#circuitpython-dev

1 messages · Page 360 of 1

still zephyr
#

Yes I did not know the name but for sure I used this in the past, do you thnk that we could use the same for CircuitPython?

vague latch
#

yes

still zephyr
#

Oh I am interested

vague latch
#

i know i can do it by swapping the kernels out but want to have it all with the one kernel

still zephyr
#

yes, with different magics

#

I see

#

now I understand 🙂

vague latch
#

so i was thinking having only magics to do python side things

#

then non magics for circuitpython

manic glacierBOT
still zephyr
#

Yes, I like the idea

vague latch
#

at the moment the magics look at softreset and upload_delay. I was thinking just adding python or python3 to the list in kernel.py

manic glacierBOT
still zephyr
#

interesting, if you need help let me know, as I said I want to use use CP with the Jupyter notebook.

#

is the nbviewer working in the web now?

vague latch
#

i havent looked

grave ledge
# vague latch then non magics for circuitpython

My Feather RP2040 arrives next week and I have an nbdev notebook with CircuitPython code (running now on Blinka on a RPi3) ready to switch to a CP kernel. I'd be happy to test things for you.

manic glacierBOT
vague latch
#

what i dont know at the moment is. How can we read a line of code in and run it as normal python

#

We catch the python code with the kernel, how do we just send it back to the standard python interpreter

still zephyr
#

The %magics, I do not think there is a way to parse the code within the cell.

manic glacierBOT
cedar moth
#

@dhalbert I see the clock stretch timeout fix going through already... that was fast! Thank you!

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

I reported the same issue in the adafruit forums ("Magtag - ConnectionError: Unknown failure 205") a few weeks ago.

In my case, when running the magtag wifi test program, or any other sample programs (from the adafruit learn circuitpython example), it can see several access points in my area. However when attempting to connect to my local wifi network, all I get is the 205 error after a 10-15 second delay. Contents of "secrets" file are correct. I have NEVER been able to connect to my netw...

tulip sleet
#

it was basically just a typo

manic glacierBOT
#
jepler
[adafruit/circuitpython] New comment on issue #4260: RP2040 \- RTC \- time\_source not working

@jerryneedell can you test this again? I think maybe this was #4379 but I didn't closely read this issue.

I think there may be a misunderstanding or an opportunity to clarify the documentation. rtc.RTC() is always the on-chip RTC, so using set_time_source() doesn't change it. However, it does change what time.time() and time.localtime() (with 0 args) use as their source of the time.

So it looks like in all your prints, what you saw was only rtc.RTC() misbehaving, and pos...

cedar moth
tulip sleet
#

i'll be merging it in a few minutes when it finishes building, so it will be available shortly and in the next beta

cedar moth
#

Great, I will give it a shot later tonight then... have a good evening.

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

If you haven't already, try connect in a loop with try/except. After connecting the first time, it's a little more resilient to re-establishing the connection if it fails.

Again, if you haven't already, try to connect to an alternate Access Point, and see if proximity to the AP makes a difference (a scan in CircuitPython will show RSSI for each AP; unfortunately a connection is probably needed to see the RSSI of the device from the AP perspective).

vague latch
#

@still zephyr i got somewhere. I have got a inline magics running python code

#

however it must be inline and not cell. I will look further. But i was successfully able to import nbdev

still zephyr
#

@vague latch That is amazing, so now when could run CP and python in the same file? I would like to take a look at that code

vague latch
#

yeah, it isnt actually all that hard. As the circuitpython kernel is still running python and just packaging up the messages i just needed a way to block and execute the message before sending it to the serial port

manic glacierBOT
manic glacierBOT
vague latch
#

Does anyone know an alternative to exec() in python

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

You want to have a look at the descriptors we use for https://github.com/keyboardio/KeyboardioHID
The idea is that you should offer up both a Boot Protocol descriptor and a bitmap descriptor.

On Thu, Mar 11, 2021, at 1:18 AM, akaenner wrote:

I also need a keyboard that works at boot time. I'm currently working on a software for a DIY split keyboard with each part having a Raspberry Pi Pico in it. All features I planned are now implemented and I will share the code on GitHub in the comm...

manic glacierBOT
#
jerryneedell
[adafruit/circuitpython] New comment on issue #4260: RP2040 \- RTC \- time\_source not working

@jepler This si what I get now -- is it the expected result?

Adafruit CircuitPython 6.2.0-beta.4-78-g8d3217cbf on 2021-03-12; Raspberry Pi Pico with rp2040
>>> import gps_time_source
Set GPS as time source
No time data from GPS yet
Fix timestamp: 03/12/2021 10:39:18
RTC timestamp: 01/01/2020 00:04:20
Local time: 03/12/2021 10:39:18
Fix timestamp: 03/12/2021 10:39:19
RTC timestamp: 01/01/2020 00:04:21
Local time: 03/12/2021 10:39:19
Fix timestamp: 03/12/2021 10:39:20
RTC tim...
vague latch
#

@still zephyr and others. I have the first alpha/beta version of python magics working. The code takes in %python and %%python magics and runs the code following. I have included a small example in the examples section, this will be further fleshed out. At the moment i have two issues that i need to work through. The magics dont print the result to the jupyter notebook and i have to figure out how to keep a scope from any %python or %%python command if it is necessary. Otherwise take a look at it https://github.com/qwertimer/circuitpython_jupyter_kernel

Finally i can say that with this functionality i can run nbdev and produce code.py files from the circuitpython script. I will add the example of this shortly

GitHub
qwertimer/circuitpython_jupyter_kernel

Jupyter kernel to interact with CircuitPython board over USB - qwertimer/circuitpython_jupyter_kernel

manic glacierBOT
vague latch
manic glacierBOT
#
dhalbert
[adafruit/circuitpython] New comment on pull request #4387: Esp32\-S2 I2C fixes

@dhalbert I think we can get the idf patch in v4.3 since its a pre-release at the moment.

I agree, it would be good to PR upstream, but I am still somewhat bothered by the whole thing because I don't see why the IDF fix works. I did some more experiments yesterday and re-read a bunch of our code and their code. I came to no new conclusion. I haven't gotten feedback from igrr about it. Also I inquired with FreeRTOS: https://forums.freertos.org/t/vsemaphoredelete-immediately-after-xsemap...

manic glacierBOT
#
jepler
[adafruit/circuitpython] New comment on issue #4260: RP2040 \- RTC \- time\_source not working

The "RTC timestamp" line is now consistently advancing by about 1 second per second. I don't see where your program set the current time of the rtc.RTC() object by assigning its datetime property, so it's also expected that it starts counting from some arbitrary time in the past. So, yeah, it looks right to me. Is there something we can improve about the docs so that you think others would understand why this behavior is the anticipated behavior?

manic glacierBOT
lone sandalBOT
manic glacierBOT
#
dhalbert
[adafruit/circuitpython] New comment on pull request #4387: Esp32\-S2 I2C fixes

No clue at the moment, I'll try to replicate this behavior in an IDF program.

When I look at the multiple threads running in a "crashed" program vs one running this fix, they look the same. In other words, it is not a hard crash, but some kind of loop or other problem. The clue with my simple tests above is that the REPL stops working, and eventually disconnects. So perhaps some program that does some USB serial activitiy would hang up. Perhaps also some wifi or I2C activity would fail...

#
dhalbert
[adafruit/circuitpython] New comment on pull request #4387: Esp32\-S2 I2C fixes

So the essence of the above is:

// From busio.I2C().
i2c_driver_delete(NULL);   // does nothing
i2c_param_config();
i2c_driver_install();
// i2c.deinit().
i2c_driver_delete();

// Below from common_hal_wifi_init() and common_hal_wifi_radio_set_enabled().
esp_netif_init();               
esp_event_loop_create_default();
 xEventGroupCreateStatic();
esp_event_handler_instance_register();
esp_event_handler_instance_register();
esp_wifi_init();
esp_wifi_set_mode();
esp_wifi_s...
manic glacierBOT
#
Gadgetoid
[adafruit/circuitpython] New comment on pull request #4315: RP2040: Implement short I2C writes \(2 bytes or less\) using bitbangio

Not got as much testing done as I'd like.

Using the raspberry_pi_pico port compiled with this change:

  • [x] msa301_tap_example
  • [x] msa301_simpletest
  • [x] bme280_normal_mode
  • [x] bme280_simpletest

PA1010D remains a mystery, it seems to perform a few writes and then lock up, attempting to init I2C again (even after a soft reset of the RP2) will fail with a pull-up error.

However, to add pure unadulterated mystery this problem...running i2c.scan() between reads will unlock...

#
dhalbert
[adafruit/circuitpython] New comment on pull request #4315: RP2040: Implement short I2C writes \(2 bytes or less\) using bitbangio

You're not going to like this, but I replaced the whole I2C read implementation with bitbangio and now reads from my PA1010D are rock solid.

That it works with bitbangio is expected, and that's the workaround for devices that don't work. Just use bitbangio.I2C() instead of busio.I2C() for such devices. We don't want to throw out the hardware I2C completely because it's more efficient, etc. It appears to be a hardware or SDK peculiarity, but I need to test the non-working devices...

#
Gadgetoid
[adafruit/circuitpython] New comment on pull request #4315: RP2040: Implement short I2C writes \(2 bytes or less\) using bitbangio

Had to freeze outside to get the teeny GPS to fix, but with the above (albeit very nuclear) patch:

========================================
Fix timestamp: 3/12/2021 16:45:03
Latitude: 52.XXXXXX degrees
Longitude: 0.XXXXXX degrees
Fix quality: 1
# satellites: 5
Altitude: 15.3 meters
Speed: 0.46 knots
Track angle: 0.0 degrees
Horizontal dilution: 1.86
Height geo ID: 47.0 meters
========================================
Fix timestamp: 3/12/2021 16:45:04
Latitude: 52.XXXXXX de...
still zephyr
manic glacierBOT
hollow gazelle
manic glacierBOT
rose cargo
#

Trying to create a socket on CPY for my feather S2, but I get an error when I try to try to issue either a "bind" or "listen" and the doc says they should be there, but when I do dir(socketpool.Socket) they are not there? What's up with that?

idle wharf
solar whale
tulip sleet
#

I think it just draws as much power as it can. It has a constant-current drive circuit internally, but the pin will only put out so much. It doesn't fry the pin because of the pin circuitry.

solar whale
#

AH -- That makes sense -- thanks.

#

A question about it s power source came up in another channel #help-with-arduino and I wanted to make sure I gave the correct information.

tulip sleet
manic glacierBOT
#
Gadgetoid
[adafruit/circuitpython] New comment on pull request #4315: RP2040: Implement short I2C writes \(2 bytes or less\) using bitbangio

Yeah PA1010D is an odd duck, I think I ended up referring to some Adafruit code to discover that it emits spurious /r or /n or something to that effect, totally breaking sentences if you don't treat only the combination of both as a valid line ending.

In a mostly SMBus world, these low-level i2c implementations always feel weird.

PA1010D results.

Hardware I2C fails miserably:

ret = i2c_read_blocking(i2c0, addr, rxdata, 10, false);

PIO "bitbang" works fine:

...

idle owl
#

Hello, all! A reminder to please send any CircuitPython/Python on hardware related projects and news to cpnews@adafruit.com to be featured in the newsletter. We're always looking for more!

manic glacierBOT
#
dhalbert
[adafruit/circuitpython] New comment on pull request #4236: sleep and wakeup for nrf52

@jun2sak: Thanks! I have started testing the code. Timed light sleep works exactly as I expected, both real and simulated. In the case of timed deep sleep, I am losing the serial REPL connection when it goes to simulated deep sleep even for a couple of seconds. On the ESP32-S2, this does not happen, so it will need investigation.

I'm testing on a CLUE board with an integral display, so I have not tried power measurements yet, since the display will swamp everything else.

