[*.py] Upgrade to latest version of pandas; Python 2/3 compatibility

.gitignore

@@ -1,3 +1,4 @@
+.idea
 *.pyc
 *.swp
 *.DS_Store

README.md

@@ -1,9 +1,18 @@
-# ggplot
+# ggplot - a working, maintained fork
 
-<img src="./examples/example-34d773b9-ec68-40b1-999b-7bb07c208be9.png" width="400px" />
-<img src="./examples/example-8f4fbffe-2999-42b0-9c34-de6f0b205733.png" width="400px" />
+## Why this fork?
+`ggplot` is a great python library. However, it is no longer maintained by its owner and still has a bunch of issues which remain unsolved. Some of these include incompatibility with newer versions of `pandas` and Python 3.
 
-### What is it?
+Many projects still rely on `ggplot` and many have to either move to alternatives or manually update `ggpy`'s code after installing it  to fix compatibility issues. To get a better context of the issue you can refer to [#654 Is this project dead?](https://github.com/yhat/ggpy/issues/654)
+
+This fork is a working copy of `ggplot` which is readily maintained and is open to updates and fixes so that developers do not have to make fixes manually.
+
+## Installation
+```bash
+$ pip3 install git+https://github.com/sushinoya/ggpy
+```
+
+## What is ggpy?
 `ggplot` is a Python implementation of the grammar of graphics. It is not intended
 to be a feature-for-feature port of [`ggplot2 for R`](https://github.com/hadley/ggplot2)--though 
 there is much greatness in `ggplot2`, the Python world could stand to benefit 
@@ -19,31 +28,3 @@ ggplot(diamonds, aes(x='price', color='clarity')) + \
     facet_wrap('cut')
 ```
 ![](./docs/example.png)
-
-### Installation
-```bash
-$ pip install -U ggplot
-# or 
-$ conda install -c conda-forge ggplot
-# or
-pip install git+https://github.com/yhat/ggplot.git
-```
-
-### Examples
-Examples are the best way to learn. There is a Jupyter Notebook full of them. 
-There are also notebooks that show how to do particular things with ggplot 
-(i.e. [make a scatter plot](./docs/how-to/Making%20a%20Scatter%20Plot.ipynb) or [make a histogram](./docs/how-to/Making%20a%20Scatter%20Plot.ipynb)).
-
-- [docs](./docs)
-- [gallery](./docs/Gallery.ipynb)
-- [various examples](./examples.md)
-
-
-### What happened to the old version that didn't work?
-It's gone--the windows, the doors, [everything](https://www.youtube.com/watch?v=YuxCKv_0GZc). 
-Just kidding, [you can find it here](https://github.com/yhat/ggplot/tree/v0.6.6), though I'm not sure why you'd want to look at it. The data grouping and manipulation bits were re-written
-(so they actually worked) with things like facets in mind.
-
-### Contributing
-Thanks to all of the ggplot [contributors](./contributors.md#contributors)!
-See *[contributing.md](./contributing.md)*.

docs/examples.py

@@ -103,7 +103,7 @@
 df = pd.DataFrame({"x": np.arange(1000)})
 df['y_low'] = df.x * 0.9
 df['y_high'] = df.x * 1.1
-df['thing'] = ['a' if i%2==0 else 'b' for i in df.x]
+df['thing'] = ['a' if i % 2 == 0 else 'b' for i in df.x]
 p = ggplot(df, aes(x='x', ymin='y_low', ymax='y_high')) + geom_area()
 p.save("./examples/example-" + str(uuid.uuid4()) + ".png")
 # # area w/ facet
@@ -131,7 +131,7 @@
 #
 df = pd.DataFrame({"x": np.arange(100)})
 df['y'] = df.x * 10
-df['z'] = ["a" if x%2==0 else "b" for x in df.x]
+df['z'] = ["a" if x % 2 == 0 else "b" for x in df.x]
 #
 # # polar coords
 p = ggplot(df, aes(x='x', y='y')) + geom_point() + coord_polar()
@@ -158,7 +158,7 @@
 p.save("./examples/example-" + str(uuid.uuid4()) + ".png")
 #
 # # # x dates formatting faceted
-pageviews['z'] = ["a" if i%2==0 else "b" for i in range(len(pageviews))]
+pageviews['z'] = ["a" if i % 2 == 0 else "b" for i in range(len(pageviews))]
 p = ggplot(pageviews, aes(x='date_hour', y='pageviews')) + geom_line() + scale_x_date(labels=date_format('%B %-d, %Y')) + facet_grid(y='z')
 p.save("./examples/example-" + str(uuid.uuid4()) + ".png")
 #

ggplot/aes.py

@@ -8,11 +8,10 @@
 
 from patsy.eval import EvalEnvironment
 
-from . import utils
-
 import numpy as np
 import pandas as pd
 
+
 class aes(UserDict):
     """
     Creates a dictionary that is used to evaluate
@@ -72,7 +71,7 @@ def __init__(self, *args, **kwargs):
         self.__eval_env__ = EvalEnvironment.capture(1)
 
     def __deepcopy__(self, memo):
-        '''deepcopy support for ggplot'''
+        """deepcopy support for ggplot"""
         result = aes()
         for key, item in self.__dict__.items():
