
djairjr
[adafruit/circuitpython] New comment on issue #9528: Filesystem READ ONLY on Circuitpython 9.1.1 and 9.2.0 builds for Seeed Xiao ESP32S3_Sense

No. It was not the problem.
I am using many other boards here. They all work.
When I am in linux, using fsck I got a message saying something about a dirty byte.
Again, in my Ubuntu boot, the CIRCUITPY is mounted as usual.
In Windows, only Serial port appears, but no Disk at all.

Thonny was able to access files in Circuitpython File system. But the Windows Explorer no. This happens only in windows.

#
djairjr
[adafruit/circuitpython] New comment on issue #9528: Filesystem READ ONLY on Circuitpython 9.1.1 and 9.2.0 builds for Seeed Xiao ESP32S3_Sense

It's been a long time since I compiled Circuitpython directly from source. The last time was with Seeed's other card, the RP2040.
I have several boards here, including some ESP32C3, which are working perfectly using WebESPTool and the firmware available on your website.

I'm trying to make the Seeed Xiao ESP32 S3 Sense board compatible with Circuitpython, a project I had already started some time ago.

My system compiles the firmware properly, without any errors. And I can save the .bin ...

tough night
#

[adafruit/circuitpython] Pull request op...

manic glacierBOT
manic glacierBOT
manic glacierBOT
slender iron
manic glacierBOT
#
dhalbert
[adafruit/circuitpython] Pull request review submitted: #9526 Fix Funhouse and update to IDF 5.3

Testing:

Metro ESP32-S3: simple BLE and WiFi HTTPS requests worked fine.

FunHouse: originally I couldn't get it to come up with serial or CIRCUITPY. I erased the FunHouse and reloaded the UF2 bootloader, and at first it was fine. But then it failed again. To reproduce, erase and load .bin, or load TinyUF2 and the .uf2. It will come up, but if you do import wifi, it hangs or crashes, with no output. If a settings.toml is present, the status line will come up with No IP, and a...

#
dhalbert
[adafruit/circuitpython] New comment on pull request #9526: Fix Funhouse and update to IDF 5.3

More testing:

Metro ESP32-S2: this has the same ESP32-S2 WROVER module as the FunHouse. Erased completely, loaded .bin. import wifi does not crash. Copied over settings.toml with CIRCUITPY_WIFI_*, values, comes up, and reports an IP in the status line.

So again the problem seems to be FunHouse specific.

I could go ahead and merge this and you could keep working on FunHouse, if that makes sense.

tulip sleet
#

@slender iron weird -- FunHouse is still failing when wifi is started. ESP-IDF 5.3 seems fine for other boards I tested.

manic glacierBOT
slender iron
manic glacierBOT
#
MrPatben8
[adafruit/circuitpython] New comment on issue #9136: storage.usb_drive_enabled()

I can't get storage.disable_usb_drive() to work. I'm running CircuitPython 9.x on a Pi Pico and I've got a boot.py with the following code:

import board
import digitalio
import storage

button = digitalio.DigitalInOut(board.GP24)
button.direction = digitalio.Direction.INPUT
button.pull = digitalio.Pull.UP

if button.value:
    storage.disable_usb_drive()

But regardless of what I try, the USB drive inevitably pops up. Any ideas what could be the problem?

tulip sleet
slender iron
#

Fine with me! It's an alpha

tulip sleet
#

will do, and will leave the FUnHouse bug open

manic glacierBOT
slender iron
manic glacierBOT
slender iron
#

(Though I'll have bundle script changes too)

slender iron
#

@devout jolt which pimoroni 2350 boards do you have?

#

I think I'll do board defs during my deep dive

devout jolt
manic glacierBOT
#
konstantint
[adafruit/circuitpython] New comment on pull request #9529: M5Stack Cardputer keyboard handling improvements.

Would you prefer to do this dynamically instead?

This is a reasonable idea. I didn't go for it immediately as I was not sure what would be the best interface for it and simply tried the simplest thing that came to mind first.

