#circuitpython-dev

1 messages Β· Page 398 of 1

idle owl
#

I think...?

#

No

#

That wouldn't work.

#

One set is a string, the other is not

#

Nevermind that.

idle wharf
#

The name or key of the feed is a string.

#

I don't know if it is simpler...

But if you wrote a setup_feed(feed_name) function and called it for each feed, it would read better.

#

And a send_io_data(key, value) function

idle owl
#

Hmm ok

idle wharf
#

But that's a very "style" \ personal choice

idle owl
#

No, I'm with you, I like it.

#

Now if I can figure out how to do it. πŸ˜„

idle wharf
#

I'm looking through some stuff I've writen...

#

I was using MQTT

#

So nothing to cut n paste ...

idle owl
#

Ok, no worries, thanks

idle wharf
#

If it helps.... I've seen you build more complicated things. You'll totally get it.

idle owl
#

Heh.... thanks. πŸ˜„

#

@idle wharf ```py
def setup_feed(feed_name):
try:
# Get the temperature feed from Adafruit IO
return io.get_feed(feed_name)
except AdafruitIO_RequestError:
# If no temperature feed exists, create one
return io.create_new_feed(feed_name)

def send_io_data(feed_name, value):
return io.send_data(feed_name["key"], value)

#

Or am I wrong on what you had in mind for the second one?

#

The first one definitely simplified things. And it worked, I renamed all the feeds, and it generated all new ones.

#

I'm a little confused about needing the second one because the io.send_data lines are all only one line themselves.

#

Is it simply to clean it up a bit?

idle wharf
#

I’m away from keyboard back in a bit.

idle owl
#

Ok!

#

This doesn't seem to like to come out of the REPL cleanly.

#

it hangs until I reset it.

#

I keep dumping it into the REPL to have it stop sending while I iterate.

#

Hangs, and no keyboard interrupting works.

#

Sometimes.

#

Oh!

idle wharf
#

back

idle owl
#

Hello

#

I did just figure out one thing that I think you may have been getting at.

#

Use the setup_feed() line in the io.send_data() line. Don't set the feed up separately. Or is that a bad idea?

#

It works. But that doesn't make it a good idea.

idle wharf
#

I would setup feeds once not every time you try to send

idle owl
#

Ok

#

I'm not sure how to do that with the function then.... if I set the function = to a variable name, isn't it running the function every time?

#

Every time I call the variable

idle wharf
#

You can put the setup outside your While True and send inside the loop;

idle owl
#

There's no loop because it's using deep sleep.

#

So it runs one time every 5 minutes.

idle wharf
#

I don't understand your qustion here "if I set the function = to a variable name, isn't it running the function every time?"

idle owl
#

Ok so I have the function I pasted above.

#

Then I use it here ```py
bme280_temperature_feed = setup_feed("bme280-temperature")

#

Then if I do io.send_data(bme280_temp_feed, temperature), isn't it setting it up again on every send?

idle wharf
#

no

tulip sleet
#

Fixed - thanks

idle wharf
#

bme280_temperature_feed is the return from setup_feed which is your feed you created

idle owl
#

Ahh.

#

But io.send_data(setup_feed(feed_name), temperature) is a bad idea?

#

As in using the function directly in the send line

idle wharf
#

It just doesn't seem necessary, because you only need to setup feed once.

tulip sleet
#

Right, there's an implied "loop", because the program restarts completely.

idle owl
#

It tries to set it up every time it runs, every 5 minutes though. That was the discussion earlier with regards to maybe telling folks to set it up manually in the UI.

idle owl
idle wharf
#

I think the python answer is if you only use something once, you don't need to assign it and then use it.

idle owl
#

Ok

idle wharf
#

But in your case, you're creating it and using it over and over

idle owl
#

So give it a name, and use the name instead? Even though the whole program is what is "looping", imports and all?

#

I think I understand... it's the concept of being Pythonic when deep sleep is a bit wonky as a Python concept?

idle wharf
#

I've not used Deep Sleep ...

I didn't know the entire file would run each time.

