I'm getting an error when I use bowtie.cache.load. As far as I can tell, the correct value is loading and everything functions normally. I'm not sure where this error comes from. I'm also using Schedule, but I managed to get that to work fine when not loading from the cache.

As I make more changes, I may be able to give more details about this.

  File "/Users/rsm5139/anaconda3/envs/learning-bowtie/lib/python3.6/site-packages/eventlet/greenthread.py", line 218, in main
    result = function(*args, **kwargs)
  File "./build/src/server.py", line 59, in foo
    func()
  File "/Users/rsm5139/Git/learning-bowtie/step_1.py", line 23, in timed_event
    ticker = cache.load('scheduled')
  File "/Users/rsm5139/anaconda3/envs/learning-bowtie/lib/python3.6/site-packages/bowtie/cache.py", line 58, in load
    return msgpack.unpackb(bytes(event.get(timeout=10)), encoding='utf8')
  File "/Users/rsm5139/anaconda3/envs/learning-bowtie/lib/python3.6/site-packages/eventlet/queue.py", line 313, in get
    return waiter.wait()
  File "/Users/rsm5139/anaconda3/envs/learning-bowtie/lib/python3.6/site-packages/eventlet/queue.py", line 141, in wait
    return get_hub().switch()
  File "/Users/rsm5139/anaconda3/envs/learning-bowtie/lib/python3.6/site-packages/eventlet/hubs/hub.py", line 295, in switch
    return self.greenlet.switch()
queue.Empty
^Cwsgi exiting

Hi, I'm trying to use Schedule, but it doesn't seem to do anything. I had it call a function to create a plot, but that didn't work. Then I tried having it just print('hello'), but still nothing. I would like to add a timer to the app and I figured Schedule would help, but maybe not.

I'm stuck trying to run the example code: https://github.com/jwkvam/bowtie-demo/blob/master/example.py

I believe I have all dependencies installed via conda and pip (plotlywrapper), and I've tried running it in several environments and consoles.