Intuitively it currently seems to me that a reasonably clean solution would be to have an M5StackCardputerKeyboard class, exposed as board.KEYBOARD, which could a DemuxKeypad with the extra functions:
- attach_serial/detach_serial that would enable/dis...

slender iron
devout jolt
slender iron
#

πŸŽ‰

manic glacierBOT
manic glacierBOT
willow totem
#

Hey @slender iron , I was going to add the links to the modules lists on boards pages (after imaginging they existed earlier 😁 ). I've got that sorted but the frozen modules are more tricky, mainly removing adafruit_ from the name and sticking after https://docs.circuitpython.org/projects/ works, e.g. https://docs.circuitpython.org/projects/display_text but not for connection_manager as it's page is without the underscore. Can RTD do aliases to fix that or do I need a more cunning plan?

orchid basinBOT
#
tyeth
[adafruit/circuitpython-org] Pull request opened: #1453 Turn libraries list on board pages into links

This turns the library list on board pages into clickable docs links.
It supports having class links like busio.SPI and frozen modules.
There were a few modules that don't have docs (_asyncio for example), and one frozen module (adafruit_connection_manager) that doesn't follow the expected path, but ignoring that it's working well.
Tested on the pico page (for normal modules), and pyportal (for frozen modules)

willow totem
#

My circuitpython day contribution πŸ˜‰

manic glacierBOT
manic glacierBOT
#
Sola85
[adafruit/circuitpython] Pull request opened: #9534 nordic: reduce sleep current by using SENSE feature for PinAlarm when possible

This PR decreases the power consumption of sleep modes with active pin interrupts by ~10x by using the SENSE feature, when possible.
"Normal" GPIOTE interrupts increase sleep current consumption by ~20Β΅A, but using SENSE (aka PORT events) instead can reduce this to ~2Β΅A.

This behaviour is documented as an Erratum here and discussed for Zephyr [here](https:/...

manic glacierBOT
manic glacierBOT
manic glacierBOT
manic glacierBOT
#
djairjr
[adafruit/circuitpython] New comment on pull request #9530: Config that works on Seeed Xiao ESP32S3

Okay, guys. I have now got a working version for the board. I tested the camera with the Xiao Round Display and everything works great.
The problem with the USB_PID and USB_VID still needs to be addressed and this is preventing my changes from being included.

But I'll see about that with Seeed.

Running USB VID/PID and Creator/Creation ID Duplicate Checker...
Duplicate VID/PID usage found!
- VID/PID: 0x303A:0x80D1
 Boards: seeed_xiao_esp32_s3_sense, unexpectedmaker_tinys3
If yo...
lone sandalBOT
manic glacierBOT
#
icebreaker87
[adafruit/circuitpython] Issue opened: #9535 ESP32-S3 I2C timeout

CircuitPython version

Adafruit CircuitPython 9.1.1 on 2024-07-22; Adafruit Feather ESP32S3 No PSRAM with ESP32S3

Code/REPL

import time
import busio
import board
from micropython import const
from adafruit_seesaw.seesaw import Seesaw

BUTTON_X = const(6)
BUTTON_Y = const(2)
BUTTON_A = const(5)
BUTTON_B = const(1)
BUTTON_SELECT = const(0)
BUTTON_START = const(16)
button_mask = const(
    (1 << BUTTON_X)
    | (1 << BUTTON_Y)
    | (1 << BUTTON_...
manic glacierBOT
#
deshipu
[adafruit/circuitpython] New comment on issue #9535: ESP32-S3 I2C timeout

It's hard to be sure without looking at the signals on the bus, but it looks like the seesaw device is crashing leaving the SCL line pulled low, that's why you get a timeout when the CircuitPython side waits on clock stretching, and it fails the test for pull-up resistors (because the line is kept low by the other side). Might be some combination of a bug in the seesaw device code, and different timings on different ports?