idle owl
#

I didn't know until today πŸ˜„

onyx hinge
# idle owl <@!624005998696333331> ```py def setup_feed(feed_name): try: # Get t... 
def send_io_data(feed_name, value):
    return io.send_data(feed_name["key"], value)
``` in this snippet I wonder if there's a better thing to call the parameter than `feed_name` , because it's not the name of a feed, but (I think) something returned from setup_feed, which is probably some kind of feed object.  Could the parameter just be called `feed`?
tulip sleet
#

it's a loop the same way that pressing the reset button over and over is a "loop"

idle owl
idle owl
onyx hinge
#

okay, thanks for accepting the nit-pick

idle wharf
#

Good catch @onyx hinge ! 100% right.

... thinking about the file running every time...

idle owl
idle wharf
#

I guess in that case my general thoughts aren't as helpful. I was thing there was some "run once" and then a loop.

tulip sleet
idle owl
idle owl
tulip sleet
#

that is not in flash, it's in a special place that is preserved during sleep. So wont' wear out flash.

#

but it's accessed the waynvm is.

idle owl
#

Ah

tulip sleet
#

they share some code internally

idle owl
#

Hmm. Not sure it's worth it.

#

I don't think it's doing damage trying to generate the feed every time, if it's not there.

#

It checks the sensor value before doing anything else, so it's not like it matters that the MCU is doing a little more later.

tulip sleet
#

it's harder to store an object in sleep memory, it's more like for storing counts, or state, or something

idle owl
#

yeah nah.

crimson ferry
#

but alarm.sleep_memory only survives reloads, not resets as in deep sleep, right?

idle owl
#

No it persists during deep sleep.

#

Power loss wipes it.

#

According to the docs (I had up right now)

crimson ferry
#

oh, cool, I thought deep-sleep was more like a reset than a reload, that's handy

idle owl
crimson ferry
#

I guess that's why it's alarm sleep_memory, d'oh

idle owl
#

I don't think it's causing issues to do what I'm doing in this code.

#

Brent would know best, but he's not around the rest of the week.

#

Not sure who else knows AIO that well. Mark said earlier he's done a lot, but Brent gets the backend and could tell me if doing this every 5-10 minutes is a Bad Idea.

idle wharf
#

Is there an API Call limit ... or only for sends (looking)

idle owl
#

I want to say it's no biggie. But I have no idea what I'm talking about.

#

The eventual wait time will be 10 minutes in this example.

#

And I was reloading the code a ton eariler while iterating and never hit an error....

#

So if there is, it must be higher than what I'm doing.

idle wharf
idle owl
#

Wow yeah.

#

No way I hit that.

#

This stuff is pretty neat. Not sure why I never did anything with it before.

#

Daunting I guess.

idle wharf
#

I think all the LED stuff you do is daunting

#

but now I'm curious about sleep and how to run once or if you can ...

idle owl
#

For a second there I thought I had reproduced the weird hanging thing.

#

But no, it isn't happening again now with the same actions.

idle wharf
#

When it works... I try not to question it.

idle owl
#

It's that it keeps happening randomly is what's irritating.

#

Anyway, I now have a working thing to leave running overnight. And it's time of late here.

#

Thanks for your help! @idle wharf

idle wharf
#

Glad to help... have a great night

proven garnet
#

Catching up from the day, let me know if you want some extra help applying patches or doing any clean up πŸ™‚

#

Also, couldn't listen in on the meeting live today, but whenever someone gets a chance, could I get the CircuitPythonista role? Want to stay on top of meeting times with holidays creeping up dreidel_parrot

solar whale
proven garnet
manic glacierBOT
manic glacierBOT
manic glacierBOT
#
jepler
[adafruit/circuitpython] New comment on issue #5418: RP2040 hang when running specific code\.py with an exception

When an RGBMatrix is initially constructed, its framebuffer (as well as other data used to do the low level "scan the matrix out" activity) is placed in the GC heap. Because displays are 'kept' over soft resets, there is a moment when it is necessary to destroy the RGBMatrix and construct a fresh one, placing the framebuffer in supervisor storage instead. This is what the common_hal_rgbmatrix_rgbmatrix_reconstruct function does when it's called during the interpreter teardown process. (the...

idle owl
idle owl
#

@onyx hinge Do you understand ADC values in CircuitPython, and potentially the non-linear values returned by the ESP32-S2/nRF52840?

onyx hinge
#

I mean, who really understands anything

idle owl
#

I'm trying to figure out how to explain the 51375 max value on the ESP32-S2. It's not the full 16 bit range anymore. So the wording has to be changed.

idle owl
onyx hinge
#

In a perfect ADC, an integer output of 0 indicates 0V and an integer value of 65535 means it's the full "reference voltage"

#

but they're all imperfect

#

the ESP32-S2 is a lot more imperfect than the others

idle owl
#

In this case it's that above that voltage range, it's non-linear

onyx hinge
#

what's the starting (guide?) text we're trying to make it better?

ornate breach
#

Quick question, did Circuitpython pave the way for SAMD support in MicroPython?

idle owl
#

so they stopped it answering

#

at that level.

#

Hold on, I don't think you can preview templates.

#

What do these values mean? In CircuitPython ADC values are put into the range of 16-bit unsigned values. This means the possible values you’ll read from the ADC fall within the range of 0 to 65535 (or 2^16 - 1). When you twist the potentiometer knob to be near ground, or as far to the left as possible, you see a value close to zero. When you twist it as far to the right as possible, near 3.3 volts, you see a value close to 65535. You’re seeing almost the full range of 16-bit values!

onyx hinge
#

@ornate breach I didn't look at the micropython code that was recently added so I don't know if it's literally "derived from" the code in CircuitPython or not.

ornate breach
#

Some of the stuff seemed really similar from what I looked at

idle owl
#

2.58698V is the ESP32-S2 max.

#

Because above that it's "non-linear"

#

so something something non-standard AREF something something

#

(which is to say, no idea how to explain this)

ornate breach
#

Obviously there is MicroPython optimizations and features

onyx hinge
#

I wouldn't put so many digits in the number

idle owl
#

That's just what I grabbed from the serial console

#

not in the guide

#

Though the original template has a serial console screenshot with that many in it

#

Β―_(ツ)_/Β―

#

More even.

#

Could fix that in code I guess.

#

Anyway, I have to make this template copy more dynamic to allow for non-standard numbers, which is the easy part

#

the hard part is changing that paragraph above to make sense

#

with this.

onyx hinge
#

What do these values mean? In CircuitPython ADC values are put into the range of 16-bit unsigned values. This means the possible values you’ll read from the ADC fall within the range of 0 to 65535 (or 2^16 - 1). When you twist the potentiometer knob to be near ground, or as far to the left as possible, you see a value close to zero.
This much is still true everywhere

idle owl
#

Ok

onyx hinge
#

When you twist it as far to the right as possible, ...
This is where things get murky

idle owl
#

Make the rest follow the numbers and not say "you're seeing the full range"?

onyx hinge
#

When you twist it to the right, the value gets bigger up to some value that is dependent on the microcontroller. Many microcontrollers get a value very close to 65535. Some, such as ESP32-S2, have a smaller limit of about 50,000 or 2.5v.

idle owl
#

Oooh.

onyx hinge
idle owl
#

I mean, Limor kind of explained it

#

said we documented it somewhere

onyx hinge
#

the limits on esp32-s2 are a bit more severe than they were on the esp-32

idle owl
#

nRF is the same apparently

onyx hinge
#

I wanted to improve the docs but I didn't come up with specific text for it

idle owl
#

ah hmm.

onyx hinge
#

I appreciate the humor in this image

stuck elbow
#

analog circuitry is hard

idle owl
#

Whee ok template updated I think.

lone axle
#

@idle owl updating the rtd config file looks to be working correctly on a small subset and the whole bundle. I think I'm ready to add the final step of pushing the changes afterward. I'll do the small subset again first when I get that added.

Should I push directly to main or make a PR for each? If push directly, do you have a preference on the commit message to use?

I'm also going to make a releases on pcf8563 and ticks before I do it so those get updated as well (they were skipped my test runs so far due to the submodule being behind main slightly)

manic glacierBOT
#
jepler
[adafruit/circuitpython] Pull request opened: #5611 Don't use reset reason to skip writing boot\_out\.txt

As @neradoc and @anecdata noted, we can't depend on the reset reason.

Instead, just rely on the existing safe mode logic (the write already is skipped anytime boot.py was skipped), and otherwise potentially write boot_out.txt anytime it differs from the version on disk (or if the file doesn't exist in the first place)

Testing performed: On qtpy m0 haxpress, removed boot_out.txt, projectly ejected, then clicked reset button. verified boot_out.txt was recreated.

Closes #5588

#
b-blake
[adafruit/circuitpython] Issue opened: #5612 CP 7\.1\.0 RP2040 Starts automatically after Ctrl\-C

CircuitPython version

Adafruit CircuitPython 7.1.0-beta.0-8-g28b5733bd on 2021-11-12; Adafruit Feather RP2040 with rp2040
Board ID:adafruit_feather_rp2040

Code/REPL

# SPDX-FileCopyrightText: 2021 ladyada for Adafruit Industries
# SPDX-License-Identifier: MIT
# adafruit_circuitpython_adafruitio usage with an esp32spi_socket
import board
import busio
import displayio
from   digitalio import DigitalInOut, Direction, Pull
import gc
import microcontroll...
idle owl
manic glacierBOT
tidal kiln
#

hey. have question on new CP 7.x RGB status blinks. who's in the know of that?

slender iron
#

I know

tidal kiln
#

the blue blinks are BLE builds only?

slender iron
#

yes

tidal kiln
slender iron
#

that's an older version of cp

tidal kiln
#

pre 7.x?

slender iron
#

that's the blue in the line number blink

#

yup

manic glacierBOT
tidal kiln
#

so the yellow would be "other error"

#

or is that safe mode?

slender iron
#

the yellow solid is the exception and the blink is part of the line number

tidal kiln
#

ah. good eye. i was seeing the steady as just the first blink.

#

so "other error"

#

line 430

idle owl
#

@tidal kiln Can I borrow your brain for a minute? Or probably a few.

tidal kiln
#

urgent?

idle owl
#

No

tidal kiln
#

can i ping you in a bit?

idle owl
#

Yes

#

Thank you

tidal kiln
#

will do. want to get through some forum stuff first.

idle owl
#

No worries.

tulip sleet
tidal kiln
#

@tulip sleet thanks. i think it's 6.x blinks in this case.

tulip sleet
#

those are further down that page, maybe you know that already

blissful pollen
#

@idle owl If you need to review any of the IO stuff still I'll have some time later today. Home from a medical appointment and kinda loopy from some stuff they gave me but should be good πŸ™‚

idle owl
tidal kiln
#

@tulip sleet yep. thanks. i think sorted for now.

#

@slender iron also thanks!

idle owl
#

It is not commented yet. But it probably is all familiar to you anyway.

blissful pollen
#

Sure, what's the basic purpose of the program? Read some values, save them to IO and sleep?

idle owl
slender iron
#

np @tidal kiln

blissful pollen
idle owl
#

Brilliant, thanks!

slender iron
#

@tulip sleet or @onyx hinge: will one of you do my rpi review?

tulip sleet
#

I started on it last night but haven't finished yet

slender iron
#

thanks! just wanted to make sure someone was looking at it

manic glacierBOT
idle owl
#

Hmm.... My Feather crashed.

#

Red NeoPixel blink intermittently, no data being sent.

#

Plugged it in, and now it's fine. So I never got to see the error.

#

It didn't register that it had crashed initially because the little red LED is on when data is being sent, so it has an "intermittent red LED" when it's running and when it crashes. πŸ˜„

lone axle
manic glacierBOT
idle owl
lone axle
#

Yep, you're welcome. I'm also going to include a few seconds of sleep in each one to hopefully avoid any github rate limits. So it will take a few minutes to get through everything.

idle owl
blissful pollen
#

@idle owl Code looks good to me and similar to the code I use in my weather monitor. I was using the HTTP API too though the python library is limited in how much of the API it uses (I modified it for my use)

idle owl
blissful pollen
#

I did use a try/except block around the whole thing as I wanted mine to run forever, and if for some reason it crashes (and with wifi it will happen) then it just went to deep sleep for 30 seconds and retried

