Allow Bars to be plotted on continuous axes

examples/reference/elements/bokeh/Bars.ipynb

@@ -59,7 +59,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "A ``Bars`` element can be sliced and selecting on like any other element:"
+    "A `Bars` element can be sliced and selected on like any other element:"
    ]
   },
   {
@@ -88,7 +88,41 @@
     "\n",
     "# or using .redim.values(**{'Car Occupants': ['three', 'two', 'four', 'one', 'five', 'six']})\n",
     "\n",
-    "hv.Bars(data, occupants, 'Count') "
+    "hv.Bars(data, occupants, 'Count')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "`Bars` also supports continuous data and x-axis."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data = pd.DataFrame({\"x\": [0, 1, 5], \"y\": [0, 2, 10]})\n",
+    "hv.Bars(data, [\"x\"], [\"y\"])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "And datetime data and x-axis."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data = pd.DataFrame({\"x\": pd.date_range(\"2017-01-01\", \"2017-01-03\"), \"y\": [0, 2, -1]})\n",
+    "hv.Bars(data, [\"x\"], [\"y\"])"
    ]
   },
   {

examples/reference/elements/matplotlib/Bars.ipynb

@@ -17,6 +17,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "import pandas as pd\n",
     "import numpy as np\n",
     "import holoviews as hv\n",
     "hv.extension('matplotlib')"
@@ -80,6 +81,40 @@
     "hv.Bars(data, occupants, 'Count') "
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "`Bars` also supports continuous data and x-axis."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data = pd.DataFrame({\"x\": [0, 1, 5], \"y\": [0, 2, 10]})\n",
+    "hv.Bars(data, [\"x\"], [\"y\"])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "And datetime data and x-axis."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data = pd.DataFrame({\"x\": pd.date_range(\"2017-01-01\", \"2017-01-03\"), \"y\": [0, 2, -1]})\n",
+    "hv.Bars(data, [\"x\"], [\"y\"])"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},
@@ -169,5 +204,5 @@
   }
  },
  "nbformat": 4,
- "nbformat_minor": 2
+ "nbformat_minor": 4
 }

examples/reference/elements/plotly/Bars.ipynb

@@ -17,6 +17,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "import pandas as pd\n",
     "import numpy as np\n",
     "import holoviews as hv\n",
     "hv.extension('plotly')"
@@ -80,6 +81,40 @@
     "hv.Bars(data, occupants, 'Count')"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "`Bars` also support continuous data and x-axis."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data = pd.DataFrame({\"x\": [0, 1, 5], \"y\": [0, 2, 10]})\n",
+    "hv.Bars(data, [\"x\"], [\"y\"])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "And datetime data and x-axis."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data = pd.DataFrame({\"x\": pd.date_range(\"2017-01-01\", \"2017-01-03\"), \"y\": [0, 2, -1]})\n",
+    "hv.Bars(data, [\"x\"], [\"y\"])"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},

holoviews/plotting/bokeh/chart.py

@@ -2,13 +2,13 @@
 
 import numpy as np
 import param
-from bokeh.models import CategoricalColorMapper, CustomJS, FactorRange, Range1d, Whisker
+from bokeh.models import CategoricalColorMapper, CustomJS, Whisker
 from bokeh.models.tools import BoxSelectTool
 from bokeh.transform import jitter
 
 from ...core.data import Dataset
 from ...core.dimension import dimension_name
-from ...core.util import dimension_sanitizer, isfinite
+from ...core.util import dimension_sanitizer, isdatetime, isfinite
 from ...operation import interpolate_curve
 from ...util.transform import dim
 from ..mixins import AreaMixin, BarsMixin, SpikesMixin
@@ -780,10 +780,6 @@ class BarPlot(BarsMixin, ColorbarPlot, LegendPlot):
     _nonvectorized_styles = base_properties + ['bar_width', 'cmap']
     _plot_methods = dict(single=('vbar', 'hbar'))
 
-    # Declare that y-range should auto-range if not bounded
-    _x_range_type = FactorRange
-    _y_range_type = Range1d
-
     def _axis_properties(self, axis, key, plot, dimension=None,
                          ax_mapping=None):
         if ax_mapping is None:
@@ -862,10 +858,10 @@ def _add_color_data(self, ds, ranges, style, cdim, data, mapping, factors, color
 
         # Merge data and mappings
         mapping.update(cmapping)
-        for k, cd in cdata.items():
+        for i, (k, cd) in enumerate(cdata.items()):
             if isinstance(cmapper, CategoricalColorMapper) and cd.dtype.kind in 'uif':
                 cd = categorize_array(cd, cdim)
-            if k not in data or len(data[k]) != next(len(data[key]) for key in data if key != k):
+            if k not in data or (len(data[k]) != next(len(data[key]) for key in data if key != k) and not i == len(cdata) - 1):
                 data[k].append(cd)
             else:
                 data[k][-1] = cd
@@ -889,6 +885,7 @@ def get_data(self, element, ranges, style):
             grouping = 'grouped'
             group_dim = element.get_dimension(1)
 
