circuitpython-dev

[timber mango]

I am figthing with it the last 2 hours.

[tulip sleet]

do you have a voltmeter?

[slender iron]

@idle owl @tiny oriole not sure why the permissions changed but I changed them back

[tiny oriole]

sweet! thanks 🙂

[timber mango]

do you have a voltmeter?
@tulip sleet not at hand, sorry

[slender iron]

np, thanks for requesting access. I saw the email

[tulip sleet]

that would be a good thing to get for troubleshooting in the future. A cheap one from the hardware store is fine. In the meantime, try setting relay.drive_mode = digitalio.DriveMode.PUSH_PULL after setting its direction to output. But that should be the default. I don't have my RPi4 set up at the moment. It will take a bit to do. @gilded cradle do you know if there's any issue with simple DigitalInOut with blinka on the RPI4?

[gilded cradle]

I'm not aware of any issues

[tulip sleet]

i'll bring my RPi4 up to date and try

[idle owl]

@slender iron Thanks.

[tulip sleet]

@timber mango you could also just try an LED with resistor on that pin instead of a voltmeter, and confirm that setting relay.value = True lights the LED. Test the LED on one of the constant 3.3V pins to confirm it's working

[timber mango]

@timber mango you could also just try an LED with resistor on that pin instead of a voltmeter, and confirm that setting relay.value = True lights the LED. Test the LED on one of the constant 3.3V pins to confirm it's working
@tulip sleet LED works, same code. True = LED on ; FALSE = LED off

[tulip sleet]

did you try the PUSH_PULL with the relay?

[timber mango]

yes, no change (actually, it was also set for the LED, forgot to remove it)

[tulip sleet]

so, just to confirm, same RPi 4, same relay board, same pin, relay works with GPIO program, but not with CIrcuitPython program? Is that right?

[idle owl]

@slender iron I may need you to record the meeting, my internet went down a bit ago. I'm currently on our backup internet which isn't great.

[timber mango]

That is correct..

[tulip sleet]

ok, I have to update my RPI4 with the latest everything. It will take a little while. I'll ping you again after testing

[timber mango]

I start python3 relay.py or python3 cp_relay.py

Sure! Thank you.

[tulip sleet]

Are you running the latest version of adafruit-blinka, etc.?

[timber mango]

Here: Adafruit-Blinka 4.0.0

[simple pulsar]

I have a question on bootloader moder with the nnnBOOT (using x character is problematic in discord!). On the double press does the board stay there forever or is there a timeout?

[idle owl]

@simple pulsar I feel like there's a timeout. But that's simply from experience where it eventually drops out of the bootloader on its own for me pretty regularly, not me knowing that for certain.

[simple pulsar]

I've never left the boards for long but I just noticed the CLUE does it as I got distracted.

[solar whale]

IIRC it is different for nrf52 vs ATMEL -- at least I think so.

I think nrf52 does time out, not sure about samd

[tulip sleet]

@solar whale is right; nrf times out; atmel does not (or it's really long)

[idle owl]

<@&356864093652516868> CircuitPython Weekly is in about an hour and 20 minutes. Everyone is encouraged to attend. Please add your status updates and hug reports to the notes doc before the meeting, even if you'll be attending - it's super helpful! Remember if you're lurking or have no mic, you can still add notes to the doc and I'll read them off during the meeting. Looking forward to seeing everyone there! https://docs.google.com/document/d/1bGuche1SWmuFUABTWpuwapGnSaz4pG9yWR_Y0JOzdbw/edit#

[tough flax]

@timber mango and @tulip sleet Just a heads up, I spent forever on rPi debugging because I forgot there are two (three?) numbering schemes for the various pins. You might make double sure CP uses the same scheme as RPi.GPIO

[timber mango]

@timber mango and @tulip sleet Just a heads up, I spent forever on rPi debugging because I forgot there are two (three?) numbering schemes for the various pins. You might make double sure CP uses the same scheme as RPi.GPIO
@tough flax
What does CP use? BCM?

[tough flax]

I'm not the one to answer (@tulip sleet, @gilded cradle , or @idle owl will) - but here's a pinout if you want to try the other two: https://pinout.xyz/

There's a tooltip on each pin that shows the Wiring.pi pin

And from the guides, yes, @odd vector it looks like they use BCM

Seems you can check which one you're using in rPi.GPIO https://sourceforge.net/p/raspberry-gpio-python/wiki/BasicUsage/

[tulip sleet]

@timber mango I did test on my own RPi4, and I see no difference between using RPi.GPIO and digitalio.DigitalInOut. However, you're trying to drive a 5V input. There is a drive-strength setting on the pins and maybe RPI.GPIO is using a higher drive strength than adafruit-blinka. I'll check into this further but it will take a while (other things to do for several hours)

[serene warren]

@timber mango https://www.adafruit.com/product/2263 saved my bacon/Pi. and T=this https://www.adafruit.com/product/3577

[manic glacier]

mesa177

[adafruit/circuitpython] Pull request opened: #2652 Creat ar\.po

Arabic translation, incomplete, needs further editing. Suggest testing before continuing translation

[simple pulsar]

Why does the CLUE screen turn off under bright light? It does this on the demo code that comes with it and with CircuitPython 5.0.0 beta5

[lapis hemlock]

@slender iron unless you have a way to get a toehold on fixing how packages made of built-in C modules work, I think this or "ulab_xxx" parallel modules are the way to go. If you have a direction to point me about the underlying problem I'm happy toc hase it down though
@onyx hinge @slender iron If I understand the issue correctly, this is not solved in micropython: https://github.com/micropython/micropython/pull/4731 (unumpy has never materialised. At least, this is what I can infer from Jim's comments elsewehere.)

[idle owl]

@simple pulsar I'm running a working example using the PyBadger library and mine doesn't dim with bright light on it. Something in the code you're running perhaps?

[simple pulsar]

Try moving it gradually close to a desk lamp

At about two inches from my tungsten desk lamp it turns off. Torch on my smartphone won't do it.

[idle owl]

I only have LED bulbs in here. All the light are Hue.

[simple pulsar]

A humble cateye bike light won't do it. Curious.

[half sedge]

What kind of "interference" would a tungsten lamp produce???

[modern wing]

I will be happily lurking today 🙂

[old smelt]

Lurking

[simple pulsar]

I'd assume it's just a bug with the light reading that the continuous spectrum including IR end of things triggers it but I may be wrong...

[half sedge]

I'll be Text Only. (I am not really lurking since I filled the meeting document)

[simple pulsar]

I'm finding the 4097 max value on demo app to be odd too in both senses of the word.

[serene warren]

Lurking today (as usual)

[modern wing]

loud and clear @idle owl

[turbid radish]

Lurking

[tidal kiln]

lurking
(until in the weeds)

[timber mango]

Text only

[onyx hinge]

Hi all! Nearly home from my vacation but can't be at a computer or mic in time. Group hug, hug to danh for release candidate. Will look at some issues assigned to me and then ulab this week.

[lapis hemlock]

most probably lurking today.

[timber mango]

Is the port of Circuitpython for the Teensy 4.0 solid yet?

[errant grail]

visiting from the lurk-zone today

[slender iron]

@timber mango no, consider it alpha

we haven't been putting a ton of polish into it yet

[timber mango]

Ah, OK, durn.

[slender iron]

it is usable though

[timber mango]

Does "usable" mean that most of it works as expected?

[half sedge]

Maybe we should have a recording of the introduction to meeting? It feel lenghty now that I heard that a few time. 😉

[river quest]

CircuitPython library parsing the data output of a Mitutoyo Digimatic instrument, like calipers or micrometers
https://github.com/vifino/CircuitPython-mitutoyo

For EngineersWeek check out Digi-Key's events with Circuit Playgrounds and more!
https://twitter.com/digikey/status/1230971271275597825
https://twitter.com/Kwalseth/status/1230927356174512129

Digi-Key Electronics (@digikey)

Our annual Girl's Night Out was a #success! The girls learned to program a #CircuitPlayground from @adafruit, and made their own night-lights.

HUGE shout out to everyone who helped bring the event to life! #STEM #GirlsWhoCode #EngineersWeek https://t.co/rpHubMHJBq

▶ Play video

Twitter

Kevin Walseth (@Kwalseth)

These are just a few reason I love my job. Working with these girls @nationaleweek girls night out @digikey proved to be a lot of fun. We focused on @MSMakeCode programming of the @adafruit #circuitplaygroundexpress @GirlsWhoCode members gave us a hand!

Twitter

[slender iron]

@timber mango you can check out the issues to know what is missing. no promises that everything works

[river quest]

Morgan Stanley Makerspace Program (Circuit Playgrounds with CircuitPython and more)
https://twitter.com/MorganStanley/status/1231217016985604098

Morgan Stanley (@MorganStanley)

Our Technology team recently hosted teens from @winnyc_org to teach them coding fundamentals using the @Adafruit Circuit Python Express. Participants created light sensors, accelerometers and more as part of our Morgan Stanley Makerspace Program. #MSGivesBack

Twitter

[slender iron]

@half sedge I think it's worth doing because we may have new folks join us each week

[river quest]

Feather previews!
https://twitter.com/arturo182/status/1229500410664280064
https://www.instagram.com/p/B8rpJbFHovw/?igshid=1qlgbwy8rjjkp

arturo182 (@arturo182)

The gang's all here

Likes

116

Twitter

HackSpace Issue 28: Python on Hardware (Drew) and more!
https://hackspace.raspberrypi.org/issues/28
https://blog.adafruit.com/2020/02/18/live-chat-hackspace-magazine-chat-on-wednesday-feb-19th-at-1pm-et-hackspacemag-ben_everard/
https://hsmag.cc/adafruit

[half sedge]

@slender iron no doubt about that, but it could be recorded once and "playout" at each meeting... Just the day something change, I might not notice. 😉

[river quest]

CLUE on Hackaday
https://hackaday.com/2020/02/24/new-part-day-ooh-the-things-you-can-do-with-a-clue/

Good thread "what is the most intimidating part of hardware development to you, if you've ever had doubts about starting a first project with microcontrollers, etc?"
https://twitter.com/noopkat/status/1229547642343972864

Suz Hinton (@noopkat)

what is the most intimidating part of hardware development to you, if you've ever had doubts about starting a first project with microcontrollers, etc?

Likes

145

Twitter

[timber mango]

@slender iron I am very interested, but I do not have one. I was just wondering how Circuitpython for the super-fast micros were coming along.

[manic glacier]

tannewt

[adafruit/circuitpython] New comment on pull request #2652: Creat ar\.po

Thank you! This is perfect for testing our pipeline to make sure everything works. I think the first step we need to do is turn off the terminalio output until we can do the rendering natively.

Is there a reason you started with the German translation? I'd expect the untranslated messages to be empty and fall back to English.

[slender iron]

@timber mango it is definitely interesting and we'll circle back to it sometime after 5.0.0 is out

[timber mango]

Is there going to be a port of Circuitpython for the PyBoard-D series someday?

[slender iron]

probably. the stm32h7 series is pretty interesting

imx rt is probably a higher priority though

(imx rt has high speed usb phy and the stm32h7 doesn't)

[timber mango]

@slender iron I am the sort where there is no MCU/CPU fast enough, no hard disk storage large enough, and never enough memory. 😉 😎

[slender iron]

agree 🙂 the fast CPUs have their own challenges though.

mostly how you get them instructions fast enough

[inland tusk]

lurking for hug reports

[timber mango]

Oh, I know, and they are typically much more complex.

[slender iron]

@lapis hemlock are you lurking?

[timber mango]

Text only

[lapis hemlock]

@slender iron sort of.

[slender iron]

do you plan on speaking when we get to you?

[lapis hemlock]

@slender iron no, I wasn't planning on that. I think, not too much has happened on my end in the past week.

[slender iron]

kk

[half sedge]

@idle owl and then make it work on the CPB+Gizmo?

[slender iron]

@tidal kiln do you have in the weeds topics?

[tidal kiln]

yep

[slender iron]

want to add them to the doc?

[tidal kiln]

sure...one sec...

[slender iron]

thanks

[ionic elk]

a last minute wedding?

were shotguns involved?

[modern wing]

Have fun in Colorado!

[half sedge]

I am interested in battery level for stuff like PyGamer. Is this possible? I would like to show on the thermal camera what is the status.

[slender iron]

@half sedge I believe so. I think most of our boards have a resistor divider hooked up to an analog pin

[half sedge]

Thanks, I'll have a look, that make it so much professional to have a battery indicator on screen. 😉

[slender iron]

🙂

[half sedge]

Thanks for the reading @idle owl.

[slender iron]

@ionic elk I put a comment about your notes

in the doc

👍

[graceful heart]

super cute!

[half sedge]

@lone axle That patchwork is interesting. I plan to display "top 3" colors + my own advertised color (or maybe keep it on my own neopixel). Maybe because I only have 4 BLE capable board (but I can cheat with an extra one from a phone). Maybe making the size dependant on the RSSI value.

[slender iron]

very cool @solar whale !

[modern wing]

@solar whale That project is for the birds 🐦

[half sedge]

And if it's early in the live of the library... how old is it?

[errant grail]

@half sedge Here's a DisplayIO TileGrid batt mon icon implementation I'm working on for a PyBadge clock. https://github.com/CedarGroveStudios/Clock_Builder/tree/master/code/clock_display

[lone axle]

@gilded cradle Oh, neat. I hadn't noticed that thank you. I see on this docs page https://circuitpython.readthedocs.io/projects/ht16k33/en/latest/api.html. that marquee is only listed under Seg14x4. Is that usable on the 16x8 matrix's as well?

[gilded cradle]

Right, I noticed you said 16x8 and removed my comment

[lone axle]

Ah, I see. sorry stepped away for a moment.

[half sedge]

Yeah that page is great, with the image.

[gilded cradle]

No problem, but maybe look at the IS31FL3731 library for inspiration

[lone axle]

I'm glad you mentioned it anyhow, is really good to know because if I do make an attempt to port it for use with 16x8 I'll look at how it's been done before.

[half sedge]

The way I learned BLE is getting to each of those video and example.