One thing about...

rose cargo
#

Build for esp32s2 failed with following errors: Use make V=1, make V=2 or set BUILD_VERBOSE similarly in your environment to increase build verbosity.
IDF_PATH=/home/jobla/circuitpython/ports/esp32s2/esp-idf cmake -S . -B build-unexpectedmaker_feathers2/esp-idf -DSDKCONFIG=build-unexpectedmaker_feathers2/esp-idf/sdkconfig -DSDKCONFIG_DEFAULTS="esp-idf-config/sdkconfig.defaults;esp-idf-config/sdkconfig-opt.defaults;esp-idf-config/sdkconfig-16MB.defaults;boards/unexpectedmaker_feathers2/sdkconfig" -DCMAKE_TOOLCHAIN_FILE=/home/jobla/circuitpython/ports/esp32s2/esp-idf/tools/cmake/toolchain-esp32s2.cmake -DIDF_TARGET=esp32s2 -GNinja
CMake Error: The source directory "/home/jobla/circuitpython/ports/esp32s2/build-unexpectedmaker_feathers2/esp-idf" does not appear to contain CMakeLists.txt.
Specify --help for usage, or press the help button on the CMake GUI.
Makefile:257: recipe for target 'build-unexpectedmaker_feathers2/esp-idf/config/sdkconfig.h' failed
make: *** [build-unexpectedmaker_feathers2/esp-idf/config/sdkconfig.h] Error 1

#

Sorry about previous post...I did follow directions and run the appropriate pre-req commands for esp-idf. CMakeLists.txt is indeed missing.

manic glacierBOT
slender iron
#

@rose cargo what version of cmake do you have?

manic glacierBOT
#
tannewt
[adafruit/circuitpython] New comment on issue #4377: Feather RP2040 CIRCUITPY folder disappears after reset

Hi folks, please try this with builds after aec03a4. I found we were clocking the flash faster than it's rated for. I think this caused it to sometimes work if the clocks started up without overshooting their speed. If it did, it'd try to load code off the flash and it'd be corrupt. The corrupt code would cause a hard fault and things would stop. #4392 lowers the speed to 40mhz-ish which should be much more reliable even if the starting clocks overshoot.

echo jacinth
#

I'm having an issue on main with CircuitPython and the ESP32S2 port. I'm using the ESP32S2 WROVER. If I build the firmware using make BOARD=<board> all and then copy the firmware.uf2 to the bootloader drive, it boots into CircuitPython, but if I make BOARD=<board> clean and build with make BOARD=<board> all DEBUG=1 to get some debugging output, it spits out this and boots back into the UF2 bootloader:

E (929) esp_image: Checksum failed. Calculated 0xb3 read 0xff
E (935) boot: OTA app partition slot 0 is not bootable
#

I did notice the DEBUG .uf2 is roughly ~2-300KB larger than the non-DEBUG one, but that may be a red herring.

idle wharf
slender iron
#

@echo jacinth I don't usually have an all in my command

echo jacinth
slender iron
#

I usually do make BOARD=<board> DEBUG=1

#

it's been a little while since I built S2 so it could be a different problem

manic glacierBOT
#
jposada202020
[adafruit/circuitpython] New comment on pull request #4367: Draft I2C FeatherRP2040 Peripheral

@tannewt Thank you for reviewing this, and sorry for the delay reply.
2. You are right, I guess I never pushed the file changes including the TODO. the problem is in this line.
https://github.com/adafruit/circuitpython/blob/ed7cb6de7c0e31356739f2e824aeee2ba1769731/shared-bindings/i2cperipheral/I2CPeripheral.c#L63
The raised error is that this i2cperipheral_i2c_peripheral_type is unknown. I tried to change to the object but it did not work. To be honest I do not what to use here,
4. Agr...

manic glacierBOT
echo jacinth
slender iron
#

it works without DEBUG though?

echo jacinth
#

It does! Boots and everything, for the most part. I'm running into some bugs and want debug symbols, but DEBUG=1 simply doesn't boot at all. Checksum failed each time, and it kicks back to the UF2 bootloader.

onyx hinge
#

Is the DEBUG= firmware too big and we're missing the firmware size check or checking against a wrong value?

echo jacinth
#

That's my first guess, unless something is seriously wrong with the way debugging symbols are formed in the binary, but unfortunately I'm not familiar enough with the toolchain yet to make that kind of call.

idle wharf
#

I just built debug for the metros2 and it booted make BOARD=adafruit_metro_esp32s2 DEBUG=1 flash

#

Oh you're doing UF2...

#

Sorry... I always flash the bin file

echo jacinth
#

I've never been able to get flash to work directly, unfortunately. It also loops into a bootloader being unable to enter the main program. Starting to wonder if these modules are cursed, but I did get them from Digikey. 😅

manic glacierBOT
#
DavePutz
[adafruit/circuitpython] New comment on pull request #4262: Issue 4111 \- Implement pulseio\(pulsein\) for RP2040

Yes, git got me (again). Correct files should be uploaded now.

On Fri, Mar 12, 2021 at 4:11 PM Jeff Epler @.***> wrote:

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

In ports/raspberrypi/common-hal/pulseio/PulseIn.c
https://github.com/adafruit/circuitpython/pull/4262#discussion_r593479374
:

+uint16_t pulsein_program[] = {
+<<<<<<< HEAD

  • 0x4001, // 1: in pins, 1
    +=======

This looks incorrect, like a merge conflict was...

echo jacinth
#

Okay, so bizarre. Doing a full esptool.py ... erase_flash before flashing CircuitPython directly using make BOARD=<board> flash works, so I can make it boot that way. That's the first time it's worked for me.

But doing make clean followed by DEBUG=1 results in a similar bootloop. This one provides perhaps a better error message, at least:

E (140) esp_image: image at 0x2d0000 has invalid magic byte
W (147) esp_image: image at 0x2d0000 has invalid SPI mode 255
W (153) esp_image: image at 0x2d0000 has invalid SPI size 15
E (159) boot: Factory app partition is not bootable
...
E (590) esp_image: Image length 1444768 doesn't fit in partition length 1441792
E (597) boot: OTA app partition slot 0 is not bootable
E (602) esp_image: image at 0x170000 has invalid magic byte
E (609) boot: OTA app partition slot 1 is not bootable
E (614) boot: No bootable app partitions in the partition table
tulip sleet
#

alias flash-esp32s2='esptool.py --chip esp32s2 -p /dev/ttyACM0 -b 460800 --before=default_reset --after=no_reset write_flash 0x0000'

#

esptool.py is v3.0.

#

the older versions can be problematic or just not work

#

i am using the esptool.py in esp-idf/components/esptool_py/esptool/esptool.py

#

i have just been building the latest on main and flashing to a Metro ESP32-S2, so that's very similar to the setup you have. You might want to start with a fresh clone and do esp-idf/install.sh and then esp-idf/export.sh to make sure the toolchain is in fine shape

echo jacinth
#

Yep, I'm using the one that ships with the repo, i.e: circuitpython/ports/esp32s2/esp-idf/components/esptool_py/esptool/esptool.py, so that seems right.

$ which esptool.py
> /Users/aliced/Code/circuitpython/ports/esp32s2/esp-idf/components/esptool_py/esptool/esptool.py
$ esptool.py --chip esp32s2 -p /dev/cu.usbmodem01 erase_flash
> Chip erase completed successfully in 17.7s
$ esptool.py --chip esp32s2 -p /dev/cu.usbmodem01 -b 460800 --before=default_reset --after=no_reset write_flash 0x0000 build-starwitch_wifi_imu/firmware.bin
> Wrote 1510304 bytes (907780 compressed) at 0x00000000 in 17.1 seconds (effective 704.7 kbit/s)...

Gets this in the UART, which I think might be the real issue here:

E (590) esp_image: Image length 1444768 doesn't fit in partition length 1441792
tulip sleet
#

what is the board in your BOARD= ?

echo jacinth
#

It's my own custom board, but the behaviour seems to be identical among adafruit_metro_esp32s2 and adafruit_feather_esp32s2_nopsram as well.

tulip sleet
#

so if you try to load the adafruit_metro_esp32 .bin, it's too big? I'm rebuilding it now with debug and will get the numbers

echo jacinth
#

esp32s2, specifically, but yep.

tulip sleet
#

-rw-rw-r-- 1 halbert halbert 1493120 Mar 12 17:46 build-adafruit_metro_esp32s2/firmware.bin

echo jacinth
#

One sec, I'm doing a fresh wipe of my entire toolchain to weed out any gremlins and it's taking a bit to do the git submodules.

manic glacierBOT
tulip sleet
# 
esptool.py --chip esp32s2 -p /dev/ttyACM0 -b 460800 --before=default_reset --after=no_reset write_flash --flash_mode dio --flash_freq 40m --flash_size 4MB 0x0000 build-adafruit_metro_esp32s2/firmware.bin
esptool.py v3.0
Serial port /dev/ttyACM0
Connecting....
Chip is ESP32-S2
Features: WiFi, ADC and temperature sensor calibration in BLK2 of efuse
Crystal is 40MHz
MAC: 7c:df:a1:00:46:a3
Uploading stub...
Running stub...
Stub running...
Changing baud rate to 460800
Changed.
Configuring flash size...
Compressed 1493120 bytes to 898075...
Wrote 1493120 bytes (898075 compressed) at 0x00000000 in 12.8 seconds (effective 932.8 kbit/s)...
Hash of data verified.

Leaving...
Staying in bootloader.
#

debug uart:

ESP-ROM:esp32s2-rc4-20191025
Build:Oct 25 2019
rst:0x1 (POWERON),boot:0x0 (DOWNLOAD(USB/UART0/1/SPI))
waiting for download
ESP-ROM:esp32s2-rc4-20191025
Build:Oct 25 2019
rst:0x3 (RTC_SW_SYS_RST),boot:0x0 (DOWNLOAD(USB/UART0/1/SPI))
Saved PC:0x400288c1
waiting for download
echo jacinth
#

Yeah, it will upload just fine, but if I hit the reset pin to actually boot the program, it'll start looping.

tulip sleet
manic glacierBOT
echo jacinth
#

Yep... complete fresh toolchain, still get an error on reboot. Except now it's a new one:

E (611) psram: PSRAM ID read error: 0x00ffffff
E (616) cpu_start: Failed to init external RAM!

I saw this error before, but I'm also 100% certain these are WROVER chips with the PSRAM on them.

#

Do you figure it's a bad module at this point? I'm tempted to just assemble another board to see what the problem is.

tulip sleet
#

There is a PSRAMCS pin exposed on the module. I don't know what it does, but we leave it unconnected. Did you ground it or anything like that? I was going to ask two things: 1. Does it say "WROVER" on the can 🙂 2. check the "strapping pins" which control how it boots

echo jacinth
#

Yep, it does say WROVER, and now I'm looking at the datasheet, which does say IO26 is reserved... and I've apparently hooked it up to something on this revision. Doh.

tulip sleet
#

💡 👍

#

good luck cutting the traces or whatever 🙂

echo jacinth
#

Cutting that trace did indeed fix the bootloader issue! Thank for very much for your time and help.

However, now I've got a new problem, and I think it might be related to your work with the I2C bus on the ESP32S2 port. The following script gets the proceeding error:

import busio
import board

# Initialize and lock the I2C bus.
i2c = busio.I2C(board.IO36, board.IO35)
while not i2c.try_lock():
    pass

# Find the first I2C device available.
devices = i2c.scan()
while len(devices) < 1:
    devices = i2c.scan()
device = devices[0]

print('Found device with address: {}'.format(hex(device)))

E (391478) i2c: ../../esp-idf/components/driver/i2c.c:362 (i2c_driver_delete):i2c driver install error
#

This is actually what I was trying to get to this whole time. 😆

supple gale
#

I'm wondering if the current CPY build with DEBUG=1 exceeds the OTA partition on the 4MB ESP32S2 devices. I ran into this when i was attempting some tests and pushed if off to 'look at this later' file

manic glacierBOT
echo jacinth
#

@supple gale If it's not just me, then my bet is "yes." The PSRAM issue was unrelated, I think.

manic glacierBOT
supple gale
#

@echo jacinth I tried your code and it worked (with some changes for pins being used) and it worked. I am using some code from the I2C/wifi PR artifacts for the UnexpectedMaker FeatherS2

manic glacierBOT
#
johncblacker
[adafruit/circuitpython] Issue opened: #4394 socketpool\.Socket is missing getaddrinfo

Documentation includes:
socket.getaddrinfo(host: str, port: int) → Tuple[int, int, int, str, str]
Gets the address information for a hostname and port

Returns the appropriate family, socket type, socket protocol and address information to call socket.socket() and socket.connect() with, as a tuple.
But it has not been implemented on the esp32-s2 port. Just wanted to insert this as a reminder.

idle wharf
# 
I (53) boot: ESP-IDF v4.3-beta1-24-g0bbc721a6-dirty 2nd stage bootloader
I (53) boot: compile time 16:13:59
I (54) boot: chip revision: 0
I (57) boot.esp32s2: SPI Speed      : 40MHz
I (62) boot.esp32s2: SPI Mode       : DIO
I (67) boot.esp32s2: SPI Flash Size : 4MB
I (71) boot: Enabling RNG early entropy source...
I (77) boot: Partition Table:
I (80) boot: ## Label            Usage          Type ST Offset   Length
I (88) boot:  0 nvs              WiFi data        01 02 00009000 00005000
I (95) boot:  1 otadata          OTA data         01 00 0000e000 00002000
I (103) boot:  2 ota_0            OTA app          00 10 00010000 00160000
I (110) boot:  3 ota_1            OTA app          00 11 00170000 00160000
I (118) boot:  4 uf2              factory app      00 00 002d0000 00040000
I (125) boot:  5 user_fs          Unknown data     01 81 00310000 000f0000
I (133) boot: End of partition table
I (137) boot: Defaulting to factory image
E (142) esp_image: image at 0x2d0000 has invalid magic byte
W (148) esp_image: image at 0x2d0000 has invalid SPI mode 255
W (154) esp_image: image at 0x2d0000 has invalid SPI size 15
E (161) boot: Factory app partition is not bootable
I (166) esp_image: segment 0: paddr=00010020 vaddr=3f000020 size=446e8h (280296) map
I (248) esp_image: segment 1: paddr=00054710 vaddr=3ffd1220 size=048a8h ( 18600) load
I (254) esp_image: segment 2: paddr=00058fc0 vaddr=40028000 size=00404h (  1028) load
I (255) esp_image: segment 3: paddr=000593cc vaddr=40028404 size=06c4ch ( 27724) load
I (272) esp_image: segment 4: paddr=00060020 vaddr=40080020 size=fd314h (1037076) map

Hmm... My Metro is not happy with my debug build...