idle owl
#

Yeah, Justin and I were wishing there was a "create_and_get_feed" type thing that did all of that at once.

idle owl
#

No idea how or why. Never got to see the error.

#

What do you except? Pylint hates bare except

blissful pollen
idle owl
#

But since it reconnects every time, yeah, I can see that. Jerry had the wifi connection code in a try/except, I presume for that reason.

blissful pollen
#

I excepted it all. Like a global if anything goes wrong sleep for 30 seconds and let's try this again. Pylint may not like it but seemed the easiest way to do it

idle owl
#

So bare except.. Yeah hmm.

#

I can disable Pylint, I simply try to avoid it.

blissful pollen
idle owl
#

But the wifi stuff doesn't throw errors, it gives nice messages.

#

So I can't except anything specific for that.

#

Wow. A lot going on there.

blissful pollen
#

Mine is not a trivial example. Its three main things are reading the current time/date, reading the current forecast from an API and then reading my own IO feeds for temp/pressure. But I modified the IO HTTP API to read X values and average (or sum) them

idle owl
#

Ahh ok

blissful pollen
#

There are a lot of features in the REST API that are not in the python library

tidal kiln
#

@idle owl ok. whatup?

manic glacierBOT
idle owl
# tidal kiln <@!330227457296957440> ok. whatup?

Ok... I have this table in a datasheet for the LC709203 battery monitor. That allows you to specify mAh size on the battery you're using for more accurate readings. It has "typical" sizes. But I wanted to try to add another size. I can't figure out a pattern or how they came up with the values. Page 11, Table 7 https://cdn-learn.adafruit.com/assets/assets/000/094/597/original/LC709203F-D.PDF?1599248750 and here is a screenshot of a spreadsheet I made so I had the decimal values to go with it, though I bet you don't need those.

Screen_Shot_2021-11-23_at_14.02.41.png
#

I wanted to add 400mAh. But.... those values do not make sense to me.

#

That list is in the library.

tidal kiln
#

guessing table is just values taken from this

idle owl
#

I tried to match it up!

#

They don't really match up πŸ˜„

#

Or I guess they mostly do?

#

now that I see that plus my spreadsheet

#

hadn't compared the two since I wrote out the decimal values

#

So at that point, I'd be picking an arbitrary number from that graph, and it might not be that accurate anyway. Hmm.

#

Google didn't produce anything useful for me.

tidal kiln
#

do you know the "type"?

idle owl
#

The library uses 01 or 03, because those are the values in the lib.

#

So no I don't know really. Going based on numbers.

lone axle
#

If you want to get into pixel counting you could take a chunk of that graph into gimp or similar image editor and determine the pixel that is 80% of the way to 500 along the bottom (which should be 400mah) and then also count pixels on the left to determine more precisely where it falls between 10 and 20 on the left.

There must be a formula that the values could be plugged in to in order to get the output (I assume they used one to create the graph) but it doesn't look like they shared the formula and my algebra is definitely not up to where it would need to be in order to solve and find it.

idle owl
tidal kiln
#

yah, i'm not sure either. in terms of what "type" is referring to.

#

but if you know type, can interpolate to other values.

#

or use graph to pick off value at 400

idle owl
#

Well 01 and 03 are apparently the same

tidal kiln
#

right. but what about the other types? i'm not sure what those are.

idle owl
#

How do I interpolate to the other values though? They don't really follow a pattern.

tidal kiln
#

the trend isn't linear unfortunately

#

but you could linear interpolate between the table's 200maH and 500mAh values

#

probably close enough

idle owl
#

Do math..........

tidal kiln
#

yep

#

easy math though

idle owl
#

Easier than counting pixels I think. But then again, I'd be doing that some kludgy manual way, so...

tidal kiln
#

i think simpleio has something

idle owl
#

I should ask Noe (not Ruiz) in #general-chat ... They just did some massive proof the other day πŸ˜„

tidal kiln
#

yah, can just use map_range

idle owl
#

But... how would that work, the ranges are different for each gap

#

simple_math would work too then, that's all that's in simple_math I think.

tidal kiln
#

you'd make different calls depending on which range you are in

#

let me look at that... i'm probably behind in my knowledge of these helper libs.

idle owl
#

Eh, it was that no one was using simpleio for anything but the map_range for the most part, so Dan made it its own thing.

tidal kiln
#

ok, yah, it's essentially same thing

#

so, for example, if you're interested in 400mAh, then use the values from 200 and 500.

#

if you wanted 800mAh, use the values from 500 and 1000

#

etc.

idle owl
#

Ok

#

Thanks πŸ™‚

#

Figured it would end up being longer than a moment.

tidal kiln
#

"piecewise linear approximation" is what you're doing

idle owl
#

Good to know.

tidal kiln
#

if they had provided a formula for the curves, then could have just used that

#

APA = f(mAh)

idle owl
#

What?

#

Did you find it or did you make that up

tidal kiln
#

it's not there

idle owl
#

ok

tidal kiln
#

like whatever the actual functions for those curves are

idle owl
#

The answer is 10.

#

For 400mAh. Apparently.

#

If I did it right.

tidal kiln
#

10 = decimal, hex, or binary?

#

πŸ™‚

idle owl
#

decimal

#

I'll convert it to hex for the lib.

tidal kiln
#

doesn't seem right. should be between 11 and 16.

idle owl
#

ohwait

#

I put in the wrong values.

#

I put in 100mah

#

and 200

#

I think.

#

14.3 which means 14.

tidal kiln
#

that seems more correctish

idle owl
#

Yeah, I picked the wrong range initially.

#

100-200 is what I had.

#

So is it worth adding when it's not specifically accurate?

tidal kiln
#

"table lookup" is another term for this

#

are you adding a feature to the library for the LC??

idle owl
#

Can you do decimals in hex? Seems like it goes weird

#

There's already this feature in there, using the existing table from the datasheet

#

I was going to add 400mAh as an option to that list

idle owl
#

Or was looking into whether I even could add it

tidal kiln
#

ah. i see. yah. you could.

#

just add a new entry with that value you came up with

idle owl
#

Right.

tidal kiln
#

a fancier approach would be to incorporate the lookup into the library code, so could then specify arbitrary sizes

#

instead of the current fixed options

idle owl
#

🀯 heh

#

I'm supposed to be doing the guide for the Feather πŸ˜„

#

This was a detour because I only figured out what the PackSize even was earlier today

#

And realised those 400mAh batteries we have designed for the Feathers are going to be common to Feather projects

tidal kiln
#

ok. yah. just add a new explicit 400 entry.

#

can get fancier later. if needed, wanted, etc.

idle owl
#

Agreed.

#

@tidal kiln 0x0E right? (I mean I checked it, but still...)

tidal kiln
# 
>>> 0x0e
14
>>> hex(14)
'0xe'
>>>
idle owl
#

huzzah

#

Thanks for the Python reminder.

tidal kiln
#

i don't have one of those brains that can think in arbitrary bases, like hex. so i always have to check also. πŸ™‚

idle owl
#

Here I thought you were some kind of hex wizard.

stuck elbow
#

hex is easier if you think about bits

crimson ferry
#

Deep Sleep: is it fair to say that when a device enters Deep Sleep, it does the usual deinit things as if a Control-C is entered in the serial console (e.g., GPIO get switched to high-Z generally except on never-reset and pin alarm pins), and when a device wakes from Deep Sleep, it does the usual init things as if a Control-D is entered in the REPL? What kinds of things fall into the "not always the case" category: Sometimes these on-board devices are controlled by the microcontroller pins, and may be turned off as a consequence of the pins being turned off, but this is not always the case? https://learn.adafruit.com/deep-sleep-with-circuitpython/power-consumption

idle owl
#

Just had it hang coming out of the REPL again. πŸ˜–

#

OK, snack time. bbiab.

blissful pollen
crimson ferry
#

there's alarm.sleep_memory which survives deep sleep (and reloads), but not hard resets / power cycles

#

what's different about a power cycle (or deep sleep) than a reload, oither than the code.py is restarted? boot.py doesn't run for waking from deep sleep, does it?

blissful pollen
#

I'm not sure how much is but something like pins for displays or anything set as "never reset" would be reset I believe after deep sleep. Been a while since I read the specs

tidal kiln
#

@idle owl anyone specific reviewing type hint PRs?

idle owl
tidal kiln
#

i'm not up to speed on them 😦

idle owl
# tidal kiln i'm not up to speed on them 😦

Who's the author? Because at this point I'd merge Tekktrik's with a basic readthrough. They've done enough of them. I think FoamyGuy was importing them into an IDE and testing them, but I'm not sure he's doing that anymore.

tidal kiln
#

that's the one

#

there was a dedicated PR, but things got a little confused

crimson ferry
#

oh, boot.py does seem to run on waking from deep sleep, at least in simulated (USB connected), so I'd expect that to be true for real ...but sleep_memory didn't behave as I expected

idle owl
#

Oh I see, a whole new example there.

#

Eh, if the example's been tested, and the type hints make sense, go for it.

tidal kiln
#

yah. that also, which i'm thinking about. but for the type hints, are there CI checks?

idle owl
#

Maybe I'm not understanding your question

tidal kiln
#

if there was an issue with the type hint stuff, would CI catch it?

idle owl
#

I have an unrelated question (tangentially related to earlier).... this is the LC709203, right? I can provide the schematic and board file if useful. (They're also on GitHub, but I have them here already.)

Screen_Shot_2021-11-23_at_15.03.26.png
idle owl
#

That's why you want to read through and make sure, yeah that's an int, etc.

tidal kiln
idle owl
#

Oh hmm

#

Let me try that.

#

HOLY CARP.

tidal kiln
#

good skill build opportunity

idle owl
#

I have been clicking on a part of the part with the eyeball this whole time

#

and guessing based on what the connections are

#

I had no idea you could click on the WHOLE part.

#

mind == blown.

#

(It is the right part)

tulip sleet
manic glacierBOT
#
DavePutz
[adafruit/circuitpython] New comment on pull request #5613: clear out interrupt when freeing the timer

In testing, this patch seems to fix the initial issue in #5418. However, the second hang mentioned in that issue; where a similar hang can be generated when running the script from https://learn.adafruit.com/rgb-led-matrices-matrix-panels-with-circuitpython/connecting-with-feather-rp2040-featherwing by hitting CTRL-C during startup; still exists. I captured a stack from that hang:

#0  0x1008cd78 in blast_long (data=0xbc, core=0x2000391c <displays+20>) at ../../lib/protomatter/src/core.c...
crimson ferry
lone axle
# tidal kiln if there was an issue with the type hint stuff, would CI catch it?

There are not currently any CI checks for type hints. Eventually we could implement one using mypy but there is discussion to be had about how strict we want it to be configured. And at present we don't have a majority of the type hints in yet so a large percentage of the libraries would fail the check if it did exist today.

For the existing PRs I've been looking over them to confirm the types make sense and testing some of them by viewing the hint popup in PyCharm (some are harder than others to do this with depending on library structure and what code is used in the examples).

tidal kiln
#

@lone axle thanks. does the type hint stuff look OK in that PR? (see above)

idle owl
#

Any reason to include the extra 0 in a hex value, such as 0x0B, versus 0xB?

#

Or not include it?

lone axle
tulip sleet
idle owl
#

Like the default I2C address listed on the back of a breakout is 0x0B, but the I2C scan returns 0xb.

idle owl
tulip sleet
#

in that case, maybe you need to say something about leading zeros, if it's not obvious

#

"0x0b" is the same as "0xb: it just has leading zeros"

idle owl
idle owl
#

Wait..... .format is the preferred string format right?

jaunty juniper
#

yup on Circuitpython because f-strings are not available everywhere

idle owl
#

Keen thanks

jaunty juniper
#

for libraries in particular, but of course you can do what you want if the code doesn't need to run on samd21

idle owl
#

This is for an example. It is from a while ago and uses %, I wanted to update my adaptation.

#

@tulip sleet Do we have an ETA on another 7.1 release? It occurs to me that all these examples in this guide will only work with latest. Because of the I2C_POWER_INVERTED rename. At least I'm about 80% sure it's not in the current beta (I thought it failed for me until I went to an S3 version).