manic glacierBOT
wraith crow
#

@brazen hatch I'm not having any luck flashing the Lilygo T-Watch 2020 v3 from circuitpython.org. CP flash/bootloader organization seems like black magic to me so I'm probably doing something obviously wrong but the first thing I did was flash the combined.bin for the S3 chip but I realized there isn't a UF2 for the watch, so I then flashed the circuitpython.org BIN file to address 0x0000 but that leads to a boot loop "invalid header". I've tried erasing the flash before attempting the firmware and I can get the default flash back on the device, but no luck with CP. Any pointers 😁?

brazen hatch
#

You could try to set -fm qio and -ff 80m

#

but idk if the flash params are the same as in the S3 version

wraith crow
#

Oh, I didn't look close enough 😁, I thought the circuitpythong.org binary was from your recent PR. I just got the S3 version and was trying to flash that, I guess I need to go to the artifacts.... Thanks!!!!

brazen hatch
#

oh lmao, no .org is not updated yet

#

wait one moment

#

here is the s3

#

The uf2 should also work, i'd give it another try with the boot button held down

#

if you have trouble, you could also try from CP with microcontroller.RunMode

#

it's a tricky board to reset

manic glacierBOT
wraith crow
brazen hatch
wraith crow
wraith crow
brazen hatch
manic glacierBOT
manic glacierBOT
manic glacierBOT
#
ZodiusInfuser
[adafruit/circuitpython] New review comment on pull request #9531: Add four Pimoroni RP2350 boards

GP26, GP27, GP28, do not offer Analog functionality on the B chip.

These should instead just be:

    { MP_ROM_QSTR(MP_QSTR_GP26), MP_ROM_PTR(&pin_GPIO26) },
    { MP_ROM_QSTR(MP_QSTR_GP27), MP_ROM_PTR(&pin_GPIO27) },
    { MP_ROM_QSTR(MP_QSTR_GP28), MP_ROM_PTR(&pin_GPIO28) },

For ADC access these additional lines are needed in relevant place:

    { MP_ROM_QSTR(MP_QSTR_GP40_A0), MP_ROM_PTR(&pin_GPIO40) },
    { MP_ROM_QSTR(MP_QSTR_GP40), MP_ROM_PTR(&pin_GP...
#
ZodiusInfuser
[adafruit/circuitpython] New review comment on pull request #9531: Add four Pimoroni RP2350 boards

This board has our new SP/CE connector, so can have these pin definitions adding:

    { MP_ROM_QSTR(MP_QSTR_SPICE_CS), MP_ROM_PTR(&pin_GPIO33) },
    { MP_ROM_QSTR(MP_QSTR_SPICE_SCK), MP_ROM_PTR(&pin_GPIO34) },
    { MP_ROM_QSTR(MP_QSTR_SCK), MP_ROM_PTR(&pin_GPIO34) },
    { MP_ROM_QSTR(MP_QSTR_SPICE_TX), MP_ROM_PTR(&pin_GPIO35) },
    { MP_ROM_QSTR(MP_QSTR_MOSI), MP_ROM_PTR(&pin_GPIO35) },
    { MP_ROM_QSTR(MP_QSTR_SPICE_RX), MP_ROM_PTR(&pin_GPIO32) },
    { MP_ROM_QSTR(M...
brazen hatch
#

Oh btw, i've extensively tested idf5.3, no breakage.

#

It's on the watch now too, gotten about a day of uptime, jumping between mcu frequencies.

manic glacierBOT
#
bill88t
[adafruit/circuitpython] New comment on pull request #9529: M5Stack Cardputer keyboard handling improvements.

I think the ideal api would be booting with it as stdin, then whenever the user calls board.KEYBOARD() the keyboard is detached from stdin.
Then when that object is deinitialized (or expires from scope, finalizers?), it's attached back to stdin.

I don't think it's wise to try to get it under shared-bindings since every keyboard board will need different hacks.
It'd be more function arguments than actual code. (Just look at a cardputer, the t-deck and your keyboard, you'll get what I ...