#

Seems like this is exactly what @echo jacinth was just talking about ...

echo jacinth
#

That looks like what I was getting, yep.

idle wharf
#

My additional variable is I'm on the M1 Mac...
I'm going to erase and flash again

#

yep... same thing.

#

And I'm on the updated-idf branch

# 
Partition table binary generated. Contents:
*********************************************
# ESP-IDF Partition Table
# Name, Type, SubType, Offset, Size, Flags
nvs,data,nvs,0x9000,20K,
otadata,data,ota,0xe000,8K,
ota_0,app,ota_0,0x10000,1408K,
ota_1,app,ota_1,0x170000,1408K,
uf2,app,factory,0x2d0000,256K,
user_fs,data,fat,0x310000,960K,
**********************************************
#

RETAIL Build

Compressed 1366480 bytes to 872394...
Wrote 1366480 bytes (872394 compressed) at 0x00000000 in 13.0 seconds (effective 837.9 kbit/s)...

ll build-adafruit_metro_esp32s2/*.bi
n
-rw-r--r--  1 patrickwalters  staff  1300944 Mar 12 16:24 build-adafruit_metro_esp32s2/circuitpython-firmware.bin
-rw-r--r--  1 patrickwalters  staff  1366480 Mar 12 16:24 build-adafruit_metro_esp32s2/firmware.bin

DEBUG Build

Compressed 1508704 bytes to 911808...
Wrote 1508704 bytes (911808 compressed) at 0x00000000 in 13.8 seconds (effective 872.2 kbit/s)...

ll build-adafruit_metro_esp32s2/*.bi
n
-rw-r--r--  1 patrickwalters  staff  1443168 Mar 12 16:28 build-adafruit_metro_esp32s2/circuitpython-firmware.bin
-rw-r--r--  1 patrickwalters  staff  1508704 Mar 12 16:28 build-adafruit_metro_esp32s2/firmware.bin
#

@supple gale I think you nailed it..

tulip sleet
manic glacierBOT
supple gale
#

it was unfortunate that it sent me down a wrong path when i was debugging something else. And DEBUG is my only method, unless i break out the JTAG stuff. Also unfortunate that the uf2 loader doesnt give any indication when the image wont fit.

manic glacierBOT
manic glacierBOT
#
jun2sak
[adafruit/circuitpython] New comment on pull request #4236: sleep and wakeup for nrf52

@dhalbert Thank you for your testing. In my environment I don't see the USB-lost problem you mentioned. In my test of timed deep sleep with USB connection, when CY executes alarm.exit_and_deep_sleep_until_alarms(), it displays following lines and stops.

Code done running.
Auto-reload is on. Simply save files over USB to run them or enter REPL to disable.

Press any key to enter the REPL. Use CTRL-D to reload.
Pretending to deep sleep until alarm, CTRL-C or file wr...
idle wharf
#

@supple gale and @echo jacinth This obviously is a hack, but I messed with the 4mb partitions and I can load debug now.
In ports/esp32s2/esp-idf-config/partitions-4MB.csv I edited the size of ota_0, the offset and size of ota_1

# ESP-IDF Partition Table
# Name,   Type, SubType, Offset,  Size, Flags
# bootloader.bin,,          0x1000, 32K
# partition table,,         0x8000, 4K
nvs,      data, nvs,      0x9000,  20K,
otadata,  data, ota,      0xe000,  8K,
ota_0,    0,    ota_0,   0x10000,  1608K,
ota_1,    0,    ota_1,  0x200000,  508K,
uf2,      app,  factory,0x2d0000,  256K,
user_fs,  data, fat,    0x310000,  960K,
supple gale
#

yep, more or less the same here. I'm using my UM FeatherS2 for now (w 16MB flash). Not really hacky as most arent using the ota_1 partition for anything.

manic glacierBOT
#
jun2sak
[adafruit/circuitpython] New comment on pull request #4236: sleep and wakeup for nrf52

@dhalbert On nRF REPL connection, I also have a similar problem. That is, when I connect my nRF board to a host PC, a serial console app (TeraTerm) cannot establish USB-serial connection for a while. After 30 seconds, the connection is established.

This problem had already been seen before I started modification for my sleep and wakeup.

After trying different releases of CY firmware, I found
adafruit-circuitpython-pca10059-en_US-6.0.0.uf2 OK
adafruit-circuitpython-pca10059-en_US-6....

echo jacinth
#

What's weird with my I2C scanner script is that it works on the standard SCL and SDA pins, but if I move them to SCK (IO36, for SCL) and MOSI (IO35, for SDA), it actually freezes CircuitPython entirely, not responding to Ctrl+C over serial. The drive continues to be accessible, though. The "detect valid pins" script does say that those two pins are a valid pair, so I'm not sure why it can't continue.

manic glacierBOT
#
jun2sak
[adafruit/circuitpython] New comment on pull request #4236: sleep and wakeup for nrf52

@nitz Thank you for your comments. I also was surprised and a bit disappointed to find that nRF RTC is not a real-time clock and does not keep running while System OFF. There are some retained registers but their bytes are limited and shared with several purposes, so I think there should be kind of a guideline or protocol among them. I don't know if there are already some.
As for the state-transition diagram, I drew it with LibreOffice Draw.

manic glacierBOT
manic glacierBOT
#
jposada202020
[adafruit/circuitpython] New comment on issue #321: Audit and fix native API for methods that accept and ignore extra args\.

@tannewt
If this is still open and valid after two yeas I am willing to work on this for the long run. I made a list, and there are approx 680 occurrences of the macros.
list is here https://github.com/jposada202020/mp_func_macro/blob/master/test.txt

Develop the code to verify this automatically will be good, but at first sight I am not seeing as a trivial thing, maybe that will change along the way. The easiest way is to create check rules parsing the MP_ and then verifying the docs....

solar whale
#

FYI -- I tried a DHT22 on a Raspberry Pi Pico now that pulseio has been added ... ```Adafruit CircuitPython 6.2.0-beta.4-121-g550eab2dd on 2021-03-13; Raspberry Pi Pico with rp2040

import dht_simpletest
Checksum did not validate. Try again.
Temp: 71.1 F / 21.7 C Humidity: 23.9%
Checksum did not validate. Try again.
Checksum did not validate. Try again.
Checksum did not validate. Try again.
Checksum did not validate. Try again.
Temp: 71.1 F / 21.7 C Humidity: 23.8%
Checksum did not validate. Try again.
Checksum did not validate. Try again.
Checksum did not validate. Try again.
Temp: 71.1 F / 21.7 C Humidity: 24.1%
Temp: 71.1 F / 21.7 C Humidity: 24.1%
Temp: 71.1 F / 21.7 C Humidity: 24.3%
Checksum did not validate. Try again.
Temp: 71.1 F / 21.7 C Humidity: 24.4%

#

I see in the PR adding pulseio that the DHT22 was tested early on. For my test I used GP22 and sampled every 2 seconds.

manic glacierBOT
#
ArielWolle
[adafruit/circuitpython] Issue opened: #4396 Digitizer HID

Hey I know that the digitizer is written into your HID devices but not enabled as it say here https://github.com/adafruit/circuitpython/blob/550eab2ddda780adb939693a51c8c07c4a540c36/tools/gen_usb_descriptor.py#L24.

I've been trying to build circuitpython with this feature enabled but can seem to get the rp2040 to build properly (with or without the changes). Whether its a uf.2 build or advice on how to get a digitizer working anything would be useful thanks!

manic glacierBOT
manic glacierBOT
manic glacierBOT
manic glacierBOT
still zephyr
lone axle
#

is there a place where all the built UF2 files from a PRs actions run are available to download?

#

I found the list of artifact links actually.

manic glacierBOT
#
FoamyGuy
[adafruit/circuitpython] New comment on pull request #4393: Speed up a bit more

I had the issue from #4377 on both of my Feather RP2040 devices. I tried out the build from the artifact of this PRs actions and can confirm it does seem to resolve the issue for me.

This is my boot_out after loading this version:

Adafruit CircuitPython 6.2.0-beta.3-144-g737063c19 on 2021-03-13; Adafruit Feather RP2040 with rp2040

I tested on both devices and have successfully reset several times with the button, as well as unplug / replug. I have about a dozen trials between b...

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

I've spent some time experimenting with location & orientation of the magtag and have FINALLY got it to connect to my wifi router.
It looks like it could possibly be signal corruption due to parts layout on the magtag pcb causing interference to the wifi signal.
Results of my testing in the adafruit forums ("Magtag - ConnectionError: Unknown failure 205").

manic glacierBOT
#
important0
[adafruit/circuitpython] Issue opened: #4400 Circuitpython code\.py not work anymore

My OS Arch linux
Item ItsyBitsy M0 Express

What do I did ?

https://circuitpython.org/board/itsybitsy_m0_express/

  1. Update > adafruit-circuitpython-itsybitsy_m0_express-en_US-6.2.0-beta.3.uf2
  2. Update > update-bootloader-itsybitsy_m0-v3.13.0.uf2

↓

Inside ItsyBitsy M0 Express

https://learn.adafruit.com/welcome-to-circuitpython/creating-and-editing-code

code.py

import board
import digitalio
import time
 
led = digitalio.DigitalInOut(board.D13)
led.direction = d...
manic glacierBOT
manic glacierBOT
amber isle
#

I was translating some strings from en to ger when i had an idea but i dont know where to put it, so maybe here is a good place:

I would love to see error codes added to cp!

When you have to google your error its quite hard if you have to translate the error to find something

manic glacierBOT
orchid basinBOT
manic glacierBOT
manic glacierBOT
#
dhalbert
[adafruit/circuitpython] Issue opened: #4401 ESP32\- S2: Investigate unexpected I2C / wifi problems

It is not clear why the fix in pr #4387 prevents i2c_driver_delete() from interfering with wifi. Rather than continue to discuss that in a closed PR, I'm opening a new issue about this for further discussion.

@dhalbert I think we can get the idf patch in v4.3 since its a pre-release at the moment.

I agree, it would be good to PR upstream, but I am still somewhat bothered by the whole thing because I don't see why the IDF fix works. I did some more experiments yesterday and re-read...

#
dhalbert
[adafruit/circuitpython] New comment on issue #4401: ESP32\- S2: Investigate unexpected I2C / wifi problems

No clue at the moment, I'll try to replicate this behavior in an IDF program.

When I look at the multiple threads running in a "crashed" program vs one running this fix, they look the same. In other words, it is not a hard crash, but some kind of loop or other problem. The clue with my simple tests above is that the REPL stops working, and eventually disconnects. So perhaps some program that does some USB serial activitiy would hang up. Perhaps also some wifi or I2C activity would fail...

#
dhalbert
[adafruit/circuitpython] New comment on issue #4401: ESP32\- S2: Investigate unexpected I2C / wifi problems

It prints >>> after the import wifi but then the REPL hangs.

Yes, I get the same behavior... it hangs and no backtrace. From the debug print I see that wifi starts initializing but stops abruptly.
This also happens on V4.3. I am going to replicate the sequence of code that is executed with these commands.

_Originally posted by @microDev1 in https://github.com/adafruit/circuitpython/issues/4387#issuecomment-797573191_

#
dhalbert
[adafruit/circuitpython] New comment on issue #4401: ESP32\- S2: Investigate unexpected I2C / wifi problems

So the essence of the above is (you can double check this):

// From busio.I2C().
i2c_driver_delete(NULL);   // does nothing
i2c_param_config();
i2c_driver_install();
// i2c.deinit().
i2c_driver_delete();

// Below from common_hal_wifi_init() and common_hal_wifi_radio_set_enabled().
esp_netif_init();               
esp_event_loop_create_default();
 xEventGroupCreateStatic();
esp_event_handler_instance_register();
esp_event_handler_instance_register();
esp_wifi_init();
esp_...
manic glacierBOT
#
dhalbert
[adafruit/circuitpython] New comment on issue #4401: ESP32\- S2: Investigate unexpected I2C / wifi problems

I worked some more on this today. I put some logging in before and after the call to esp_wifi_start() in Radio.c. What's interesting is that in the "fixed version", there is an extra logging entry after that call:

...
I (8988) wifi_init: tcp tx win: 2880
I (8998) wifi_init: tcp rx win: 2880
I (8998) wifi_init: tcp mss: 1440
I (11693178) wifi_init: WiFi IRAM OP enabled
I (11693178) wifi_init: WiFi RX IRAM OP enabled
I (11693188) common_hal_wifi_radio_set_enabled: > esp_wifi_sta...
solar basin
#

Where might one find PIO examples for CP?

manic glacierBOT
cedar lark
solar basin
#

Sweet! Thanks, I don't know why I didn't think to check learn!

cedar lark
#

@solar basin if you need more details check the appendix of the Get Started with MicroPython on Raspberry Pi Pico book for the pio syntax https://hackspace.raspberrypi.org/books/micropython-pico or chapter 3 of the official datasheet for the nitty gritty.

