I know this is a 3+ year old thread but I still haven't been able to get this to work correctly with Dropbox (or several other apps that expose files through the External Files API).

I have confirmed that overall the process works as I've been able to get Working Copy to show up as expected as a file provider when I click Pythonista > External Files > Open... However Dropbox always seems to show up disabled and I am unable to browse it to select files.

Working Copy and Dropbox are both configured as external file providers in the Files app, and I'm able to browse their contents from the Files app. However when trying to browse them from Pythonista, Dropbox remains disabled.

Is there any other configuration required to fully expose a 3rd party file provider in Pythonista?

Forgot to reply to close this out: using the get_image version as suggested by @mikael did the trick, at least in terms of saving a second or so in runtime (which is noticeable). Thanks all!

Wow! This concept is great! This definitely seems like a way better solution then what I was doing. I am going to mess around with this and I will update it with what I got to see how far I have gotten.

Thanks for this!!

@alexandrsemenov if @JonB 's solution is too complex for your request, please try

import ui tf = ui.TextField() tf.frame = (0,0,300,40) tf.name = 'Window of TextField' tf.placeholder = 'type here, then press X' tf.begin_editing() tf.present('sheet') tf.wait_modal() var = tf.text # put the TextField text in a variable def next_function(v): print(v) next_function(var) # pass the variable to the next function

@Ahab3030 for the fun

def segment_action2(sender): sender.superview['CM Value'].text = str([4107,6530,10383,16509,26251,41740,66369,1005518,133056,167780,211566] [sender.selected_index])

You cannot. https://forum.omz-software.com/search/pyglet

@mikael awesome, thanks

@JonB thanks so much! Will try stash. I understand now.

okay Thanks @ccc

thank you so much. I had foolishly forgotten ...

@JonB said:

I haven't tried you code yet, but I have recently been helping my son debug a similar issue on a javascript game he is trying to write.

What I suggested was to have a debugging mode, where you display a label on each wall block showing it's bbox. Likewise, attach a label to your player node,and display it's bbox. The color of the label for each wall should change depending on whether it interests the player bbox (and maybe a color to indicate whether it interests the top, bottom, left or right), or draw lines showing which edges are touching, etc)

One thing that you do need to watch out for is that, IIRC, bbox or frame is with respect to the parent node. So if your walls are all children to a ground node, but your player is child of the root scene, their coordinate systems might have an offset that you have to account for. Also, if you are using scale or other tranaforms, that can change what these return...

Is your character walking through all walls, or does it eventually stop, say after it is halfway or all the way inside the wall? There could be issues with where you have your anchor point set.

I am now seeing this. Maybe I just wasn’t moving the sprite around enough to see it but there are definitely places where my left button wont work for large swathes of the screen and then it will start to work again once I get above a certain point. I have my sprite scaled down to 80%, but that can’t account for the position being off by 1/3 of the screen...how can i display the bbox’s? I’m not sure I know how to even do that...

That worked! Thanks everyone I can’t believe how fast you all provided help.

And try the last script of this topic to get a Blurr effect.

Thank you @mikael, that was helpful. Certainly as a workaround, then a combination of on_main_thread, my debugging print() calls, and the saving/restoring of sv.image_view.image all (when used together) seem to solve the instance of corruption. Any two of them without the third is less successful, which is all very odd: I’m guessing there’s a race condition somewhere, or that this is some kind of heisenbug.

I’ll run with this workaround for now, thank you once again for your insights and suggestions.

Brilliant @mikael thanks so much for sharing.

I'll dig into this and post back if I run into trouble.

Well, it is important to look at the traceback, and try to understand what it is telling you.

AttributeError 'NoneType' has no attribute .text means you tried to access a .text attribute on an object which is None.

So, if you look at where you try to access .text, you will see it is date['label1'] -- so for some reason, that lookup is returning None. That must mean there is no subview named label1 within the view you are accessing.

Since you didn't provide the contents of your pyui or code that creates your view, it is hard to debug remotely. You can rename your .pyui to .json, and copy/paste to the forum. Common mistakes include:

If you have views that have subviews that have othe subviews, you might be referring to a label1 which is a subview of a different subview, not the date subview. you may have spelled the name wrong. maybe you were copying code from the forums, but you haven't actually created the same view structure. This code seems to expect a button within the main view, the main view has a subview called datepicker, and the datepicker view has a subview called label1. Is that what you intended? Or maybe you meant sender.superview['label1'] ?

Also, as a suggestion, you should usually create a new post, rather than replying to someone else's post when you have an unrelated question.

@DavinE 😂