#
bill88t
[adafruit/circuitpython] Issue opened: #9536 Espressif 20Mhz operation breaks usb operation till next reset

CircuitPython version

Any > #9342, discovered on S3

Code/REPL

import microcontroller
microcontroller.cpu.frequency = 20_000_000 # 20MHz

Behavior

Usb goes away. This is fine as by itself, as usb cannot operate with the board being at 20MHz.
However when switched back to a higher clock, usb does not start back up, even if the board is unplugged and plugged back in (assuming it's battery powered).

Description

No response

Additional in...

lone axle
#

it looks like there was some exception raised during the generation of the 'state of circuitpython' report. I'm trying to run it locally with some additonal print statements to try to get a better idea about what the specific issue is.

#

It traces to here: https://github.com/adafruit/adabot/blob/main/adabot/lib/bundle_announcer.py#L70 basically comps is None. It's not clear at all to me why though. I added some print statements wrapping the line that creates comps

print(f"lib: {lib}")
comps = parse.parse("[{name:S}]({link_comp:S})", lib)
print(f"comps: {comps}")

They show the value of lib and that indeed comps is None:

lib: [rfm](https://github.com/adafruit/Adafruit_CircuitPython_RFM/releases/1.0.0)
comps: None
Exception Occurred! ...

But it doesn't really make sense to me why comps ends up None In a REPL when I run the essence of this parse operation it seems to succeed and get a legimate value that isn't None

>>> import parse
>>> lib = "[rfm](https://github.com/adafruit/Adafruit_CircuitPython_RFM/releases/1.0.0)"
>>> comps = parse.parse("[{name:S}]({link_comp:S})", lib)
>>> print(comps)
>>> <Result () {'name': 'rfm', 'link_comp': 'https://github.com/adafruit/Adafruit_CircuitPython_RFM/releases/1.0.0'}>
spare jacinth
#

what if you print like {lib!r} (or {repr(lib)}, which is the same thing)?

#

perhaps whitespace (or some other invisible char) messing tthings up

#

FYI, using {lib =} would print both variable name and value, and you wouldn't need to write "lib: " on the string πŸ˜‰

lone axle
#

Thanks, I've never seen that syntax.

spare jacinth
#

πŸ€“ ```py

lib = "hello"

f"{lib = !r}"
"lib = 'hello'"

f"{lib=!r}"
"lib='hello'"

lone axle
#

I also started wondering about whitespace causing trouble. I'm running it again now with some additional printing. I did confirm in a REPL that if there were mis-matched whitespace that parse.parse() does return None so it does seem like a plausible theory.

#

Does circuitpython support that syntax too?

spare jacinth
#

idk

#

it's quick to check, though :p

#

(if you have a device handy, unlike me, that is)

lone axle
#

Do you know if there is a name for this syntax? Or where to find reference to it in the docs. I'm curious now if there are other helpers that I don't know about.

spare jacinth
#

if you look for something like "f-strings cheatsheet" there should be nice images around

#

i have no idea how the !r or var = constructs are named (if they have a name to begin with)

#

if i were to give them a name, i would call most of the things you can do in f-strings as "format specifier" but that's a very vague and generic name

lone axle
#

it does appear that whitespace, specifically a \r might be the issue.

lib_components: ['[rfm](https://github.com/adafruit/Adafruit_CircuitPython_RFM/releases/1.0.0)\r']
spare jacinth
#

perhaps safer to .strip() the value before using parse on it

austere acorn
#

Anyone know what changed in memory management between CircuitPython 8.2.9 and 9.1.1, and the respective versions of adafruit_display_shapes lib, that would make code fail on CP 9.1.1 with

File "code.py", line 103, in <module> File "adafruit_display_shapes/roundrect.py", line 64, in __init__ MemoryError: memory allocation failed, allocating 26624 bytes

Code is here: https://gist.github.com/snkYmkrct/a4cb0d6d44bf6e2350225036db4f2197

crimson ferry
#

I honestly don't understand the new model fully (just when I'd gotten my arms around the old one), maybe someone with knowledge can help

austere acorn
lone axle
#

I pasted in the report info for today and added the parts that got cutoff of the automated one by the exception. It did run successfully locally with after stripping the whitespace. There were 2 of the strings that contained \r and the rest didn't. Not sure about why the difference, but this does resolve it.

lone axle
# austere acorn Anyone know what changed in memory management between CircuitPython 8.2.9 and 9....