HackSpace magazine
Get Started with MicroPython on Raspberry Pi Pico — HackSpace magaz...

Raspberry Pi Pico is a new low-cost, high-performance microcontroller board with flexible digital interfaces. Microcontrollers are computers stripped back to their bare essentials. You don’t use monitors or keyboards, but program them to take their input from, and send their output to the input/output pins. Using these programmable connections, ...

tulip sleet
amber isle
#

Hmm i think i need to rephrase: if i get an error i dont understand i like to google it, since nearly everything on the internet is in english it is quite hard to find someone with the same problem. But if the error had a code attatched to the text i could google "circuitpython error 1234" it would be simpler.

#

Sure it is possible to translate the german error to english and search for it, but it is hard to get the exact same wording to find relevant results

#

For me personaly it is no problem, i use the english version because of that reason. But i did a bit of translating for the german version and realised that it could be a problem

stuck elbow
#

you can always search for the German error in the German translation file, and you will see the English error there then

jaunty juniper
#

I have encountered that kind of issue, not in CPY, but in general with error messages given by apps or operating systems. When trying to lookup info, or when trying to ask about it. "I'm getting an error, well it's in french but basically it says this"

manic glacierBOT
solar basin
#

@cedar lark thanks, I was already looking at the RPi Pico docs, I just wanted to compare examples. 😄

solar basin
#

Does anyone know how rel is encoded into the irq instruction?

#

The compiler doesn't know which state machine the code will run on at compile time, right? So it's calculated at runtime, right? Because you could choose to run a program on all SMs

hushed folio
#

Is there a helper that will convert strings into hid keycodes nevermind I just made a switch for it

manic glacierBOT
amber isle
manic glacierBOT
manic glacierBOT
tulip sleet
#

@hybrid scarab I am testing BME280 as I type this.

tulip sleet
#

@modest hazel We saw the PlayPiper announcement and makezine video - congrats! It looks like most of the code is closed source, except for your CPy helper library. Is that the long-term plan?

Your successful use of Web Serial for the REPL connection is encouraging. Is that code available? Having a generic USB terminal available from the browser would be great for many purposes. I have tried the sample terminal that Google supplies, but it doesn't play well with the REPL: it seems to keep disconnecting and reconnecting.

drowsy geyser
#

@tulip sleet Thank you for the CircuitPython build guide and your pointers a few days ago. Turns out I had an outdated version of the gcc arm cross compiler in my path. Fixed, and builds are progressing just fine. (Context: cast boundary error on make.)

manic glacierBOT
#
jepler
[adafruit/circuitpython] Pull request opened: #4403 bitmaptools: Add readinto

When reading uncompressed bitmap data directly, readinto can work much more quickly than a Python-coded loop.

On a Raspberry Pi Pico, I benchmarked a modified version of adafruit_bitmap_font's pcf reader which uses readinto instead of the existing code. My test font was a 72-point file created from Arial.

This decreased the time to load all the ASCII glyphs from 4.9 seconds to just 0.44 seconds.

While this attempts to support many pixel configurations (1/2/4/8/16/24/32 bpp; swapped w...

manic glacierBOT
#
bitboy85
[adafruit/circuitpython] Issue opened: #4404 Firmware configuration file

To customize the board i think a firmware configuration file could come handy.
The file should stay in the botsel part of the storage and should allow some basic configuration of the device.
The configuration should be applied on each firmware update so the configuration will always be the same even if the defaults in the firmware itself changes.

It could look like this
`
[GENERAL]
FileSystem="FS Label"

[USB]
USB_VID = 0x239A
USB_PID = 0x804C
USB_PRODUCT = "PyRuler"
USB_MANUFAC...

manic glacierBOT
#
DavePutz
[adafruit/circuitpython] Pull request opened: #4405 Issue 4237 \- Changed initial setting in program\_struct

Although I cannot find it documented anywhere; looking at _pio_find_offset_for_program() in sdk/src/rp2_common/hardware_pio/pio.c, it is checking if the program origin is >= 0 and, if so, using that origin to look for space. As a result, with the default origin of 0 then pio_can_add_program() will fail on the second attempt to add a program. This limits CP to only adding one program per PIO. This was the cause of issue #4237, where the third call to rp2pio.StateMachine() would fail. The fix...

#
FoamyGuy
[adafruit/circuitpython] New comment on pull request #4403: bitmaptools: Add readinto

I get this error when I attempt to make the uf2 for PyPortal with this branch:

❯ make BOARD=pyportal
Use make V=1, make V=2 or set BUILD_VERBOSE similarly in your environment to increase build verbosity.
QSTR updated
../../shared-module/bitmaptools/__init__.c: In function 'common_hal_bitmaptools_readinto':
../../shared-module/bitmaptools/__init__.c:365:35: error: comparison of integer expressions of different signedness: 'int' and 'size_t' {aka 'unsigned int'} [-Werror=sign-compare]...
manic glacierBOT
modest hazel
# tulip sleet <@!766963928679841824> We saw the PlayPiper announcement and makezine video - co...

Hi! Thanks! It's Piper's call ultimately, but I and Shree are both advocating for it. It's not currently because when we started a year ago, we started with a different internal project and just kept it closed while we developed it. And we've been so busy trying to get it finished an polished that we haven't had a chance to even have those conversations! At a minimum we need to go through the whole repo and add in all of the appropriate licenses! I'd really like to publish the web REPL we developed. It's actually pretty simple, and it's mostly a matter of carefully tracking states and using the paste-mode. There are a couple of weird things with a couple of characters that need to be compensated for, but nothing serious.

I have a repo for a in-browser terminal that I wrote when I was at Parallax that I adapted for web-serial use here: https://matzelectronics.github.io/uTerm/web-serial.html
It's a little rough and doesn't do any of the state management that the Piper Make project does, but it does work 🙂

Also, we realized the FileSystem Access API in Chrome is really good for writing files/firmware to a device, too - so that lets us provide a way to install CircuitPython right from the site.

marble hornet
#

are circuitpython dicts still unordered?
Did some searching in PRs, issues, and docs but have missed a mention about it if there is one. if not where should one PR a change to reflect that?

#

from testing on 6.1.0-rc it seems like they are but it's hard to test as if they aren't the ordered-ness could be coincidental...

fathom shell
#

I think the orderednes would be coincidental?

#

at least last time I checked but that was before 6.1

marble hornet
#

how would you suggest testing? I've been trying **kwargs and manually writing literal dicts and thee always mirror the entry order but as you said...

#

ah. nm, took fiddling but found a case that is out of entry order.

fathom shell
#

it was on my first try heh, I got lucky I guess.

manic glacierBOT
manic glacierBOT
tidal kiln
#

you can not rely on any kind of ordering for a dictionary, different implementations are free to do different things in terms of ordering

marble hornet
#

riiiiight🤦‍♂️ cp is 3.4 not 3.7

#

thank you

manic glacierBOT
#
bitboy85
[adafruit/circuitpython] New comment on issue #2791: Add option to suppress terminal output on display \(if present\)

I would go further and think this should be disabled by default. This behaviour is unexpected and a small display like the 128x32 shows only two and a half lines. Of course with no useful information.
If someone really needs this output it should be explicitly enabled.
Maybe like this:
display = adafruit_displayio_ssd1306.SSD1306(display_bus, width=WIDTH, height=HEIGHT, rotation=180, console=True)

manic glacierBOT
manic glacierBOT
#
kmatch98
[adafruit/circuitpython] New comment on pull request #4403: bitmaptools: Add readinto

This looks like a cool addition. I haven’t tried yet but have a few questions since I couldn’t follow all the code with all the bit operations.

  • Does this require that the dimensions of the destination bitmap and the file bitmap are the same?
  • If the color palettes are different bit sizes or number of colors, how does it handle this?
  • Also how do I get the palette colors from the file into a new pixel_shader palette object?

If I’m missing something important about the main goal of ...

manic glacierBOT
#
MisterD66
[adafruit/circuitpython] Issue opened: #4409 alarm for RP2040

I would love to see some kind of powersaving / sleep functions for the RP2040 / Raspberry Pi Pico.

I think that is one of the key features missing for CircuitPython on the new Pi Pico board are some Sleep funktions!
For me the reason choosing a Pi Pico over a ESP32 is the lower powerconsumption for projects needing no WiFi!

turbid radish
#

?serverinfo

digital shoreBOT
#
adafruit
Owner

adafruit#3230

Region

us-west

Channel Categories

8

Text Channels

58

Voice Channels

6

Members

28052

Roles

34

ID: 327254708534116352 | Server Created ·
fossil gorge
#

Woah, 28k!

ornate breach
#

Woah

manic glacierBOT
#
jepler
[adafruit/circuitpython] New comment on pull request #4403: bitmaptools: Add readinto

Does this require that the dimensions of the destination bitmap and the file bitmap are the same?

Yes, it always reads all width×height pixels. (oh, and rows of pixels always go from top to bottom, which is different from at least some BMP files)

If the color palettes are different bit sizes or number of colors, how does it handle this?

Whatever displayio_bitmap_write_pixel does.

Also how do I get the palette colors from the file into a new pixel_shader pale...

manic glacierBOT
#
microDev1
[adafruit/circuitpython] New comment on pull request #4362: Code formatting

Is there a way to check the formatting in the CI via pre-commit?

@tannewt I have added formatting check to pre-commit along with translations check.

did the above comment enable you to move forward...

@jepler Thanks! for digging into this. :)
The qstrdefs issue is resolved. I disabled the formatting on it.
// *FORMAT-ON/OFF* for C and # fmt: off/on for Python can be used to limit formatting.

#
jepler
[adafruit/circuitpython] New comment on issue #4311: ESP32S2 Saola: TouchIO pins cause adjacent ones to constantly trigger

I notice that the docs state

When the TouchIn object is created, an initial raw_value is read from the pin, and then threshold is set to be 100 + that value.

I doubt that all ports do exactly this, so the docs may need to be updated, but basically I think that for other platforms we've found that adding a constant to a value that is assumed to be the 'background value' was an effective but simple way to set the initial threshold.

#
dhalbert
[adafruit/circuitpython] New comment on issue #4311: ESP32S2: TouchIO default threshold is unsatisfactory

I notice that the docs state

When the TouchIn object is created, an initial raw_value is read from the pin, and then threshold is set to be 100 + that value.

I doubt that all ports do exactly this, so the docs may need to be updated, but basically I think that for other platforms we've found that adding a constant to a value that is assumed to be the 'background value' was an effective but simple way to set the initial threshold.

The threshold is in shared-bindings,...

#
jepler
[adafruit/circuitpython] New comment on pull request #4362: Code formatting

The problem is due to the added line in cmd_showbc.py. As far as I can tell, when the expected file has something like

########
  bc=\\d\+ line=155

it searches for a line that matches this regular expression, and then checks all the subsequent lines until the next ######## for also matching as regular expressions. However, line 155 has become line 166 due to the added line, and so forth

https://gist.github.com/8293402cffa2d8dfc3bbc80786988151 fixes this test by updatin...

manic glacierBOT
orchid basinBOT
tulip sleet
opaque abyss
#

hey can someone pluck me with the circuitpythonistas role? I figure tuned in enough notifications should be helpful for me to pay attention 🙂

lone axle
opaque abyss
#

Thank you! That was the big reason, I typically always miss them and just listen on YT afterwards, but it's quiet at the office today so I figured I'd give it a shot listening live :]

idle owl
#

@opaque abyss Welcome! Looking forward to possibly seeing you in today's meeting!

tulip sleet
# modest hazel Hi! Thanks! It's Piper's call ultimately, but I and Shree are both advocating ...

btw, I just tried https://matzelectronics.github.io/uTerm/web-serial.html and it can connect, but when I see Press any key to enter the REPL. Use CTRL-D to reload. and then type anything, it disconnects. I did get it to stay connected at one point, but I'm not sure why. This is the same behavior I saw with the Google example. It may be some odd DTR thing?? I imagine you have solved this in PlayPiper.

opaque abyss
idle owl
#

Me trying to enter hug reports while @slender iron updates the notes doc. type type type type ......... where did everything go...? oh right. scroll scroll scroll

slender iron
#

sorry about that! I figured now was better than the hour before the meeting

idle owl
#

It was like my ADHD was somehow made tangible... I would look away, and then everything I was typing was gone....

slender iron
#

lol, should be stable now

#

I can try to warn here first next time

idle owl
#

I really should have expected it. I wouldn't be too concerned about warning folx.

#

It happens to all of us at some point 😄

fossil gorge
#

Scroll scroll scroll certainly beats a "This file was modified, would you like to reload it? Clicking 'Yes' will lose all your changes"

idle owl
#

@slender iron @onyx hinge FYI I can't take notes next week because there is a scheduled power outage and I will likely be tethered and working from my laptop only. So Jeff is apparently hosting, Scott if you could take notes or arrange for Melissa to do it, that would be appreciated.

slender iron
#

Ya, I can do it

idle owl
#

Cheers.