The confusing part was the "old" BLE friend things that are there... those that you control with AT on UART and SPI.

[graceful heart]

related to that^^, is there a way to actually use the BLE 'friend' boards in circuit python, the same way as using a bluefruit nrf built in board?

[lone axle]

@half sedge nice! I do like the idea of using the RSSI strength value to determine part of it. I hadn't thought of using the size though, index order is what came to mind for me.

[slender iron]

https://github.com/adafruit/Adafruit_CircuitPython_Radio

[solar whale]

@graceful heart there is a separate lib for the SPI "friend"

not compatible with bleio

[slender iron]

I haven't looked at building a _bleio compatible library based on the friend stuff but I bet it is possible

[solar whale]

https://github.com/adafruit/Adafruit_CircuitPython_BluefruitSPI

[half sedge]

@lone axle One of my problem is that I don't know the min and max value for RSSI. Now on the graphical side, it could be "dots" where the surface depend on the strength. And those dots will touch each other. If you want it squarish, then it might be easier. There are those tools that tell you how your disk space is used and display it with color that fill a square.

[timber mango]

Will there ever be a Bluetooth module that can be added to current Express or regular boards? Something like the AirLift WiFi Feather and breakout board, but for Bluetooth.

[half sedge]

There is a guide that explain how to sniff BLE protocol and communicate to your gadget by replicating that.

[slender iron]

https://learn.adafruit.com/circuitpython-nrf52840

[gilded cradle]

@timber mango maybe something like https://www.adafruit.com/product/2633 or https://www.adafruit.com/product/2479?

[slender iron]

https://learn.adafruit.com/introduction-to-bluetooth-low-energy

[modern wing]

Like an awesome list of guides?

[half sedge]

The guide from @idle owl could use the graphics from here: https://learn.adafruit.com/ble-hid-keyboard-buttons-with-circuitpython/understanding-ble

[solar whale]

a guide to using guides...

[slender iron]

@timber mango I think we'd like to get BLE going on the ESP32 at some point

[half sedge]

metaguide.

[turbid radish]

We do need some more macro guides

[half sedge]

The existence of a guide can drive hardware acquisition...

[slender iron]

@timber mango it'll probably coincide with the ESP32-S2 work

[onyx hinge]

hey I was here for the last 30 seconds!

[turbid radish]

Yay @onyx hinge

[lapis hemlock]

hey I was here for the last 30 seconds!
@onyx hinge 👍

[onyx hinge]

hi @lapis hemlock anything we need to talk about rq?

[serene warren]

Till next week

[slender iron]

👋

[modern wing]

Thanks y'all!

[turbid radish]

Thanks all

[half sedge]

@idle owl Do you want me to do the chronological thing?

[gilded cradle]

thanks

[ionic elk]

thanks all

[graceful heart]

👋

[errant grail]

Thanks!

[idle owl]

@half sedge Yes please.

[lapis hemlock]

@onyx hinge I would have a couple of questions, but they are not particularly urgent, we can discuss them later.

[onyx hinge]

@lapis hemlock ok, I just walked in the door from vacation, so it would be better to do it later on anyway

[lapis hemlock]

@onyx hinge what about tomorrow?

[onyx hinge]

and I think there's another issue that jumped my queue to look at, because it's relative to the 5.0.0 release

[lapis hemlock]

OK.

[onyx hinge]

but yes! I'd love to chat tomorow.

[timber mango]

@gilded cradle Yes, like those, and fully compatible with the nRF52840. I have one of the Bluefruit LE SPI Friends, but do not know if it would be compatible.

[gilded cradle]

I'm not sure either

[half sedge]

@idle owl I think it is done.

[modern wing]

Awesome 🙂

[onyx hinge]

@lapis hemlock I am in a US time zone (GMT+6 right now) and usually work in the mornings my time.

[timber mango]

@gilded cradle I am thinking about connecting it up to one of my Feather M4 Express boards or my ItsyBitsy M4.

[onyx hinge]

so, anywhere from 18 to 26 hours from now would be pretty good for me

@slender iron HTTP supports it, but hardly anybody does it that way

http accept-language

[slender iron]

👍

[half sedge]

@idle owl Translated learn guide ??? I am helping in a Coder Dojo, but the language issue make it difficult to have kids work independently when they don't know english (most of them).

Of course.

[lapis hemlock]

@onyx hinge I am 7 hours ahead of you, so I could make it sometime between noon and 3 pm tomorrow.

[half sedge]

You need to identify a core of guide that are the most important.

[lapis hemlock]

@onyx hinge your time, that is.

[half sedge]

I would focus on CPX or CPB.

[onyx hinge]

@lapis hemlock okay, I will try to be around during those hours and we can talk when you're around. Or if you want to give me a narrower range just mention me

I'm going to go afk again and start unpacking.

👋

[half sedge]

And that is to what language?

In Belgium, there is a adafruit reseller that translate a lot of your guide

McHobby or something.

Available to all

With Adafruit bleshing.

[lapis hemlock]

@onyx hinge 👋

[half sedge]

https://shop.mchobby.be/en/

That is the shop, but there is a blog somewhere.

He wrote the book on Micropython too...

400+ pages. all in french.

I learned english thanks to ZX81 documentation... but I might be an exception. 😉

That was the only way

See you, good by.

[gilded cradle]

later

[timber mango]

@half sedge Do we have any fluent French speakers who would be interested in translating it? 😃 😉 😎

[idle owl]

The other nice thing about starting the translations on GitHub is collaboration on the translations.

CLUE badge demo:

[turbid radish]

I love that

[onyx hinge]

@slender iron since you are opinionated about the module issue and it may have some technical sticking points it would be convenient if you could chat with us tomorrow. Assuming you didn't figure out the issue already

[timber mango]

@idle owl Do you have any interesting 160x128 images? I need a few I can use in testing my PyGamer.

[slender iron]

@onyx hinge what time? I'm having lunch with a friend around 12:30 my time. I'm free otherwise

[idle owl]

@timber mango I have a series of kittens that I think are meant for PyBadge/PyGamer. And some Blinka iterations somewhere. But that's about it.

[onyx hinge]

You are gmt+8?

Eerr -8

[idle owl]

All the kittens are square. So they're not quite right. Not sure what I had them for.

[timber mango]

I would love to get some Blinka images!

[idle owl]

So I'm not sure I do. Let me see if the Blinka images are proper.

[turbid radish]

@idle owl, there is the Adafruit image dropbox folders for art to mine if you like

[idle owl]

@turbid radish If you can link it, that would be great. But I think geekguy is looking for bitmaps that already work on PyGamer

[lapis hemlock]

@onyx hinge if necessary, I can stay till 10 pm GMT tomorrow.

[onyx hinge]

"same hour as weekly meeting", @lapis hemlock @slender iron ?

[turbid radish]

ok, I'll send linky on Slack

[lapis hemlock]

@onyx hinge It is OK with me.

Actually, would be perfect.

[idle owl]

@timber mango This is all I'm seeing that looks accurate.

[onyx hinge]

That will naturally time box it to about an hour so tannewt can make his lunch event

(likely an hour is not necessary)

[timber mango]

@idle owl @turbid radish I can take images I can resize. 🙂

[lapis hemlock]

@onyx hinge I think we should be able to sort out the conceptual issues in 15 minutes, the rest we can discuss without Scott.

That is not to say that Scott should be excluded, but he clearly has an engagement.

[timber mango]

@idle owl Thank you!

[slender iron]

@lapis hemlock @onyx hinge same as weekly is perfect for me (11am)

[idle owl]

@timber mango This might be right too, not sure.

[lapis hemlock]

@slender iron @onyx hinge OK, I can confirm.

[slender iron]

👍

lunchtime for me 🙂

[meager fog]

@indigo wedge hihi

[indigo wedge]

Hi 🙂

[meager fog]

checkin' if you were able to ship the feathers?

[indigo wedge]

I tried to get it going Friday but I require more info, I sent an email to Alicia on Friday, waiting for her reply.

[meager fog]

alicia? or mary

or stella?

[indigo wedge]

Sorry, Alisa

Stella is also CC

[meager fog]

ok! thanx

[timber mango]

@idle owl I will check it out. 🙂

[idle owl]

@slender iron Meeting is up on YouTube, video successful. Thanks again for backup recording.

[manic glacier]

dhalbert

[adafruit/circuitpython] New comment on issue #2639: Sanity check memory use in 5\.0\.0

I started looking at this. I get the identical memory issue when the program runs automatically as code.py. If I import code.py in the REPL it runs fine.

otter_data.py has a hideously large set of nested lists of small integers and could be made a lot more efficient.

Here is an instrumented view of memory consumption:

import board
import gc
print("before most imports", gc.mem_free())
import displayio
print("a", gc.mem_free())
from time import sleep
import otter_data
p...

[half sedge]

@timber mango I am native French, but my spelling is really poor, even when I try very hard. I try to introduce Circuit Python to other coach in the Coder Dojo where I go. Just because Scratch is boring to me and I'd like to go physical. Coming from Scratch, kids might prefer to go to Maker Code (I guess this is currently English only). Edublocks could be a way to get from Scratch to writing code in Python: https://edublocks.org/circuitpy.html (not responding, try https://edublocks.org/ )

@lone axle I found the squarish algorithm name and this in your search engine could help: "squarified algorithm treemap" ... That make nice visualisation. Now I don't know how to get from RSSI to surface/percentage. RSSI are in the negative range, maybe it is a logarithmic scale. Maybe I need to experiment a bit to figure out what is the best way. And I don't know the complexity of that algorithm...

[manic glacier]

tannewt

[adafruit/circuitpython] New comment on pull request #2652: Creat ar\.po

@mesa177 I removed a trailing \n the build didn't like to see what happens if it is correct. So, make sure to pull the latest version of the branch from GitHub if you have a local copy.

[half sedge]

Hi @plucky flint did you meet @heavy pasture at that LEGO stuff? I wonder how many UK maker that I am following on Twitter were present. That's a place I would love to have been.

[timber mango]

Well this is not amusing. I have run out of AA batteries. My keyboards are dead, so I cannot use my computers. 😕🤔

[simple pulsar]

@timber mango Does your mouse still work?

Does anyone else have a CLUE and a non-LED desklamp? I've been puzzling over the display turning off with bright lights. It's all bit mysterious.

[tiny oriole]

Does anyone else have a CLUE and a non-LED desklamp? I've been puzzling over the display turning off with bright lights. It's all bit mysterious.
@simple pulsar i might have both of those things at home. 🙂

although, the display in mine is non-stock

[simple pulsar]

You've "pimped" your CLUE?

[tiny oriole]

and curiously does have a truckload of black masking over where the driver IC is

more like i broke my display, and had to order one from the product page 😛

[simple pulsar]

Ah, i see. I have two CLUEs and both turn off their displays if placed about 5 cm from a desklamp

[tiny oriole]

do you have a camera with a flashbulb? and does IT cause them to blank out or reset?

[simple pulsar]

I have one. Are you suggesting its the same thing that cause Rasperry Pi 2's to reset when a flash went off?

[tiny oriole]

reminds me of an old Raspberry Pi issue involving an improperly capped power supply IC that was having some kind of weird photovoltaic reaction with super bright light, triggering the Pi to reboot. I know the Driver IC seems to be just outside of the pixel on those screens, and on my stock screen that came wiht the clue it seemed to just be masked with blue translucent tape

maybe a similar thing is happening?

the one i replaced mine with though came with thick black epoxy covering the driver IC

[half sedge]

That was trigger by a flash I believe. But getting near a lightbulb?

It seems that in my house, I mostly have Ikea Tradfri. 😦

[simple pulsar]

The normal obvious answer would be a bug in dimming code but it does it with display.auto_brightness = False too

[tiny oriole]

It seems that in my house, I mostly have Ikea Tradfri. 😦
@half sedge a felow TRÅDFRI user huh?

[half sedge]

I was wondering if there would be a way to control Tradfri from Circuit Python...

Maybe I would need a gateway for heavy lifting.

[simple pulsar]

It does not like my T32 flash. That made the display go permanently off! @tiny oriole you may be onto something there.

[tiny oriole]

what kind of control are we talking? like a light switch or something more automatic

[half sedge]

@tiny oriole Color for the RGB bulb would be great. On/Off/Color/Brightness

I used Home Assistant with TradFri... and I guess with MQTT it is possible to talk from CircuitPython + AirLift to the Home Assistant. So maybe there is a way (that require a Pi) but that would work.

There is a learn guide about displaying from Home Assistant to a PyPortal. So one day, if I have time, I could try to control my home from a ... PyPortal: PyPortal => Raspberry Pi (running Home Assistant) => TradFri gateway => Light Bulb

[tiny oriole]

borrowed my friend's clue at the office and can confirm i can power the LCD off (backlight stays lit) by pointing my mini maglight focused on roughly the center to left side of the driver IC

[simple pulsar]

Tis lucky your friend has a clue. And unlucky for him/her that you have a maglight 🙂

Thanks.

[tiny oriole]

yup.

[onyx hinge]

heads to tindie to offer reasonably priced squares of electrical tape, precision cut

[tiny oriole]

@half sedge i think those Tradfri devices are just Zigbee, you could probably trick a TRÅDFRI Hub into thinking your CP device is just a tradfri remote

[tidal kiln]

is it somehow possible to have a PR create a new branch? for example - to add a documentation branch to a repo that does not have one yet?

[onyx hinge]

https://learn.pimoroni.com/tutorial/sandyj/controlling-ikea-tradfri-lights-from-your-pi

Or you can talk to thehub over WiFi

[tiny oriole]

🤔

maybe some room for a CircuitPython Tradfri library? might be fun to play around with 😛

[arctic violet]

Oh reeeaaaaally, I'm much more interested in those Ikea lights now.

[tiny oriole]

me too actually

since my house is filled with them

could use a lux sensor as a light switch

obviously not in direct view of the lamp in question... unless you like strobe lights

[manic glacier]

dhalbert

[adafruit/circuitpython] New comment on issue #2639: Sanity check memory use in 5\.0\.0

I bisected and got to this commit as the first one that causes the problem: d32dc814d3f0d3e845e54a043467f37a5908f5f0, which is #2330, which fixes a GC bug in displayio, but supposedly only for epaper. I repeated the bisect and got the same thing. Moving back one commit, I don't see the issue. So this would seem to make no sense, because I'm testing with a pyportal, which is not epaper. But maybe there is a mixup somewhere in displayio about epaper vs not?