#!/usr/bin/env python ... # -- coding: utf-8 -- ... ... from bowtie import App, command ... from bowtie.control import Dropdown, Slider ... from bowtie.visual import Plotly, Table ... from bowtie.html import Markdown ... ... import numpy as np ... import pandas as pd ... import plotlywrapper as pw ... ... from sklearn.kernel_ridge import KernelRidge ... ... iris = pd.read_csv('./iris.csv') ... iris = iris.drop(iris.columns[0], axis=1) ... ... attrs = iris.columns[:-1] ... ... description = Markdown("""Bowtie Demo ... =========== ... ... Demonstrates interactive elements with the iris dataset. ... Select some attributes to plot and select some data on the 2d plot. ... Change the alpha parameter to see how that affects the model. ... """) ... ... xdown = Dropdown(caption='X variable', labels=attrs, values=attrs) ... ydown = Dropdown(caption='Y variable', labels=attrs, values=attrs) ... zdown = Dropdown(caption='Z variable', labels=attrs, values=attrs) ... alphaslider = Slider(caption='alpha parameter', start=10, minimum=1, maximum=50) ... ... mainplot = Plotly() ... mplot3 = Plotly() ... linear = Plotly() ... table1 = Table() ... ... ... def pairplot(x, y): ... print('hellox') ... if x is None or y is None: ... return ... x = x['value'] ... y = y['value'] ... plot = pw.Chart() ... for i, df in iris.groupby('Species'): ... plot += pw.scatter(df[x], df[y], label=i) ... plot.xlabel(x) ... plot.ylabel(y) ... mainplot.do_all(plot.to_json()) ... ... ... def threeplot(x, y, z): ... if x is None or y is None or z is None: ... return ... x = x['value'] ... y = y['value'] ... z = z['value'] ... plot = pw.Chart() ... for i, df in iris.groupby('Species'): ... plot += pw.scatter3d(df[x], df[y], df[z], label=i) ... plot.xlabel(x) ... plot.ylabel(y) ... plot.zlabel(z) ... mplot3.do_all(plot.to_json()) ... ... ... def mainregress(selection, alpha): ... if len(selection) < 2: ... return ... ... x = xdown.get()['value'] ... y = ydown.get()['value'] ... ... tabdata = [] ... mldatax = [] ... mldatay = [] ... species = iris.Species.unique() ... for i, p in enumerate(selection['points']): ... mldatax.append(p['x']) ... mldatay.append(p['y']) ... tabdata.append({ ... x: p['x'], ... y: p['y'], ... 'species': species[p['curve']] ... }) ... ... ... X = np.c_[mldatax, np.array(mldatax) ** 2] ... ridge = KernelRidge(alpha=alpha).fit(X, mldatay) ... ... xspace = np.linspace(min(mldatax)-1, max(mldatax)+1, 100) ... ... plot = pw.scatter(mldatax, mldatay, label='train', markersize=15) ... for i, df in iris.groupby('Species'): ... plot += pw.scatter(df[x], df[y], label=i) ... plot += pw.line(xspace, ridge.predict(np.c_[xspace, xspace**2]), label='model', mode='lines') ... plot.xlabel(x) ... plot.ylabel(y) ... linear.do_all(plot.to_json()) ... table1.do_data(pd.DataFrame(tabdata)) ... ... ... @command ... def main(): ... app = App(rows=2, columns=3, background_color='PaleTurquoise', debug=False) ... app.columns[0].fraction(2) ... app.columns[1].fraction(1) ... app.columns[2].fraction(1) ... ... app.add_sidebar(description) ... app.add_sidebar(xdown) ... app.add_sidebar(ydown) ... app.add_sidebar(zdown) ... app.add_sidebar(alphaslider) ... ... app[0, 0] = mainplot ... app[0, 1:] = mplot3 ... app[1, :2] = linear ... app[1, 2] = table1 ... ... app.subscribe(pairplot, xdown.on_change, ydown.on_change) ... app.subscribe(threeplot, xdown.on_change, ydown.on_change, zdown.on_change) ... app.subscribe(mainregress, mainplot.on_select, alphaslider.on_change) ... ... return app ... ImportError: No module named html ImportErrorTraceback (most recent call last) in () 5 from bowtie.control import Dropdown, Slider 6 from bowtie.visual import Plotly, Table ----> 7 from bowtie.html import Markdown 8 9 import numpy as np ImportError: No module named html

I'm working on a dashboard where i'd like to stack 3 widgets, but on my screen they render with very small heights... is there a way to forece them to occupy a minimum height?

ERROR in ./src/app/nouislider.jsx Module not found: Error: Cannot resolve module 'nouislider/src/nouislider.css' in /Users/wesselhuising/Documents/triplea/bowtie/test/build/src/app @ ./src/app/nouislider.jsx 15:0-40

Using a virtualenv, followed the documentation about installing and setting up a bowtie app. Tried a lot, but can't figure out how to make the npm use the compiler of less to create a css. This happens while building the example app in the quickstart of the documentation.

Any thoughts?

It would be handy to have a function that get's called on initial page load from the client. Here's a sketch of how this can work:

1. Add an emit from index.jsx
2. Add a handler template in server.py
3. Add functionality to Layout class to call a python function on the initial signal.

Put server.py.j2 functions into App class.

• [x] move all routes to App
• [x] remove server.py.j2
• [x] test notebook integration still works
• [x] different endpoint for each route
• [x] move authentication
• [x] document authentication
• [x] decorator for upload and load
• [x] run layout at build
• [x] test upload
• [x] protect routes in serve
• [x] ~document layout building function~
• [x] test pager
• [x] test init

Fixes #234 #247

For dynamic layouts, the app will need to be global or at least the views will be. I would rather not confuse the matter and encourage the app to be global. As a side benefit, users can then use the listen decorator instead of the subscribe function which should be more readable. This may also allow us to be more compatible with Flask apps if someone wants to have a bowtie dashboard sitting with some other flask stuff.

