The previous update to HyDE.py has replaced space-indents and tab-indents with two different, reserved, special characters. And it has added an option to choose which one gets inserted when "tab" is pressed.

This functionality is a little bit confusing, because space indents and tab indents are kind of hard to tell apart in the current version. It also adds some unneeded complexity to the code by requiring any code that deals with space characters to check for both reserved indent characters.

I'm thinking this could be an improvement:

• Remove distinction between tabs and space indentation.
• Add a check to find the first occurrence of tabs, or space indentation. Use this to automatically set the tab-option to the appropriate mode
• simplify code to only care about the one reserved indentation character
• "save" method should replace all indentations with the automatically set tab mode

This would have the added benefit of giving HyDE the ability to convert tabs to spaces (and vice-versa) by changing the option in the menu.

Copy-pasting files between locations work.
But it doesn't propose renaming/renames automagically when pasting in the same location

is there any reason why you're not using the compiled mpy version instead of the interpreted one? (other than it's extra steps to install)

I have seen no sd card or and flash pop up
I dont know is im just not doing something or im missing something
I flashed it on to my cardputer from the 5m burner and made a folder for the py files and notes
But It does not show up anywhere and i dont know how to derectly accese the flach to just add it to the flash
Some help would be nice thank you

oh and if you dont mine a seggestion
people showed a software similar to this booting and launching bin files or applications
Maybe the capabillity to boot up bins like other launchers would expand the capbilties of the laucher
:) also its cools stuff what yall are doing

A module for creating simple menus like the settings.py menu the launcher uses, would be very useful for many apps.

It is super annoying to click arrows multiple times to navigate around big files.
Holding fn or ctrl modifiers is also painful :)
I propose to:

1. Implement key repeats (holding a key will repeat the effect)
2. Implement modifier-keys locks (e.g by triple click). That would require adding some visual indicator (I'd propose a popup in top-right corner with 4 small squares to indicate active locks)

Same question as in title.
Tried pressing all buttons but ive not came across a right combination.

A potentially useful feature would be to create a method in both st7789py and st7789fbuf, which could control the display brightness easily.

This could open the door to adding a display brightness setting to the config at a later date, as well.

In MicroHydra 1.0, the API for controlling the hardware is highly specific to the Cardputer. This means that, to create apps that work on multiple devices, you would need to completely rewrite large parts of the code for every single device.

In order for MicroHydra to support multiple platforms (such as the Cardputer II and the original Cardputer) simultaneously, the API for all of the built-in modules must be redesigned such that apps no longer need to specify exact Pins or other device-specific parameters in order to control things like the display, speaker, keyboard, etc.

For example, this:

# init object for accessing display
tft = st7789fbuf.ST7789(
    machine.SPI(
        1,baudrate=40000000,sck=machine.Pin(36),mosi=machine.Pin(35),miso=None),
    _DISPLAY_HEIGHT,
    _DISPLAY_WIDTH,
    reset=machine.Pin(33, machine.Pin.OUT),
    cs=machine.Pin(37, machine.Pin.OUT),
    dc=machine.Pin(34, machine.Pin.OUT),
    backlight=machine.Pin(38, machine.Pin.OUT),
    rotation=1,
    color_order=st7789fbuf.BGR
    )

Will not work on anything except the Cardputer.

MicroHydra can use device-specific modules to simplify this process, and make the display initialization code in each app would look like this:

tft = Display()

Hi,

I have a basic app which I have written using uiFlow 2 for the cardputer(using the envro IV sensor), the app runs fine on the cardputer and I can upload it to the device and it will run fine when I unplug it from my computer. I downloaded the main.py from the cardputer (this is my sensor code), renamed it and then uploaded it to the apps folder on my sd card for microhydra, microhydra sees it but I get an error message saying the M5 lib is unavailable, however it is on the device somewhere as the app will run separately from microhydra . I have tried one of the example programs and this works fine. So everything works independantly but when launching the app from Microhyrda it doesn't.

Do you have any ideas on this please?

Many thanks,

Matthew

This is an idea I'd like to work on. The ability for main.py to handle multiple arguments stored in RTC memory. Then, apps can be launched the same way, but apps could also send data to each other on startup through the RTC memory. I had this idea while thinking of how a file browser might be implemented, and here's how I envision it working:

• File browser can show list of files and allow user to select a file.
• File type is compared against a dictionary of file types and paths to file handlers.
• File browser writes to RTC memory both the path to the file handler, and the path to the file. These would need to be joined by a predetermined character or set of characters that is disallowed from regular file paths. Currently I'm imagining something like "|//|", but I'm open to suggestions.
• If main.py finds "|//|" in file path, uses str.split() to separate into a list, and launch the first element as an app, writing the rest back into RTC memory.
• File handler app is now open, and can read RTC memory to find the path to the file it is meant to open.