tulip sleet
idle owl
#

I am ok with putting a warning in to use unstable, but telling folks to use S3 is kinda meh.

#

Ah fair enough. I knew they weren't supported, but I was second guessing .format vs %

tulip sleet
idle owl
#

It's the main board guide for the Feather ESP32-S2.

crimson ferry
#

deleting the above, n/m, boot.py doesn't seem to run in (simulated) deep sleep

idle owl
#

I'm out Thursday and Friday though, so I have to get as far as I can tomorrow. At this point, it definitely won't make the AaE cutoff tomorrow morning.

tulip sleet
#

another release could be any time. It takes me about an hour of work. I'm not sure there are many substantial fixes so far, which is why I was holding off. But board support is an OK reason.

#

I could certainly do one early next week

idle owl
#

Hmm.

#

Alright.

#

That'll have to do. I'll let you know how far I get on it today. Are you here tomorrow?

tulip sleet
#

yes, I am working a regular day tomw

#

the shorter the interval, the shorter the list of changes, so the shorter time

idle owl
#

Ok, if it becomes urgent, I'll let you know tomorrow.

tulip sleet
#

after that, mostly it's babysitting the builds

idle owl
#

Whee found another spot to simplify!

crimson ferry
#

Deep Sleep (simulated) gives microcontroller.ResetReason.POWER_ON & supervisor.RunReason.STARTUP, but boot.py doesn't run. Do we need a new reason for deep sleep? It's a little confusing to know if we had a (possibly unexpected) power cycle, or an (expected) sleep cycle . I may be missing something.

#

I guess if alarm.wake_alarm is None, then it was a power cycle.

#

Maybe microcontroller.ResetReason.DEEP_SLEEP_ALARM just isn't set for simulated?

idle owl
#

Any reason why this Feather won't mount when plugged in....? Wait! If it's deep sleeping, does it have to wait to wake up to mount?

crimson ferry
#

it should stay mounted if it's simulated deep sleeping

idle owl
#

No, it was real deep sleep

blissful pollen
#

yeah if it was not plugged in and you plugged it in while deep sleeping you would have to wait (or reset it)

idle owl
ember iris
#

I believe if it's in deep sleep, then plugged in, then wakes and goes back to deep sleep, it'll go into simulated deep sleep because it notices it's powered from the USB

#

the esp deep sleep wake alarm reason bug gave me a ton of time to explore that one πŸ˜…

idle owl
#

Heh fair enough

crimson ferry
ember iris
crimson ferry
#

oh, I'm seeing the wrong reason for simulated

ember iris
#

Deep sleep wake reason worked in circuit python 6.x, but didn't (doesn't?) in 7.x, I haven't worked on tracking it down much past that though because the core is a bit difficult to dig into

ember iris
idle wharf
lunar gull
#

I just spoke with @lone axle on my other account. Can I please have the circuitpythonistas role?

idle owl
crimson ferry
#

If I've been in simulated since power-on, I get: None microcontroller.ResetReason.POWER_ON supervisor.RunReason.STARTUP on power-on, then on subsequent sleep cycles: <TimeAlarm> microcontroller.ResetReason.POWER_ON supervisor.RunReason.STARTUP But if I come back to connected from unconnected, I get: None microcontroller.ResetReason.DEEP_SLEEP_ALARM supervisor.RunReason.STARTUP the first time

#

oh, but once that happens, on subsequent sleep cycles, I continue to get simulated: <TimeAlarm> microcontroller.ResetReason.DEEP_SLEEP_ALARM supervisor.RunReason.STARTUP

ember iris
#

I think I'm going to be a bit slower to give it a shot--I just upgraded esptool to v3.2 and am running into protocol errors when I'm trying to interface with it. I'll be more helpful shortly I hope

main furnace
ember iris
tulip sleet
ember iris
#

Thank you! I downgraded but wanted to be mindful that a lot of users would be on v3.2 and wanted a solution that's easier than, "Downgrade a version"

ember iris
#
  • I am using a magtag for this test
crimson ferry
#

MagTag, latest S3: I print a bunch of other stuff, but then these are the key lines:```py
print(alarm.wake_alarm, microcontroller.cpu.reset_reason, supervisor.runtime.run_reason)
al = alarm.time.TimeAlarm(monotonic_time=time.monotonic() + 10)
alarm.exit_and_deep_sleep_until_alarms(al)

manic glacierBOT
#
anecdata
[adafruit/circuitpython] New comment on issue #5343: alarm\.wake\_alarm not set upon wake from deep sleep

I don't know if this is a different issue, or if the current issue is more complex.

Testing with:
Adafruit CircuitPython 7.1.0-beta.0-85-g387a8a46b on 2021-11-22; Adafruit MagTag with ESP32S2

in code.py:

print(alarm.wake_alarm, microcontroller.cpu.reset_reason, supervisor.runtime.run_reason)

If I've been in simulated since power-on, I get:
None microcontroller.ResetReason.POWER_ON supervisor.RunReason.STARTUP
on power-on, then on subsequent sleep cycles:
`<TimeA...

ember iris
crimson ferry
#

it was

ember iris
#

I have no idea why my counter is sitting at 1 when in deep sleep if yours is incrementing

crimson ferry
#

that was in simulated

ember iris
#

Ohh, ok. Deep sleep my counter is staying at 1, but in simulated it increments

crimson ferry
#

I'll put the counter back in and try more disconnected testing

ember iris
#

If you have it disconnected and it's in full deep sleep and the counter stays at 1 after multiple wake cycles, that might help narrow the issue down to the sleep memory

crimson ferry
#

when I plugged it back in, it actually restarted... boot.py and all, so it makes sense that it would reset alarm_memory

#

I'll keep it disconnected but send the count to the cloud

ember iris
#

Adjusting the text display to state reset reasons:

# Display the current battery voltage and the count.
magtag.set_text(
    "battery: {}V    count: {} mrsr:{}    srr:{}".format(
        magtag.peripherals.battery, alarm.sleep_memory[5],
        str(microcontroller.cpu.reset_reason).split('.')[-1],
        str(supervisor.runtime.run_reason).split('.')[-1]
    )
)
crimson ferry
#

I tried just flashing the NeoPixels, but unplugged they don't reflect incrementing alarm_memory

ember iris
#

Ok--here's what I've got
from reset Powered by battery or by usb:
microcontroller.cpu.reset_reason: Power On
supervisor.runtime.run_reason: startup
Count is 1 (it increments prior to printing)
alarm.wake_alarm: None

From Simulated Sleep (Has at least one cycle of 'reawakening' after reset)
microcontroller.cpu.reset_reason: Power On
supervisor.runtime.run_reason: startup
Count is > 1 (incrementing as expected)
alarm.wake_alarm: <TimeAlarm>

From Deep Sleep:
microcontroller.cpu.reset_reason: DEEP_SLEEP_ALARM
supervisor.runtime.run_reason: startup
Count is 1 regardless of number of times it's been in deep sleep
alarm.wake_alarm: None

crimson ferry
#

yeah, that's pretty much what I'm seeing, I added in the MagTag text to make sure

ember iris
#

I'm willing to bet this is all tied to the same bug. microcontroller.ResetReason. is super helpful.

#

Is there a way I can test circuit python on either side of a specific pull request?

crimson ferry
#

yeah, each merge should have a build on S3

ember iris
#

Sweet, thank you! I'll start digging

crimson ferry
#

at least going back until the last release or so

ember iris
#

To go further back, would I need to fork the repo and build for a target board? (I probably should learn to use that process anyway, so it's not a bad thing to do it that way)

crimson ferry
#

there should be UF2 and .bin files out there for recents, but before that, yeah I guess build though I'm not really sure how to do an old build

#

hm, if someone wants to throw all of this into a thread, that would be fine

#

I don't know if that's possible after-the-fact

ember iris
#

Deep Sleep vs Simulated Deep Sleep and Memory

manic glacierBOT
jaunty juniper
#

in fact with the gh command, you can checkout pull requests (open or closed) with gh pr checkout 5613 for example, which I use to test PRs (it's faster to build locally than wait for CI to finish, and/or dive on the site to find the right file in the artifacts)

ember iris
#

thank you! I will have to look into this tomorrow. I think that's exactly what I'll need. I have a pull request I want to try on either side of so that's exactly what I need!

idle owl
ember iris
#

sweeet! Thank you both!

jaunty juniper
#

in this case it will pull the PR branch from the submitter's fork, which would otherwise be a little more annoying to do manually, having to lookup the user, the name of the branch...

manic glacierBOT
manic glacierBOT
manic glacierBOT
manic glacierBOT
manic glacierBOT
lone sandalBOT
#
prcutler
[adafruit/awesome-circuitpython] Issue opened: #33 Video File does not display at https://circuitpython\.org/awesome

The video that is embedded in the README on line 11 displays fine if you're viewing the Awesome List on Github.

But if you view it at https://circuitpython.org/awesome it displays the URL and not the video file:

https://user-images.githubusercontent.com/1685947/115119719-d6e21f00-9f77-11eb-84bf-3f7af59948a3.mov

We should decide which site is the main site and whether the video should be included or not. Or if it can be embedded in HTML instead?

manic glacierBOT
manic glacierBOT
#
dcalvo
[adafruit/circuitpython] Issue opened: #5614 Adafruit Feather ESP32S2 I2C \- OSError: Unsupported Operation

CircuitPython version

Adafruit CircuitPython 7.1.0-beta.0-91-gaf2ac0aa6 on 2021-11-23; Adafruit Feather ESP32S2 with ESP32S2

Code/REPL

import time
import board
import digitalio
import adafruit_icm20x

ip = digitalio.DigitalInOut(board.I2C_POWER_INVERTED)
ip.switch_to_output()
ip.value = True

i2c = board.I2C()
icm = adafruit_icm20x.ICM20649(i2c)

while True:
    print("Acceleration: X:%.2f, Y: %.2f, Z: %.2f m/s^2" % (icm.acceleration))
    pri...
timber mango
#

@languid whale my ANO encoder arrived and got to test your code.
Works great!

manic glacierBOT
manic glacierBOT
manic glacierBOT
lone sandalBOT
manic glacierBOT
lone sandalBOT
manic glacierBOT
#
dhalbert
[adafruit/circuitpython] New comment on pull request #5610: Initial broadcom port for Raspberry Pi

@tannewt Making a commit from the GitHub web interface seems to have upset the build cloning optimizations you added. From https://github.com/adafruit/circuitpython/runs/4313459204?check_suite_focus=true:

Run git describe --dirty --tags
  git describe --dirty --tags
  echo >>$GITHUB_ENV CP_VERSION=$(git describe --dirty --tags)
  shell: /usr/bin/bash -e {0}
  env:
    pythonLocation: /opt/hostedtoolcache/Python/3.8.12/x64
fatal: No tags can describe 'e98223cbae0c9dfe636badcc49bb...
manic glacierBOT
lone sandalBOT
manic glacierBOT
#
jepler
[adafruit/circuitpython] Pull request opened: #5616 Bitmaptools bugfixes

Previously, Floyd-Stenberg dither couldn't be accessed, because its constant had the wrong value.

When it was enabled, there were occasional artifacts in all-white areas. Switching from an under-parenthesized shift to a well-parenthesized division makes the artifacts go away.

There was also an off-by-one problem loading image data that caused row 2 to be duplicated. You could see this most clearly by repeatedly dithering an already-dithered image back into itself.

manic glacierBOT
lone sandalBOT
lone sandalBOT
#
FoamyGuy
[adafruit/awesome-circuitpython] New comment on issue #33: Video File does not display at https://circuitpython\.org/awesome

I think the page is rendered from this file here: https://github.com/adafruit/circuitpython-org/blob/master/awesome.html int he CircuitPython_Org repo.

It might make sense to move this issue over to that repo.

Only took a quick peek into it, but looks to me like the gets rendered via the Jekyll build process. The template loads in the contents of the md file and renders it using a markdownify template tag. I'm not sure if that functionality specifically comes from Jekyll or is a feat...

crimson ferry
#

This board is a Raspberry Pi Pico RP2040 with built-in Ethernet (uses SPI and a couple of pins for that). It runs the CircuitPython for Pico UF2 fine. Should we encourage the manufacturer to add the board to CP code & .org due to the dedicated pins, or just use it like a Pico? https://www.wiznet.io/product-item/w5100s-evb-pico/

W5100S-EVB-Pico | WIZnet Co., Ltd.

W5100S-EVB-Pico is a microcontroller evaluation board based on the Raspberry Pi RP2040.

manic glacierBOT
slender iron
#

@crimson ferry ideally they'd add it. I get the impression they want to freeze the drivers in when they do

manic glacierBOT
orchid basinBOT
#
FoamyGuy
[adafruit/circuitpython-org] Pull request opened: #792 change with video URL to element on awesome page

This is meant to resolve: https://github.com/adafruit/awesome-circuitpython/issues/33

I looked into it a bit and markdownify is a template filter provided by the Jekyll project. The templating language used is called Liquid, but Jekyll adds some functionality that isn't in the upstream Liquid project. Markdownify is documented on this page: https://jekyllrb.com/docs/liquid/filters/