Tagging @tannewt in case this ri...

[manic glacier]

tannewt

[adafruit/circuitpython] Pull request opened: #2653 Fix Mac crash when waking up with an ejected CIRCUITPY

We now correctly set the reason for the unit not being ready and
always start the unit.

Fixes #2567

[slender iron]

@tulip sleet sent out a pr for the usb issue I have. also found a safe mode crash exiting the repl with ctrl-d

which I'll look into tonight

[fallen anvil]

@slender iron is the fact that the app restarts about 10 seconds after I've saved the App an intended behaviour or is there something a bit off happening on imxrt in that regard? Once it's restarted it's compeltely stable. I thought it was down to my OS making some kind of secondary save but it does it even when the PC is disconnected....

[slender iron]

it should only delay under a second

I think it might be a tinyusb thing. I feel like I've seen that on nrf too

[fallen anvil]

No, it runs for perhaps 10 and then does a restart

Its one of those things thats likely to have been there a while and we've all ignored it

[slender iron]

yup

[fallen anvil]

Here's an example, numbers are in seconds;

$ cu -l /dev/ttyACM0
Connected.
Starting

Code done running. Waiting for reload.
soft reboot

Auto-reload is on. Simply save files over USB to run them or enter REPL to disable.
code.py output:
Starting
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
soft reboot

Auto-reload is on. Simply save files over USB to run them or enter REPL to disable.
code.py output:
Starting
1
2
3
4
5
6
7
8

OK, just run that 3-4 times and it always gets to 23seconds. There's a 5 second boot delay and a second or two writing the screen so I'll hazard a guess at it being 30 secs on total.

[slender iron]

hrm

how are you saving to the drive?

[fallen anvil]

hang on, removed all the crud for more accurate timing...

linux session, notepad, CTRL-S

yep, smack on 30 secs.

[slender iron]

it could be the host os flush

windows linux session?

[fallen anvil]

no

real one

Once its done it, it never resets again

[slender iron]

try sync (iirc) on linux

it should reset once after the file writes are done

[fallen anvil]

ok, I'll save then do a sync in another windoiw

Yep, that resets it too

[tidal kiln]

@idle owl has there been any talk of adding shake to the clue lib?

[fallen anvil]

so, that's the problem...just got to figure out how to solve it. I'll dig into it. Bedtime now.

[slender iron]

@fallen anvil some editors are better than others. they are documented somewhere

[fallen anvil]

Yeah, I read that, but this seems independent of editor. emacs does it too.

[slender iron]

hrm, weird

[idle owl]

@tidal kiln No talk. Shake is built into the LIS3DH library and the accelerometer on the CLUE is not the same. It would have to be added to the lib for the CLUE accel, or a lot of code added to the CLUE lib to do it.

[slender iron]

https://learn.adafruit.com/welcome-to-circuitpython/creating-and-editing-code#1-use-an-editor-that-writes-out-the-file-completely-when-you-save-it-7-16

[tidal kiln]

@idle owl tap is special to the LIS3DH, but shake is just math on accel values, and maybe not a ton of code? could just add this to clue lib?
https://github.com/adafruit/Adafruit_CircuitPython_LIS3DH/blob/master/adafruit_lis3dh.py#L168

[fallen anvil]

So, if I cp the file to the drive I don't get the issue, only if I save it...but I'm seeing it with emacs which is supposed to be a 'safe' option. I'll look further.

Well, here's the reason, this will happen independent of editor;

$ cat /proc/sys/vm/dirty_expire_centisecs 
3000
$

We can switch that off, but only at the expense of upsetting the write performance of the whole system. I'll see if there's any mechanism for doing it on one drive.

[manic glacier]

tannewt

[adafruit/circuitpython] Pull request opened: #2654 Fix crash after empty REPL session

We were trying to reset bluetooth when it was off and then trying
to raise an exception without the heap.

[manic glacier]

tannewt

[adafruit/circuitpython] New comment on issue #2639: Sanity check memory use in 5\.0\.0

Nothing jumps out at me. My next step would be to see if I can break in that collect via a debugger. Let me know if you'd like me to take a look.

[bright aspen]

@fallen anvil I'm coming in late, so I have missed much of this. I get a reboot after 10s on Feather Mn Express on Mac. I forgot which one, maybe both. This occurs after I save into /lib. If this is not applicable, then "never mind." I forgot what version, a 5 beta, I think.

[manic glacier]

dhalbert

[adafruit/circuitpython] New comment on issue #2639: Sanity check memory use in 5\.0\.0

Yes, I'll look at this tomorrow morning: set some breakpoints and see what's going on. It shouldn't be calling the epaper code at all.

dhalbert

[adafruit/circuitpython] Pull request review submitted: #2654 Fix crash after empty REPL session

This looks good. The build failure is just one subjob that was a GitHub actions problem.

dhalbert

[adafruit/circuitpython] Pull request closed: #2654 Fix crash after empty REPL session

[manic glacier]

dhalbert

[adafruit/circuitpython] Pull request review submitted: #2653 Fix Mac crash when waking up with an ejected CIRCUITPY

This looks good, though the test and mpy-cross subjobs are crashing. I re-ran the actions, but the same thing happened. Merge as you wish.

hathach

[adafruit/circuitpython] Pull request review submitted: #2653 Fix Mac crash when waking up with an ejected CIRCUITPY

macOS is so persistent to get the sense code, look perfect :D

[trim elm]

@fallen anvil @slender iron I've had the same thing happen to me, just infrequently enough I personally never found it to be an issue, when editing it locally (on linux) with vim and using cp to copy it to CIRCUITPY

I'd estimate it happens once every 30 times I copy code.py over

[manic glacier]

fgallaire

[adafruit/circuitpython] New comment on pull request #2638: Arduino Nano 33 IoT support

lgtm - please be aware that there's not enough memory in circuitpy to talk to the wifi module and do internetty things :)

Hello, nevertheless I would like to test it, so I need to upgrade the NINA firmware. So I need a Serial Passthroug firmware for the Nano IoT but I don't find the repo to fork/how to do it.

dhalbert

[adafruit/circuitpython] New comment on pull request #2638: Arduino Nano 33 IoT support

Check out this thread: https://forum.arduino.cc/index.php?topic=654833.0
The Arduino forums and website are probably best for inquiring about this.

[manic glacier]

fgallaire

[adafruit/circuitpython] New comment on pull request #2638: Arduino Nano 33 IoT support

Thanks @dhalbert I'm so focused on CircuitPython that I forgot that it's an Arduino board ;-)
Flashing the firmware with the Arduino IDE works fine:
https://www.arduino.cc/en/Tutorial/WiFiNINA-FirmwareUpdater

mesa177

[adafruit/circuitpython] New comment on pull request #2652: Creat ar\.po

@tannewt Hey Scott! It's Alaa from Lamba Labs.

So I used the German copy for 2 reasons: 1) was learning the lingo for a while before I stopped, so wanted to sink my teeth back into it. 2) saved me time adding necessary annotations. 2 birds, 1 stone.

Just make sure that what's being printed on screen characters wise and orientation wise matches the comments that I wrote in Arabic. If that's ok, then will proceed with the translation and pull a later version from my local copy.

PS: tha...

[manic glacier]

fgallaire

[adafruit/circuitpython] New comment on pull request #2638: Arduino Nano 33 IoT support

I have done my best:

• adafruit_esp32spi as frozen module
• Adafruit NINA firmware 1.6 flashed via Arduino IDE
• MOSI pin changed to PB09 to match the pin change https://github.com/adafruit/nina-fw/commit/9e76479a7e1e11682009078c45ad307cb96c1066

But always the same error:

RuntimeError: Timed out waiting for SPI char

[lone sandal]

fgallaire

[adafruit/awesome-circuitpython] Fork created: fgallaire/awesome-circuitpython

fgallaire

[adafruit/awesome-circuitpython] Pull request opened: #20 Update README\.md

https://github.com/adafruit/circuitpython/pull/2638

[plucky flint]

@half sedge 👋 yeah... I know Les (@heavy pasture). He's a wonderful person, a kind and thoughtful influence in the UK community and a great writer/coder/maker. We need more Les's in the world. He was there and we got a chance to catch up all too briefly. You should drop him a line and see how he is..!

[hollow mountain]

Hi there, iam completely new to Circuitpython, is there a working tutorial out there on how to get Circuitpython running on the MKR Zero?

[fallen anvil]

Oh, for goodness sake, the language robot is a bit draconian. I just typed out a whole explaination about the sync problem and it's thrown it away because I said something about the current mount being screwed up!!!

Anyway, let's have another go.

@slender iron @bright aspen @trim elm the language bot needs to calm down. Here's what I wrote...if anyone knows why it keeps throwing it away I'm all ears. Anyway, the above graphic is the fix as a sort of inverted capacha.

I'm not a fan of these automated checkers....they do daft things like this.

[simple pulsar]

I think talking about Vin Diesel's yYy (replace with previous character in the alphabet) will also get you into trouble 🙂

[solar whale]

the bot does not like strings of "x"s

[fallen anvil]

I'm all for civil discourse but politicians from around the world prove time and again that you can say excessively nasty things without swearing and the collorary to that is any performance by Billy Connelly who uses swearing, without malace, as a whole new toolset extension to the English language (look him up, but not at work!) .... and no bot can really understand if you're behaving with the right intention.

[solar whale]

There are some things the bot does well (that only the moderators see, thankfully) and some things that are quite annoying that we all have to live with. At this point, it is what it is.

[fallen anvil]

Yeah, I suspect there's some pretty horrible stuff it turfs automatically...I just wish it told you what it didn't actually like though. If you're writing something truly nasty you already know, and if it's catching you by accident you can rephrase.

[solar whale]

The moderators do have access to the rejected post and I looked at yours and other than the one with the "x"s I could not figure out what it was complaining about...

It does not tell us specifically what it rejected the post for, just that it did.

It will reject strings of "x"

[fallen anvil]

I figured I'd maybe put some inappropriate expression in the first version, but was really careful in the second, and they both got turfed. Oh well, avoid gid and uid in future :-)

Nope, not that...

Ah, it'll be the X's to replace with your gid and uid!

[solar whale]

Use “n” instead of “x” that will avoid one common issue.

[lone sandal]

mardab

[adafruit/awesome-circuitpython] New star added

[lone sandal]

jepler

[adafruit/awesome-circuitpython] Pull request review submitted: #20 Update README\.md

jepler

[adafruit/awesome-circuitpython] Pull request closed: #20 Update README\.md

jepler

[awesome-circuitpython:master] 2 new commits

efa7d02 Update README.md - fgallaire
57c2bc9 Merge pull request #20 from fgallaire/master - jepler

[manic glacier]

dhalbert

[adafruit/circuitpython] New comment on issue #2639: Sanity check memory use in 5\.0\.0

This was not working on beta.5, but is working on tip of master now. :upside_down_face: More later.

[old smelt]

I was on the road yesterday and unable to attend the call. Tried, but it didn't work... Anyway, you may have covered this already, but @idle owl got some love from Kojo Idrissa on episode 168 of the PythonBytes podcast about her upcoming PyCon tutorial. I checked the hug reports from yesterday's call and did not see a reference to it, so I thought I'd mention it today.

https://pythonbytes.fm/episodes/show/168/race-your-donkey-car-with-python

[lone axle]

@gilded cradle thank you for pointing me toward the IS31FL3731 library. With a bit of tweaking I was able to get one of the examples there: https://github.com/adafruit/Adafruit_CircuitPython_IS31FL3731/blob/master/examples/is31fl3731_text_example.py to work with the HT16K33 16x8 LED Matrix.

[idle owl]

<@&356864093652516868> Here is the notes document for next Monday’s CircuitPython Weekly meeting. Everyone is encouraged to attend! Please add your hug reports and status updates even if you’ll be attending the meeting - it’s super helpful! If you are unable to attend but would still like to include updates, feel free to include them in the notes and we’ll read them off during the meeting. Hope to see you there! https://docs.google.com/document/d/1Lw_aT4nXU4RpuuiQophR502XjFh7n-YcI9xUG6JFY_o/edit

[idle owl]

@old smelt Thank you for the mention. 😊 Feel free to add it to hug reports for next week!

[old smelt]

Done!

[idle owl]

@old smelt Thank you!

[onyx hinge]

oh this is infuriating. I initially reproduced https://github.com/adafruit/circuitpython/issues/2639 including that displayio_epaperdisplay_collect_ptrs was reached.. but I made a trivial change and it stopped doing that. and now when I revert my changes it's STILL not doing that

    self=0x20001718 <displays+80>)
    at ../../shared-module/displayio/display_core.c:311
311    void displayio_display_core_collect_ptrs(displayio_display_core_t* self) {
(gdb) p self.base.type
$7 = (const mp_obj_type_t *) 0x41f1c <displayio_display_type>

[simple pulsar]

Has displayio changed from a user's point of view between 4.x and 5.x or is it just internals?

For https://circuitpython.readthedocs.io/en/latest/shared-bindings/displayio/__init__.html there's no 5.x link available, btw

[idle owl]

There are definitely user changes - keywords and some functionality differences, as far as I know.

[simple pulsar]

Thanks, I'll proceed with caution.

[crimson ferry]

latest should be up to date with the 5.0.0 beta. I'd expect that 5.x would show up once it's released as stable

[tidal kiln]

"latest" is essentially 5.x, so for example, compare:
https://circuitpython.readthedocs.io/en/latest/shared-bindings/displayio/Display.html
with 4.x:
https://circuitpython.readthedocs.io/en/4.x/shared-bindings/displayio/Display.html

refresh_soon and wait_for_frame are going away as one example

[meager fog]

@tidal kiln ok how about shake goes in CLUE

at some point we will have a 'sensorlab' equivalent for all these generic sensor calculations

[idle owl]

Please welcome our new CircuitPython Helpers! When they're available, they can help with CircuitPython related questions!