I don't know any more about the memory inner-workings. but it might be worth a try to replace adafruit_display_shapes usage here:

game_board_frame = RoundRect(INGROUPXOFFSET-8, INGROUPYOFFSET-8,
                            HTILES*TILESIZE+16, VTILES*TILESIZE+16, 8, fill=BLACK, outline=PINK, stroke=3)

with an inflated tilegrid (https://github.com/adafruit/Adafruit_CircuitPython_ImageLoad/blob/main/adafruit_imageload/tilegrid_inflator.py#L32).

I haven't done rigorous testing, but my understanding is that shapes will make a bitmap of the full size, while Tilegrid gets represented internally in a more compact way storing indexes for each tile rather than raw graphical data.

austere acorn
#

I vote for a memory management deep dive show! πŸ˜†

crimson ferry
#

there's an optional arg to .load_font to load only the glyphs you use ...may trim things slightly

austere acorn
#

Yea, for sure, I will try some memory saving tips.
I was just curious what changed that made the code stop working between CP 8 and 9

slender iron
#

@lone axle thanks for fixing the report!

manic glacierBOT
#
djairjr
[adafruit/circuitpython] New comment on issue #9528: Filesystem READ ONLY on Circuitpython 9.1.1 and 9.2.0 builds for Seeed Xiao ESP32S3_Sense

I'll leave the observation here, in case anyone has the same type of problem. The folks at Seeed Studio still need to answer some questions regarding USB device registration. My Windows system was mounting the CIRCUITPY drive as read-only because it was not recognizing the drive as a drive. Probably due to a problem in the registry, which identified the USB device incorrectly.

I added values ​​from other boards (from ESPRESSIF itself) and managed to perform a first compilation. With the he...

lone axle
#

I'm tempted to try to figure out why some of the tests take an hour when they run in actions. I did see messages about API rate limit requests per hour in the adabot print outs though. I wonder if the tests and other actions add up to over the per hour limit so it waits for them to get replenished.

manic glacierBOT
slender iron
#

<@&356864093652516868> We'll have our weekly meeting in about 50 minutes from now in this text channel and in the circuitpython voice channel. Please take the time to add your notes in advance to the document: https://docs.google.com/document/d/13Q0vqnzMAVGwjdVG3-lXMxt62ouymSQ134-o-ROgXBM/edit -- I look forward to everyone's updates!

Google Docs
CircuitPython Weekly Meeting for August 19, 2024

CircuitPython Weekly Meeting for August 19, 2024 Welcome to the CircuitPython Weekly meeting notes! Feel free to add your Hug Reports and Status Updates before the meeting, alphabetically by your username. During the meeting, we go through them in order. If you can’t make the meeting and woul...

manic glacierBOT
#
konstantint
[adafruit/circuitpython] New comment on pull request #9529: M5Stack Cardputer keyboard handling improvements.

Given that for a few of the other boards that I've seen board.XXXXXX seems to always be either a pin or a display object, making board.KEYBOARD a function would feel rather inconsistent. I'd also find the behavior of detaching from stdin on object access pretty unintuitive and even more intutiive if anything would happen to it automagically on scope expiration.

