Passing An Image Back To Workflow
Those of you who use Workflow know it can handle images.
Any idea how a Pythonista "stage" in the pipeline might pass an image back?
@MartinPacker , I am not really sure about Editorial vrs Pythonista for the graphing, they could be equivalent. But my guess would be that Pythonista would be the superset of functionality.
But just really wanted to mention the 'Pie Chart Demo.py' in the Examples/Plotting dir in Pythonista. Great demo how to easily create a pie chart with some nice features. It's so easy to understand. Then I am pretty sure if you use a ui.ImageContext to draw into you could return that image for the clipboard. Possibly the matplotlib gives a more direct way. Just interesting
I had checked this page but they only support iCloud file storage, that will say Apple apps storage, like Notes, but not iCloud Drive storage of all other apps, what does Workflow app.
In Workflow, you can save a file to the ICloud Drive of another app, like Documents, FileBrowser, Pages, Numbers etc...
It's exactly like you can manually do in Pythonista by
- select a file
- add to ICloud Drive
All this talk about iCloud illustrates one of the reasons why I'm not looking for a "write to file" solution.
Clipboard would be fine, though "trashing" the clipboard should only be done if necessary. It seems that it is.
Something for y'all to pick the bits out of...
A reply from @WorkflowHQ on Twitter contained the following...
"Also regarding sending images, Workflow can decode base64 encoded images."
Does this help us here? Can we encode images base64?
@MartinPacker Encoding images as base64 should be fairly straightforward in Python. Do you have any more details?
I figured out how to do it without using the clipboard. The trick is basically to use "Get Contents of URL" with a
data:URI that gets passed to the workflow via the
workflow://scheme, containing base64-encoded image data.
Here's the script:
import webbrowser import base64 from PIL import Image import io from urllib.parse import quote img = Image.open('test:Lenna') buffer = io.BytesIO() img.save(buffer, 'PNG') data_uri = 'data:image/png;base64,' + base64.b64encode(buffer.getvalue()).decode('ascii') webbrowser.open('workflow://run-workflow?name=EditBase64Image&input=' + quote(data_uri, ''))
...and the workflow should be something like this:
invalid base64 encode.
remove 'data:image/png;base64,' like this:
data_uri = base64.b64encode(buffer.getvalue()).decode('ascii')
now it works