[tidal kiln]

fine by me. i'll move the issue back.

did someone mention in yesterday's meeting that the Radio library currently does not work with latest CP release?
https://github.com/adafruit/Adafruit_CircuitPython_Radio

[slender iron]

@tidal kiln ya, someone mentioned it. I can't remember who

@fallen anvil sorry for the frustration with the bot. if you ever want text of yours to revise, the mods have access to what has been deleted. I'd rather have false positives instead of missing stuff. I can also look into the moderated words and tweak the settings. (It used to flag subportions of a word but we can set it better now.)

[simple pulsar]

Are the docs for Label class supposed to be found with this: https://circuitpython.readthedocs.io/en/latest/search.html?q=Label&check_keywords=yes&area=default# ?

[fallen anvil]

I wouldn't stress on it too much...I use ripcord as the client and it ate the long and thought through message I'd constructed about the sync options and I felt a bit peeved. I'd rather have a full false positives that a constant stream of adverts for extensions or reductions to parts of my body. It might be worth looking at allowing multiple X's though though...there are lots of valid use cases for those.

[slender iron]

@simple pulsar is will be a subproject since label is in the display text library

easiest way to find it is probably the github page for the library

https://circuitpython.readthedocs.io/projects/display_text/en/latest/api.html

thanks @fallen anvil, will see what other mods think about removing xs

[fallen anvil]

I would imagine X's would be accompanied by other 'indicative factors' for those messages we didn't want to see....

[simple pulsar]

Is that a deficiency in readthedocs?

[lone axle]

I use the release page in github to get the docs a fair amount. There is usually a line like "Read the docs for info on how to use it." that is linkified. Another thing I've found myself doing though is in google using a term like "circuitpython label docs" using both circuitpython and docs usually gets the correct docs page somewhere in the top results.

I don't think it's deficiency so much as organization. The search box is looking within one project. Since adafruit_display_text is a separate project it's not finding it from the base project.

once you get to the docs page for the adafruit_display_text the search box will look within that project

[simple pulsar]

It feels like it's going to confuse the average user.

[fallen anvil]

@simple pulsar speaking as a newbie on the python side....trust me, it does.

[manic glacier]

jepler

[adafruit/circuitpython] Pull request opened: #2655 samd51 thing plus: new port

This adds support for the SparkFun Thing Plus - SAMD51

testing performed:

• successfully store and retrieve a 500kB file on the SPI flash
• a square wave output on each pin appears on o'scope
• double checked silk vs names in pins, but you know how that goes
• board.SPI(), board.SERIAL(), board.I2C() all construct without an exception

[onyx hinge]