I couldn't find a way to instruct it to render video elements. Theoretically that functionality could be...

lone sandalBOT
lone sandalBOT
manic glacierBOT
slender iron
#

any intuition on the value of reworking the table structure used by map to use 16 bits for the key when all keys are qstr?

#

I think it'd drop the sizes by 25%

#

8 bytes an entry to 6 bytes an entry

manic glacierBOT
tulip sleet
#

also, do you understand what's going on with the broadcom build failures at all?

slender iron
#

the tables that define objects

#

qstr -> function object usually

#

@tulip sleet no, it's weird. I added a thing to print git log if the describe fails

tulip sleet
#

the commits the logs complained about make no sense to me

slender iron
#

I think the CI runs on a merge commit

#

not of the tip of the branch

tulip sleet
#

i happen to have more full access to your repo than is usual (you gave it to me a while ago), but I don't know why that would make a difference here. The PR now has merge conflicts, btw, if you did not see that already

#

@slender iron, you mean like the STATIC const mp_rom_map_elem_t busio_i2c_locals_dict_table[] = { etc.?

slender iron
#

I figured it would

#

yup, those tables

#

I'm trying to reduce the build size so we can merge translations in

tulip sleet
#

there are 1497 non-pin-table entries like that, so it would save 750 bytes, but only if every module was in every build

#

which builds are not fitting?

slender iron
#

my arch build has 2k free on arduino_zero ru

tulip sleet
#

and it was overflowing before?

slender iron
#

(I canceled the 726 build)

tulip sleet
#

CIRCUITPY_ONEWIREIO = 0 has worked for me for a lot of tight builds

slender iron
#

ya, I'd rather not remove it from existing boards though

tulip sleet
#

the only boards where it was an issue were ones that saw very little use. Since it's just a compatibility thing for now until we get to 8.0.0, I thought it was the least annoying

slender iron
#

πŸ‘

#

I'd do the table optimization if I knew a C macro trick to have the key and value on the same line still

tulip sleet
#

not sure what you mean, what do you want to output?

#

for grepping purposes?

slender iron
#

right now it's KVKVKV in memory but I'd want it to be KKKVVV going forwards

#

so that K can be 16 bit, not 32

tulip sleet
#

ah, got it

slender iron
#

but maintaining it with K, V on a line is nice

#

bonus points for reordering keys to speed up lookup πŸ™‚

tulip sleet
#

cpp is just too weak to do anything like that, I think

slender iron
#

ya, I think so too. not the first time I've been tempted to post-process a build artifact πŸ™‚

#

or... we have templates that generate c

#

MP_QSTR ordering depends on the generator though too

#

MP_REGISTERED_MODULES kinda does that

tulip sleet
#

i think this is interesting, but perhaps deferrable πŸ™‚

slender iron
#

ya, the three boards that need it are arduino_mkr1300, arduino_zero and stringcar_m0_express

tulip sleet
#

the first two have no external flash anyway, so not that interesting, and the last doesn't need onewireio compatibility πŸ™‚

idle owl
#

@blissful pollen Greetings! Are you around?

blissful pollen
#

Sorta right now. I’ll be home in an hour or so

idle owl
# blissful pollen Sorta right now. I’ll be home in an hour or so

Maybe you can answer this now, or maybe later. My question is, what is going on here, which is to say, if I'm doing a walk-through of the code, what sentence can I put to explain this. Right now it's "Connecting to Wi-Fi" but I think this is not quite part of that? Or is it? pool = socketpool.SocketPool(wifi.radio) requests = adafruit_requests.Session(pool, ssl.create_default_context())

#

It's in my Adafruit IO sending code.

#

I can gloss over it if it's bonkers to explain.

blissful pollen
#

It’s not about connecting to the wifi (that’s happened already) but about setting up for communication using the wifi.
The first line sets up a pool of socket objects to use. The second initializes the requests library with said pool.

That doesn’t really say much on its own I know

idle owl
#

Ooh no that's good.

#

I can use that. Thanks!

blissful pollen
#

Welcome! It confuses me as well. I’d have to look back into the libraries to see exactly what is set up. But for the average user just steps you need to take

daring pumice
#

Hello everyone,

I hope I'm not interrupting. I've been corresponding with @tulip sleet on the adafruit/curcuitpython forums, and was invited to join this discord to contribute to help find issues with the build for the nrf52833 dk board.

Is this the right place to ask questions and/or ask for help?

idle owl
daring pumice
#

Thank you, Kattni πŸ™‚ Pleased to be here.

Is it alright if I ask for some help with building the firmware? I've been slowly progressing towards what I think is the intended build, but have encountered some problems and wanted to run them by you guys to see what you make of them.

That ok?

idle owl
#

Absolutely! Ask away.

tulip sleet
daring pumice
#

@idle owl Thanks again πŸ™‚
@tulip sleet I did, and thanks again! I've posted my latest issue on the forums, but I realize people are in the middle of work, so I thought I'd come here to see if anyone has some free time and a kind point in the right direction.

tulip sleet
#

i will try to reproduce your build. On what system are you building?

daring pumice
#

As it stands, I can build the .bin, .hex., and .uf2 files - no issue. However, the next step in the make appears to try and create firmware.combined.hex (if I'm reading that correctly - I may not be), using hexmerge.py.

Hexmerge.py doesn't exist on my box. Have I missed somehting?

#

Win 10

tulip sleet
#

with wsl?

daring pumice
#

Nope, native.

#

I have a mac at my disposal, so can try there - either works - but right now I've made more progress on win10 (if that can be believed).

tulip sleet
#

you're building the bootloader or circuitpython?

daring pumice
#

curcuitpython

#

I can post a SShot if you like - not sure if that's helpful

#

can also paste the text from the make attempt

tulip sleet
#

we don't have instructions for building on windows except for using WSL, so I'm not sure how you got started. The paths in your errors have forward slashes, so it didn't look like windows

#

you can upload a file with the + to the left

daring pumice
#

Sounds good - I'll post

tulip sleet
#

if the paste is too big

#

or put it in pastebin or gist, anything is fine

daring pumice
#

Gotcha - should be ok

#

To get this error, I mirrored the repository for adafruit/circuitpython from git

#

ran the submodule command that appears in the page you sent, Dan

#

and then ran "make V=1 BOARD=PCA10100"

#

It runs quite far, and can build the files I mentioned above, but gets stuck here immediately after

tulip sleet
#

that doesn't look like windows, I'm confused

daring pumice
#

I'm using the terminal app running cmd

tulip sleet
#

that will basically not work at all. do you have some Unix utilities installed somehwere

#

i'm surprised it got this far

daring pumice
#

I have a bunch installed

#

mingw, etc.

#

which are you referring to?

#

(and thanks, Dan, appreciate your time)

tulip sleet
#

it really won't work. It's sort of possible to do with mingw, msys, etc, but it's quite painful. It would be much easier if you installed WSL or used a Linux VM or a native Linux box

daring pumice
#

ok, so you're saying I should try this on my mac?

tulip sleet
#

no, I'm not saying that either πŸ™‚ . The mac has its own issues. A clean Linux system like Ubuntu 20.04 is what the rest of us use for day-to-day development

#

you can get it working on Mac more easily, but there's still a lot of homebrewing to do, and making sure you have the right python, etc.

#

for instance, on Mac, you need to create a case-sensitive filesystem

#

to do the work in

daring pumice
#

Ok. I admit, I'm not looking forward to installing WSL - it doesn't forward resources attached via USB in my experience... perhaps I'm just doing it wrong

#

Regardless, your point is well made and well taken. I'll set up an Ubuntu install and try it there

#

Good idea?

idle owl
#

I am able to build on MacOS, but Dan's right, it takes a bit of work to get there.

tulip sleet
#

The USB business is an issue, but you can do the testing on the Windows side. Note the instructions about using the filesystem inside WSL, not the Windows filesystem, again due to case sensitivity and symbolic links

#

if you have any other old box, even an old laptop, and can get Ubuntu on it easily, you may be more productive.

#

Use WSL2. The filesystem performance on WSL has been an issue in the past, making for slow builds, but they have been working on it.

daring pumice
#

I hear you - I'll try both. I've installed brew and a bunch of build tools, so I'll see how that goes as I install WSL2 (I see that advice, and will follow it).

#

Thanks again - very kind πŸ™‚

#

@tulip sleet and @idle owl πŸ™‚

tulip sleet
#

sure, np, follow the steps carefully in the guide, and there may yet be problems with new homebrew or the latest macOS

idle owl
#

And don't use Homebrew to install Python!!

#

It leads to bad things later because of how updating it is designed.

daring pumice
#

Ok. I'm really new to MacOS - last time I wrote for mac was OS 8 or 9 (1993/4?)... what would you suggest, @idle owl ?

#

@tulip sleet will do

idle owl
#

Hah yeah, been a while there. I would recommend pyenv. It's well documented, and you definitely sound like you're up on things enough to get it sorted out with the docs.

daring pumice
#

@idle owl Good deal, and thanks for the confidence πŸ™‚

#

Cheers!

idle owl
#

Cheers!

#

@daring pumice Also, pyenv-virtualenv and pyenv-virtualenvwrapper. Long story short regarding Brew to install Python, is that it will update out from under you and break all your virtual envs. Along with updating dependencies that maybe you didn't want to or intend to update.

daring pumice
#

Interesting - I didn't realize that. I admit, I'm a little confused at all the package management paradigms these days - my first linux install (in ~1999) didn't really have that... I think apt-get came in a little while after and everyone loved it.

#

Appreciate the insights.

idle owl
#

Yeah, brew is meant to be a package manager for MacOS, as there is not a native one to be had.

#

But it handles Python really awkwardly. They kind of wrote their Python implementation for themselves, not for others. And it breaks thing if you're not paying attention.

#

So, either install Python directly from python.org , or use pyenv to manage multiple Python versions without a lot of nonsense.

#

Bear in mind, I got help with most of this, so don't feel bad about asking for assistance. Frankly, if you run into issues, I'll be asking someone else to help out πŸ˜„

daring pumice
#

Ok. I'll def do that. Would you suggest rolling back all of the stuff I've installed with brew thus far? Hasn't been that much, and the projects I've been trying to build with the packages it has installed haven't worked yet.

#

No issue - I'm very glad for the help! πŸ™‚

idle owl
#

Depends on what you've installed, but I wouldn't be too concerned.

#

But if nothing has worked yet, you could start fresh.

daring pumice
#

Alright. I'll pull out brew for now and start over. Thanks again!

idle owl
#

You're welcome!

#

@blissful pollen Wait, you said it's probably best to wrap this whole thing in a try/except because Wi-Fi is sketchy? Does it make more sense to put only the Wi-Fi connection into a try/except?

#

Jerry had it that way, I took it out not realising it might be needed.

#

If I'm going to add that back in, I need to make the deep sleep a function because it's used twice in that case.

blissful pollen
#

I wrapped the whole thing just in case anything failed I wouldn’t stay awake wasting battery power.

The io requests like send/get can fail too. Internet hiccups or anything

idle owl
#

Ahhhhh.

#

OK. Fair enough.

#

I guess it's worth it then.

#

Since this tiny battery will last at least 2 days, so a bigger one could last a week or something.

#

And I doubt the code will run for a week without some kind of exception.

#

Not consistently anyway.

blissful pollen
#

I ran into that in my magtag project. Something would fail and then the battery would die. It’s not perfect but worked well enough. I was tracing a weird wifi issue but never found the root cause

idle owl
#

It works... but I'm not sure whether it's right.

ember iris
#

I would change it so your try/excepts are covering smaller ranges of the code, that way if something goes wrong you know what went awry. This will safely wake up every 10 minutes and try again regardless of what happened last time but if you're getting an error it can make debugging harder