Something like an explicit attach_keyboard_serial / detach_keyboard_serial would be a way more straighforward interface for th...

onyx hinge
#

According to plan I'll be back (but jet lagged) next week. Group hug. Photo: spice market in Mombasa Kenya

PXL_20240818_092747665.jpg
solar whale
#

Nice! I can smell it! I have fond memories of several trips to Mombassa for work in 1976 and 1979.

slender iron
manic glacierBOT
gilded cradle
#

Waiting on plumber who could show up anytime, so I'm text-only this week.

lone axle
turbid radish
manic glacierBOT
slender iron
#

@spare jacinth want to talk on voice about your in the weeds or should I read it?

hoary moat
#

🐈

spare jacinth
#

i dont mind either way

#

my english will be worse than yours, that's for sure πŸ˜›

#

yeah, i meant out-of-sync issues, will trust what the stubs say then

random junco
spare jacinth
#

it's fine, thanks!

lone axle
#

Thanks for hosting Scott, have a great week everyone!

slender iron
#

np. thanks for fixing the stats!

slender iron
#

Here is the notes document for next Monday’s CircuitPython Weekly Meeting. It is at the normal time of 11am Pacific / 2pm US Eastern here on Discord. Add your hug reports and status updates to the document before the meeting. If you are unable to attend but would still like to contribute, feel free to add your notes and we’ll read them off during the meeting. Hope to see you there! <@&356864093652516868> https://docs.google.com/document/d/1AlQdjcpzMwWRF_2UsoXykCqwJVCPx5kSR14VG4ozEOo/edit?usp=sharing

Google Docs
CircuitPython Weekly Meeting for August 26, 2024

CircuitPython Weekly Meeting for August 26, 2024 Welcome to the CircuitPython Weekly meeting notes! Feel free to add your Hug Reports and Status Updates before the meeting, alphabetically by your username. During the meeting, we go through them in order. If you can’t make the meeting and would...

slender iron
manic glacierBOT
manic glacierBOT
spare jacinth
#

was taking a look at the issues for "type annotations are missing", first one in the list is Adafruit_CircuitPython_BLE_Eddystone on which i dont get mypy to error on the lines specified on the issue because it bails out as soon as it sees the imports coming from adafruit_ble not being typed (stubs or py.typed)

so... is there a reason why such marker is not in there alredy? adding it on my machine's copy of the lib gets mypy to go further but it's a pretty ugly hack around

#

though, BLE's typing seems to be very broken, when running mypy . in the repo, it yells about a bunch of stuff

manic glacierBOT
#
tannewt
[adafruit/circuitpython] New comment on issue #9498: Qualia / DotClockFrameBuffer unexpectedly renders a few lines of pixels at once in the most bizarre way

I experienced really slow updates on Qualia w 2.1 round display when using any display rotation. Even calling rotation=0 seemed to cause slow loading if I recall.

Is this new? Loading from flash into PSRAM is going to be limited by the speed of their shared memory bus.

Rotation can be bad for loading since it may cause skipping through the file system to go column by column instead of row by row which wouldn't. This isn't new and I'd suggest rotating the asset to align with the displa...

orchid basinBOT
manic glacierBOT
#
mrdalgaard
[adafruit/circuitpython] New comment on pull request #9534: nordic: reduce sleep current by using SENSE feature for PinAlarm when possible

@Sola85
Have you done any tests on what the reduced accuracy of using SENSE instead of GPIOTE is? (or have you found any documentation that states it)
I'm using light sleep with pin alarms to detect events in the ~30ms range, so just wondering if I will have reduced accuracy with this change.
Otherwise awesome with more powersavings on an already very low power chip.

slender iron
#

@tulip sleet do you have a script to update board file headers?

tulip sleet
#

it's in a PR, hold on

slender iron
#

πŸ‘

tulip sleet
slender iron
#

thanks!

manic glacierBOT
orchid basinBOT
manic glacierBOT
spare jacinth
manic glacierBOT
#
konstantint
[adafruit/circuitpython] New comment on pull request #9529: M5Stack Cardputer keyboard handling improvements.

... so I looked a bit into the directory structure and it feels to me that creating something like a espressif/bindings/m5stack_cardputer_keyboard, along with the need to introduce the pretty niche makefile variable and effectively require every build to indirectly be aware of this, is not nice at all. Things would then become especially ugly if the approach starts proliferating for any other boards that might want to implement a similar board-specific module.