 ! [remote failure]      thing51 -> thing51 (remote failed to report status)
error: failed to push some refs to '[e-mail]:jepler/circuitpython.git'
``` well that's weird

[simple pulsar]

How many built-in fonts are there in displayio on CLUE? Is it just the (tiny) terminalio.FONT ?

[onyx hinge]

oh github is having problems at the moment.

@simple pulsar yes afaik

[simple pulsar]

Ta. The C/Arduino world of Arcade appears to have the ability to modify the size of the font. That's set to 2 in the demo code: https://github.com/adafruit/Adafruit_Arcada/blob/master/examples/full_board_tests/arcada_clue_test/arcada_clue_test.ino#L71

[manic glacier]

ladyada

[adafruit/circuitpython] Pull request review submitted: #2655 samd51 thing plus: new port

good enuf to get in before next release!

[lone axle]

@simple pulsar you can modify the size of the font in circuit python by scaling the TileGrid or Group.

[tulip sleet]

@onyx hinge I am wondering if there's something uninitialized somewhere

[slender iron]

pretty battery voltage curve

[stuck elbow]

@slender iron which board is it?

[slender iron]

feather nrf52840 reporting internal temp on a 100mah battery

gonna run it again with the neopixel off 🙂

[stuck elbow]

that should help :P

[slender iron]

ya, this is the baseline

[stuck elbow]

I assume it's 4h and not 4 minutes

[slender iron]

yup, it is hours. voltage is in millivolts

[stuck elbow]

though with neopixels you never know

[slender iron]

I think it is broadcasting every 10 seconds

[solar whale]

@slender iron so I should be able to do xxx.yyy

yay!

[slender iron]

🙂

[solar whale]

but not yyy.xxx

hmmm

[manic glacier]

ladyada

[adafruit/circuitpython] Pull request review dismissed: #2655 samd51 thing plus: new port

good enuf to get in before next release!

tannewt

[adafruit/circuitpython] New comment on issue #2639: Sanity check memory use in 5\.0\.0

Are we using the inline limits at all on this board? I wonder if the compiler is improperly dropping the type check sometimes.

dhalbert

[adafruit/circuitpython] New comment on issue #2639: Sanity check memory use in 5\.0\.0

No, pyportal has no tricky stuff in mpconfigboard.*

dhalbert

[adafruit/circuitpython] New comment on issue #2639: Sanity check memory use in 5\.0\.0

I am doing some more bisects, but I'm suspicious of something uninitialized somewhere.

ladyada

[adafruit/circuitpython] Pull request review submitted: #2655 samd51 thing plus: new port

still good :D

[onyx hinge]

I dunno if we want to do this as text or as voice.

[lapis hemlock]

I dunno if we want to do this as text or as voice.
@onyx hinge We can try over voice.

[manic glacier]

tannewt

[adafruit/circuitpython] Issue closed: #2567 Leaving CIRCUITPY plugged into Mac when it sleeps crashes the Mac

tannewt

[adafruit/circuitpython] Pull request closed: #2653 Fix Mac crash when waking up with an ejected CIRCUITPY

[onyx hinge]

Sorry I couldn't make it for in the weeds, and thanks for making time today. I think the main thing we need to decide before merging a ulab PR is how it is presented to users. Here are the options I thought about:
Option 1. multiple ulab_mmm modules
Pros: Easiest
Cons: Less pythonic than dotted names

Option 2. fix underlying code for ulab.mmm native modules
Pros: Most pythonic, no extra glue
Cons: less chance of acceptance at micropython

Option 3. Wrap multiple _ulab_mmm modules into ulab.mmm modules via frozen ulab package
Pros: most pythonic
Cons: extra glue in the form of frozen package (but can transition later to option 2)

we can step over to the voice room once @slender iron joins us, I just wanted to get that block of text out of the way 🙂

[lapis hemlock]

@onyx hinge I think the mike on my computer doesn't play along. I will try to install the app on the phone. I might need 5 minutes.

[onyx hinge]

@lapis hemlock text is fine too

[lapis hemlock]

Actually, I don't hear Scott.

[onyx hinge]

<module ''>
>>> import microcontroller.pin as pin
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ImportError: no module named 'microcontroller'

[slender iron]

@lapis hemlock are you using the app? it tends to work better

[lapis hemlock]

No, I am in browser.

OK, let's do that.

[half sedge]

Is there an "in the weed" right now? I have something that I planned for next week, but could use @slender iron or @gilded cradle opinion: https://gist.github.com/dglaude/9a158af4b0332417e34abc5f8ff6b16f

[slender iron]

@half sedge we're talking about ulab

[half sedge]

Nothing urgent, I don't have the hardware yet (except for the Pi and a lot of HAT)... here is a list of the Pi hardware I have (or should have): https://gist.github.com/dglaude/e5070d1f5737e80aa246ccf7db3b7e32

[slender iron]

@lapis hemlock I'm taking look at why the import microcontroller.pin doesn't work

[lapis hemlock]

In the meantime, I am installing the app.

[half sedge]

Just lurking on the voice channel.

[onyx hinge]

@lapis hemlock text will be fine if you are tired of messing with it!

@lapis hemlock we cannot hear you

[lapis hemlock]

This is not tiredness, this is embarrasment.

I can hear you both.

[onyx hinge]

@lapis hemlock what do you want to do?

[lapis hemlock]

It's up to you.

I have a list that I wanted to discuss, I can send it by mail?

[onyx hinge]

let's shift to text, then, instead of listening to us try to run a debugger over voice chat

@lapis hemlock you can go by e-mail too if you want

[lapis hemlock]

I can hear you.

[onyx hinge]

we can work in all the ways

[half sedge]

Do you want ulab to work with Blinka on Raspberry Pi? Or is this not possible because this is C code and build into Circuit Python kernel? Are we going to be like for displayio?

[lapis hemlock]

E-mail is out.

[half sedge]

Asking for a friend. 😉

[lapis hemlock]

I think the problem is that this concerns py-core, so Damien will take his time.

Sorry Scott, I don't have your e-mail.

[slender iron]

@lapis hemlock it's [e-mail]

[lapis hemlock]

OK, then let's take the beauty plaster.

So, Jeff, then I will implement that, and then update the PR.

But whatever is in .attr, overrides the local dicts.

Right.

[manic glacier]

dhalbert

[adafruit/circuitpython] New comment on issue #2639: Sanity check memory use in 5\.0\.0

I did a bisect from 5.0.0-beta.5 to the tip of master. git bisect assumes you are going from "good" to "bad", not the other way, so I had to mark good results as bad and vice versa. (I didn't know this at first and got into some confusing situations.) The problem got "fixed" at 36e6cc8febf7f744f61772d29eed68d9ebb57200 (#2614), which makes sense because it allocated memory more efficiently, but it didn't necessarily fix any underlying bug. Will do a little more looking on beta.5.

[lapis hemlock]

The problem is that https://github.com/micropython/micropython/pull/4731 old, and it is probably old of a reason.

Andrew needs 2 dims, but openmv needs 3.

Exactly, slices are difficult.

And broadcasting.

I mean, not conceptually, but the implementation.

https://github.com/v923z/micropython-ulab/issues/2

Someone needed 4 dims.

OK, so are you saying that speedtest is not worth it?

FFT is another example.

I haven't yet had time.

Right.

It is easy to find out: you just have to run the code with two different types.

Float will call your code, uint16 the other.

The problem with floats is that sersor data are uint16s, right?

Treating identical types is fine with me.

But is your filter uint16?

[ionic elk]

hey @tulip sleet you work on Ubuntu right?

[tulip sleet]

@ionic elk yes

[ionic elk]

@tulip sleet could you run which arm-none-eabi-gdb for me real quick and tell me where it is?

[tulip sleet]

you need to use the gcc9 download from arm, and install its bin's in some directory you add to your path. The standard arm-none toolchain is out of date from our point of view

[lapis hemlock]

OK, I can relatively quickly polish the internals of 2D arrays, so you can use it.

[ionic elk]

hrrm

[tulip sleet]

download 9-2019-q4 from here: https://developer.arm.com/tools-and-software/open-source-software/developer-tools/gnu-toolchain/gnu-rm/downloads

unzip it in some directory, and add its bin directory to your path. For example. I unzip it in ~/bin as

[lapis hemlock]

A question about complexes: as far as I heard, complex is not implemented for all micropython platforms. WHat is the situation in CP?

[tulip sleet]

~/bin/gcc-arm-none-eabi-9-2019-q4-major, and have it on my PATH as /home/halbert/bin/gcc-arm-none-eabi-9-2019-q4-major/bin

[onyx hinge]

mkdir: cannot create directory ‘sys’: File exists
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
AttributeError: 'module' object has no attribute 'boo'

[tulip sleet]

there is no PPA any more for recent versions of the arm-none toolchain

[lapis hemlock]

OK, so then FFT must be implemented as it is.

I would like to steer clear of them.

STM32F405 also has complexes.

pyboard.

Right, but then binary operators will again go up in size.

[onyx hinge]

mpconfigport.h:#define MICROPY_PY_CMATH (1)

[lapis hemlock]

In complexity, too.

What about floats vs. doubles?

Some hardware has doubles.

Right.

OK, then we stay with mp_float_t only, without differentiating between floats and doubles.

I am that is up to the micropython pre-processor flag/hardware.

The question really was, whether you would want to have floats and doubles side by side, if the hardware allows it.

It is not only that: matrix inversion is such an example.

OK, we would have to explore it.

@onyx hinge I think we have discussed all my questions, so I will update the the PR, and then we can merge it. After that I can clean up the internals, but that should not change the interface.

Right.

The only thing missing is the treatment of properties.

Where is the .attr overriden?

The local dict.

If the imports are sorted out, do you plan on pushing it upstream?

[idle owl]

@tulip sleet We've had a similar discussion before, but I'd like some advice on what type of error to throw. If it were to throw the error itself, it would be a TypeError but that's because it defaults to None and none-type objects can't be converted to strings or whatever the error is. I'm trying to throw an error if an image_name is not provided to a bitmap background feature. Right now it defaults to something else because of the code, and I'd rather it fail if someone calls the image_background feature without providing an image name.

[lapis hemlock]

What is the exact path of attr's resolution?

[onyx hinge]

after: ```155672 bytes free in flash firmware space out of 499712 bytes (488.0kB).
164584 bytes free in ram for stack and heap out of 196608 bytes (192.0kB).

before: ```156520 bytes free in flash firmware space out of 499712 bytes (488.0kB).
164584 bytes free in ram for stack and heap out of 196608 bytes (192.0kB).

metro m4 express

[tulip sleet]

@idle owl usually if an argument is wrong in some way we throw ValueError, with explanatory text

[idle owl]

@tulip sleet That's what I have in there now. Excellent. Thank you.

[lapis hemlock]

But sorting and things like that have to be implemented for complexes.

[onyx hinge]

sorted((1+0j, 1j, -1j))
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: '<' not supported between instances of 'complex' and 'complex'

[lapis hemlock]

I think math function for complexes are an extra module in micropython.

On the other hand, it is not too much: https://github.com/micropython/micropython/blob/master/py/modmath.c

[manic glacier]

jepler

[adafruit/circuitpython] Pull request closed: #2655 samd51 thing plus: new port

[lapis hemlock]

Thanks for Scott

SOrry for the trouble.

[onyx hinge]

see you later @lapis hemlock !

[lapis hemlock]

Cheers!

[manic glacier]

ladyada

[adafruit/circuitpython] New comment on issue #2141: Add board support for SAMD51 Thing Plus

added and merged - thanks @jepler :)

ladyada

[adafruit/circuitpython] Issue closed: #2141 Add board support for SAMD51 Thing Plus

[onyx hinge]

I'm going to take a break. subsequently to that pr being merged in micropython-ulab I'll update my PR in circuitpython on top of it.

It'll take a bit more time because I'll have docs to reformat as well, and it might end up having this dependency on scott's work too

[lapis hemlock]

OK. I will try to fix the issue with the properties by tomorrow night.

[onyx hinge]

@lapis hemlock if you want to try voice chat again sometime just to figure out if your tech is working, I'm often around so don't hesitate to ask

[lapis hemlock]

Oh, thanks!

I will take up the offer.

[idle owl]

@lapis hemlock I recommend not remaining in the circuitpython voice channel unmuted on the off chance that your mic starts working and you begin transmitting sound without realising it.

[lapis hemlock]

OK, thanks for the hint!

[idle owl]

You're welcome 🙂

[half sedge]

I did not notice the voice ended. 😉

[lapis hemlock]

@lapis hemlock I recommend not remaining in the circuitpython voice channel unmuted on the off chance that your mic starts working and you begin transmitting sound without realising it.
@idle owl Actaually, the mike was OK. At least, it passed the mike test under settings, but for some reason, the voice didn't end up in the channel.

Weird.

[idle owl]

That is odd. Discord can be frustrating at times.

[solar whale]

@idle owl was there a breaking change for the CLUE recently https://forums.adafruit.com/viewtopic.php?f=60&t=162679 or any suggestions for this?

[idle owl]

@solar whale https://github.com/adafruit/Adafruit_CircuitPython_CLUE/releases/tag/2.0.0 was the breaking change. The rest were not breaking since then.

Without seeing their code, I have no idea what is going on.

[solar whale]

Thanks!

[manic glacier]

dhalbert

[adafruit/circuitpython] New comment on issue #2639: Sanity check memory use in 5\.0\.0

5.0.0-beta.5 gets the memory error only if compiled with -flto and with inlining enabled. Disabling -flto or enabling -fno-inline causes the problem to go away. And the breakpoint addresses are very fishy in the problem case:

(gdb) bre displayio_display_collect_ptrs
Breakpoint 3 at 0x15280: file ../../shared-module/displayio/display_core.c, line 311.
(gdb) bre displayio_epaperdisplay_collect_ptrs
Breakpoint 4 at 0x15282: file ../../shared-module/displayio/display_core.c, line 3...

[fallen anvil]

@indigo wedge responded to your PR comments (you're probably already aware). Will push PR update tomorrow AM after testing and , then I think it can be pulled.

[indigo wedge]

Yes, reading now, thanks so much 🙂

[fallen anvil]

One question...should UF2_MAGIC_START1 be different per board? That would stop an incompatible flash being submitted, and there's no protection against that at the moment.

I can push the PR now if you want, but I want to test against the boards before it's ready for pull, and I'm too tatered tonight to do that now.

[indigo wedge]

no this is part of the spec https://github.com/microsoft/uf2#file-format

there is the family ID

so you can't send a samd uf2 file to a imx board

you can still send a rt1011 to a rt1062 but i guess that's good enough

[fallen anvil]

Yeah, that fixes what I was concerned about, I thought your comment about 'Randomly selected' meant it was our choice!

[indigo wedge]

yeah the comment came from https://github.com/microsoft/uf2/blob/master/uf2.h#L39

[fallen anvil]

OK

[manic glacier]

creston-bob

[adafruit/circuitpython] New comment on issue #2635: I2C Bus error leaves board unrecoverable without power down

In a project using the Microchip PIC32MZ microcontroller I discovered that the first two versions (v1 & v2) have a hardware bug in the silicon that causes the I2C to hang occasionally. I dug into the matter and found the errata sheets indicated that v1 & v2 had this flaw but it was apparently fixed in v3. After trying out all Microchip's suggestions for fixing the hung I2C from within my program code (and failing), I eventually concluded that the ONLY possible fix was a power reset (which wor...

[fallen anvil]

@indigo wedge Is the syntax for the build.yaml like this boards: ['mimxrt1010_evk','mimxrt1020_evk','versiboard2'] ?

[indigo wedge]

i believe so, yes

[fallen anvil]

k

well, if not, we'll find out

[indigo wedge]

only tested with 1 board so we'll see

[fallen anvil]

shall I push these changes? I'll message you tomorrow when they're ready for pulling (i.e. tested on boards), but means I've got a backup :-)

[manic glacier]

creston-bob

[adafruit/circuitpython] New comment on issue #2635: I2C Bus error leaves board unrecoverable without power down

I forgot to mention that the Arduino board I mentioned in my previous comment energizes a relay to power reset the PIC32MZ board. The power is connected through the relay's NC contact which is opened for a few seconds to cause the power reset.

[simple pulsar]

Is there a recommended technique for scrolling things efficiently in displayio? I'm pondering shuffling stuff around in a Bitmap.

[manic glacier]

ladyada

[adafruit/circuitpython] Pull request opened: #2656 move red LED to D13 and make D3 lsm6ds IRQ

to match rev B

dhalbert

[adafruit/circuitpython] Pull request review submitted: #2656 move red LED to D13 and make D3 lsm6ds IRQ

Looks good, and consistent!

[slender iron]

@simple pulsar not really, refreshing the whole screen is slow regardless. if you are moving a bunch of individual things then I recommend disabling the refresh as you do

[manic glacier]

ladyada