#

Not happy about this at all. They're repairing or installing something that requires the entire building to be offline. Including elevators.

fossil gorge
#

Yuk. I hope it's not too cold, or that heating isn't fully dependent on electricity

idle owl
fossil gorge
idle owl
manic glacierBOT
fossil gorge
orchid basinBOT
onyx hinge
#

wait, I'm hosting?

#

@slender iron I can, I just .. did I forget a whole week?

orchid basinBOT
onyx hinge
#

oh wait Kattni said next week.

idle owl
#

Yeah. I actually remembered something ahead of time for once.

idle owl
orchid basinBOT
idle owl
#

@tulip sleet Are there any CircuitPython boards that wouldn't have board.LED except for QT Py (and anything without a tiny LED)? Or is that pretty much standard?

tulip sleet
#

i think there are some that are just board.D13, or maybe board.L. We haven't been as consistent as we should be about the names. Also sometimes there's board.RED_LED and board.BLUE_LED or something like that, because there's more than one and we wanted to distinguish them. Maybe Feather nRF52840.

#

i'd have to look

idle owl
#

@tulip sleet Hmm ok. That's what I was worried about. D13 doesn't work for all of them, but if they all had LED, that would work. So I figured I would ask.

tulip sleet
#

yea, feather 840 is

{ MP_ROM_QSTR(MP_QSTR_L), MP_ROM_PTR(&pin_P1_15) },
    { MP_ROM_QSTR(MP_QSTR_RED_LED), MP_ROM_PTR(&pin_P1_15) },
    { MP_ROM_QSTR(MP_QSTR_D3), MP_ROM_PTR(&pin_P1_15) },

    { MP_ROM_QSTR(MP_QSTR_BLUE_LED), MP_ROM_PTR(&pin_P1_10) },
idle owl
#

hmph.

tulip sleet
#

so L or RED_LED

idle owl
#

😕 Bleh.

tulip sleet
#

it doesn't mean we shouldn't clean these up, but right now the consistency is not good

idle owl
#

Trying to deal with the templating stuff, and realising it's still limited because of that inconsistency.

#

How hard would it be to fix, do you think?

#

I guess I don't even know how many boards we have anymore 😄

tulip sleet
#

we have to make a pass over all the board pin.c's to add LED or L, and we have to decide

#

we have like 188 boards

#

or something ridiculous like that

idle owl
#

Oi fair enough.

thorny jay
#

Did you sleep one hour more or one hour less? This is the week where I will be two hours too early or two hours too late... Hopefully my notes are ready.

#

Is the meetin in 1H20 from now?

tulip sleet
#

spring forward, lost an hour, but I went to bed early

#

Yes, meeting is at 2pm New York time, so you can ask google what time it is in NY

idle owl
idle owl
thorny jay
#

Relative time is the best...

idle owl
#

@tulip sleet I think L is carryover from Arduino, isn't it?

#

LED is at least descriptive.

tulip sleet
#

yeah, I think so, at some point someone said "name it L", but I don't remember the circumstance; this might be a brief mtg topic

idle owl
#

Ok.

tulip sleet
#

for internal mtg

idle owl
#

Right. Will do.

manic glacierBOT
modest hazel
# tulip sleet btw, I just tried <https://matzelectronics.github.io/uTerm/web-serial.html> and ...

I hope we solved it, too! 🙂 That tool was one of the earliest proof-of-concepts and is not much more than Google's implementation. The one we are using for Piper Make is a bit more robust. I am curious, though - if you go to https://make.playpiper.com/?python=true&verbose=true and click New Project, then you can use the PYTHON tab at the bottom just like an IDE, and the CONSOLE tab is there too. The add-on to the URL is for debugging, so it won't save anything (changing the block code in any way will wipe out whatever python you type in). I'd be curious to know if you have the same sorts of connection issues there.

ornate breach
#

@modest hazel invalid ssl cert

#

might want to make sure that it's properly signed

#

a lot of browsers don't like self signed SSL

lone axle
#

There is a free service called "Let's Encrypt" that lets you get signed SSL certificates issued. https://letsencrypt.org/

modest hazel
thorny jay
#

Let's Encrypt is cool if you automate the renewal of certificate... or be ready for a problem every 3 months.

ornate breach
#

ah yes, that's better

modest hazel
#

Sorry about that!

ornate breach
#

but the web page keeps crashing in safari

fossil gorge
ornate breach
#

much better in chrome

modest hazel
#

Yeah - only the Chorme-based browsers support the APIs we need for it. I'm still working on the Safari specific problem.

analog bridge
#

can we get #4362 merged in first thing today to avoid conflicts
the conflicts shouldn't be that big of a deal but I would like to keep the pr clean due to the scale of changes

#

@slender iron @onyx hinge ^^^

manic glacierBOT
slender iron
#

@analog bridge yup, that seems like a good idea to me

manic glacierBOT
#
jposada202020
[adafruit/circuitpython] New comment on issue #321: Audit and fix native API for methods that accept and ignore extra args\.

I did a preliminary review of some of the docs vs code:
I take a look at these libraries. order dictated by commit and oswalk function in python

  • touchio
  • rtc
  • alarm
  • display
  • pulseio
  • bitmaptools

I found that MP_DEFINE_CONST_FUN_OBJ_KW is used differently in depending on the library

bitmaptools: https://github.com/adafruit/circuitpython/blob/3cbff45f9aac5ea2855bbc888083d356a91c80fe/shared-bindings/bitmaptools/__init__.c#L243

bitmap: https://github.com/adafruit/c...

idle owl
#

brb.

slender iron
#

<@&356864093652516868> Meeting is about to begin

modern wing
#

Good afternoon all you wonderful folks 🙂

still zephyr
#

lurking text only

fossil gorge
#

Make sure you mark it in the doc @still zephyr

analog bridge
#

the time's changed?

fossil gorge
#

Greetings, good sir Mr. of Certainly!

modern wing
#

@fossil gorge @still zephyr You don't need to mark lurking anymore in the notes doc 🙂

slender iron
#

@analog bridge ya, the US changed timezones this weekend

cobalt grail
#

Oh, DST already?

lone axle
#

Daylight savings time change in the US moved it.

modern wing
#

Only if you have something to say, you can then add it to the notes.

cobalt grail
#

I will be lurking anyway. 🙂

sterile bronze
#

just lurking

onyx hinge
lone axle
onyx hinge
#

Self-promotion is totally legitimate too! Your project, or your friend's project!

modern wing
#

🍬 🥨 🥧

onyx hinge
#

breaking the glassmalloc ceiling

fossil gorge
#

yes

#

I think I've resolved my PTT issue. Again.

old smelt
#

lurking

ornate breach
onyx hinge
#

looks comfy

iu.png
ornate breach
#

the official chair of circuitpython?

onyx hinge
#

if you say so

sick creek
#

with blinka it could give some data too

#

how comfortable you are

modern wing
#

@ornate breach That actually sounds like a fun science experiment 🙂

ornate breach
#

i love science 🙂

fossil gorge
modern wing
#

I'm thinking a crust made from/similar to the Butter Waffle Cookies from Trader Joes.

manic glacierBOT
ornate breach
#

I made a cheesecase with a pretzel crust that was fabulous.

onyx hinge
solar whale
#

💉

slender iron
#

@onyx hinge want to finish reviewing the formatting PR for @analog bridge ?

lapis hemlock
#

Sorry, I am a bit late.

idle owl
#

@lapis hemlock No worries. Welcome!

onyx hinge
#

I think I rambled a bit 🙍

lapis hemlock
#

I totally missed the daylight savings.

manic glacierBOT
modern wing
#

An engineer is never late, nor are they early. They arrive precisely when they mean to.

slender iron
idle owl
#

@lapis hemlock We all wish we could miss it. In that it would be great if it didn't exist. 🙄

onyx hinge
#

oh yes belated hug report to @still zephyr for the doc improvements. Each small change makes it better

errant grail
tulip sleet
#

@solar whale if you are using Linux for BLE, the bluez library used underneath bleak and adafruit-blinka-bleio treats the battery service specially (big 😦 ), or at least it used to. This is fortunately not true for CircuitPython native or AirLift BLE.

onyx hinge
#

@errant grail thanks! Part of the WWVB signal are 2 bits of information about DST: 0 = DST not in effect, 3 = DST in effect, 2 = DST begins this UTC day, 1 = DST ends this UTC day. (I think, from memory). So in principle you don't need DST rules (aside from an "is DST observed at all" flag, and of course your standard time offset) .. HOWEVER, the bug was: when the UTC day changes (by counting up using the local RTC), but you don't have a new successful WWVB signal reception yet to get the DST flag value of 3, you still calculate whether DST is in effect while assuming "DST begins this UTC day", i.e., you revert to non-Daylight time..

#

so I WANT to keep using the bits from WWVB, and the fix is to change the DST flags from 2->3 or 1->0 when the UTC day changes..

solar whale
#

@tulip sleet Thanks -- First attempt is with an nrf52840 -- I'm sure I'll have questions...

onyx hinge
#

(oh and of course you assume which local hour DST begins or ends)

idle owl
#

@tulip sleet Is there even remotely a way to automate updating the board defs, or is that going to need to be manual across the board....(no pun intended, but I'll let it stand)

errant grail
idle owl
#

Basically I don't mind if manual is the answer, but I want to make sure I'm not wasting time doing it manually.

tulip sleet
lone axle
#

I wonder if something could be added to adabot to check the pins file and modify it if it doesn't contain the correct pin name?

onyx hinge
#

and of course we'll work through it with you here on Discord if you run into trouble with the new formatting checks!

idle owl
analog bridge
#

having pre-commit is highly recommended

idle owl
onyx hinge
#

we'll have to figure out how our mac and windows devs can set up the new stuff, there may be complications too

tulip sleet
#

@idle owl i would look for all pins.c with MP_QSTR_*LED*, MP_QSTR_L, and MP_QSTR_D13, and then edit those files.

idle owl
thorny jay
#

colorsys from Cpython would be great...

idle owl
thorny jay
#

And adapting example and learn guide to use that (or fancy LED)

still zephyr
#

Yes we are talking hee about styling, not the conversion functions

thorny jay
#

If you take rainbow example from Pimoroni for Pi, they use colorsys.

#

So having it work the same way would mean reuse of those example directly.

#

CPython way.

lapis hemlock
still zephyr
#

cool thanks

onyx hinge
idle owl
#

Nice!

lapis hemlock
#

Sorry, I don't have mike at the moment.

slender iron
#

np, I can read off your idea

lapis hemlock
#

Which other objects could benefit from adding the buffer protocol?

#

I can help with that.

#

We have ulab.array.tobytes, which is a buffer.

#

Right, I do.

idle owl
#

@onyx hinge Yeah, the latter lists are what I need. Ones that don't have LED. Thanks! Part of the issue is that some use a pin other than D13, but we'll stumble on them eventually. This is a perfect start.

lapis hemlock
#

I would call out to kmatch98 🙂

#

I am also open to adding tools on my side,

#

@thorny jay could you ping me in the issue?

mental nexus
solar whale
#

clone and use grep ....

lapis hemlock
#

Well, if there is a constraint, we have to find a workaround.

modern wing
onyx hinge
#

"git grep"

thorny jay
#

Seach was also important for German error message... 🙂

onyx hinge
#

it is a built-in sub-command of git

lapis hemlock
#

@onyx hinge OK, thanks!

#

Searching locally hinders over-the-web collaboration.

#

I can't just send a link to someone else.

#

Thanks!

#

Permalinks is what I meant.

mental nexus
#

@thorny jay one minor speed up for array functions is to replace the tuple indexing into a calculated direct index.

 image_bitmap[h, w] = inta[index]

Would be something like:

 image_bitmap[h * image_bitmap.width + w] = inta[index]

But please double check my math calculates the right index before using! For some reason accessing by tuple is slower than direct.

lapis hemlock
#

@mental nexus But that still iterates on inta, and that is slow.

analog bridge
#

Thanks! 🙂

onyx hinge
manic glacierBOT
#
dglaude
[adafruit/circuitpython] Issue opened: #4414 Fast copy from ulab to displayio\.Bitmap

Playing with ulab, I found it was costly to copy from an array to a display.bitmap.

ulab can build a buffer: ulab.array.tobytes

Here is the current fastest way I was able to copy:

image_bitmap = displayio.Bitmap(24, 32, number_of_colors)
inta=ulab.array((npframe-min_t)*factor,dtype=ulab.int8)
index = 0
for h in range(24):
    for w in range(32):
        image_bitmap[h, w] = inta[index] 
        index+=1
onyx hinge
#

so maybe the answer is find us 10,000 more stars on github

lapis hemlock
#

Only 7700 are missing.