I'm therefore pretty convinc...

manic glacierBOT
#
Casa-Machinalia
[adafruit/circuitpython] New comment on issue #9535: ESP32-S3 I2C timeout

I'm having a very similar issue with I2C on QT-PY-ESP32-S2. I've come at it all kinds of ways, with one or multiple I2C devices and varying the I2C bus speed from 10_000 to 400_000.

With these i2c devices:

  • adafruit_ssd1306

  • adafruit_seesaw (rotaryio: product number 4991)

When I do a i2c.scan() it never shows any devices, but I can "speak" to the rotaryio. When I enable the ssd1306, I get:

main.py output:
sys.implementation=(name='circuitpython', version=(9, 1, 1), ...
manic glacierBOT
#
dhalbert
[adafruit/circuitpython] Pull request review submitted: #9538 Espressif: I2C timeouts calculated based on tick rate.

In V5.3, i2c_master_write_to_device() is deprecated and replaced with i2c_master_transmit(), which takes the timout in ms. I think it would be good to upgrade to that.

We changed CONFIG_FREERTOS_HZ from 100 to 1000 for most boards in #9134. So the old timeout was 100 10ms ticks = 1 second, not 100ms, which is what it is with the new CONFIG_FREERTOS_HZ . That may be the problem in #9535. Your change in this PR keeps it at 100 ticks (100 ms).

harsh rock
#

Hello all, is there a way to more quickly create a circuitpy module out of an existing Arduino/C++ library (that hopefully avoids full reimplementation)?

tulip sleet
harsh rock
#

It's in C++. I'm hoping for something like just creating a similar circuitpy API, but executes the underlying compiled c++ code.
Desktop Python has something like this, but I'm not sure with circuitpy (or even micropython)

#

Basically creating a circuitpy shim over the C++ code.

tulip sleet
#

what is the C++ code doing?

harsh rock
#

MicroPython has a somewhat easier way of doing this

What is the feature called, if I may, so I can look more about it?

tulip sleet
#

It would be far easier to rewrite the i2C code in CIrcuitPython. Our I2C API is fine.

harsh rock
#

Thank you! I'll look into these methods first, but if that becomes too cumbersome, I'm also not writing off rewriting in CircuitPy/MicroPy.

tulip sleet
#

I2C transactions are straightforward. You might try using an AI engine to do the intial translation if it's coming from Arduino.

#

Rewriting it in CircuitPython will also make it easier to change later.

tulip sleet
tulip sleet
manic glacierBOT
harsh rock
#

Ooh, thank you very much!

orchid basinBOT
spare jacinth
#

Idea would be to add py.typed to the ones that pass MyPy checks, and fix the ones that dont (thats a heck ton of work, i may bail out at any point)

manic glacierBOT
#
Sola85
[adafruit/circuitpython] New comment on pull request #9534: nordic: reduce sleep current by using SENSE feature for PinAlarm when possible

@mrdalgaard I currently don't have the ability to do any tests (apart from simple current consumption measurements) myself. Regarding documentation, I have only found Table 5.3.8.3 "Device startup times" in the datasheet. This does not differentiate between GPIOTE and PORT events and only mentions general HW events (which might mean that there is no difference?).

In fact, the datasheet does not mention the word "accuracy" when it comes to interrupts and wake events, this only seems to appe...

#
marcobrianza
[adafruit/circuitpython] New comment on issue #9507: please support XIAO ESP32S3

Hi I have written an email to techsupport@seeed.io asking for VID PID and below is the replay I got
I think this helps the official support ;-)
...

Hi Marco,

Thank you for reaching out. Here are the VID and PID for the XIAO ESP32-S3:

VID: 0x2886
PID: 0x8056
These details should assist with the process. If you have any more questions or need further assistance, feel free to ask.

Regards!

