mini-kep/frontend-dash

Dash application with interactive controls and time series plotting.

frontend-dash's Issues

current version cannot start locally

  File "D:\Continuum\Anaconda3\lib\site-packages\werkzeug\serving.py", line 691, in run_simple
    s.bind((hostname, port))

OSError: [WinError 10013] Сделана попытка доступа к сокету методом, запрещенным правами доступа

show annotation on dash plot

Proposed code here, but it is not working:
https://community.plot.ly/t/annotation-not-showing-on-dash-dcc-graph/6660

use tables in variable information layout

group selector for variable list

frontend-dash/app.py

Line 26 in 8c744b4

1. Group selector added before each variable name selector.

We now have 2 variables selectors. For each of these selectors I would like to have a two selectors(dropdown menus) on one line: one chooses varibale group (eg GDP components, Prices, Foreign Trade) and the other chooses a ariable from a smaller list.

Need a) loyout for this on a separate branch b) data structure, c) callbacks.

add 1 integration test

Am integration test must:

• show the application displays first page
• run on travis (minimal setup on travis is preferred)

Make table for variable information

frontend-dash/app.py

Lines 276 to 313 in f9fea1b

	# --------------------------------------------------------------------------
	# FIMXE: this is a call back to update variabl einformation on the screen, it is
	# a draft / placeholder
	#
	# Needed following change
	#
	# a. Layout: make every variable information block an html table
	#
	# Variable BRENT
	# Frequency: d
	# Start: 1987-05-20
	# End: 2017-10-30
	# Latest value: 60.65
	# Download: <api/datapoints?freq=d&name=BRENT&format=csv>
	# Short link: <oil/series/BRENT/d>
	# More info: <api/info?name=BRENT>
	#
	# b. content - retrieve this data from api/info?name=BRENT
	#
	# c. NOT TODO: update with slider change ?
	# start placeholders for variable information
	@app.callback(output=Output('var1-info', 'children'),
	inputs=[Input('frequency', component_property='value'),
	Input('name1', component_property='value')
	])
	def update_varinfo1(freq, name):
	return varinfo(freq, name)
	@app.callback(output=Output('var2-info', 'children'),
	inputs=[Input('frequency', component_property='value'),
	Input('name2', component_property='value')
	])
	def update_varinfo2(freq, name):
	return varinfo(freq, name)
	# end placeholders for variable information
	# ------------------------------------------------------------------------

change `Download data` to one link

We currently have:

Download data: q_GDP_yoy.csv q_CPI_rog.csv. 

This should link to:

Download data: csv

with proper start date.

Links to individual files should appear in variable information.

current version crahes in heroku

2017-11-01T15:07:36.729843+00:00 heroku[router]: at=info method=POST path="/_dash-update-component" host=macrodash.herokuapp.com request_id=796fd55c-29b8-4178-acf1-ce3f90e1010e fwd="91.220.93.136" dyno=web.1 connect=0ms service=3ms status=200 bytes=419 protocol=https

2017-11-01T15:07:36.918710+00:00 heroku[router]: at=info method=POST path="/_dash-update-component" host=macrodash.herokuapp.com request_id=1277318b-79b8-490e-bb8f-45d4771191d9 fwd="91.220.93.136" dyno=web.1 connect=0ms service=191ms status=500 bytes=456 protocol=https

2017-11-01T15:07:36.914720+00:00 app[web.1]: [2017-11-01 15:07:36,914] ERROR in app: Exception on /_dash-update-component [POST]

2017-11-01T15:07:36.914737+00:00 app[web.1]: Traceback (most recent call last):

2017-11-01T15:07:36.914740+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.6/site-packages/flask/app.py", line 1982, in wsgi_app

2017-11-01T15:07:36.914741+00:00 app[web.1]:     response = self.full_dispatch_request()

2017-11-01T15:07:36.914742+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.6/site-packages/flask/app.py", line 1614, in full_dispatch_request

2017-11-01T15:07:36.914742+00:00 app[web.1]:     rv = self.handle_user_exception(e)

2017-11-01T15:07:36.914743+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.6/site-packages/flask/app.py", line 1517, in handle_user_exception

2017-11-01T15:07:36.914744+00:00 app[web.1]:     reraise(exc_type, exc_value, tb)

2017-11-01T15:07:36.914744+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.6/site-packages/flask/_compat.py", line 33, in reraise