#

It should be doable.

modern wing
#

Thanks everyone!

gilded cradle
#

Thanks

still zephyr
#

Thanks

mental nexus
#

Thanks all!

thorny jay
#

Thanks.

lapis hemlock
#

@modern wing Thanks for playing the court jester!

modern wing
#

@lapis hemlock ?

lapis hemlock
#

@modern wing What is exactly the question?

fossil gorge
idle owl
#

Dropping off. Thanks folx!

modern wing
#

@lapis hemlock Court jester?

lapis hemlock
#

You always tend to pop jokes, that is what I meant.

#

Good ones, by the way.

thorny jay
#

That fail for me. Because the bitmap is two dimension.

idle owl
#

@slender iron Please ping me when you put in the PR for the newsletter updates today.

slender iron
#

👍

onyx hinge
#

@ionic elk also the request for you to review that pwmio came via the internal meeting

#

signing off, thanks all!

fossil gorge
#

SourceGraph is nice, not gonna lie. I don't know what their licensing/pricing is like though. There's some free "up to a point"

manic glacierBOT
thorny jay
#

Is this deep dive in the Voice channel? 🙂

#

Deep Dive without the cat cam I guess.

fossil gorge
#

lol

cobalt grail
#

I do enjoy the deep dive.

solar whale
#

Have a good week. Dog is demanding a walk.... 👋

idle owl
#

@onyx hinge Does git ls-files -z "ports/*/boards/*/pins.c" | xargs -0 grep -l MP_QSTR_.*LED | sort > has-led.txt catch .RED_LED and .LED properly?

#

I mean this is a perfect start and we can catch the oddities as they come up, but I wanted to make sure.

#

Because there is at least one board with RED_LED and BLUE_LED.

slender iron
onyx hinge
#

@idle owl yes I think it should

idle owl
#

Right on.

onyx hinge
#

While the other two were exact word searches

idle owl
#

Right

#

I'll start with your lists. Better than I would have done, I'm sure. (I would had had help though.)

#

@onyx hinge Even if there's work on the core, as long as no one is editing pins.c, there won't be conflicts, right? I think that's right.

onyx hinge
#

Should only get conflicts when two people work in the same file

idle owl
#

Ok that's what I thought.

#

Rebasing still messes me up every time, so I try to avoid it.

onyx hinge
#

I hear you

idle owl
#

Ok, so here is my git status: ```1800 kattni@robocrepe:circuitpython [67m main *= 05ed179e1]$ git status
On branch main
Your branch is up to date with 'adafruit/main'.

Changes not staged for commit:
(use "git add <file>..." to update what will be committed)
(use "git restore <file>..." to discard changes in working directory)
(commit or discard the untracked or modified content in submodules)
modified: extmod/ulab (new commits)
modified: frozen/Adafruit_CircuitPython_CircuitPlayground (modified content)
modified: frozen/Adafruit_CircuitPython_FocalTouch (new commits)
modified: frozen/Adafruit_CircuitPython_NeoPixel (modified content)
modified: frozen/Adafruit_CircuitPython_Register (new commits)
modified: frozen/pew-pewpew-standalone-10.x (new commits)
modified: lib/mp3 (new commits)
modified: lib/protomatter (new commits)
modified: lib/tinyusb (new commits)
modified: ports/esp32s2/esp-idf (new commits)
modified: ports/raspberrypi/sdk (new commits)
modified: tools/usb_descriptor (new commits)

no changes added to commit (use "git add" and/or "git commit -a")```

#

I did not make those changes (except maybe to CircuitPlayground) and would like to not have to deal with them. Do I have to discard them individually? Or is the git restore listed above what I need to get rid of them?

#

I did a git pull adafruit main before this.

mental nexus
# thorny jay for i in range(768) image_bitmap[i] = inta[i]

That's strange. Here is a gist with the example of a "rotozoom" function written in python. It uses the calculated "direct" indexing into a bitmap: https://gist.github.com/kmatch98/27bc30076e015e8acd0ec7bae1af9c7a

Here it is at line 145:

                if c != skip_index:  # ignore any pixels with skip_index
                    # place the pixel color (c) into the destination bitmap at (x,y)
                    destination[
                        x + y * destination.width
                    ] = c  # direct index into bitmap is faster than tuple
                    # destination[x,y] = c
lapis hemlock
#

@mental nexus Adding the buffer keyword to the bitmap constructor would have the added benefit that you could "display" various things, e.g., the data in an ADC buffer.

idle owl
onyx hinge
#

@idle owl If you don't have any intentional changes to submodules, then first fix to try is "git submodule update" in the top directory

fossil gorge
lapis hemlock
idle owl
#

@onyx hinge oh, of course. sigh. 🤦

#

This whole thing? git submodule sync --quiet --recursive && git submodule update --init

onyx hinge
#

first try with the pre-commit code formatting check was unsatisfactory. It is not running on the right set of files.

#

entry: sh -c "git diff --staged --name-only | xargs python3 tools/codeformat.py"

#

if the git diff names no files, then tools/codeformat is invoked without any files, and it does a recursive operation including submodules and ignored files

slender iron
idle owl
#

@slender iron Thanks!

#

@onyx hinge That worked. It's only the two that I think I manually changed that are showing changes now. Thank you.

slender iron
#

Here is the notes document for Monday’s CircuitPython Weekly meeting. It is at the normal time of 11am Pacific / 2pm Eastern here on Discord. Everyone is encouraged to attend! Please add your hug reports and status updates even if you’ll be attending the meeting - it’s super helpful! If you are unable to attend but would still like to include updates, feel free to include them in the notes and we’ll read them off during the meeting. Hope to see you there! <@&356864093652516868> https://docs.google.com/document/d/11uksHbpF2KT8sBC9Bov9XLtSgM62HXAHUKVxKhjr5cE/edit?usp=sharing

Google Docs
CircuitPython Weekly for March 22nd, 2021

CircuitPython Weekly for March 22nd, 2021 Welcome to the CircuitPython Weekly meeting notes! Feel free to add your Hug Reports and Status Updates early. During the meeting, we go through them as a round robin sorted by username. If you can’t make the meeting and would still like to participate...

slender iron
onyx hinge
#

I think pre-commit has its own ways of knowing what files to work on, unfortunately this rule doesn't pay attention to it. Or maybe it's only for determining whether to run the check or not? In any case, I think the way it's written doesn't work with pre-commit run --all

ornate breach
#

I thought it was pre-commit run --all-files ?

#

Or is it the same command

onyx hinge
#

--all or --all-files, the thing is that this uses git to say 'what files are "staged for commit"' and run on just those files: entry: sh -c "git diff --staged --name-only | xargs python3 tools/codeformat.py"

#

so it won't run on main.c unless main.c is "staged", which is git parlance for "has a change and you've requested it to be commited with git add or other means"

ornate breach
#

Oh I see

onyx hinge
#

@analog bridge what version of uncrustify is required? I get a whole lot of messages which imply that maybe I have something wrong. ```$ python3 tools/codeformat.py main.c
/home/jepler/src/circuitpython/tools/uncrustify.cfg:41: unknown symbol 'disable_processing_nl_cont'
/home/jepler/src/circuitpython/tools/uncrustify.cfg:70: unknown symbol 'sp_do_brace_open'
/home/jepler/src/circuitpython/tools/uncrustify.cfg:73: unknown symbol 'sp_brace_close_while'
/home/jepler/src/circuitpython/tools/uncrustify.cfg:76: unknown symbol 'sp_while_paren_open'