+        data = defaultdict(list)
         xdim = element.get_dimension(0)
         ydim = element.vdims[0]
         no_cidx = self.color_index is None
@@ -906,25 +903,39 @@ def get_data(self, element, ranges, style):
         hover = 'hover' in self.handles
 
         # Group by stack or group dim if necessary
+        xdiff = None
+        xvals = element.dimension_values(xdim)
         if group_dim is None:
             grouped = {0: element}
+            is_dt = isdatetime(xvals)
+            if is_dt or xvals.dtype.kind != 'O':
+                xdiff = np.diff(xvals)
+                if len(np.unique(xdiff)) == 1 and xdiff[0] == 0:
+                    xdiff = 1
+                if is_dt:
+                    width = (xdiff.astype('timedelta64[ms]').astype(np.int64) * width)
+                    width = np.min(np.abs(width))
+                else:
+                    width = width / xdiff
+                    width = 1 - np.min(np.abs(width))
         else:
             grouped = element.groupby(group_dim, group_type=Dataset,
                                       container_type=dict,
                                       datatype=['dataframe', 'dictionary'])
 
+        width = abs(width)
         y0, y1 = ranges.get(ydim.name, {'combined': (None, None)})['combined']
         if self.logy:
             bottom = (ydim.range[0] or (0.01 if y1 > 0.01 else 10**(np.log10(y1)-2)))
         else:
             bottom = 0
+
         # Map attributes to data
         if grouping == 'stacked':
             mapping = {'x': xdim.name, 'top': 'top',
                        'bottom': 'bottom', 'width': width}
         elif grouping == 'grouped':
-            mapping = {'x': 'xoffsets', 'top': ydim.name, 'bottom': bottom,
-                       'width': width}
+            mapping = {'x': 'xoffsets', 'top': ydim.name, 'bottom': bottom, 'width': width}
         else:
             mapping = {'x': xdim.name, 'top': ydim.name, 'bottom': bottom, 'width': width}
 
@@ -956,7 +967,6 @@ def get_data(self, element, ranges, style):
             factors, colors = None, None
 
         # Iterate over stacks and groups and accumulate data
-        data = defaultdict(list)
         baselines = defaultdict(lambda: {'positive': bottom, 'negative': 0})
         for k, ds in grouped.items():
             k = k[0] if isinstance(k, tuple) else k
@@ -995,7 +1005,7 @@ def get_data(self, element, ranges, style):
                     ds = ds.add_dimension(group_dim, ds.ndims, gval)
                 data[group_dim.name].append(ds.dimension_values(group_dim))
             else:
-                data[xdim.name].append(ds.dimension_values(xdim))
+                data[xdim.name].append(xvals)
                 data[ydim.name].append(ds.dimension_values(ydim))
 
             if hover and grouping != 'stacked':
@@ -1027,7 +1037,7 @@ def get_data(self, element, ranges, style):
 
         # Ensure x-values are categorical
         xname = dimension_sanitizer(xdim.name)
-        if xname in sanitized_data:
+        if xname in sanitized_data and isinstance(sanitized_data[xname], np.ndarray) and sanitized_data[xname].dtype.kind not in 'uifM' and not isdatetime(sanitized_data[xname]):
             sanitized_data[xname] = categorize_array(sanitized_data[xname], xdim)
 
         # If axes inverted change mapping to match hbar signature

holoviews/plotting/bokeh/plot.py

@@ -154,7 +154,6 @@ def _postprocess_data(self, data):
         new_data = {}
         for k, values in data.items():
             values = decode_bytes(values) # Bytes need decoding to strings
-
             # Certain datetime types need to be converted
             if len(values) and isinstance(values[0], cftime_types):
                 if any(v.calendar not in _STANDARD_CALENDARS for v in values):

holoviews/plotting/mixins.py

@@ -160,8 +160,9 @@ def get_extents(self, element, ranges, range_type='combined', **kwargs):
         s0 = min(s0, 0) if util.isfinite(s0) else 0
         s1 = max(s1, 0) if util.isfinite(s1) else 0
         ranges[vdim]['soft'] = (s0, s1)
+        l, b, r, t = super().get_extents(element, ranges, range_type, ydim=element.vdims[0])
         if range_type not in ('combined', 'data'):
-            return super().get_extents(element, ranges, range_type, ydim=element.vdims[0])
+            return l, b, r, t
 
         # Compute stack heights
         xdim = element.kdims[0]
@@ -173,14 +174,15 @@ def get_extents(self, element, ranges, range_type='combined', **kwargs):
         else:
             y0, y1 = ranges[vdim]['combined']
 
+        x0, x1 = (l, r) if util.isnumeric(l) and len(element.kdims) == 1 else ('', '')
         if range_type == 'data':
-            return ('', y0, '', y1)
+            return (x0, y0, x1, y1)
 
         padding = 0 if self.overlaid else self.padding
         _, ypad, _ = get_axis_padding(padding)
         y0, y1 = util.dimension_range(y0, y1, ranges[vdim]['hard'], ranges[vdim]['soft'], ypad, self.logy)
         y0, y1 = util.dimension_range(y0, y1, self.ylim, (None, None))
-        return ('', y0, '', y1)
+        return (x0, y0, x1, y1)
 
     def _get_coords(self, element, ranges, as_string=True):
         """