2017-11-01T15:07:36.914745+00:00 app[web.1]:     raise value

2017-11-01T15:07:36.914746+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.6/site-packages/flask/app.py", line 1612, in full_dispatch_request

2017-11-01T15:07:36.914747+00:00 app[web.1]:     rv = self.dispatch_request()

2017-11-01T15:07:36.914747+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.6/site-packages/flask/app.py", line 1598, in dispatch_request

2017-11-01T15:07:36.914748+00:00 app[web.1]:     return self.view_functions[rule.endpoint](**req.view_args)

2017-11-01T15:07:36.914749+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.6/site-packages/dash/dash.py", line 541, in dispatch

2017-11-01T15:07:36.914750+00:00 app[web.1]:     return self.callback_map[target_id]['callback'](*args)

2017-11-01T15:07:36.914750+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.6/site-packages/dash/dash.py", line 498, in add_context

2017-11-01T15:07:36.914751+00:00 app[web.1]:     output_value = func(*args, **kwargs)

2017-11-01T15:07:36.914751+00:00 app[web.1]:   File "/app/app.py", line 132, in update_graph_parameters

2017-11-01T15:07:36.914752+00:00 app[web.1]:     data_list = [get_time_series_dict(freq, name1)]

2017-11-01T15:07:36.914753+00:00 app[web.1]:   File "/app/app.py", line 81, in get_time_series_dict

2017-11-01T15:07:36.914753+00:00 app[web.1]:     data = get_from_api_datapoints(freq, name)

2017-11-01T15:07:36.914754+00:00 app[web.1]:   File "/app/app.py", line 74, in get_from_api_datapoints

2017-11-01T15:07:36.914754+00:00 app[web.1]:     data = requests.get(url).json()

2017-11-01T15:07:36.914755+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.6/site-packages/requests/models.py", line 892, in json

2017-11-01T15:07:36.914756+00:00 app[web.1]:     return complexjson.loads(self.text, **kwargs)

2017-11-01T15:07:36.914757+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.6/json/__init__.py", line 354, in loads

2017-11-01T15:07:36.914758+00:00 app[web.1]:     return _default_decoder.decode(s)

2017-11-01T15:07:36.914758+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.6/json/decoder.py", line 339, in decode

2017-11-01T15:07:36.914759+00:00 app[web.1]:     obj, end = self.raw_decode(s, idx=_w(s, 0).end())

2017-11-01T15:07:36.914760+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.6/json/decoder.py", line 357, in raw_decode

2017-11-01T15:07:36.914760+00:00 app[web.1]:     raise JSONDecodeError("Expecting value", s, err.value) from None

2017-11-01T15:07:36.914765+00:00 app[web.1]: json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)

move proposed enhancements to another page/tab

need add travis tests

See test example at https://github.com/jackdbd/dash-earthquakes/blob/master/tests/test_basics.py

show latest value in variable information

add css

Fixed by 35f0daa

right pane must be aligned vertically to the top

frontend-dash/app.py

Lines 204 to 209 in c741624

	# FIXME: must align to top
	right_window = html.Div([
	html.Div(["Varibale information"], style={'fontWeight':'bold'}),
	html.Div(id='var1-info'),
	html.Div(id='var2-info'),
	dcc.Markdown("""

URLs to access specific plot in browser

I have an app in plotly/Dash, that draws some charts. Dash produces JavaScript code,
which is rendered by browser. In is working fairly nicely, but has one limitation:
once it is browser-side Javascript code, there is just one address for the whole application.

App: http://macrodash.herokuapp.com
Code: https://github.com/mini-kep/frontend-dash/blob/master/app.py

My desired behaviour is the following: I want to have URLs that link to
to specific charts, eg:

• http://macrodash.herokuapp.com/q/GDP_yoy+CPI_rog
• http://macrodash.herokuapp.com/d/BRENT

It is a bit tricky to do that with Dash and Flask.

In Dash there are callbacks to update/reload browser address:

https://plot.ly/dash/urls

This functionality is ok for changing the addresses once you are at http://macrodash.herokuapp.com and further navigate by selecting variable names and frequencies.

What I'm lacking now is the landing page functionality: how can I type and addrress like http://macrodash.herokuapp.com/q/GDP_yoy+CPI_rog in the browser and further find myself on a page with these indicators plotted?

This functionality involves wrapping Dash with Flask, and I could not find an easy way to do that. Th possible routes are outlined here: plotly/dash#38 (comment)

show custom api url for selected variables + fix if url not working