fossil gorge
onyx hinge
# 
Uncrustify-0.68_f
``` this is the version in Debian stable (buster)
fossil gorge
#

Basically, this for the hook (with name, etc)

entry: /usr/bin/env bash -c
args: ['([[ ! -d "examples" ]] || for example in $(find . -path "./examples/*.py"); do pylint --disable=missing-docstring,invalid-name $example; done)']
language: system
#

So you put the call to the shell in the "entry" field, and the script/commands/etc in the 'args' array

onyx hinge
#

I don't think it's a quoting problem or anything like that, the command is simply .. not doing what is desired.

manic glacierBOT
#
jepler
[adafruit/circuitpython] Pull request opened: #4415 pre\-commit: only run uncrustify if changed files are detected

This adds the "no-run-if-empty" flag to the xargs invocation. Otherwise, if git diff names no files, the command is run once with no agument specified which leads to running on a default list of files that appears to include ignored files and files in submodules.

I am uneasy about how this works (it means that pre-commit run --all doesn't actually check all files) but that's a separate issue.

#
tannewt
[adafruit/circuitpython] New branch created: tannewt\-patch\-1
onyx hinge
#

OK, so to answer my own question you must have a SPECIFIC version of uncrustify, which is NOT IN DEBIAN OR UBUNTU, but only in a "ppa".

Get:6 http://ppa.launchpad.net/pybricks/ppa/ubuntu focal/main amd64 uncrustify amd64 0.71.0-0pybricks1 [547 kB]

thorny jay
onyx hinge
#

0.72 (debian testing) doesn't format the same either

slender iron
#

@ionic elk you could do sleep on the rp2040

idle owl
#

@thorny jay Thank you for posting an example to that colorsys issue.

onyx hinge
#

OK I remain SUPER puzzled, because code format is reformatting a bunch of files for me, but it's "clean" on github in actions. So either it's not running during CI as expected OR I still have not replicated the setup

#

I wish I had looked more closely before accepting the PR.

# 
Uncrustify-0.71.0_f
slender iron
#

@onyx hinge I think this inevitable with such a large change. It's worth biting the bullet and working through it IMO

onyx hinge
#

@slender iron I am willing to give the auto-code-formatting a try, but I need to be able to replicate the setup where it doesn't introduce 18,000 changes if I can expect others to do it as well

slender iron
#

true

#

want me to help you look at it too?

onyx hinge
#

if you're available that would be great.

#

I'm on Debian but I installed the specific (intended for ubuntu focal) version of uncrustify that the CI is using

slender iron
#

I'm in the valentina woice

onyx hinge
#

ok give me 2 minutes to get to my computer w/mic!

slender iron
#

kk

onyx hinge
#

ps uww | grep uncrust

# 
 static void reset_devices(void) {
-#if CIRCUITPY_BLEIO_HCI
+    #if CIRCUITPY_BLEIO_HCI
     bleio_reset();
-#endif
+    #endif
 }```
manic glacierBOT
onyx hinge
#

pre-commit run --all && git commit -a -m"run pre-commit"

# 
-        files = list_files(args.files)
+        files = list_files(args.files, EXCLUSIONS)
manic glacierBOT
#
mintakka
[adafruit/circuitpython] New comment on pull request #4411: added board configuration for Sparkfun ThingPlus RP2040

Yes it should be, I missed that!

On Mon, Mar 15, 2021 at 6:26 PM falkenad @.***> wrote:

@mintakka https://github.com/mintakka Shouldn't the flash size be 16 as
they're 25Q128 flash chips?
IE: #define TOTAL_FLASH_SIZE 8 * 1024 * 1024
Should be:
#define TOTAL_FLASH_SIZE 16 * 1024 * 1024 ?

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/adafruit/circuitpython/pull/4411#issuecomment-799795954,
...

#
kmatch98
[adafruit/circuitpython] New comment on issue #4414: Fast copy from ulab to displayio\.Bitmap

I'm interested in helping out on this and stretching my skills a bit to try to understand better how the bitmap structure is formed and modified. Sorry I didn't chime in during today's CircuitPython call, but I'm not very well versed in the terminology y'all use, so I have to do some digging to be partially intelligible when discussing this. Here goes:

As best I understand, to save memory the bitmap structure uses the minimum number of bits necessary to hold the maximum palette indices (...

onyx hinge
# 
args.files= ['shared-bindings/neopixel_write/__init__.c', 'drivers/wiznet5k/ethernet/socket.c', 'tests/cmdline/repl_basic.py']
files= ['/home/jepler/src/circuitpython/shared-bindings/neopixel_write/__init__.c']
manic glacierBOT
#
ArielWolle
[adafruit/circuitpython] Issue opened: #4417 Can't disable mass storage for the Pimoroni tiny 2040

<!-- Thanks! for testing out CircuitPython. Now that you have got a problem...
you can file a bug report for it. Feel free to modify the below format to better
suit your issue. :) --

Lastest firmware (git clone https://github.com/adafruit/circuitpython.git)

Error when building

Code/REPL

adding the line: CIRCUITPY_USB_MSC = 0
to /ports/raspberrypi/boards/pimoroni_tiny2040/mpconfigboard.mk results in a build error

Behavior

![image](https://user-ima...

still zephyr
#

I am seeing both USB ports in Adafruit CircuitPython 6.2.0-beta.3 on 2021-03-04; Adafruit Feather RP2040 with rp2040 that I just downloaded from the Webpage. Would be good to point the link to a build that does not have both ports.

lavish steppe
lavish steppe
still zephyr
#

oh no

#

@lavish steppe I am answering an issue in github related to the Jupyter notebook, and the problem is this

#

If you download the lastest firmware form the S3 you would solve this

onyx hinge
# 
True
# 
False
>>> Path("a/b/c.h").match("./*.h")
True
slender iron
# 
>>> fnmatch.filter(["a/c.h", "a/b/c.h"], "a/*.h")
['a/c.h', 'a/b/c.h']
manic glacierBOT
onyx hinge
#

@slender iron if the problem comes back ... it was also because of the running-in-parallel that the first time we put check-translate in pre-commit, some people got empty files sometimes.

#

.. I bet

manic glacierBOT
#
kevinjwalters
[adafruit/circuitpython] New comment on pull request #4405: Issue 4237 \- Changed initial setting in program\_struct

I've not used the C SDK. The documentation does not state what -1 would do: https://raspberrypi.github.io/pico-sdk-doxygen/group__hardware__pio.html#ga3571e3c7c6e645e92b8d438203a911b9 & https://raspberrypi.github.io/pico-sdk-doxygen/structpio__program.html (this is the matching data structure even though it's missing the _t suffix, the quantity is signed making -1 a legal value).

The actual source code can be viewed, I don't know if this matches the SDK in use but it does appear to d...

orchid basinBOT
onyx hinge
manic glacierBOT
#
mintakka
[adafruit/circuitpython] New comment on pull request #4411: added board configuration for Sparkfun ThingPlus RP2040

Thank you, I will make these changes and re-submit

On Mon, Mar 15, 2021 at 8:43 PM Scott Shawcroft @.***>
wrote:

@.**** requested changes on this pull request.

Thanks for the definition! Please rename the folder to
sparkfun_thingplus_rp2040 because thingplus is a name used for the form
factor. You'll also need to add the board to .github/workflows/build.yml

In ports/raspberrypi/boards/sparkfun_thingplus/mpconfigboard.h
<https://gi...

#
tannewt
[adafruit/circuitpython] New comment on issue #2791: Add option to suppress terminal output on display \(if present\)

I would go further and think this should be disabled by default. This behaviour is unexpected and a small display like the 128x32 shows only two and a half lines. Of course with no useful information.
If someone really needs this output it should be explicitly enabled.
Maybe like this:
display = adafruit_displayio_ssd1306.SSD1306(display_bus, width=WIDTH, height=HEIGHT, rotation=180, console=True)

I strongly disagree. It's useful to know things are running and for most boards ...

onyx hinge
#

@slender iron do you think it's worth defining a minimum number of WxH characters for a display to show the console? Right now you get the console on a 32x32 rgbmatrix and I think you can see an entire 2x2 character region of the console in all...

slender iron
#

I'm open to it but it's still nice to see its working

onyx hinge
#

we shift to showing the blinka logo if we can't show a single character, I think

#

idk, it's not a big deal to me either

manic glacierBOT
#
jepler
[adafruit/circuitpython] New comment on pull request #4403: bitmaptools: Add readinto, arrayblit

I've now added arrayblit.

You can blit the whole bitmap:

>>> b = displayio.Bitmap(8, 8, 256)
>>> a = array.array('H', range(64))
>>> bitmaptools.arrayblit(b, a)
>>> print([b[i] for i in range(64)])
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63]

You can blit just part ...

manic glacierBOT
manic glacierBOT
#
jposada202020
[adafruit/circuitpython] New comment on issue #321: Audit and fix native API for methods that accept and ignore extra args\.

I propose this:

  1. I will review the shared bindings in order to verify, the class definitions, a python "lexer" house-made could help here to compare with the documentation and the enum, and mp_check..
  2. Review the docs at the same time to see if we need add usage examples or correct return types.
    whether the work is completed or not we close this in the summer 2021. I will do commits with the progress.
manic glacierBOT
manic glacierBOT
manic glacierBOT
#
wildestpixel
[adafruit/circuitpython] New comment on issue #4226: rp2040: implement rotaryio

Can confirm this now working in latest nightly builds (cannot build myself due to M1 silicon config error). Might be that assignation of encoder pins A & B is flipped. I have :

encoder = rotaryio.IncrementalEncoder(board.GP3, board.GP4)

working in correct orientation (clockwise increment, counter-clockwise decrement), however,

GP3 is connected to B and GP is connected to A.

Switch the wires, the encoder goes the wrong way, or code as

`encoder = rotaryio.IncrementalEncoder(b...

manic glacierBOT
#
bitboy85
[adafruit/circuitpython] New comment on issue #4404: Firmware configuration file

If it is possible to edit those settings within boot.py i'm fine with it. I thought, the configuration needs to be available at the time circuit python is installed/updated.
The idea behind the bot-sel storge was that this configuration cannot be accidentally deleted and it would be easier to deploy because you just need to copy the configuration an the firmwarefile to the same directory.

plucky flint
manic glacierBOT
onyx hinge
manic glacierBOT
blissful pollen
#

If the pre-commit check fails on a PR the PR should fail to build right? I think mine last night still ran the build after. Just don't want to slow down other builds accidentally

analog bridge
blissful pollen
analog bridge
#

I think a lot of CI time can be saved if we can somehow cancel previously running checks on a pr when new commits are pushed.

blissful pollen
#

Yup, I can imagine (especially for people like me who this is still somewhat new to)

#

So I may need help from someone here, my PR has been around a while (due to my lack of time) and pre-commit formatting is now failing on formatting some files that aren't even part of the change.

My guess is with the new formatting checks these were fixed already but I don't have them? Not sure my best course of action at this point

manic glacierBOT
manic glacierBOT
#
ThomasAtBBTF
[adafruit/circuitpython] Issue opened: #4420 Sockets do not work reliable on a ESP32\-S2 / carried over from "uPy socket vs cPy socket need help" topic in the forum on request of Dan

Firmware

With 6.2.0-beta.2 and 6.2.0-beta.3 on a MagTag

Code/REPL


import sys
import ipaddress
import time
import wifi
import socketpool
import adafruit_requests
import ssl
import board
import busio
import json

MYSERVER = "HTTP/1.1 200 OK\nServer: WebSocket Server\nContent-Type: text/html\n"
MYHTML_LEN = "Content-Length: {}\n\n"
MYHTML = """


	 
		
			title {
				text-align: left;
			}
			body {
				background-co...
thorny jay
#

My Feather RP2040 does not behave very well. If it is unplug for 10-30 minutes, I can plug it, boot into Firmware upload mode, upload CircuitPython UF2. But I don't the CircuitPython drive, at best I see Windows complaining about "Unrecognised USB device". I sometime get a green LED and previously I was able to push code in the Feather. Is there a know problem? If I try to use it without that gap of time, then I don't even get light and I cannot go to UF2 mode. Is there a factory UF2 I could upload, or a known good one?

manic glacierBOT
onyx hinge
#

@thorny jay assuming you are using the "hold bootsel and plug in USB" steps to access the bootloader, then the problem cannot be CircuitPython, because the previously-uploaded UF2 data is not accessed in that case. Note that when you enter the bootloader, NO LEDS WILL LIGHT due to limitations of the RP2040 microcontroller. The only way to tell is that an "RPI-RP2" mass storage device becomes available on your computer. Also, if you have a battery plugged in to the feather, these steps won't work because the RP2040 microcontroller is looking at the BOOTSEL button when it is first powered on, and with a charged battery connected it would already be powered on.

thorny jay
#

It must be hardware issue, I can get to the RPI-RP2. Sometime I can upload UF2 there. If I wait, Windows will lose the drive and complain about the USB device.

onyx hinge
#

Of course it can be all sorts of things like a dodgy USB cable or hub, etc.

orchid basinBOT
manic glacierBOT
#
kmatch98
[adafruit/circuitpython] New comment on pull request #4403: bitmaptools: Add readinto, arrayblit

I spent some time on the readinto function and it's working great. It shows a big speedup when loading PCF fonts, and a similar speedup for loading other BMP files. As @jepler mentioned above, some bitmap files are loaded in different row direction. I submitted a PR to give the option to reverse the row loading and an associated PR in the adafruit_imageload library to use this new .readinto function.

It looks good to me! I'll spend some time on the .arrayblit function later t...

idle owl
simple pulsar
#

Hello. Is PULL working properly now in PIO on RP2040 boards?

manic glacierBOT
idle owl
#

@tough flax For the Feather RP2040 pinout image, I think it's good to go. My one question about the location of the CP pins I decided was fine, the board is labeled with the proper pins already, so it really doesn't matter where they fall in the diagram. I'll get it added to the guide. Thanks again!

signal pasture
thorny jay
# signal pasture sorry for pinging you. there is a "bug" where parts of the behavior you see will...

No I did not write a file from CP, only from Windows when it was possible. I was definitively capable of copying code on it at least once, and use Mu and see the console, and upload CircuitPython at least twice too. But to give this a it a try, I need to keep it without power for let's say 30 minutes. It is as if after being powered, some capacity(?) charge and then the behaviour change, and not using it make it "usable" for a few minutes.

signal pasture
#

it does indeed sound like an hardware issue

jaunty juniper
#

could it be some weird USB cache on the windows side ?

thorny jay
signal pasture
#

have you tried micropython? just to see if the behaviour changes?

thorny jay
#

But yes, the "not recognised USB device" is not consistent, you do not always get it, as if Windows was like: "I already told you that device was bad."

thorny jay
signal pasture
#

i meant for you to try to put micropython instead of circuitpython in your device that's giving you problems

#

just to see if anything changes

slender iron
#

@thorny jay what version of circuitpython are you using?

thorny jay
slender iron
#

hrm, that should have my speed fix

thorny jay
#

Those are the two I have downloaded, so I believe yesterday I did a try and today the latest.

thorny jay
tough flax
slender iron
idle owl
#

@tough flax That would be brilliant. I use Illustrator but I'm absolutely not comfortable with it. I'd greatly appreciate the assistance.

tough flax
#

Sure. Send me any logos or I’ll just grab them from other files.

idle owl
#

I'll get you something. Give me a moment.

manic glacierBOT
idle owl
#

@tough flax SVG?

tough flax
#

Yep! If you have them!

idle owl
#

@tough flax Hopefully this will work for what you need.

adafruit_logo.svg
tough flax
#

Yep! BTW if anyone else sees any flaws in this pinout (or things I missed) nows a great time to let me know 🙂

idle owl
#

It looked right to me, but I'm going crosseyed from going through that table in the datasheet so many times.

signal pasture
#

@tough flax sorry for the ping, but, the pin on the bottom left side is labeled as D4, but the purple legend says 4

tough flax
#

@idle owl do you have a feather logo or the rp2040 logo?

signal pasture
#

is that intentional?

idle owl
#

@tough flax No I don't know where to find those logos. If you can grab them from elsewhere, great, If not, that's ok too.

tough flax
#

Ok.

tough flax