             # don't make a deepcopy of the env!
@@ -122,7 +121,7 @@ def _get_discrete_aes(self, df):
         for aes_type, column in self.data.items():
             if aes_type in ['x', 'y']:
                 continue
-            elif aes_type=="group":
+            elif aes_type == "group":
                 discrete_aes.append((aes_type, column))
             elif column not in non_numeric_columns:
                 continue

ggplot/chart_components.py

@@ -46,13 +46,13 @@ class xlim(object):
     >>> ggplot(mpg, aes(x='hwy')) + geom_hisotgram() + xlim(0, 20)
     """
     def __init__(self, low = None, high = None):
-        if low != None :
+        if low is not None:
             try:
                 _ = low - 0
             except TypeError:
                 raise Exception("The 'low' argument to", self.__class__.__name__,
                                 "must be of a numeric type or None")
-        if high != None :
+        if high is not None:
             try:
                 _ = high - 0
             except TypeError:
@@ -83,13 +83,13 @@ class ylim(object):
     >>> ggplot(mpg, aes(x='hwy')) + geom_hisotgram() + ylim(0, 5)
     """
     def __init__(self, low = None, high = None):
-        if low != None :
+        if low is not None:
             try:
                 _ = low - 0
             except TypeError:
                 raise Exception("The 'low' argument to", self.__class__.__name__,
                                 "must be of a numeric type or None")
-        if high != None :
+        if high is not None:
             try:
                 _ = high - 0
             except TypeError:
@@ -140,7 +140,7 @@ class ylab(object):
 
     Examples
     --------
-    >>> ggplot(mpg, aes(x='hwy')) + geom_hisotgram() + ylab("Count\n(# of cars)")
+    >>> ggplot(mpg, aes(x='hwy')) + geom_hisotgram() + ylab('''Count\n(# of cars)''')
     """
     def __init__(self, ylab):
         if ylab is None:
@@ -169,7 +169,7 @@ class labs(object):
 
     Examples
     --------
-    >>> ggplot(mpg, aes(x='hwy')) + geom_hisotgram() + labs("Miles / gallon", "Count\n(# of cars)", "MPG Plot")
+    >>> ggplot(mpg, aes(x='hwy')) + geom_hisotgram() + labs("Miles / gallon", '''Count\n(# of cars)''', "MPG Plot")
     """
     def __init__(self, x=None, y=None, title=None):
         self.x = x

ggplot/facets.py

@@ -20,7 +20,7 @@ def __init__(self, data, is_wrap, rowvar=None, colvar=None, nrow=None, ncol=None
         # assign subplot indices to rowvars and columnvars
         self.ndim = ndim = self.calculate_ndimensions(data, rowvar, colvar)
 
-        if is_wrap==True:
+        if is_wrap:
             if self.nrow:
                 self.ncol = ncol = int(math.ceil(ndim / float(self.nrow)))
                 self.nrow = nrow = int(self.nrow)
@@ -47,9 +47,9 @@ def __init__(self, data, is_wrap, rowvar=None, colvar=None, nrow=None, ncol=None
                     value = next(facet_values)
                 except Exception as e:
                     continue
-                if ncol==1:
+                if ncol == 1:
                     self.facet_map[value] = (row, None)
-                elif nrow==1:
+                elif nrow == 1:
                     self.facet_map[value] = (None, col)
                 else:
                     self.facet_map[value] = (row, col)
@@ -119,12 +119,13 @@ def __init__(self, x=None, y=None, nrow=None, ncol=None, scales=None):
         self.scales = scales
 
     def __radd__(self, gg):
-        if gg.__class__.__name__=="ggplot":
+        if gg.__class__.__name__ == "ggplot":
             gg.facets = Facet(gg.data, True, self.x_var, self.y_var, nrow=self.nrow, ncol=self.ncol, scales=self.scales)
             return gg
 
         return self
 
+
 class facet_grid(object):
     """
     Layout panels from x and (optionally) y variables in a grid format.