For example if you don't have a secrets.py file, the code won't raise the error and break because that error is caught, so in 10 minute's it'll try again

idle owl
#

You can't see what went wrong anyway.

#

Oh you mean before getting it going

#

I see.

#

Yeah fair enough.

#

Ok, let me try that.

ember iris
#

Right. So if your sensor isn't connected, or your credentials are bad or missing, or you didn't add a library, it'll quietly fail. Wrapping the things that interact with the wifi only will help keep the program running during wifi issues, without being quiet about the other parts

idle owl
#

I understand now

#

May still not get it right, but I get where you're going with it πŸ˜„

ember iris
#

The wifi stuff is a pain! There are so many parts that could go wrong in the middle you have no control over. I have a 'retry x times' if I get an exception because sometimes it breaks and trying again works

idle owl
#

The Wi-Fi doesn't fail with an exception, it fails with a nice message. Which is useless for excepting anything useful. Not sure about AIO, so I left it bare as well.

#

Hate doing that, but eh.

#

Better to have this code keep running.

ember iris
#

That looks contained. You can add,

except Exception as e:
    print(e)

that way if any of the exception blocks occur while the micro is plugged in you can see what's failing

idle owl
#

Ok

idle owl
ember iris
# 

# Fetch the feed of the provided name. If the feed does not exist, create it.
def setup_feed(feed_name):
    try:
        # Get the feed of provided feed_name from Adafruit IO
        return io.get_feed(feed_name)
    except AdafruitIO_RequestError:
        # If no feed of that name exists, create it
        return io.create_new_feed(feed_name)

I don't know what the io stuff is--if there's no wifi, could this last line io.create_new_feed(feed_name) break? or is it just created a new feed name--a completely on the device no wifi kind of thing?

idle owl
#

It gets called inside the second try/except block

#

I assume the function doesn't do anything until you call it?

#

Or am I misunderstanding how all of that works

#

Because that definitely requires a network connection. But I don't do anything with it until inside another try.

ember iris
#

Oh there it is!

#

I missed the spot it was called in

idle owl
#

Ah ok!

#

You had me there for a minute πŸ˜„

ember iris
#

I keep reopening it and tracing everything and missing a ton because of it πŸ™‚

#

Looks like the code should be stable from what I can see

#

if you have i2c errors it'll break, but I think that's a good thing?

orchid basinBOT
idle owl
#

Yeah, I don't foresee those happening once the code is working.

#

The sensors are either going to work or not.

#

Because they're built in.

#

so if they don't work, you'll know when you're setting up to begin with.

ember iris
#

and if a wire gets cut over it's life, it is actually broken, so the code shouldn't pretend it's working

idle owl
#

True.

#

You'll get an intermittent blinking red NeoPixel if the code crashes.

#

If it's on battery and you can't see the errors.

#

Dinner time!

ember iris
#

Well, one way to test if it can handle radio failure is to throw it in a microwave (don't turn it on). That'll cut off the radio and if it comes back and works 10 minutes after you pull it out then the radio code is stable

ember iris
manic glacierBOT
blissful pollen
orchid basinBOT
manic glacierBOT
daring pumice
#

@tulip sleet you still around?

tulip sleet
#

yes

slender iron
#

just ask your questions. multiple folks can help

daring pumice
#

haha glad. Let me know if/when you're done for the night - don't want to bother you during off-hours

#

@slender iron thanks πŸ™‚

#

I've installed Ubuntu over WSL2 as you advised

#

and have gotten to precisely the same spot with this process

#

exact same errors

tulip sleet
#

i will try it myself

daring pumice
#

Much appreciated. In the mean time I'll post a shot of my terminal - maybe that'll help?

#

Is it (perhaps) a pathing error?

idle owl
blissful pollen
#

Quick question if someone knows, was going to open a tracking issue for both implementing the eye glass rings as pixelbuf objects (which is obvious core) and also a python helper library for IS31. For the helper library does it make sense to include it with the current IS31 python library?

idle owl
slender iron
#

@daring pumice do pip install intelhex

#

hexmerge is a tool from that package

daring pumice
#

@slender iron already satisfied

blissful pollen
daring pumice
slender iron
#

can you run hexmerge.py yourself?

daring pumice
#

But much appreciated πŸ™‚

idle owl
daring pumice
#

just tried [code]python3 hexmerge.py[/code] (hope the braces work)

#

lol nm

slender iron
#

the other trick can be what python is setup. sometimes you want pip3 install intelhex

#

It's backticks around code

idle owl
daring pumice
daring pumice
#

this is what I ran. perhaps I should just do pip install intelhex?

blissful pollen
#

The new IS31 core module could use some helper python functionality (the glasses LED mapping for example is passed in). Figured if the helper is another file in https://github.com/adafruit/Adafruit_CircuitPython_IS31FL3741 then it is all togeher.
So you could include adafruit_ledglasses.py for the python library and say native_ledglasses.py (just made that up) for the native library?

tulip sleet
# daring pumice Cheers πŸ™‚

The issue is probably that in that terminal instance, it was the first time that .local/bin was created, so it was not in your path. If you started a new terminal, it would notice the .local/bin and add it to your path

idle owl
#

@blissful pollen We could do Adafruit CircuitPython LED Glasses or something?

tulip sleet
#

there is a conditional in .bashrc or someplace that does that

daring pumice
#

@tulip sleet trying now - @slender iron thanks a lot for jumping in, btw πŸ™‚

slender iron
#

np πŸ™‚

blissful pollen
idle owl
blissful pollen
#

I can always write some code, put it somewhere and decide a home for it later

idle owl
#

That also works.

#

I would have to see some of it, I think, to make a more informed decision.

#

So that might make sense.

blissful pollen
#

I'll ping you when I have something or bring it up on Monday for discussion if I have an example to share

idle owl
daring pumice
#

@tulip sleet @slender iron this is weird - now make BOARD=pc10100 gives me there are no targets specified and no makefile found...

tulip sleet
#

cd ports/nrf

daring pumice
#

yup, I'm there πŸ™‚

#

sorry, was in /boards

#

from /nrf I get "Invalid BOARD specified."

tulip sleet
#

do make -j<n> where n is the number of cores you have, to speed things up

#

pca10100 not pc10100

daring pumice
#

my bad

tulip sleet
#

we have a meeting in 5 minutes, so you'll be on your own for a while πŸ™‚

#

echo $PATH

daring pumice
#

@tulip sleet understood - much appreciate

#

d

blissful pollen
#

I'm half around @daring pumice if you have a question and can try to help with what I know

daring pumice
#

/home/mountain/gcc-arm-none-eabi-10-2020-q4-major/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/usr/lib/wsl/lib:/mnt/c/Program Files/Python310/Scripts/:/mnt/c/Program Files/Python310/:/mnt/c/Windows/system32:/mnt/c/Windows:/mnt/c/Windows/System32/Wbem:/mnt/c/Windows/System32/WindowsPowerShell/v1.0/:/mnt/c/Windows/System32/OpenSSH/:/mnt/c/Program Files (x86)/NVIDIA Corporation/PhysX/Common:/mnt/c/Program Files/Git/cmd:/mnt/c/Program Files/PuTTY/:/mnt/c/Program Files/CMake/bin:/mnt/d/Ginkgo/Development/ninja:/mnt/d/Ginkgo/Development/MinGW/bin:/mnt/c/Program Files (x86)/GNU Arm Embedded Toolchain/10 2021.10/bin:/mnt/c/Users/scath/AppData/Local/Microsoft/WindowsApps:/mnt/c/Users/scath/AppData/Local/Programs/Microsoft VS Code/bin:/mnt/d/Ginkgo/Eclipse IDE/arm_tools/bin:/mnt/d/Ginkgo/Development/MinGW/msys/1.0/bin:/snap/bin

daring pumice
tulip sleet
#

you may need to start a shell from the top-level, not from inside the previous shell

#

e.g. exit this shell and start naother one from windows

daring pumice
#

trying now

#

no issue if you have to bounce

#

same issue

tulip sleet
#

you did pip3 install, not sudo pip3 install, right

daring pumice
#

I didn't, but have done so now - 1 sec

tulip sleet
#

no DONT

#

dont' do sudo pip3 install

daring pumice
#

yikes

#

lol

#

ok

tulip sleet
#

just uninstall it

daring pumice
#

I'll uninstall

#

done

#

still installed without sudo - I still see it

#

Go to your meeting, Dan πŸ™‚ You've certainly done enough for right now and don't want to make you late πŸ™‚

#

With my thanks πŸ™‚

ember iris
#

How is python installed on your WSL2--did you use pyenv or do you have a virtual environment option available?

tulip sleet
#

the stock python should be ok

daring pumice
#

no venv, just straight

#

however, I've added the dir that hexmerge is in to the path

#

this appears to succeed, but only renders mpversion.h

slender iron
#

try cleaning and then make again

daring pumice
#

You got it - running now /w V=2

#

Interesting. Failed at at the hexmerge step, despite that dir being in my path

#

No rush if you're busy, guys. I appreciate it, but don't want to distract anyone from their day jobs

slender iron
#

can you run hexmerge yourself? if so, you can do which hexmerge.py

tulip sleet
#

This is in .profile, which should get run, I think, when you start a terminal, or start WSL:

# set PATH so it includes user's private bin if it exists
if [ -d "$HOME/.local/bin" ] ; then
    PATH="$HOME/.local/bin:$PATH"
fi
daring pumice
#

I added the wrong dir to the path

#

@tulip sleet will check - reasonably sure you're correct

#

the build succeeded

#

@tulip sleet @slender iron @idle owl thanks so much

tulip sleet
#

only the beginning πŸ™‚

idle owl
#

You're quite welcome!

slender iron
#

np, I'm happy someone is looking to fix it πŸ™‚

daring pumice
#

@tulip sleet @slender iron completely - I'm going to look into how to get this thing on my board now. The uf2 file is currently too big, as I previously explained to @tulip sleet. Looking for a file I can drag and drop onto the bootloader (<= 128k) or some other methodology.

#

Won't bother you with this just now, I'll take a look and come back if I get lost or go nowhere.

#

Thank you again for all your kindness and patience πŸ™‚

slender iron
#

the bootloaders drive is fake so I'm surprised it's too small

daring pumice
#

@slender iron honestly, so am I - this one issue has caused me no end of problems; have been hard-stuck trying to work around this for 2 days and counting. lol

manic glacierBOT
#
CaribouJohn
[adafruit/circuitpython] New comment on issue #5614: Adafruit Feather ESP32S2 I2C \- OSError: Unsupported Operation

I am getting an issue after my script runs for some period of time - fairly arbitrary time period

from adafruit_bme280 import basic as adafruit_bme280
i2c = board.I2C() # uses board.SCL and board.SDA
bme280 = adafruit_bme280.Adafruit_BME280_I2C(i2c)

Traceback (most recent call last):
File "code.py", line 123, in <module>
File "adafruit_bme280/basic.py", line 200, in temperature
File "adafruit_bme280/basic.py", line 115, in _read_temperature
File "adafruit_bme280/basic.py...

daring pumice
#

Question: is the firmware.hex file the complete build?

#

If so, seems like I should be able to drag that file onto the mounted drive - not the bootloader - and reset.

blissful pollen
#

Does anyone know of an example of a native C implementation of a python class having a super class?

slender iron
#

@daring pumice the hex is as well but the bootloader won't know what to do with it. uf2 is a different encoding of the hex contents

#

@blissful pollen I don't know of a native class with a native super class

blissful pollen
slender iron
#

ah, that's true

manic glacierBOT
manic glacierBOT
#
scath999
[adafruit/circuitpython] New comment on issue #5498: nRF52833 DK \(PCA10100\) CircuitPython UF2 does not work on Adafruit\_nRF52\_Bootloader

Ok. With @dhalbert's help, I'm now up to speed and at the same point as @jpconstantineau.

The dk board has a built in segger jlink, and I'm able to successfully executemake BOARD=pca10100 flash

This appears to complete properly, but when the device restarts there is no change - the board reboots and appears as a drive to the OS, but with the same volume label and contents as before.

I have left a message on the nordic dev boards asking for some insight here - perhaps there's some k...

daring pumice
#

@tulip sleet Update: I now understand the issue #5498 on git. I have posted an update.

lone sandalBOT
lone sandalBOT
onyx hinge
# 
Submodule 'ports/broadcom/peripherals' (https://github.com/adafruit/broadcom-peripherals.git) registered for path '../broadcom/peripherals'
Cloning into '/home/jepler/src/circuitpython/ports/broadcom/firmware'...
``` does this take a verrry long time the first time?  It's been running for minutes and minutes.
#