• [x] refactor example code
• [x] refactor bowtie-demo
• [x] move server.py.j2 functionality into app.py or wherever appropriate.

TODO

• [x] remove captions from controllers
• [x] ~~move markdown back to visual :/~~
• [x] move react link to html
• [x] ~~functional style for html widgets? (no events or commands)~~
• [x] make it easy to create captions
• [x] list of components for each cell
• [ ] embed links in other html? children in div's?
• [ ] put server.py template into bowtie proper
• [ ] change convention to putting app as global variable

Fixes #236

Issue Type

[x] Bug (Typo)

Steps to Replicate and Expected Behaviour

• Examine bowtie/tests/test_components.py and observe instatiation, however expect to see instantiation.
• Examine bowtie/tests/test_tags.py and observe instantation, however expect to see instantiation.

Notes

Semi-automated issue generated by https://github.com/timgates42/meticulous/blob/master/docs/NOTE.md

To avoid wasting CI processing resources a branch with the fix has been prepared but a pull request has not yet been created. A pull request fixing the issue can be prepared from the link below, feel free to create it or request @timgates42 create the PR. Alternatively if the fix is undesired please close the issue with a small comment about the reasoning.

https://github.com/timgates42/bowtie/pull/new/bugfix_typos

Thanks.

yarn install v1.22.4
warning package.json: No license field
warning No license field
[1/4] Resolving packages...
success Already up-to-date.
Done in 0.45s.
'.' is not recognized as an internal or external command,
operable program or batch file.
Traceback (most recent call last):
  File "example.py", line 126, in <module>
    def main():
  File "C:\Users\sabar\.conda\envs\Drug_Analysis\lib\site-packages\bowtie\_command.py", line 103, in command
    sys.exit(cmd(arg))
  File "C:\Users\sabar\.conda\envs\Drug_Analysis\lib\site-packages\click\core.py", line 829, in __call__
    return self.main(*args, **kwargs)
  File "C:\Users\sabar\.conda\envs\Drug_Analysis\lib\site-packages\click\core.py", line 782, in main
    rv = self.invoke(ctx)
  File "C:\Users\sabar\.conda\envs\Drug_Analysis\lib\site-packages\click\core.py", line 1259, in invoke
    return _process_result(sub_ctx.command.invoke(sub_ctx))
  File "C:\Users\sabar\.conda\envs\Drug_Analysis\lib\site-packages\click\core.py", line 1066, in invoke
    return ctx.invoke(self.callback, **ctx.params)
  File "C:\Users\sabar\.conda\envs\Drug_Analysis\lib\site-packages\click\core.py", line 610, in invoke
    return callback(*args, **kwargs)
  File "C:\Users\sabar\.conda\envs\Drug_Analysis\lib\site-packages\bowtie\_command.py", line 61, in build
    app._build()
  File "C:\Users\sabar\.conda\envs\Drug_Analysis\lib\site-packages\bowtie\_app.py", line 949, in _build
    raise WebpackError('Error building with webpack')
bowtie.exceptions.WebpackError: Error building with webpack

Main use case is to have links in divs so the links can be embedded in text. Using links like this will help with single page apps with multiple views.

By default there is 0 border around the edge of the screen. This is visually unattractive and there is no way to change it with the bowtie API. This is to note that this can and should be improved.

Thoughts:

1. Should this be combined with the row/column gaps?
2. Take inspiration from other APIs, such as matplotlib's margins?

not sure how easy this will be, but it will add assurance that refreshing hopefully works or at the very least doesn't break the app.

Basic test structure:

1. Build app
2. Use selenium to interact with widget.
3. Refresh web page
4. Make sure the page loaded successfully.
5. Extra credit: make sure the state is same as it was before refresh.

widgets tested

• [ ] dates
• [ ] dropdown etc.