[adafruit/circuitpython] Pull request closed: #2656 move red LED to D13 and make D3 lsm6ds IRQ

tannewt

[adafruit/circuitpython] Pull request opened: #2657 Support importing native modules in native packages\.

This only fixes the import portion. It doesn't actually change
reference behavior because modules within a package could already
be referenced through the parent package even though an error should
have been thrown.

[manic glacier]

tannewt

[adafruit/circuitpython] New comment on pull request #2652: Creat ar\.po

Based on my eye it looks like it is partially working in iTerm with screen on Mac OSX. (I don't expect it to work on built in displays yet.)

I believe the line in the traceback is correct but the code.py : output line needs the filename in the translated string itself and the Arabic portion is reversed.

I think translating these two would be hel...

[manic glacier]

tannewt

[adafruit/circuitpython] New comment on pull request #2657: Support importing native modules in native packages\.

@jepler @v923z This should allow for importing submodules of ulab.

I've linked @jimmo to it and I bet @dpgeorge would be interested too.

[analog surge]

Latest being "adafruit-circuitpython-metro_m0_express-en_US-4.1.2.uf2"

[manic glacier]

dhalbert

[adafruit/circuitpython] New comment on issue #2639: Sanity check memory use in 5\.0\.0

I spent time looking at the assembly code for 5.0.0-beta.5, and for e560b419f, which is the last commit before beta.5 that does not show the problem. I don't see anything wrong with the code. It did a ton of inlining: in both, all of displayio_gc_collect() is inlined, and all the single-use functions it calls are inlined. In e560b419f, even more lower-level functions are inlined.

I think the difference here may have something to do with how gc handles pointers in existing stack frames. P...

[slender iron]

@analog surge that is the latest stable but we recommend 5.0.0 beta. (we're releasing a release candidate for stable in the next day or so)

[manic glacier]

tannewt

[adafruit/circuitpython] New comment on issue #2639: Sanity check memory use in 5\.0\.0

@dhalbert I'm fine closing it. I've seen similar with pointers in stack frames living longer than intended. It's all part of the imprecision in the GC process.

tannewt

[adafruit/circuitpython] Issue closed: #2639 Sanity check memory use in 5\.0\.0

[lone sandal]

faverolles

[adafruit/awesome-circuitpython] New star added

[meager fog]

@tidal kiln hayy

i didnt have a break till now, if ya wanna chat about the clue

[tidal kiln]

sure.

you saw basecamp stuff?

[meager fog]

yah

but i was in meatings all day 😄

[tidal kiln]

ok. no worries.

[meager fog]

ok so i like that it has the cheat sheet

cause its tough to know morse code

so this is a nice trainer

incoming on top, outgoing on bottom is good

another thing...

maybe it sends the characaters immediately?

you can see the history but its still like, all chars

cause there's no spaces and punctuation anyhow

[tidal kiln]

it sends dot/dash immediately? or letter?

[meager fog]

letter

you have to pause between letters anyways right

so it can detect end of one char

so maybe 1 second pause?

[tidal kiln]

or double buttons - that was working nice on that other example

[meager fog]

ohh hmm

well how do people do it in real morse?

[tidal kiln]

so left=dot, right=dash, both=done

[meager fog]

maybe both == space?

or double both == space

[tidal kiln]

same thing? space would mean you're done with that letter?

[meager fog]

i mean if you press both

then dont click left or right before hitting both again

its a space

both == end char (try to decode it)

both pause both == space

maybe 3 times == stop

(period)

if that isnt hard to detect

[tidal kiln]

or the light sensor for that?

[meager fog]

risky

you could move your head and light sensor dims

could do proximity sensor

but then you have to move your hands

there's cap touch pads

[tidal kiln]

that also seemed ricky - they're right there while you're holding it

oh - i was using proximity, not light

while clue.proximity < 255:

[meager fog]

proximity is ok

[tidal kiln]

i was using that for "send". worked pretty well.

[meager fog]

that's pretty reliable

ok cool

downside is ... it only works on CLUE then

isnt as portable

[tidal kiln]

hmmm. you thinking CPB also?

[meager fog]

yeha

and other future boards 🙂

what about shake() instead of prox

then its most microbitty

or did you not like shake

[tidal kiln]

so far, only been using clue, so no shake

[meager fog]

could add shake v easily

i could do that now if its holding u up 🙂

[tidal kiln]

i can PR that. just need to do some housekeeping in my repo. i had to do some quick mods to make a no-LIS3DM version for my one clue

but i remember using shake for something else. it worked, but was sort of GRRRRRR vs. just pressing a button.

[lone axle]

I've used cap sense on pin 2 as a "third button" for some things recently and found it pretty nice.

0 and 1, like you said I found that I touched it accidentally too easily but 2 I don't hit accidentally and it feels naturally to reach down and touch it with a thumb for me at least.

[meager fog]

@tidal kiln can change the shake calibration

[tidal kiln]

the defaults?

PR pretty much ready to send

oh - you mean for less GRRR

@lone axle yah, good idea about using 2. i should try that at least.

[meager fog]

either way!

ok im gonna 💤 for a bit

back l8r

[tidal kiln]

@meager fog ok, i'll keep playing with the cheat sheet version, put the other one on back burner. and general summary:

left = dot
right = dash
both = end char / decode (nothing = space)
shake/cap touch/prox/TBD = send message
---
upper area = incoming message
lower area = outgoing message

[meager fog]

yep

xactly

i know you'll do perfect 🙂

[tidal kiln]

ha! thanks. will at least try. l8r!

[crimson ferry]

I gotta ask, @rare geode, did you somehow get #1337 on purpose?

[ivory yew]

👀 new circuitpython eurorack board coming soon: https://twitter.com/theavalkyrie/status/1232534946696187904

Stargirl (@theavalkyrie)

Panel dimensions are good to go.

Time to put some pretty art on this. :)

Twitter

[manic glacier]

mesa177

[adafruit/circuitpython] New comment on pull request #2652: Creat ar\.po

Good eye!

Yes indeed the Arabic portion is reversed. On the other hand, the nature of the written Arabic letters retain their proper forms (so the middle lam has the correct writing shape, the initial alef, and so forth...) Hope that makes sense :)

On another note, I tried continue translating some other lines and hit a snag on one of them (line 141 to be exact, 0x%x apparently doesn't like having Arabic writing within the same sentence hehe). For now I recommend just commenting it unt...

[manic glacier]

v923z

[adafruit/circuitpython] New comment on pull request #2657: Support importing native modules in native packages\.

So the claim is that

from ulab import fft

fft.fft(...)

would work with this fix, right? If so, it is really great, thanks for taking care of it!

@dpgeorge Damien, do you think this could be merged into micropython proper on a reasonable time scale? With that we could maintain a single version of ulab, ant that would save some resources on our end.

[fallen anvil]

@indigo wedge Good to go!

[fallen anvil]

Updated the push to remove the release bins. Only real remaining issue is to tidy up VID:PIDs for the individual boards, but you know what you've got allocated so it's easier for you to do it. Versiboard2 doesn't have its own VID/PID so will use the generic one unless someone wants to start making some (which won't be me...I've no interest in the pain of mass production).

[manic glacier]

dpgeorge

[adafruit/circuitpython] New comment on pull request #2657: Support importing native modules in native packages\.

@tannewt thanks for the note, I'll look into incorporating this or similar. I didn't yet look in detail but this might be related: https://github.com/micropython/micropython/pull/4731

[timber mango]

I am getting the following error: Auto-reload is on. Simply save files over USB to run them or enter REPL to disable. main.py output: Traceback (most recent call last): File "main.py", line 41, in <module> File "adafruit_sht31d.py", line 169, in __init__ File "adafruit_sht31d.py", line 181, in _reset File "adafruit_sht31d.py", line 173, in _command File "adafruit_sht31d.py", line 173, in _command File "adafruit_bus_device/i2c_device.py", line 99, in write TypeError: can't convert NoneType to int The source that produces it is:```python
import board
import busio
import digitalio
import math
import time

#from micropython import const

import adafruit_lis3dh
import adafruit_sht31d
import neopixel

Configuration for the LIS3DH:

ACCEL_RANGE = adafruit_lis3dh.RANGE_16_G # Accelerometer range.
TAP_THRESHOLD = 20 # Accelerometer tap threshold. Higher values
# mean you need to tap harder to start a spin.

i2c = board.I2C()
int1 = digitalio.DigitalInOut(board.D7) # Set to correct pin for interrupt!
lis3dh = adafruit_lis3dh.LIS3DH_I2C(i2c, int1=int1)

Set range of accelerometer (can be RANGE_2_G, RANGE_4_G, RANGE_8_G or RANGE_16_G).

lis3dh.range = adafruit_lis3dh.RANGE_2_G

Set accelerometer range.

lis3dh.range = ACCEL_RANGE

Enable single click detection, but use a custom CLICK_CFG register value

to only detect clicks on the X axis (instead of all 3 X, Y, Z axes).

lis3dh.set_tap(1, TAP_THRESHOLD, click_cfg=0x01)

Create library object using our Bus I2C portLine 41

Line 41***>>>sht31d = adafruit_sht31d.SHT31D(i2c)```I have no idea what is causing this. I can not create the sht31d object.

[fallen anvil]

Hmm....mimxrt SPIs (at least when used with ESP32SPI) are getting created with double speed clocks...anyone aware of this I do I go fishing?

[indigo wedge]

Yes Limor mentioned that some time ago but I haven't looked into it yet

there might be a issue created for that

[fallen anvil]

The MISO/MOSI were also reversed

[indigo wedge]

oh?

[fallen anvil]

OK, I'll brew a patch

[indigo wedge]

ah there it is https://github.com/adafruit/circuitpython/issues/2483#issuecomment-573267916

[fallen anvil]

I've only checked 1021, which I doubt anyone else has used

[indigo wedge]

i tested on my 1011 feathers and it worked without reversing

but maybe it's consistently reversed everywhere and that's why it works

[fallen anvil]

Yeah, it's probably a 1021 specific thing, but I'll look, since I've got the manuals now.

(Makes me happy, since my ESP-32 WIFI now works :-) )

[indigo wedge]

@fallen anvil merged the PR, thanks again 🙂

[solar whale]

@timber mango what MCU board are you using?

[fallen anvil]

@indigo wedge Yep, saw the merge. No problem. Every little helps :-)

[solar whale]

@timber mango I tried your sequence on a PyGamer with an sht31d attached and it worked OK for me - bot the lis3dh and the sht31d are working.

sorry -- no idea what is going wrong for you.

[manic glacier]

jimmo

[adafruit/circuitpython] New comment on pull request #2657: Support importing native modules in native packages\.

See https://github.com/micropython/micropython/pull/5701 for the equivalent change upstream. Thanks for suggesting this, I think it's a much better approach than micropython#4731

Some minor notes on this PR:

• Might need to check when looking at the globals dict that the result is actually a module (i.e. type == mp_module_type).
• You can avoid the change in objmodule.c by just setting a flag in builtinimport, then avoid calling mp_store_attr at all.

[solar whale]

@stable aspen i suggest just trying the sht31d_simpletest to see if you can read from the sensor.

[simple pulsar]

What's the purpose of the minimum refresh feature on displayio.refresh() ```>>> import board

import time
display = board.DISPLAY
display.auto_refresh = False

print("hello") ; display.refresh() ; time.sleep(2) ; print("world") ; display.refresh()
hello
True
world
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
RuntimeError: Below minimum frame rate

[solar whale]

@simple pulsar https://circuitpython.readthedocs.io/en/latest/shared-bindings/displayio/Display.html#displayio.Display.refresh does this help?

it works with this ```>>> print("hello") ; display.refresh(minimum_frames_per_second=0) ; time.sleep(2) ; print("world") ; display.refresh(minimum_frames_per_second=0)
hello
False
world
False

[manic glacier]

mubes

[adafruit/circuitpython] Pull request opened: #2658 Fix SPI clock speed on mimxrt10xx family & mimxrt1020 pinmux fixups

Make SPI dependant on the divisor register rather than a constant, and fix more mixup in the 1020 pinmux table relating to SPI and I2C.

[fallen anvil]

...@indigo wedge That should help with a few bits and pieces.

[simple pulsar]

@solar whale You're setting the value to 0. I'm on a CLUE with 5.0.0-beta.5. I was wondering what the purpose of the feature is.

[manic glacier]

mubes

[adafruit/circuitpython] New comment on issue #2483: mimxrt10xx: Verify the busio module

SPI Frequency is corrected in https://github.com/adafruit/circuitpython/pull/2658

[lone axle]

@simple pulsar from the docs it looks like setting the value to 0 will prevent it from raising the RuntimeError that you posted.

the purpose I believe is to give you a mechanism to allow your app to refresh the screen at a desired rate 60 FPS for instance, without having to "manually" track the time that have logic that chooses whether or not to draw a frame yet

I believe in a "real" use case where are trying to achieve a specific frame rate you would want to set the parameter to the desired rate i.e. 60 and then your code would need to try/except to catch the RuntimeError if/when it's too early to refresh the screen.

And the True/False return help you know whether the the frame actually got drawn or not.

[timber mango]

@solar whale Same error on sht31d_simpletest.py. ItsyBitsy M4 Express.```python
import time
import board
import busio
import adafruit_sht31d

Create library object using our Bus I2C port

i2c = busio.I2C(board.SCL, board.SDA)
sensor = adafruit_sht31d.SHT31D(i2c)

loopcount = 0
while True:
print("\nTemperature: %0.1f C" % sensor.temperature)
print("Humidity: %0.1f %%" % sensor.relative_humidity)
loopcount += 1
time.sleep(2)
# every 10 passes turn on the heater for 1 second
if loopcount == 10:
loopcount = 0
sensor.heater = True
print("Sensor Heater status =", sensor.heater)
time.sleep(1)
sensor.heater = False
print("Sensor Heater status =", sensor.heater)```

main.py output:
Traceback (most recent call last):
  File "main.py", line 8, in <module>
  File "adafruit_sht31d.py", line 169, in __init__
  File "adafruit_sht31d.py", line 181, in _reset
  File "adafruit_sht31d.py", line 173, in _command
  File "adafruit_sht31d.py", line 173, in _command
  File "adafruit_bus_device/i2c_device.py", line 99, in write
TypeError: can't convert NoneType to int```

[lone axle]

@timber mango if you run this code in the REPL do you see any addresses get printed?

import board
import busio
i2c = busio.I2C(board.SCL, board.SDA)
print([hex(x) for x in i2c.scan()])

Looking at this page: https://learn.adafruit.com/circuitpython-basics-i2c-and-spi/i2c-devices it looks like it's using try_lock() which I don't see in your code. I'm not certain if that could result in the error you are seeing though.

I would start with the scan and verify your hardware / wire connections to see if the circuit python device can see the sensor on the I2C bus or not.

[raven canopy]

@timber mango iirc, that was a recent bug. Are you using most current adafruit_bus_device?

[solar whale]

@simple pulsar I was on a Pybadge -- just playing with the settings -- no idea why its used.. IIRC the auto_refresh was updated not to long ago. Not sure if this is just a legacy thing or what. Best to check with @slender iron

[raven canopy]

@timber mango here is the fix to the issue: https://github.com/adafruit/Adafruit_CircuitPython_BusDevice/pull/45

[timber mango]

@solar whale I am using the Feb 22 libraries.

@raven canopy When is that fix going to get into the official library??

[raven canopy]

From the looks of it, it went in on 19 Feb...

Well, the fix was merged 18 Feb. Went into the Bundle on 19 Feb.

[solar whale]

probaly worth updating all libs -- I am using the 0226 bundle

@timber mango do other I2C devices work on yout ItsyBitsy M4 ? may be an issues with the sht31d. or try the sht31d on another board

[timber mango]

Yes. I also have an LIS3DH connected.

[solar whale]

has the sht31d ever worked?

[timber mango]

Yes, it has always worked great.

[solar whale]

@timber mango something odd with your bus device lib -- line 99 is a comment in the current release https://github.com/adafruit/Adafruit_CircuitPython_BusDevice/blob/master/adafruit_bus_device/i2c_device.py#L99

try updating

[manic glacier]

arturo182

[adafruit/circuitpython] New review comment on pull request #2658: Fix SPI clock speed on mimxrt10xx family & mimxrt1020 pinmux fixups

That's what I get for copying example code ;)

arturo182

[adafruit/circuitpython] Pull request review submitted: #2658 Fix SPI clock speed on mimxrt10xx family & mimxrt1020 pinmux fixups

arturo182

[adafruit/circuitpython] Pull request review submitted: #2658 Fix SPI clock speed on mimxrt10xx family & mimxrt1020 pinmux fixups

arturo182

[adafruit/circuitpython] Pull request closed: #2658 Fix SPI clock speed on mimxrt10xx family & mimxrt1020 pinmux fixups

arturo182