18 minutes, in fact.

stuck elbow
#

it's a big processor ;-)

onyx hinge
#

ah it finally finished. the magic of complaining on the internet

#

18GB, that's a lot of data

stuck elbow
onyx hinge
#

neat, have fun!

manic glacierBOT
manic glacierBOT
daring pumice
#

Anyone here familiar with the NRF52 family of boards?

#

Also, afternoon all πŸ™‚

blissful pollen
#

I have some familiarity with them. What about?

daring pumice
#

Hi Mark,

I'm curious about the requirements for loading a hex manually.

From what I understand, the process is as follows:

  1. Erase the chip
  2. Upload the softdevice (ANT/BLE stack)
  3. Upload the bootloader
  4. Upload the settings for the app
  5. Upload the app itself
  6. Reset the chip
#

When I run the make command for circuitpython, I only ever see step 5... could it be this is what I'm missing?

blissful pollen
#

Hmm I haven’t loaded from a hex myself only from the uf2 using the installed uf2 boatloader.

#

Is it possible for you to install the uf2 loader? I’m not sure I can be of much help offhand otherwise. Maybe if I looked at it a while

daring pumice
#

Sadly, no. Throwing the board into bootloader mode works, but the drive it exposes only has 128k - the uf2 file is ~460k

blissful pollen
#

Im not sure then unfortunately. If I get a moment I’ll take a look later today but not sure if I’ll find much on that.

onyx hinge
#

A .uf2 file is twice the size of a .bin file, but still a 460KiB uf2 file would not "fit" in a device with 128KiB of flash. Normally during the build process of circuitpython, a 'too big' flash file is detected and treated as an error, but perhaps that's not working in this case.

daring pumice
daring pumice
lone sandalBOT
#
FoamyGuy
[adafruit/awesome-circuitpython] Pull request opened: #36 Video embed

resolves: #33

This PR resolves the issue by specifically using an embedded `` tag for the video rather than relying on github markdown viewer to render the video.

I tested both on github on my branch from this PR, and on a local build of circuitpython.org/awesome page. Both are embedding the video correctly using this technique.

This is a better and more general approach to solve the problem than my first attempt from: https://github.com/adafruit/circuitpython-org/pull/792

orchid basinBOT
#
FoamyGuy
[adafruit/circuitpython-org] New comment on pull request #792: change with video URL to element on awesome page

Thanks for the review @makermelissa. I didn't look through enough of the MD file before, I didn't realize it was possible to embed HTML tags within it and have them get rendered. That is definitely a much better and more general solution than this javascript one.

https://github.com/adafruit/awesome-circuitpython/pull/36 presents a better solution to this by specifically using an embedded <video> tag in the markdown. Closing this in favor of that one.

manic glacierBOT
#
matthewjerome
[adafruit/circuitpython] Issue opened: #5619 Large Broadcom Ports Folder

CircuitPython version

When running `git submodule update --init`, the size is >8gb just for `circuitpython/ports/broadcom/firmware`

Is there a way to target a different repo or branch when building?

Code/REPL

git submodule update --init

Behavior

long module sync time, local dev image size required

Description

No response

Additional information

No response

manic glacierBOT
manic glacierBOT
#
ladyada
[adafruit/circuitpython] New comment on pull request #5615: Update espressif port

@microDev1 we tried to build but got this

οΏ½ESP-ROM:esp32s3-20210327f
Build:Mar 27 2021
rst:0x7 (TG0WDT_SYS_RST),boot:0x8 (SPI_FAST_FLASH_BOOT)
Saved PC:0x40048836
invalid header: 0xffffffff
invalid header: 0xffffffff
invalid header: 0xffffffff

maybe we did something wrong, do you mind also commiting a board we could try against like one of the ESP S3 eval boards? they all have 4MB flash and one has 2 MB PSRAM. it would help us with testing :)

manic glacierBOT
manic glacierBOT
lone sandalBOT
manic glacierBOT
manic glacierBOT
lone sandalBOT
manic glacierBOT
manic glacierBOT
jaunty juniper
#

it is

lone sandalBOT
#
FoamyGuy
[adafruit/awesome-circuitpython] New comment on pull request #36: Video embed

Thanks for the review @FlantasticDan, good catch on the width looking broken on mobile. The latest commit changes to use max-width 720px and width: 100%. I tested this change successfully with chrome in a small width window and Android phone with Firefox browser.

I see what you mean about the image at the top as well. In the interest of limiting the PR to one change I'll make a separate PR to apply a similar responsive size fix to the image.

lone axle
#

How would I go about testing a newer version of tinyusb that has already been merged on a given circuitpython device? Would it be just update submodules and then make a build from main branch in the circuitpython repo?