I'm pretty excited about the potential possibilities something like this could open up. However I'm still not totally certain on how it would be best to format the special strings for this purpose. I also imagine the passed data could be JSON data, or something, instead.

Can we run linux on Cardputer?
yt

I already made ChatGPT do coding, but it looks ugly.
You can insert this into lib/mhoverlay.py to see how it works:

    def popup_options_2d(self, options, title=None, shadow=True, extended_border=False, scrollable=False):
        """
        Display a popup message with given options arranged in 2D grid.
        Blocks until option selected, returns selection.
        """
        tft = self.display
        num_cols = 5  # Number of columns in the grid
        per_colomn = 6
        col_width = 240 // num_cols  # Width of each column
        row_height = 16  # Height of each row
        
        if scrollable:
            old_options = options
            options = old_options[:num_cols*per_colomn]

        if self.compatibility_mode:
            # draw box
            box_height = (len(options) // num_cols + 1) * row_height + 8
            box_y = 67 - (box_height // 2)
            if extended_border:
                tft.fill_rect(0, max(box_y - 10, 0), 240, min(box_height + 20, 135), self.config.palette[1])
            if title:
                self.draw_textbox(title, 120, box_y - 16, shadow=shadow, extended_border=extended_border)
        else:
            # draw box
            box_height = (len(options) // num_cols + 1) * row_height + 8
            box_y = 67 - (box_height // 2)
            if extended_border:
                tft.fill_rect(0, max(box_y - 10, 0), 240, min(box_height + 20, 135), self.config.palette[1])
            if title:
                self.draw_textbox(title, 120, box_y - 14, shadow=shadow, extended_border=extended_border)

        cursor_index = 0
        keys = self.kb.get_new_keys()
        while True:
            if self.compatibility_mode:
                # draw options
                for idx, option in enumerate(options):
                    col = idx % num_cols
                    row = idx // num_cols
                    x = col * col_width + (col_width - len(option) * 4) // 2
                    y = box_y + row * row_height + 4
                    if idx == cursor_index:
                        tft.fill_rect(x - 2, y - 2, len(option) * 4 + 4, 20, self.config.palette[0])
                        tft.text(self.font, option, x, y, self.config.palette[5], self.config.palette[0])
                    else:
                        tft.text(self.font, option, x, y, self.config.palette[4], self.config.palette[2])
            else:
                # draw options
                for idx, option in enumerate(options):
                    col = idx % num_cols
                    row = idx // num_cols
                    x = col * col_width + (col_width - len(option) * 4) // 2
                    y = box_y + row * row_height + 4
                    if idx == cursor_index:
                        tft.fill_rect(x - 2, y - 2, len(option) * 4 + 4, 20, self.config.palette[0])
                        tft.text(option, x, y, self.config.palette[5])
                    else:
                        tft.text(option, x, y, self.config.palette[4])
                tft.show()

            keys = self.kb.get_new_keys()
            for key in keys:
                if key == ";":  # Move cursor up
                    cursor_index = (cursor_index - num_cols) % len(options)
                elif key == ".":  # Move cursor down
                    cursor_index = (cursor_index + num_cols) % len(options)
                elif key == ",":  # Move cursor left
                    cursor_index = (cursor_index - 1) % len(options)
                elif key == "/":  # Move cursor right
                    cursor_index = (cursor_index + 1) % len(options)
                elif key == "w" and scrollable:
                    next_page_result = self.popup_options_2d(options=old_options[num_cols*per_colomn:], title=title, shadow=shadow,
                                                          extended_border=extended_border, scrollable=True)
                    if next_page_result != None:
                        return next_page_result
                    else:
                        cursor_index = 0
                elif key == "`" or key == "ESC" or key == "BSPC" or key == "a":
                    return None
                elif key == "ENT" or key == "GO":
                    return options[cursor_index]

            time.sleep_ms(10)

Hay is there a place with the full package of all the micropython libs used in apps or micro python used in microhydra
so i can plug the lib Package into Thonny so I can start to write up some apps and code
it would be nice to know so i can start to figure stuff out about how to make apps

also is there any network examples i can look at
I'm hopping to make a app to allow the carputer to use a ESB-NOW chat like system and some other stuff

sorry if i am just blind and couldn't find it
please correct me on anything I'm exited to learn and work on this
thank you for your time :)