[adafruit/circuitpython] New comment on pull request #2658: Fix SPI clock speed on mimxrt10xx family & mimxrt1020 pinmux fixups

Thank you!

[tulip sleet]

@timber mango the "TypeError: can't convert NoneType to int" was fixed by https://github.com/adafruit/Adafruit_CircuitPython_BusDevice/pull/45, which is in release 4.2.1 of busdevice

if you do:

>>> import adafruit_bus_device
>>>> adafruit_bus_device.__version__

what do you see?

[orchid basin]

ladyada

[adafruit/circuitpython-org] New comment on issue #410: Misleading photo for Arduino Nano 33 BLE

hi, they are the same boards / pinouts. we tested specifically with the SENSE version.

dhalbert

[adafruit/circuitpython-org] New comment on issue #410: Misleading photo for Arduino Nano 33 BLE

It might make sense (ouch) to update the description and/or board title in https://github.com/adafruit/circuitpython-org/blob/master/_board/arduino_nano_33_ble.md. Would you like to submit a PR?

[timber mango]

@solar whale Here it is after I updated libraries to 0226.```python
import time
import board
import busio
import adafruit_sht31d

Create library object using our Bus I2C port

i2c = busio.I2C(board.SCL, board.SDA)
sensor = adafruit_sht31d.SHT31D(i2c)

loopcount = 0
while True:
print("\nTemperature: %0.1f C" % sensor.temperature)
print("Humidity: %0.1f %%" % sensor.relative_humidity)
loopcount += 1
time.sleep(2)
# every 10 passes turn on the heater for 1 second
if loopcount == 10:
loopcount = 0
sensor.heater = True
print("Sensor Heater status =", sensor.heater)
time.sleep(1)
sensor.heater = False
print("Sensor Heater status =", sensor.heater) Auto-reload is on. Simply save files over USB to run them or enter REPL to disable.
main.py output:
Traceback (most recent call last):
File "main.py", line 7, in <module>
RuntimeError: SDA or SCL needs a pull up```

[solar whale]

I suspect you have a HW issue -- check/replace the jumpers

[timber mango]

I already verified my circuit wiring and it is fine.

[tulip sleet]

could you check that the SDA and SCL pins at the itsy show 3.3V?

power-cycle the SHT31D if you have not done that yet

[solar whale]

@tulip sleet ```>>> import adafruit_busdevice
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
ImportError: no module named 'adafruit_busdevice'

[tulip sleet]

sorry adafruit_bus_device with an underscore

[manic glacier]

dhalbert

[adafruit/circuitpython] New branch created: 5\.0\.x

[solar whale]

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
AttributeError: 'module' object has no attribute '__version__'

[timber mango]

@tulip sleet I do not have any way to verify the voltage coming into the circuit.

[solar whale]

How did you verify the wiring? I would suggest reseating all jumpers. may just have a bad connection.

[slender iron]

@fallen anvil adafruit gives out USB PIDs for hobby boards. https://learn.adafruit.com/how-to-add-a-new-board-to-circuitpython/customizing-the-board-files#mpconfigboard-dot-mk-5-25

[tulip sleet]

@timber mango sorry, it's this:

>>> import adafruit_bus_device.i2c_device
>>> adafruit_bus_device.i2c_device.__version__

[timber mango]

@tulip sleet The errors I am getting?

[tulip sleet]

i was referering to "None ... int" errors; the pull-up errors are new. But let's make sure your version is OK first.

[timber mango]

@solar whale @tulip sleet I reseated all the jumpers. I also reseated myself, just to be sure.

[tulip sleet]

i am reproducing your setup...

[slender iron]

@stuck elbow over 10 hours when just flashing the neopixel during broadcast (every 30 seconds)

[timber mango]

@tulip sleet OK. I have the ItsyBitsy M4, followed by the LIS3DH and then the SHT31D.

[old smelt]

A while back, I had a link to CircuitPython/Blinka logos and other such marketing fodder. Can't seem to find it. Anyone have that handy?

[stuck elbow]

@slender iron that is some useful battery life

[ionic elk]

Has anyone on here played around with Migen? It's python for FPGAs. I've got a buddy ranting about it to me over text. https://github.com/m-labs/migen

Seems neat

[slender iron]

@old smelt look in awesome-circuitpython

@stuck elbow it's a start. I think I'll make it broadcast every 60 seconds and see how that changes the lifetime

[old smelt]

Thanks @slender iron

[timber mango]

@slender iron I have connected my Bluefruit SPI Friend to my ItsyBitsy M4, but it is not powered up yet.

[tulip sleet]

@timber mango, ok, I have an SHT31D connected to an itsy m4, running 5.0.0-beta.5, with 0226 bundle, and your test program above works and gives temp and humidity

[timber mango]

Do I go MISO to MISO and MOSI to MOSI, or do I cross connect them? I have never connected an SPI device before.

[tulip sleet]

if you have the lis3dh also connected, but not powered, that can mess up the pullups for the sht31d

[stuck elbow]

@timber mango no crossing

[tulip sleet]

try just the sht31d connected

[stuck elbow]

they named them so that you don't have to cross them

[tulip sleet]

master out slave in

[timber mango]

@tulip sleet I just removed power from the LIS3DH and reloaded sht31d_simpletest.py. Same SDA/SCL needs a pull up error.

[tulip sleet]

disconnect the SDA/SCL jumpers from the LIS3DH

[orchid basin]

tannewt

[adafruit/circuitpython-org] New comment on issue #410: Misleading photo for Arduino Nano 33 BLE

Why don't we just make two board definitions?

ladyada

[adafruit/circuitpython-org] New comment on issue #410: Misleading photo for Arduino Nano 33 BLE

thats fine too :)

[tulip sleet]

if you connect just the lis3dh and not the shtd31d, does it work?

[timber mango]

@tulip sleet The LIS3DH is not powered up, and SDA/SCL are disconnected. Same SDA/SCL error.

[tulip sleet]

from the repl, a simple test is:

>>> import board
>>> i2c = board.I2c()

with one or the other connected. If it gives the "needs a pullup error", then there's a problem

you don't even need the test programs to do the basic "needs a pullup" test. As long as the breakout board is powered, it should work

[timber mango]

With just the LIS3DH connected, I have the same SDA/SCL error. It was working earlier.

[tulip sleet]

would it be easy to take a photo?

[timber mango]

Yes, I will do that, but I have no camera on my computer. I have to take the picture and email it to myself, and then download to my computer. It will take a few minutes.

[tulip sleet]

if you have a phone, you can install the discord app, and then in the discord app, you can upload a picture directly

you can be logged into discord from the phone and the computer at the same time

[timber mango]

Do you want a picture with only the I2C devices connected, or just the LIS3DH OR the SHT31D?

Oh, yes, right!

[tulip sleet]

i want to see the wiring from the itsy to one of the devices

[timber mango]

OK, this will be with just the LIS3DH then.

[tulip sleet]

sure

a multimeter is very handy if you can put it on your shopping list. a cheap one from the hw store is fine

[timber mango]

@tulip sleet Here is the picture you ordered, Sir.

The loose jumpers go to the Bluefruit SPI Friend.

Or the SHT31D.

[tidal kiln]

@timber mango SDA pin on the LIS3DH looks a little under soldered

[timber mango]

@tidal kiln It works fine though. It was working earlier.

[tidal kiln]

ok, i just showed up, so may not understand what the issue is. looked like a "no pullup" error?

[timber mango]

Yes, that is it.

[tidal kiln]

it could be marginal, so it worked for a bit, then stopped

[timber mango]

It has always worked, in several different circuits since I got it.

[tidal kiln]

if you connect only the SHT31, does the "no pullup" error go away?

[solar whale]

since it happens on both the I2c devices - you ay also want to check the soldering of the header to the ItsyBitsy M4

[timber mango]

If it was going to fail, it should have failed much sooner than this.

[tidal kiln]

oh, does it for both? yah, check the itsy headers.

"worked before" -> with something other than this itsy?

[tulip sleet]

all the header pins are soldered, right?

You can try another set of pins: i = busio.I2C(board.TX, board.RX) will work. Wire the SCL to TX and SDA to RX

I'd disconnect everything except one I2C device to eliminate any issues. Also try the female headers on the board rather than the breadboard, or try other breadboard slots in the same column. if the breadabord is getting worn out, the connections can be flaky

[tidal kiln]

^^ yep. all that. and wired directly. don't use the power rail, etc.

[tulip sleet]

remove the unused wires while testing

[timber mango]

@tidal kiln @tulip sleet Here is a picture of the bottom of the ItsyBitsy M4.

[tulip sleet]

looks ok; if you can use the flash (or a desk lamp) when taking photos that would be great

[timber mango]

The breadboard is brand new, right out of the box. The LIS3DH is the only thing connected to the ItsyBitsy M4 now.

[tulip sleet]

then i'd try alternate pins next, in case the pins somehow got fried

[timber mango]

@tulip sleet Underside on the ItsyBitsy M4.

[rare geode]

@crimson ferry the 1337 is intentional yeah lol, Discord nitro

[lapis hemlock]

@onyx hinge Jeff, I think https://github.com/v923z/micropython-ulab/pull/41 is OK, you can definitely merge it.

@onyx hinge Do you want to wait for https://github.com/micropython/micropython/pull/5701 , or do you want to go ahead with @slender iron 's solution?

[timber mango]

@tulip sleet The ItsyBitsy M4 is connected only to the LIS3DH now with no connection to the power buses. Same SDA/SCL error.

[prime flower]

@idle owl I marked the Wiznet5k library as an alpha initial release (https://github.com/adafruit/Adafruit_CircuitPython_Wiznet5k/releases/tag/1.0.0-alpha). Should I add it to the bundle if it's marked as an alpha release?

[onyx hinge]

@lapis hemlock I will leave the import resolution to @slender iron and we can move forward either way

[lapis hemlock]

@lapis hemlock I will leave the import resolution to @slender iron and we can move forward either way
@onyx hinge OK. I forgot to mention, I also updated the manual https://github.com/v923z/micropython-ulab/blob/2dim/docs/manual/source/ulab.rst with a couple CP-related sections, and I also gave credit, where credit was due.

[idle owl]

@prime flower That's up to you, I suppose. Are we prepared to support it as it is and so on? You're welcome to add it, but keep in mind what that means. More people will find it, but... more people will find it.

[onyx hinge]

@lapis hemlock thanks, we link back to ulab official documentation from ours as well.

[tulip sleet]

@timber mango Try two things:
Jumper the 3V to the SDA and SCL pins (with the LIS3DH not connected), and try:

import board
i2c = board.I2C()

This should work because it simulates pullups on the pins.
If that works, try wiring the LIS3DH SCL and SDA jumpers to the TX and RX pins, with the LIS3DH powered, and try:

import board, busio
i2c = busio.I2C(board.TX, board.RX)

[prime flower]

@idle owl Yep - you're right. I'll hold off on adding it to the bundle until I merge in DNS work or people will be confused why it's failing.

Thanks!

[idle owl]

You're welcome

[lapis hemlock]

@onyx hinge I think I have fixed the mike. We can test it sometime, but I am pretty sure it is OK. It's not urgent.

[onyx hinge]

@lapis hemlock If you would like to change the way we attribute ulab please let me know what you'd like to see

right now what I have is

//| ulab is adapted from micropython-ulab, and the original project's
//| documentation can be found at
//| https://micropython-ulab.readthedocs.io/en/latest/

[timber mango]

@tulip sleet The ItsyBitsy M4 and LIS3DH are isolated now, and everything is using alternate breadboard pins. Same error. Also when wiring to the female headers on the ItsyBitsy M4.

[lapis hemlock]

@lapis hemlock If you would like to change the way we attribute ulab please let me know what you'd like to see
@onyx hinge Oh, I am not picky at all, anything is fine. I have to say, you (adafruit) have gone out on your way, when acknowledging my contribution, so it is really OK, whatever you decide.

[tulip sleet]

@tulip sleet just to double-check, does your REPL prompt say "ItsyBitsy M4", and not some other board? And it's 5.0.0-beta.5?

[timber mango]

@tulip sleet Adafruit CircuitPython 5.0.0-beta.5 on 2020-02-05; Adafruit ItsyBitsy M4 Express with samd51g19.

[tulip sleet]

are you trying the pin stuff above?

[timber mango]

@tulip sleet 1) I have isolated the ItsyBitsy M4 and LIS3DH, 2) tried alternate breadboard pins for everything, and then 3) wired to the female headers of the ItsyBitsy M4, all in separate steps.

[tulip sleet]

did you try the tests I wrote above at xx:30 (not sure of your timezone)?

[orchid basin]

jepler

[adafruit/circuitpython-org] Pull request opened: #411 Add the SAMD51 Thing Plus

[timber mango]

@tulip sleet 1) import board i2c = board.I2C() Traceback (most recent call last): File "<stdin>", line 1, in <module> RuntimeError: SDA or SCL needs a pull up
2) ```>>> import board, busio

i2c = busio.I2C(board.TX, board.RX)```

  File "<stdin>", line 1, in <module>
RuntimeError: SDA or SCL needs a pull up```

[tulip sleet]

@timber mango so 1) means the pins on the Itsy are OK, they are not damaged. 2) looks like there is some issue with the connections to the LIS3DH or the board itself, since even with alternate pins it is not working. (I'm assuming it's fully wired up)

in 2), you did wire the LIS3DH, right?

[main meteor]

If you're not using a LIS3DH breakout with I2C pullups, you'll need to provide the pullups for it to work.

[timber mango]

@tulip sleet It is now. 😜

[tulip sleet]

@main meteor it is the adafruit breakout, which has pullups. it WAS working, and stopped

[timber mango]

@main meteor All of the Adafruit breakouts are supposed to already have pull ups on them.

[tulip sleet]

"it is now" wired, or it is now working??

[solar whale]

if test 1) failed above with 3V to SCL/SDA, then there is a problem with the ItsyBitsy

[timber mango]

@tulip sleet As in, the LIS3DH is wired now but still not working.

[tulip sleet]

oh, 1) failed, i see, that otuput was edited. Yes, @solar whale is right, there's something wrong with the itsy.

[timber mango]

@solar whale Everything has failed so far.

[tulip sleet]

do you have another board to try?

[solar whale]

do you have another MCU?

[timber mango]

@tulip sleet @solar whale I do not have anything else I can wire in place of the ItsyBitsy M4. I do have two Feather M4 Express boards, but they are in other circuits, and are not drop-in replacements for the ItsyBitsy M4. To use one of them, I would have to rewire my whole circuit.

[tidal kiln]

@timber mango are any of the hairy things conductive? where did the solder blob on the jumper pad come from (circle)?

[orchid basin]

makermelissa

[adafruit/circuitpython-org] Pull request review submitted: #411 Add the SAMD51 Thing Plus

Hi, thanks for adding. I see a couple things that could be changed.

Can you remove the Stemma QT from features? We're trying to limit the list size to what's in this guide: https://learn.adafruit.com/how-to-add-a-new-board-to-the-circuitpython-org-website/adding-to-downloads.

For the images, can you make it so they are a 13:10 ratio? The sizes I tend to target are:
Large: 780x600
Small: 293x225

Otherwise they look quite different from the other images on the page. Thanks

[timber mango]

I really do not like removing and reseating MCU breakouts any more than absolutely necessary.

[slender iron]

@onyx hinge @lapis hemlock jimmo's idea of checking that it's a module is a good idea. I don't think we need to wait from upstream since they've redone how they handle weak modules

@tulip sleet need help with rc.0? I saw the branch go by

[timber mango]

As for test equipment, I can not even afford a DMM from Adafruit.

[tulip sleet]

@slender iron I'm working the release notes; about 80% done

[slender iron]

💯

[tulip sleet]

@timber mango do you have harbor freight?

[slender iron]

@prime flower I don't think the bundle includes releases marked as pre-release

I tend to release a <1.0.0 version as stable

you should have a numeric designation after alpha as well 1.0.0-alpha.0 for example

[tulip sleet]

https://www.harborfreight.com/7-function-digital-multimeter-63759.html

$6.39

[timber mango]

@tulip sleet It would not matter if I did have Harbor Freight. I would still not be able to afford any test equipment. I was lucky just to be able to get my Hakko soldering station.

[tulip sleet]

you can use an led and resistor as a simple HIGH/LOW indicator. So see if on the LIS3DH, the 3Vo pin is high, which means the board is powered and the regulator is working. The led prob won't work on the pullups; the current may be too low. But you could use an analog input pin on the Itsy to measure the voltage on the SCL and SDA pins on the LIS3DH.

[solar whale]

@timber mango can you try test 1) again - and take a picture of the setup? I'm hoping there is just some setup problem.

[timber mango]

@tulip sleet I had not thought of using an analog pin for that. I know about using an LED for high/low checks but thought you wanted an actual voltage reading.

[tulip sleet]

you could try a bare led without resistor on the SCL and SDA pins, but the pullups are 10k, so you may not illuminate the LED

[timber mango]

@tulip sleet I saw the solder blob, but it was not causing a solder bridge so I left it alone.

[tulip sleet]

the fact that if you wire the SDA and SCL pins to 3v, board.I2C() gets a pullup error is fishy. it's like those pins have been damaged, you could try the same thing to the TX and RX pins as i mentioned above