Frank
Seeed Technical Support Team

manic glacierBOT
#
ZodiusInfuser
[adafruit/circuitpython] New review comment on pull request #9531: Add four Pimoroni RP2350 boards

As already mentioned, missing the correct ADC pins:

    { MP_ROM_QSTR(MP_QSTR_GP40_A0), MP_ROM_PTR(&pin_GPIO40) },
    { MP_ROM_QSTR(MP_QSTR_GP40), MP_ROM_PTR(&pin_GPIO40) },
    { MP_ROM_QSTR(MP_QSTR_A0), MP_ROM_PTR(&pin_GPIO40) },

    { MP_ROM_QSTR(MP_QSTR_GP41_A1), MP_ROM_PTR(&pin_GPIO41) },
    { MP_ROM_QSTR(MP_QSTR_GP41), MP_ROM_PTR(&pin_GPIO41) },
    { MP_ROM_QSTR(MP_QSTR_A1), MP_ROM_PTR(&pin_GPIO41) },

    { MP_ROM_QSTR(MP_QSTR_GP42_A2), MP_ROM_PTR(&pin_GPIO42) }...
manic glacierBOT
#
Sola85
[adafruit/circuitpython] New comment on issue #9521: Third sleep option: RAM-preserving "deep" sleep

@tannewt If I understand correctly, your preferred solution would be such that light sleep automatically determines the most power conserving, ram retaining sleep mode that is possible, given the ongoing background tasks. E.g. on an esp32, if nothing is going on in the background, actual light sleep could be used and if WiFi is still active the current behaviour would be used. Is that correct?

As far as I can tell, this would be a valid solution of this issue.
However it does sound like ...

lone axle
# spare jacinth wondering if anyone has ever run mypy on them, lol... the more i look into readi...

Any libraries that got a type annotation issue created did have mypy run once with a specific configuration to create the list that was posted in the original issue. That was a while ago and not live updating so the lists in the issue are likely to be incorrect in substance or line numbers if the libraries have changed in the interim.

I do not think we would want to add mypy into the pre-commit across all the libraries right away because I think it's likely there will still be many warnings and issues emitted by the mypy check. The original issues and first pass through the libraries was focused mostly on the "lowest hanging fruits" of adding annotations for function arguments and returns. The mypy checks can include many additional things beyond those depending on how it's configured and I think it's worth discussing in the weeds at some point what configuration we'd want to use on the libraries before we start adding it to pre-commit.

With regards to a list of libraries, the Bundle repo is the best way to get it IMO. You can use git submodule foreach for automation tasks, or if you want more fine grained control you could list out the directories or even parse the drivers.rst file in the bundle repo.

If you're interested, this repo: https://github.com/FoamyGuy/missing_type_issue_maker contains the scripts that were used to run that initial mypy check and create the issues. It utilizes the bundle for looping over the libraries with git submodule foreach.

#

and this one https://github.com/FoamyGuy/circuitpython_bundle_automation is a slightly more generalized version built upon the same ideas. I also left a bit more documentation in the readme of this one that explains how it was used, and how to use it on a subset of the bundle. It's not necessarily comprehensive docs, but it's better than the "basically nothing" in the type_issue_maker repo.

brazen hatch
#

Wow, wokwi is actually amazing for testing idf api..
I'm now half way there on switching CircuitPython's i2c api to the new idf driver.

spare jacinth
# lone axle Any libraries that got a type annotation issue created did have mypy run once wi...

i have always used mypy without giving it any configuration, i guess you could opt-out some checks but i'd say falling through the default settings the best way to use it

and yeah, maybe things got worse over the years as the code was changed, but given how broken some typing is, my feeling is that it was never really checked for type correctness but only checked whether any hint (regardless of how correct/wrong it is) was added to each(or most) variable... you can see how bad the thing is on the gist 🀣 πŸ™ƒ

slender iron
spare jacinth
brazen hatch
slender iron
brazen hatch
#

Oh, alright, i'll go check that first then, thanks!

slender iron