@@ -155,7 +156,7 @@ def __init__(self, x=None, y=None, scales=None):
         self.scales = scales
 
     def __radd__(self, gg):
-        if gg.__class__.__name__=="ggplot":
+        if gg.__class__.__name__ == "ggplot":
             gg.facets = Facet(gg.data, False, self.x_var, self.y_var, scales=self.scales)
             return gg
         return self

ggplot/geoms/geom.py

@@ -59,7 +59,7 @@ def _get_plot_args(self, data, _aes):
         for key, value in _aes.items():
             if value not in data:
                 mpl_params[key] = value
-            elif data[value].nunique()==1:
+            elif data[value].nunique() == 1:
                 mpl_params[key] = data[value].iloc[0]
             else:
                 mpl_params[key] = data[value]

ggplot/geoms/geom_area.py

@@ -43,7 +43,7 @@ def plot(self, ax, data, _aes):
         if self.last_y is None:
             self.last_y = pd.Series(np.repeat(0, len(data)))
         ymin = self.last_y
-        if self.DEFAULT_PARAMS['position']=="stack":
+        if self.DEFAULT_PARAMS['position'] == "stack":
             ymax = self.last_y.reset_index(drop=True) + data[variables['y']].reset_index(drop=True)
         else:
             ymax = data[variables['y']]

ggplot/geoms/geom_bar.py

@@ -91,9 +91,9 @@ def plot(self, ax, data, _aes, x_levels, fill_levels, lookups):
 
         xticks = []
         for i, x_level in enumerate(x_levels):
-            mask = data[variables['x']]==x_level
+            mask = data[variables['x']] == x_level
             row = data[mask]
-            if len(row)==0:
+            if len(row) == 0:
                 xticks.append(i)
                 continue
 
@@ -111,19 +111,19 @@ def plot(self, ax, data, _aes, x_levels, fill_levels, lookups):
                     height = 1.0
                     ypos = 0
                 else:
-                    mask = (lookups[variables['x']]==x_level) & (lookups[variables['fill']]==fillval)
+                    mask = (lookups[variables['x']] == x_level) & (lookups[variables['fill']] == fillval)
                     height = lookups[mask]['__calc_weight__'].sum()
-                    mask = (lookups[variables['x']]==x_level) & (lookups[variables['fill']] < fillval)
+                    mask = (lookups[variables['x']] == x_level) & (lookups[variables['fill']] < fillval)
                     ypos = lookups[mask]['__calc_weight__'].sum()
             else:
                 if fill_levels is not None:
-                    dodge = (width * fill_idx)
+                    dodge = width * fill_idx
                 else:
                     dodge = width
                 ypos = 0.0
                 height = row[weight_col].sum()
 
-            xy = (dodge + i  - fill_x_adjustment, ypos)
+            xy = (dodge + i - fill_x_adjustment, ypos)
 
             ax.add_patch(patches.Rectangle(xy, width, height, **params))
             if fill_levels is not None:

ggplot/geoms/geom_boxplot.py

@@ -34,30 +34,30 @@ def plot(self, ax, data, _aes, x_levels):
 
         xticks = []
         for (i, xvalue) in enumerate(x_levels):
-            subset = data[data[variables['x']]==xvalue]
+            subset = data[data[variables['x']] == xvalue]
             xi = np.repeat(i, len(subset))
             yvalues = subset[variables['y']]
             xticks.append(i)
 
             bounds_25_75 = yvalues.quantile([0.25, 0.75]).values
             bounds_5_95 = yvalues.quantile([0.05, 0.95]).values
 
-            if self.params.get('outliers', True)==True:
+            if self.params.get('outliers', True):