if that doesn't work either, there's something strange going on. I wouldn't expect all the pins to go bad suddenly.

[fallen anvil]

Folks, If someone has the chance, would they mind running the following snippet (or near equivalent for your pinning) on a non-imxrt CP board with a single I2C device present?...trying to figure out if I've got a platform or port problem here;

import board
import busio
import time

time.sleep(1)
i = busio.I2C(board.EXP_P28,board.EXP_P27)

i.try_lock()
j = 0
while True:
	j=j+1
	print(j, i.scan())
	

[tulip sleet]

@fallen anvil

Adafruit CircuitPython 5.0.0-beta.5 on 2020-02-05; Adafruit ItsyBitsy M4 Express with samd51g19
>>> import mubes
1 [68]
2 [68]
3 [68]
4 [68]
5 [68]
6 [68]
7 [68]
8 [68]
9 [68]
...

[prime flower]

@slender iron thx for the info. I'll remove this release and release a 1.0.0-alpha.0 instead

[fallen anvil]

@tulip sleet thanks.... I get to about 1800 before it falls over. With no devices connected I'm uip to 60K so far...

[solar whale]

@fallen anvil -- aslo works for me on bybadge

ah -- will let it run

[fallen anvil]

yeah, I'm afraid I like my punishment tests :-)

[tulip sleet]

using an SHT31D

13k and running

[fallen anvil]

So long as it's not imxrt then it's partitioning the problem.

Grr, as I suspects, it's gonna be driver :-(

[tulip sleet]

SHT31D I2C device

maybe clock speed?

[fallen anvil]

ah, sorry! I've got a MPU-65 thing here but I don't think that makes any difference

could be anything, but if its driver I can't hand it off to anyone else 😁

Dinnertime, then back into battle. Currently at 110K with no devices connected.

[solar whale]

@fallen anvil running on pybadge here -- > 6000 and running

[fallen anvil]

drat. Thanks folks, appreciate the help!

I'm pretty sure it would have gone AWOL by 6K

[timber mango]

@tulip sleet 1) ```python

import board
i2c = board.I2C() Traceback (most recent call last):
File "<stdin>", line 1, in <module>
RuntimeError: SDA or SCL needs a pull up python >>>
import board, busio
i2c = busio.I2C(board.TX, board.RX)``` NO ERRORS!!!!

[tulip sleet]

ok, so there's something wrong with one or both of those pins, whether it's soldering or electrical, we don't know.

you could hook up an LED or resistor to SCL and SDA one at a time, and use (in the REPL)

import board, digitalio
scl = digitalio.DigitalInOut(board.SCL)
scl.switch_to_output()
scl.value = True
scl.value = False

and see if you can toggle the pin, and do the same for SDA

or feed SCL back into an analog input pin and see if it goes up and down

[timber mango]

@tulip sleet Because board.I2C() does not work and busio.I2C(board.TX, board.RX)does work?

[tulip sleet]

board.I2C() is basically equivalent to busio.I2C(board.SCL, board.SDA)

it is the "default" I2C object, using the default pins

[timber mango]

@tulip sleet If both use the same pins, why does one not work and the other works fine??

[tulip sleet]

they're not the same pins: SCL/SDA, vs TX/RX

[manic glacier]

tannewt

[adafruit/circuitpython] New comment on pull request #2657: Support importing native modules in native packages\.

Thanks @jimmo and @dpgeorge. Will keep an eye on upstream and plan on replacing this PR next time we merge from upstream.

@jimmo Good idea on the type check. I've added it and confirmed we can't import microcontroller.cpu which is not a module. My preference is to keep the set_attr because I'm assuming a module will track whether it's members have been imported in the future rather than having import manage it.

[fallen anvil]

Oh great, failed at 800K...this is going to be a fun one to find 😬 Definately dinnertime.

[timber mango]

@tulip sleet I have made an error, and am untangling it now. Please sit by.

[tulip sleet]

np, i am doing other things 🙂

[timber mango]

@tulip sleet TX/RX is for board.I2C() and SDA/SCL is for busio(board.SDA, board.SCL), right??

[tulip sleet]

no, board.I2C() is similar to busio(board.SCL, board.SDA). It use the SCL and SDA pins (in that order), and it creates busio.I2C() object, but only once. So the first time you call board.I2C() it creates the busio.I2C() object, and then it returns the same one over and over (it's a singleton).

[manic glacier]

tannewt

[adafruit/circuitpython] New comment on pull request #2652: Creat ar\.po

Thank you for the additional translations! Here is an updated view. I tried adding the filename into the output translation but it still messes up the direction of the Arabic portion.

Do you know how the mixing of directions is supposed to work? Are there unicode characters to designate direction that we may be dropping? Maybe @joeycastillo knows.

[tulip sleet]

TX/RX are normally for UART, but they will also work for I2C as alternate pins (and so will various other pins). I was just suggesting them to test whether the SCL and SDA pins were working or not.

[timber mango]

@tulip sleet So, TX/RX is really the same as SCL/SDA and can be used interchangeably?

Well, different pin pairs, but same usage.

I believe I am back on track now. 🙂

@tulip sleet There should be a notice in learn guides or in some other place of the various pin pairs that can be used in place of SDA and SCL.

[idle owl]

There is.

CircuitPython Essentials guide - I2C page.

There's even a script to find the pins.

https://learn.adafruit.com/circuitpython-essentials/circuitpython-i2c#wheres-my-i2c-15-17

[timber mango]

@idle owl ACK! I did read that guide, but apparently not well enough. 😦

[orchid basin]

aidancuckoo

[adafruit/circuitpython-org] New comment on issue #410: Misleading photo for Arduino Nano 33 BLE

@tannewt that's what I was thinking! As for the LPS22, the description for the Fritzing part which was released says that it uses an LPS25 and not an LPS22. Maybe they're code compatible and @ladyada I noticed that they both use the same PCB too! So it's probably 90%/100% compatible with CircuitPython.

tannewt

[adafruit/circuitpython-org] New comment on issue #410: Misleading photo for Arduino Nano 33 BLE

@aidancuckoo Would you like to add a PR to CircuitPython for it? There is a guide on how to do it here: https://learn.adafruit.com/how-to-add-a-new-board-to-circuitpython

[tulip sleet]

i am just suggesting using those pins only because the others appear to be broken. So the test is whether the alternate pins work. I'm trying divide and conquer on the problem

in ordinary circumstances, there would be no reason to choose other pins

[timber mango]

@idle owl Thank you. I will take that dunce cap and go to the corner now. 😉 😎

@tulip sleet Have I helped any at all?

[tulip sleet]

you said that when you wired 3V to both SCL and SDA, and also TX and RX, that both still gave you pullup errors:

>>> import board, busio
>>> i2c1 = busio.I2C(board.SCL, board.SDA)
### Does the above give an error when both pins are jumpered to 3v?
>>> i2c2 = busio.I2C(board.TX, board.RX)
### Does the above give an error when both pins are jumpered to 3v?

[timber mango]

@tulip sleet I never wired 3V to TX and RX. At that time I did not fully understand what was going on. I understand it all now.

Yes, that gave an error.

[tulip sleet]

ok, got it, I was giving explicit instructions for testing

[timber mango]

I will have to try doing that with TX/RX to verify it.

[tulip sleet]

try it again with SCL and SDA, and then try it with TX and RX

[manic glacier]

joeycastillo

[adafruit/circuitpython] New comment on pull request #2652: Creat ar\.po

This might be an issue with iTerm's RTL support... displaying ar.po in the macOS terminal on the left and iTerm on the right, only iTerm seems to have the directionality issues:

<img width="582" alt="Screen Shot 2020-02-26 at 2 13 34 PM" src="https://user-images.githubusercontent.com/1765152/75378895-848cfd00-58a2-11ea-8d38-7d8af7e22274.png">

[timber mango]

Yes, you were. My brain is strange sometimes and loses data right after it absorbs it. I can fully understand something one minute and lose it all the next minute. This is extremely frustrating and has gotten me in trouble with people who do not understand many times.

[orchid basin]

jepler

[adafruit/circuitpython-org] New comment on pull request #411: Add the SAMD51 Thing Plus

Images are now:

large/sparkfun_samd51_thing_plus.jpg JPEG 780x600 780x600+0+0 8-bit sRGB 181514B 0.000u 0:00.000
original/sparkfun_samd51_thing_plus.jpg JPEG 600x600 600x600+0+0 8-bit sRGB 198979B 0.000u 0:00.000
small/sparkfun_samd51_thing_plus.jpg JPEG 293x225 293x225+0+0 8-bit sRGB 61349B 0.000u 0:00.000

[timber mango]

I think I may have a memory leak that needs to be debugged. 🙂

Bad people who I did not know or think were bad at the time have used this against me many times. Now, the whole Circuitpython world knows. 😦

[tulip sleet]

we will not take advantage of you 🙂

[timber mango]

@tulip sleet I want to believe that, but I have heard it before. I have trust issues because of that and now I have to be on guard again. 😦

[tulip sleet]

@slender iron ready to go on rc0. if you want to review the release notes, fine, else I will just go.

[timber mango]

@idle owl I have saved that I2C pin finder script so I will always have it. That is a great tool!

[slender iron]

@tulip sleet all good. I’m head out on a run

[manic glacier]

dhalbert

[adafruit/circuitpython] New release published: 5\.0\.0\-rc\.0

dhalbert

[adafruit/circuitpython] New tag created: 5\.0\.0\-rc\.0

[timber mango]

I always have my pinouts handy when I am working on projects

[orchid basin]

adafruit-blinka

[adafruit/circuitpython-org] Pull request opened: #412 Automated website update for release 5\.0\.0\-rc\.0

Automated website update for release 5.0.0-rc.0 by Blinka.

New boards:

• arduino_nano_33_iot
• pycubed
• circuitbrains_basic_m0
• circuitbrains_deluxe_m4
• sparkfun_samd51_thing_plus
• espruino_wifi
• stm32f4_discovery
• espruino_pico
• feather_m7_1011

[lapis hemlock]

@onyx hinge @lapis hemlock jimmo's idea of checking that it's a module is a good idea. I don't think we need to wait from upstream since they've redone how they handle weak modules
@slender iron @onyx hinge Do we have to change anything on my side, or does this work out of the box?

[half sedge]

Is this the time where I need to download the new version for all of my board?

[tulip sleet]

@half sedge if you want to help test, sure. It's still building. When all boards have finished building, I'll push the changes to circuitpython.org

[onyx hinge]

@lapis hemlock I don't think so

[lapis hemlock]

@lapis hemlock I don't think so
@onyx hinge You don't think that we have to change anything, or you don't think that it works out of the box?

[onyx hinge]

I think you won't have to change anything. At least, I hope not.

[manic glacier]

mesa177

[adafruit/circuitpython] New comment on pull request #2652: Creat ar\.po

@joeycastillo is right, and it seems to be a common issue:

https://gitlab.com/gnachman/iterm2/issues/1611
https://gitlab.com/gnachman/iterm2/issues/2686

Someone on the forum started to do something about it about 5 months ago, but it seems they haven't gotten far with iTerm / iTerm2:

https://gist.github.com/XVilka/a0e49e1c65370ba11c17
https://gitlab.gnome.org/GNOME/vte/issues/53
https://gitlab.com/gnachman/iterm2/-/milestones/12

Not sure how to move forward...

[lapis hemlock]

I think you won't have to change anything. At least, I hope not.
@onyx hinge Fantastic! I can then iron out a couple of internal implementation issues. I need a couple of days, but I will keep you posted on the progress.

[half sedge]

@tulip sleet Ok, thanks. Maybe make some noise here then. Maybe the announcement should only be done when it is downloadable. Someone seeing the following tweet could jump, try to download and then be disapointed: https://twitter.com/CircuitPython/status/1232752951422341120?s=20

CircuitPython (@CircuitPython)

I see there is a new release of CircuitPython on https://t.co/gyFGfqI4Sl - check it out!

Twitter

[tulip sleet]

that tweet is automatic; it jumps the gun a bit

we should make it contingent on something other than the github release

[half sedge]

All those automation (for some reason, I cannot use "D a m n" when writing here, as if I knew what word are offensive or not).

[tulip sleet]

yep, please don't use even mild explietives

[indigo wedge]

...
Feather M4 1011. Thanks @arturo182#1355.```

M7 not M4, M4 is sooo last year 😉

[tulip sleet]

i'll fix 🙂

at least I didn't spell your handle wrong again

[indigo wedge]

😄

if someone is updating the M7 1011 circuitpython.org entry, you can use this image or wait a day or two until the boards reach Adafruit and then I'm guessing you'll have more pr0 photos 🙂 https://pbs.twimg.com/media/ERAQFsPW4AgvItW.jpg

Nice that we got the spi fix for the RC 🙂

[tulip sleet]

i didn't expect that but it was good to get in

[half sedge]

This language police is so bizarre. I had to google "mild explietives" to figure out. There must be a cultural thing together with a religious thing. It could be very US centric. I must have learn some of that vocabulary by watching rated movies in original language, or is it something I got from English speaker coming from the UK.

[bright aspen]

I2S MCK: Though I created the issue, I don't think I will be able to work on getting the master clock output for I2S, digital audio. Well, not soon. (I still need MCK for my audiometer design, but I'm likely to go to C on a SAMD/E—wish I had time to get up to speed with Rust. I very much wanted do this in CircuitPython, but factors such as USB interface control, 1.8 V peripherals, and MCK for DAC are influencing me.)

[tulip sleet]

@half sedge we have a number of younger members here, and rather than try to draw a line between "mild" and "not so mild" expletives, we simply don't allow them

[half sedge]

One technical issue is that if you write a long message with one of those word, you loose it all.

[main meteor]

On another server, they modified the bot to replace (suspected) expletives with REDACTED so the gist was intact.

[tulip sleet]

that's an interesting idea

[main meteor]

On the upside, the message wasn't lost, and you could figure out which term(s) the bot objected to. On the downside, the gist could still be offensive and people might be less careful since they wouldn't lose their typing.

[solar whale]

@half sedge just for your information, that particular word, while very commonly used, is considered very offensive to some. I'm not justifying or agreeing with that sentiment, but I have witnessed an ugly scene based upon its utterance.....

[fallen anvil]

@indigo wedge are those M7 boards available on this side of the pond or do they do a round trip?

[bright aspen]

I REDACTED agree that some REDACTED text would be REDACTED offensive, and some REDACTED individuals would play it, REDACTED, and make a big deal of having "REDACTED" come up.

[stuck elbow]

5.0.0rc0 seems to be working well for me

[ionic elk]

@tulip sleet is there a convenient way to be notified of ALL changes to a section of a github project? For instance, someone changed up the STM32 port to revise ticks_ms and it took me way longer than it should have to figure out why nothing was working

[indigo wedge]

@fallen anvil I assembled them here in Sweden and should have 2 more bare boards so one assembled one could find its way to you if you want

[ionic elk]

I'd be super nice if there was some way of rigging a big bright notification of any commits to the stm32f4 directory that aren't mine

[tulip sleet]

you can just subscribe to the repo and get everything

[indigo wedge]

everything is a lot 😅

[ionic elk]

hmm but then stm32 changes just get lost in the traffic

I just need to know when people make breaking changes to the stm32 port

[indigo wedge]

same for me with imx 😄

[ionic elk]

so it doesn't bork all of my workflow

Like it's not my code, it's everyone's and I love that people are contributing but it's like coming back to your homework and finding that someone changed every x to a z

[indigo wedge]

I think that's fair, as a maintainer of that port you have a plan in your head 😄

[fallen anvil]

@indigo wedge that would be cool. I'll happily pay for it and shipping....just save a bit of import cost and perhaps a bit of tree while we're still nominally in the EU trading block. PM me when you're ready to rock.

[ionic elk]

ticks_ms isn't even a thing anymore, and I don't know when that change was made or who made it, I don't have any notes on what to replace it with, it's kind of crazy

[tulip sleet]

so you are not watching everything in circuitpython

you could have searches for comments with keywords, but no, you just have to skim everyting

[indigo wedge]

@fallen anvil cool, it takes about 2h to assemble, i can do that next week

[fallen anvil]

Ouch...if you're manual assembling then don't worry about it. Are they going to become a production item? There aren't too many 1011 boards around...

[tulip sleet]

@ionic elk now there's supervisor_ticks_ms64(), etc., is that what you mean has chagned?

[ionic elk]

I feel like there's got to be a tool out there for this... I mean we have hundreds of lines changed per day, and I won't even understand what's happening in most of them. I'll see if somebody has put together some kind of git-alerts thingy I could run when I fetch upstream

right, ticks_ms64() - how long ago did that happen?

[tulip sleet]

months ago

[ionic elk]

Oof yeah