# signal pasture is that intentional?

So all of the digital pins are just the numbers. I think that’s been standard since Arduino doesn’t use the D. However in the past we had split (A0/27) style chevrons to show the digital names. I don’t think the rp2040 has that.

#

Part of the challenge is there isn’t an Arduino port yet

signal pasture
#

oh, i was just asking because of the "D" there on the board, while the others don't have it

idle owl
#

@tulip sleet I've started adding LED to board defs. The list is long and so I think I'll do a little every day I have time instead of trying to do them all at once. The PR I'm about to put in is required for a guide to work that still isn't ready to be published, but ideally would be merged reasonably soon.

signal pasture
#

but googling it seems to show the board has the "D" on it

idle owl
#

It's implied, I guess.

tulip sleet
#

Sounds good - if you need help due to time let me know

simple pulsar
#

Feather M4 is the same, only D4 has D on the board

signal pasture
#

yes, and that makes sense

signal pasture
tough flax
#

Thinking about it @idle owl I have the schedule for that final feather pinout. Would you like me to add the onboard components section and the logos from that?

idle owl
idle owl
#

I mean I wrote the guide, but I didn't memorise the components 😄

tough flax
#

I think I can put the same columns (uart, spi, etc) on the pins for the QSPI and neopixel. I need to check if there’s a battery monitor pin

#

And I can stick the little battery connector in

manic glacierBOT
orchid basinBOT
solar whale
#

@idle owl Since I saw you were active working on the bno08x library, I thought I' point out that the guide refers to an adafruit_bno08x.mpy file, but there is none -- it is an adafruit_bno08x folder with .mpys in it -- came up on the forum https://forums.adafruit.com/viewtopic.php?f=60&t=176831

ripe cedar
#

sorry about that

idle owl
idle owl
#

@lone axle You've dealt with the new licensing stuff on the libs right?

#

I'm trying to tell someone what they need to add to their example to make it pass CI. Will the following be sufficient? ```# SPDX-FileCopyrightText: 2021 KC YUGESH

SPDX-License-Identifier: Unlicense```

lone axle
#

Yep, I believe so.

idle owl
#

Ok thank you. They're a first time contributor so I'm trying to help as much as I can.

lone axle
#

gnerally speaking I think it also needs a matching file in the LICENSES dir. but Unlicense is a commonly used one It's most likely already in there.

idle owl
#

ok Good to know thank you

stuck elbow
#

@tough flax the PWM pins are marked with a sine wave instead of a square wave

manic glacierBOT
idle owl
#

@solar whale Does that mean all the example code in the guide is wrong too?

solar whale
#

No. It should be fine. I have never used it

idle owl
#

Ok. It has import adafruit_bno08x. Does that work with packages? I'm second guessing now.

solar whale
#

It imports from adafruit_bno8x.something

idle owl
#

The code snippets don't.

solar whale
#

hmmm -- not sure why the import adafruit_bno08x is even there....

idle owl
#

Fair enough.

solar whale
#

just a sec -- I'll try it

idle owl
#

Cheers

solar whale
#

it works -- "does no harm"

#

but probably not needed

idle owl
#

ok then I'm leaving it. I don't have one to update the screenshots anyway

#

Thanks!

#

I updated the guide to reference a folder instead of a file

#

If you want to let the forum poster know, that would be great

solar whale
#

or maybe it has to be there -- I'm really not sure ....

idle owl
#

if it works, I'm ok with it regardless.

solar whale
#

I'lll add a note to the forum -- thanks!

manic glacierBOT
tough flax
manic glacierBOT
slender iron
#

@shy rain are you around?

shy rain
#

@slender iron Hey yes I am

slender iron
#

just looking at the naming of the rp2040

#

the samd51 board is sparkfun_samd51_thing_plus

#

looks like the sfe product listing has the chip last though

shy rain
#

I did see that. You’d prefer to align for consistency?

#

That makes sense I can change it

slender iron
#

I'm torn between consistency in circuitpython vs consistency with the product name on the sparkfun site

shy rain
#

Yes I named per the product page initially

slender iron
#

ok, that's ok with me

#

the nice thing about having thing plus earlier is that it'll list them all together then

shy rain
#

Haha sometimes happenstance works out nice

slender iron
#

should it be thing_plus?

#

or am I being too picky

shy rain
#

No in their documentation they are two words. I think it’s one word in the pico-sdk which is where I actually took the naming from initially

slender iron
#

ah, hrm

#

did sparkfun add it to the pico sdk?

shy rain
#

The sparkfun_thingplus.h file in the sdk i mean

#

I can check the history to see who added it

slender iron
#

looks like it was pi folks

shy rain
#

I think it’s your executive decision, whatever you want is how I’ll name it and update if needed

manic glacierBOT
slender iron
#

kk, thanks @shy rain. sorry I didn't post this yesterday

manic glacierBOT
shy rain
#

No problem! This is my first PR so I’m just pumped to be helping

slender iron
#

thanks!

manic glacierBOT
lone sandalBOT
manic glacierBOT
slender iron
#

@onyx hinge thanks again for getting the formatting polished up

onyx hinge
#

@slender iron you're welcome! I do enjoy working on tooling, even if I might cuss at it in the moment 🙂 Thanks @analog bridge for setting the stage

slender iron
#

🙂

analog bridge
#

Thanks @onyx hinge and @slender iron for taking out time to test the pre-commit changes
@onyx hinge since you mentioned your enthusiasm for tooling work, I have got another thing coming up..
I am working on port specific tests to save CI time, I can do a crude implementation and if need arises you can do the final polishing 🙂

onyx hinge
#

as we don't have any tests that we run except on the unix port, I'm interested to see what you're up to..

analog bridge
analog bridge
manic glacierBOT
onyx hinge
#

@analog bridge oh I think we have decided not to have the CI "take shortcuts" based on what kinds of paths are modified in a PR, if that's what you mean. I'd check with @slender iron before trying to change the CI like that, to verify my recollection.

analog bridge
#

ya I have this question too that whether a change in a port sub-directory will impact other ports

manic glacierBOT
blissful pollen
#

@slender iron should i merge the ParallelBus support? I saw you approved it but didn't merge it so just making sure there wasn't a reason for that

manic glacierBOT
slender iron
#

@blissful pollen I was waiting to get the translations in first

#

go ahead now

manic glacierBOT
slender iron
stuck elbow
#

Google has a whole complex system that determines which tests to run on any given change :)

idle owl
#

@gilded cradle Does the blinka displayio PR need testing or can I approve it w/o?

manic glacierBOT
gilded cradle
#

@idle owl it should probably be tested first.

idle owl
#

@gilded cradle Ok I won't merge it. Rerequest a review from cplibians because I approved it. I don't have anything to test it with.

gilded cradle
#

Ok @idle owl, you know what, I'll just test it really quick and if it works, I'll merge.

idle owl
#

Right on. That works.

manic glacierBOT
tough flax
#

From what I see in the datasheet, the QSPI pins on the RP2040 CAN be used for GPIO (including Analog input), however, they are not in the GPIO "Functions" table, so I'm assuming that they have no I2C, SPI, UART, or PIO functions... so, I don't think it makes sense to pull them out in the diagram as they inherently "can't" conflict with the pulled out pins. Thoughts? I do think it makes sense to pull out the Neopixel and voltage divider pins and even to add the "digital pin numbers" when they're known for the Arduino port.

idle owl
onyx hinge
#

There is no way exposed in CP to directly use the QSPI pins as GPIO. They're dedicated to their purpose.

tough flax
#

Do we have any idea what those digital pin #s will be? Making them the same as the underlying GPIO pins would make sense (that'd be 26-29) but that's just me (and isn't what the M4 did)

#

@onyx hinge I get that, but we pulled them out in the M4 because they DO have other functions defined on them so could cause conflicts

#

And I know that's unlikely in CP, but the diagram kind of needs to support Arduino as well (eventually)

idle owl
tough flax
#

I see no voltage divider on VBAT to check battery level. Is this intentional due to power consumption? Or am I missing it?

idle owl
tough flax
#

Yep!

#

I'll remove it

idle owl
#

I knew I had seen it somewhere. 😄

ionic elk
#

@tulip sleet would you happen to be familiar with the port_enable_tick system? It is causing problems during sleep for me that appear correlated with file writes, and I'd like to know what I can tweak without breaking filesystem/MSC stuff.

manic glacierBOT
#
lesamouraipourpre
[adafruit/circuitpython] Issue opened: #4423 Overriding and then setting a parent property doesn't work on a subclass of Group

Following on from a comment by @kmatch98 on adafruit/Adafruit_CircuitPython_Display_Text#116 about the difficulty subclassing/overriding Group, I've been trying to subclass Label with the example class GridLabel below.

# test polymorphism

import board
import displayio
import terminalio
import time

from adafruit_display_text import label

class GridLabel(label.Label):
    def __init__(self, font, **kwargs):
        super().__init__(font, **kwargs)

    @property
    def...
#
kmatch98
[adafruit/circuitpython] Pull request review submitted: #4403 bitmaptools: Add readinto, arrayblit

Testing of the readinto function worked well. Just one minor type in the docstrings.

I briefly tested the arrayblit, function and it operated as described.

Any functions that use this will want to watch out for if the array values exceed the bitmap color depth. In this case the value copied is array_value % color_depth. If used with the skip_index value and the color depth is too large, you can still get the skip_index value copied into your bitmap (since it is the result ...

tough flax
#

@tulip sleet helped with some final details 🙂

idle owl
#

Beautiful!

tough flax
#

A quick screenshot

unknown.png
idle owl
#

I'll get it added to the guide. @tough flax How would you like me to attribute you? Is your name ok, or do you have something else you'd rather have?

serene briar
#

Do the pinout diagrams normally show the stemma pins? I don’t remember seeing that on a board diagram before.

tough flax
#

"Bill Binko of ATMakers" would be ideal, but Bill Binko is fine

idle owl
idle owl
fossil gorge
#

Very nice diagram @tough flax! Crikey!

serene briar
#

The diagram looks really nice though

tough flax
#

Glad you like it!

#

I made a video about how to make them a while ago ... will get link

manic glacierBOT
tough flax
idle owl
tough flax
#

SWEET!

manic glacierBOT
grave ledge
#

I created class PPController(busio.I2C) and encountered two problems, 1) it cannot get a lock via try_lock(), e.g. to run scan(), and 2) when an object instance is destroyed, it seems that it does not deinit() as any other uses of busio.I2C fail with ValueError: SCL in use. I've tried all sorts of debugging & digging through CircuitPython source files, and I'm kinda stumped, I can't see what I'm missing. 🙂

still zephyr
tough flax
grave ledge
still zephyr
#

I see it know the first one got me confused

grave ledge
iron aspen
#

I've been beating my head against this for a while now. Does anyone know if it is possible to set the current date/time to a running board with Micropython? I know how to get set/get the current datetime with an rtc (like the Adafruit Adalogger Featherwing,) but I'd like to set it to the current time in a program, so that time.time() shows the correct value. Does anyone know if this is possible?

grave ledge
#

oh right, PPDevice class is defined first. The naming is a bit confusing, as I2CDevice refers to the I2C interface on a board. perhaps PPDevice is better renamed to PPPeripheral.

still zephyr
#

lets changd channel and go to hep with circuit python

grave ledge
grave ledge
iron aspen
# grave ledge I'm unsure that there's a concept of 'board time'. https://docs.micropython.org/...

Thanks - I was hoping to have it saved the other way around though, so that time.time() would return the real-world value, because I have another module that makes use of time.time() assuming it's right (it was written for cpython.) Since time: https://github.com/adafruit/circuitpython/blob/main/shared-bindings/time/__init__.c will return a value for time() I feel like it should be able to have an offset or something so that a realistic value can be set. Not seeing anything in there though. Maybe a pull request would be accepted?

GitHub
adafruit/circuitpython

CircuitPython - a Python implementation for teaching coding with microcontrollers - adafruit/circuitpython

grave ledge
iron aspen
#

I'm trying to set the time.time() itself - if I just turn on my board and run time.time() it thinks it's 1/1/2000. I'd like to get the value back from the RTC and set it correctly, so that time.time() returns a realistic value

#

I can already set the RTC, but it doesn't work with the time module itself - so rtc.datetime is returning the right value, but time.time() doesn't

#

So when I turn it on, I'd like to set. time.setcurrenttime(rtc.datetime) -- or something like that

grave ledge
manic glacierBOT
iron aspen
grave ledge
manic glacierBOT
iron aspen
#

Sorry, I meant that I made that up - I mean that I can't find a method that does that, looking through the C code it doesn't look like it does that (though there is some micropython code in there that looks like it can use an on-board RTC but that doesn't look relevant)

#

At the risk of really shaving the yak, if you guys accept pull requests I'll take my hand at adding it?

jaunty juniper
#

(I'm not clear if you are talking about Micropython or Circuitpython)
(and yes CPY accepts pull requests a plenty)

manic glacierBOT
iron aspen
jaunty juniper
#

I pulled that from the PortalBase library, which has a network.get_local_time() feature that I knew of

iron aspen
#

That's great, thanks @jaunty juniper

jaunty juniper
#

(it's conditional to the rtc module being implemented on the board)

#

and yeah that would be a good idea to document that there

iron aspen
#

actually I don't think so, as long as I set it every time in code.py - I just did something like this:```python

myrtc = adafruit_pcf8523.PCF8523(myI2C)
myrtc.datetime

rtc.RTC().datetime=myrtc.datetime
time.time()
1615918633```

#

(sorry one of these days I'll figure out how to format code)

jaunty juniper
#

(you can put ```python on a line to open a block of code and close with ```)

iron aspen
#

Thanks again. Setting it in code.py looks like it will work even if the board doesn't support the RTC module using that method it looks like.

jaunty juniper
#

no I mean on some boards import rtc will just not work because it's not implemented, like small M0 boards to make space, and the stm32 port

iron aspen
#

Ah, got you. Right now I'm only testing on the S2 and a Pico, might be an ok limitation though

#

Overall I'm working on a module for the Contrib library, a circuitpython version of the "schedule" library. Just released it on pypi, but doing some testing with RTC etc to make sure everything works right before doing the PR for the contrib

grave ledge
iron aspen
#

Thanks - I'm actually working on a plant monitor/watering system, but didn't want to have to implement my own scheduling system on those longer times (turn on/off lights twice a day, check and water once a day, etc). Much easier to have a library for it. Testing it out now, but now that I have the RTC figured out I can do things like: ```python
schedule.every(1).tuesday.at("20:00:00").do(telltime)

grave ledge
iron aspen
manic glacierBOT
manic glacierBOT
manic glacierBOT
manic glacierBOT
bleak spear
#

hey @steep finch sorry to bother you. had a question about your circuitpython fork. I'm updating the adafruit_hid library to work with your modification and I see you recently changed the report size from 22 to 19, but I'm still counting 22 reports. am I misinterpreting something?

Screenshot_from_2021-03-16_21-08-07.png
manic glacierBOT
manic glacierBOT
onyx hinge
#

@mental nexus thanks for being my reviewer on this one.

manic glacierBOT
manic glacierBOT
#
kmatch98
[adafruit/circuitpython] Pull request opened: #4428 Add \`reverse\_rows\` to speedy bitmaptools\.readinto function

The recent addition of the bitmaptools.readinto function helps speed loading of bitmap files. This provides an additional input variable reverse_rows to flip the row order to accommodate the format of some bitmap files.

This is required for speedups to the Adafruit_CircuitPython_Imageloadlibrary: https://github.com/adafruit/Adafruit_CircuitPython_ImageLoad/pull/45

(Note: There were some doc string changes that appeared due to my manual merging. I tried to match the formatting ...

idle owl
#

@lone axle I have a favor to ask.