                 mask = ((yvalues > bounds_5_95[1]) | (yvalues < bounds_5_95[0])).values
                 ax.scatter(x=xi[mask], y=yvalues[mask], c=self.params.get('outlier_color', 'black'))
 
-            if self.params.get('lines', True)==True:
+            if self.params.get('lines', True):
                 ax.vlines(x=i, ymin=bounds_25_75[1], ymax=bounds_5_95[1])
                 ax.vlines(x=i, ymin=bounds_5_95[0], ymax=bounds_25_75[0])
 
-            if self.params.get('notch', False)==True:
+            if self.params.get('notch', False):
                 ax.hlines(bounds_5_95[0], i - 0.25/2, i + 0.25/2, linewidth=2)
                 ax.hlines(bounds_5_95[1], i - 0.25/2, i + 0.25/2, linewidth=2)
 
-            if self.params.get('median', True)==True:
+            if self.params.get('median', True):
                 ax.hlines(yvalues.median(), i - 0.25, i + 0.25, linewidth=2)
 
-            if self.params.get('box', True)==True:
+            if self.params.get('box', True):
                 params = {
                     'facecolor': 'white',
                     'edgecolor': 'black',

ggplot/geoms/geom_density.py

@@ -49,6 +49,6 @@ def plot(self, ax, data, _aes):
         params = self._get_plot_args(data, _aes)
         variables = _aes.data
         x = data[variables['x']]
-        x = x[x.isnull()==False]
+        x = x[x.isnull()  == False]
         x, y = self._calculate_density(x)
         ax.plot(x, y, **params)

ggplot/geoms/geom_histogram.py

@@ -42,7 +42,7 @@ def plot(self, ax, data, _aes):
 
         variables = _aes.data
         x = data[variables['x']]
-        x = x[x.isnull()==False]
+        x = x[x.isnull()  == False]
 
         if 'binwidth' in self.params:
             params['bins'] = np.arange(np.min(x), np.max(x) + self.params['binwidth'], self.params['binwidth'])

ggplot/geoms/geom_line.py

@@ -37,8 +37,8 @@ def plot(self, ax, data, _aes):
         y = data[variables['y']]
 
         nulls = (x.isnull() | y.isnull())
-        x = x[nulls==False]
-        y = y[nulls==False]
+        x = x[nulls  == False]
+        y = y[nulls  == False]
 
         if self.is_path:
             pass

ggplot/geoms/geom_step.py

@@ -37,8 +37,8 @@ def plot(self, ax, data, _aes):
         y = data[variables['y']]
 
         nulls = (x.isnull() | y.isnull())
-        x = x[nulls==False]
-        y = y[nulls==False]
+        x = x[nulls  == False]
+        y = y[nulls  == False]
 
         xs = [None] * (2 * (len(x)-1))
         ys = [None] * (2 * (len(x)-1))

ggplot/geoms/geom_tile.py

@@ -55,7 +55,7 @@ def plot(self, ax, data, _aes):
         counts = data[[weight, variables['x'] + "_cut", variables['y'] + "_cut"]].groupby([variables['x'] + "_cut", variables['y'] + "_cut"]).count().fillna(0)
         weighted = data[[weight, variables['x'] + "_cut", variables['y'] + "_cut"]].groupby([variables['x'] + "_cut", variables['y'] + "_cut"]).sum().fillna(0)
 
-        if self.params['interpolate']==False:
+        if self.params['interpolate']  == False:
             def get_xy():
                 for x in x_bins:
                     for y in y_bins:

ggplot/geoms/geom_violin.py

@@ -27,8 +27,8 @@ def plot(self, ax, data, _aes, x_levels):
         variables = _aes.data
 
         xticks = []
-        for (i, xvalue) in enumerate(x_levels):
-            subset = data[data[variables['x']]==xvalue]
+        for i, xvalue in enumerate(x_levels):
+            subset = data[data[variables['x']] == xvalue]
             yi = subset[variables['y']].values
 
             # so this is weird, apparently violinplot is *the only plot that