One-two symbols in the line ends getting stripped when opening files in editor.
Seems like the issue is with the code handling line breaks (\r\n) and trimming 2 last chars afterwards, and repeats once again:
#string formatter
def remove_line_breaks(line):
if line.endswith('\r') or line.endswith('\n'):
line = line[:-2]
if line.endswith('\r') or line.endswith('\n'):
line = line[:-2]
return line

I think, only last char shoud be trimmed: line = line[:-1]

the wiki has some inaccuracies

I was trying to add 24h clock setting to settings.py (there were several iterations, ended up with this):

menu_def = [
    (HydraMenu.IntItem, 'volume', {'min_int':0,'max_int':10, 'instant_callback':update_config}),
    (HydraMenu.RGBItem, 'ui_color', {'instant_callback':update_config}),
    (HydraMenu.RGBItem, 'bg_color', {'instant_callback':update_config}),
    (HydraMenu.WriteItem, 'wifi_ssid', {}),
    (HydraMenu.WriteItem, 'wifi_pass', {'hide':True}),
    (HydraMenu.BoolItem, 'sync_clock', {}),
    **(HydraMenu.BoolItem, 'clock_24h', {}),**
    (HydraMenu.IntItem, 'timezone', {'min_int':-13,'max_int':13}),
    ]

Launching settings.py causes black screen.

'log.txt':

Tried to launch '/launcher/settings.py', but failed: '24h_clock'
Tried to launch '', but failed: ''
Tried to launch '', but failed: ''
Tried to launch '/launcher/settings.py', but failed: 'name 'false' isn't defined'
Tried to launch '', but failed: ''
Tried to launch '', but failed: ''
Tried to launch '/launcher/settings.py', but failed: 'clock_24h'
Tried to launch '', but failed: ''
Tried to launch '', but failed: ''
Tried to launch '', but failed: ''
Tried to launch '', but failed: ''
Tried to launch '/launcher/settings.py', but failed: 'clock_h'

Hi ,

Amazing library, made my entry into the world of MicroPython so easy.

Can I tinker with the code to add an option in the settings menu to enable Daylight Saving Time ?

You can let me know if that is the right way to do it or you have something else in mind

There are new features almost every day!
I'd like to get them directly to the device :)
I've tried to adapt micropython_ota: https://github.com/mattjackets/micropython_ota/blob/main/micropython_ota.py
But it fails to download 'big' files: in my case it was failing on memory alocation for 44kb
Googled a bit, and it appears that there's a mrequests lib providing chunked downloads (https://github.com/SpotlightKid/mrequests/tree/master)

Looking into it. Any inputs are welcome!

st7789fbuf gives errors easily, due to it's allocating of a large bytearray, potentially after other processes have already fragmented the heap.

Creating a large byte array and then deleting it right before creating the ST7789 object seems to work around this issue. A more elegant solution would probably be to simply pass that bytearray to the ST7789 on init. This way, the bytearray could be immediately allocated, and reused.

Brand new, just got it today. Burning the firmware image (v0.2) that you uploaded to M5Burner results in my unit being bricked. After burning finished, instead of rebooting the screen just went black and now will no longer boot. No error messages, nada. Just a black screen.

Attempted to burn the original core image back to it, but M5Burner won't write to it, says it can't connect over the ComPort.

I cannot find documentation or help for the editor or file browser here. I have tried simply installing both directly, but they crash and return to app select. It would be amazing if someone could figure out where I went wrong.

Using the wrong config.json can cause a KeyError, and prevent the launcher from starting.

lib.mhconfig.Config() should check that the required keys exist in the config,, in order to prevent this.

Last update to HyDE.py has added support for tab characters, and converted both tabs/spaces to reserved space characters to better take advantage of the small screen width.

This has broken the editor's ability to automatically indent on new lines, and it has caused the backspace on indentation to delete too many characters.

Possible solution: update newline method to check for the appropriate special characters rather than checking for spaces. Update backspace method to only delete one character.

Currently HyDE does not support tab characters.

This issue comes from the current monospaced implementation. All characters are equally spaced, and the font does not have a built-in symbol for tab characters.

Current fix I'm imagining, is replacing tab characters with a couple of rare whitespace characters on file import, which will simulate a tab indent. Then when saving the file, the specific whitespace characters can be re-converted to tabs. I would also like to add some kind of config options to HyDE to switch between tabs and spaces.

If the above fix is implemented, then there could also be an opportunity to convert space indentation to a special character too, and add an option to control the desired amount of spaces to display with.