# Copyright 2019-2020 The Lux Authors.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import lux
import pandas as pd
from typing import Callable
from lux.vislib.altair.BarChart import BarChart
from lux.vislib.altair.ScatterChart import ScatterChart
from lux.vislib.altair.LineChart import LineChart
from lux.vislib.altair.Histogram import Histogram
from lux.vislib.altair.Heatmap import Heatmap
from lux.vislib.altair.Choropleth import Choropleth
[docs]class AltairRenderer:
"""
Renderer for Charts based on Altair (https://altair-viz.github.io/)
"""
[docs] def __init__(self, output_type="VegaLite"):
self.output_type = output_type
def __repr__(self):
return f"AltairRenderer"
[docs] def create_vis(self, vis, standalone=True):
"""
Input Vis object and return a visualization specification
Parameters
----------
vis: lux.vis.Vis
Input Vis (with data)
standalone: bool
Flag to determine if outputted code uses user-defined variable names or can be run independently
Returns
-------
chart : altair.Chart
Output Altair Chart Object
"""
# Lazy Evaluation for 2D Binning
if vis.approx:
if vis.mark == "scatter" and vis._postbin:
vis._mark = "heatmap"
lux.config.executor.execute_2D_binning(vis)
else:
# Exactly recompute the selected vis (e.g., top k) to display
lux.config.executor.execute([vis], vis._original_df, approx=False)
# If a column has a Period dtype, or contains Period objects, convert it back to Datetime
if vis.data is not None:
for attr in list(vis.data.columns):
if pd.api.types.is_period_dtype(vis.data.dtypes[attr]) or isinstance(
vis.data[attr].iloc[0], pd.Period
):
dateColumn = vis.data[attr]
vis.data[attr] = pd.PeriodIndex(dateColumn.values).to_timestamp()
if pd.api.types.is_interval_dtype(vis.data.dtypes[attr]) or isinstance(
vis.data[attr].iloc[0], pd.Interval
):
vis.data[attr] = vis.data[attr].astype(str)
if isinstance(attr, str):
if "." in attr:
attr_clause = vis.get_attr_by_attr_name(attr)[0]
# Suppress special character ".", not displayable in Altair
# attr_clause.attribute = attr_clause.attribute.replace(".", "")
vis._vis_data = vis.data.rename(columns={attr: attr.replace(".", "")})
if vis.mark == "histogram":
chart = Histogram(vis)
elif vis.mark == "bar":
chart = BarChart(vis)
elif vis.mark == "scatter":
chart = ScatterChart(vis)
elif vis.mark == "line":
chart = LineChart(vis)
elif vis.mark == "heatmap":
chart = Heatmap(vis)
elif vis.mark == "geographical":
chart = Choropleth(vis)
else:
chart = None
if chart:
if lux.config.plotting_style and (
lux.config.plotting_backend == "vegalite" or lux.config.plotting_backend == "altair"
):
chart.chart = lux.config.plotting_style(chart.chart)
if self.output_type == "VegaLite":
chart_dict = chart.chart.to_dict()
# this is a bit of a work around because altair must take a pandas dataframe and we can only generate a luxDataFrame
# chart["data"] = { "values": vis.data.to_dict(orient='records') }
# chart_dict["width"] = 160
# chart_dict["height"] = 150
chart_dict["vislib"] = "vegalite"
return chart_dict
elif self.output_type == "Altair":
import inspect
if lux.config.plotting_style:
chart.code += "\n".join(
inspect.getsource(lux.config.plotting_style).split("\n ")[1:-1]
)
chart.code += "\nchart"
chart.code = chart.code.replace("\n\t\t", "\n")
var = vis._source
if var is not None:
all_vars = []
for f_info in inspect.getouterframes(inspect.currentframe()):
local_vars = f_info.frame.f_back
if local_vars:
callers_local_vars = local_vars.f_locals.items()
possible_vars = [
var_name for var_name, var_val in callers_local_vars if var_val is var
]
all_vars.extend(possible_vars)
for possible_var in all_vars:
if possible_var[0] != "_":
print(possible_var)
found_variable = [
possible_var for possible_var in all_vars if possible_var[0] != "_"
]
if len(found_variable) > 0:
found_variable = found_variable[0]
else:
found_variable = "df"
else: # if vis._source was not set when the Vis was created
found_variable = "df"
if standalone:
chart.code = chart.code.replace(
"placeholder_variable",
f"pd.DataFrame({str(vis.data.to_dict())})",
)
else:
# TODO: Placeholder (need to read dynamically via locals())
chart.code = chart.code.replace("placeholder_variable", found_variable)
return chart.code