#

or, secondarily, is there a way to check the version of tinyusb on a given device / circuitpython build?

lone axle
#

I think I managed to do it with these:

cd lib/tinyusb/
git pull origin master

Then go back and make the build as normal. Still not sure if there is a way to verify. I suspect maybe not an easy one because there are newer commits in tinyusb (than where the repo sits after a submodule update in cp repo) but no release yet. So I'm maybe not technically on a new version, just happen to have some newer commits.

blissful pollen
#

What you said should work I'd assume. If there isn't any version of tinyusb stored i the code you could add a printf to CP somewhere to output it. Not elegant but may work

manic glacierBOT
#
FoamyGuy
[adafruit/circuitpython] New comment on issue #3152: bulk pasting code extremely flaky on nRF

I do think this is much better after https://github.com/hathach/tinyusb/pull/1208 :tada:

I tested for a bit with a CLUE pasting:

print(123456789123456789123456789123456789123456789123456789123456789123456789%11)

Over 2-3 dozen tests pasting that I had only 3 times where there was a difference in what got pasted in the REPL. With a current stock build of circuitpython the ratio of different pasted values felt much higher (I didn't record stats for it but felt more like 50% or s...

manic glacierBOT
manic glacierBOT
lone sandalBOT
#
FoamyGuy
[adafruit/awesome-circuitpython] New comment on pull request #36: Video embed

@FlantasticDan I tried in a different branch making the image responsive as well, but it seems there is some issue with the way Github renders the image. You're proposed change does fix the image on circuitpython.org/awesome page for mobile layouts. But on the github side the style gets overwritten to max-width 100% even though it's definitely coded to max-width: 400px in the file. Which results in that image getting stretched and a bit fuzzy on larger displays. Tried a few different ways...

manic glacierBOT
lone sandalBOT
#
FlantasticDan
[adafruit/awesome-circuitpython] New comment on pull request #36: Video embed

@FoamyGuy Thanks for looking into the image business, I thought it would be a one-liner. I looked at the commits you made in that branch on your repo and it seems that GitHub does some unique (and undocumented) markdown -> html generation for their rendered READMEs. Luckily it seems they don't mess with CSS functions, so I've opened #37 to resolve this. If you could review that I'd appreciate it.

manic glacierBOT
#
DietmarSchwertberger
[adafruit/circuitpython] New comment on issue #1136: Unable to use HID keyboard in boot OS \(macOS boot screen\)

@akaenner : On a quick look I did not see exception handlers. See PR #1929. Without, on my Windows PCs the software would not re-connect on re-boot or when the BIOS takes a few seconds before Windows is booting. If you have exception handlers, please point me to the positions, as I would like to update the PR with the actual exception types instead of catching all.

manic glacierBOT
manic glacierBOT
#
scath999
[adafruit/circuitpython] New comment on issue #5498: nRF52833 DK \(PCA10100\) CircuitPython UF2 does not work on Adafruit\_nRF52\_Bootloader

Update:

While trying to flash firmware.hex over to the board using nrfjprog, I get the following errors:

`
d:\CircuitPython\circuitpython\ports\nrf\build-pca10100>nrfjprog -f NRF52 --program firmware.hex -erasesector --verify --log
Parsing image file.
Verifying programming.
ERROR: [ nRF52] - Data does not match in address range [0x00027000-0x00027100] (Flash)
ERROR: [ nRF52] - Expected byte value 0x02 but read 0x00 at address 0x00027002.
ERROR: [ nRF52] - Flash verification fai...

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

I used the drag/drop feature of the JLink that's embedded with the PCA10100. I didn't use the command line.
What you provide as command line is indeed progress!

From the address, it seems to fall within the flash address space.
I don't know if the memory map is similar to the nrf52832/40 (at the bottom), but if it is, this would fall well within the application space (See [here](https://learn.adafruit.com/introducing-the-adafruit-nrf52840-feather/hathach-memory-map?gclid=CjwKCAiAqIKNBhA...

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

Found a good nugget of information in the nrf makefile here

It does the flash in multiple steps, including an overwrite of a bit to bypass the validation check. I suspect this is something we don't do with the UF2 or the drag/drop of the Hex file; making the bootloader reject circuitpython. However, that doesn't explain the verify error @scath999 found.

@scath999 If you can figure out the parameters passed ...

lone sandalBOT
manic glacierBOT
manic glacierBOT
#
FoamyGuy
[adafruit/circuitpython] New comment on issue #4867: Raspberry Pi Pico hanging when writing microcontroller\.nvm

I got a backtrace of this issue from a Raspberry Pi Pico using JLink and GDB. I'm not entirely sure what it means or what to do with the information I found but here is a paste of the backtrace in-case some else a little bit more knowledgeable knows what is going on "under the hood here".

Upon crashing the first backtrace I saw was this:

Program received signal SIGTRAP, Trace/breakpoint trap.
data_cpy () at sdk/src/rp2_common/pico_standard_link/crt0.S:275
275         blo data_cpy_l...
manic glacierBOT
#
jpconstantineau
[adafruit/circuitpython] New comment on issue #5498: nRF52833 DK \(PCA10100\) CircuitPython UF2 does not work on Adafruit\_nRF52\_Bootloader

I have done that already. Something is actually broken... Somewhere...
I get this when I do my build:

Traceback (most recent call last):
  File "/usr/local/bin/hexmerge.py", line 11, in <module>
    load_entry_point('intelhex==2.3.0', 'console_scripts', 'hexmerge.py')()
  File "/usr/lib/python3/dist-packages/pkg_resources/__init__.py", line 490, in load_entry_point
    return get_distribution(dist).load_entry_point(group, name)
  File "/usr/lib/python3/dist-packages/pkg_resources...
#
xobs
[adafruit/circuitpython] New comment on issue #5498: nRF52833 DK \(PCA10100\) CircuitPython UF2 does not work on Adafruit\_nRF52\_Bootloader

Progress! Thanks! However, nrfjprog within wsl isn't found... I know I have it on the windows side... I'll try adding it to the path but I suspect I got to fix my hexmerge first.

One trick you can use is add nrjfprog to your PATH somewhere (e.g. adding /mnt/c/Program Files/... to your PATH), then call it by running nrjfprog.exe instead of nrjfprog.

You may also be able to symlink the .exe into your wsl root to get it to work. Since the interfaces are identical, the Makefile wo...

manic glacierBOT
manic glacierBOT
hidden oxide
#

There is a meeting on Monday, right

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

I tried that a few times yesterday with no luck. I am thinking of setting
up a script to setup the full environment on wsl/Ubuntu so that it's
replicable...

On Fri., Nov. 26, 2021, 23:35 scath999, @.***> wrote:

Seems like it's not installed, or not installed properly.
I'd try pip3 uninstall intelhex and see what happens.
Pending that, pip3 install intelhex?

β€”
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<https://git...

lone sandalBOT
manic glacierBOT
#
JaggerJo
[adafruit/circuitpython] New comment on issue #5461: USB HID Report \- USB Busy 
# boot.py

import time
import usb_hid

# https://stackoverflow.com/questions/36750287/two-byte-report-count-for-hid-report-descriptor
absolute_mouse = usb_hid.Device(
    report_descriptor=bytes(
        # Absolute mouse
        (0x05, 0x01)  # Usage Page (Generic Desktop)
        + (0x09, 0x02)  # Usage (Mouse)
        + (0xA1, 0x01)  # Collection (Application)
        + (0x09, 0x01)  # Usage (Pointer)
        + (0xA1, 0x00)  # Collection (Physical)
        + (0x85, 0x...
manic glacierBOT
manic glacierBOT
manic glacierBOT
#
scath999
[adafruit/circuitpython] New comment on issue #5498: nRF52833 DK \(PCA10100\) CircuitPython UF2 does not work on Adafruit\_nRF52\_Bootloader

Nordic's documentation may possibly point to the issue, which @jpconstantineau's previous comments raised: The application spaces for nrf boards are all different, and change depending on which SoftDevice is deployed.

NRF 833
S113 v7.0.x softdevice: The soft device address range is 0x0000 1000 to 0x0001 C000 (148kB). The application area is then 0x0001 C000 to 0x0007 8000 (368kb).

S140 v7.0.x softdevice: The soft device address range is 0x0000 1000 to 0x0002 6000 (148kb). The a...

manic glacierBOT
#
deshipu
[adafruit/circuitpython] New comment on issue #5476: Re\-initializing a displayio display leads to a crash

I managed to get the pybadge unlocked.

This seems to be related to backlight handling, as it seems to be stuck in this loop:

Program received signal SIGTRAP, Trace/breakpoint trap.
0x00013f98 in common_hal_pwmio_pwmout_set_duty_cycle (
    self=self@entry=0x20000840 <displays+184>, duty=duty@entry=65535)
    at common-hal/pwmio/PWMOut.c:309
309	        while (tc->COUNT16.SYNCBUSY.bit.CC1 != 0) {
manic glacierBOT
#
jpconstantineau
[adafruit/circuitpython] New comment on issue #5498: nRF52833 DK \(PCA10100\) CircuitPython UF2 does not work on Adafruit\_nRF52\_Bootloader

@scath999 Regarding virtual... I build on windows using WSL. I now have this repo to script the creation of the WSL environment, setup the dependencies, clone the repo, setup the python dependencies, then do a build of the pca10100.

That new broadcom firmware module is a beast! at 1 GiB and only 5% done. Takes forever to download...

Is WSL ultimately running in a virtual machine? I guess so... I am still stuck at trying to get ...

hidden oxide
dusk mauve
#

finally back to work on some driver PRs and it looks like Github's all-500's. 😦

#

After years working on big SaaS web apps it feels kinda ... sad to watch a site fail in real-time. sadPanda

#

Tried a git pull on a repo cloned 30 min ago and, bleh, 504 error. Guess that means we should all just enjoy a weekend day huh? πŸ˜‚

lone axle
#

Yep, looks like Github is having some issues from my end as well. Pretty wide spread outtage I guess.

dusk mauve
#

I feel bad for their SREs and/or ops team. holiday weekend failures are the worst. 😦 worked way too many 36-hour incidents in my day.

#

The status page was all-green until a minute ago, so at least the pagers are going off.

https://www.githubstatus.com/

GitHub Status

Welcome to GitHub's home for real-time and historical data on system performance.