Basic support for plotting lists of strings/categorical data. Support…

… for updating ticks/animation in progress/buggy, 'especially for scatter.
matplotlib · tacaswell · Jul 8, 2016 · Jun 16, 2016 · Jun 17, 2016 · Jul 8, 2016
commit b1848420bf80cb40211f37fb32f3a2d07e9b5f7f
diff --git a/lib/matplotlib/__init__.py b/lib/matplotlib/__init__.py
@@ -1489,6 +1489,7 @@ def _jupyter_nbextension_paths():
     'matplotlib.tests.test_backend_svg',
     'matplotlib.tests.test_basic',
     'matplotlib.tests.test_bbox_tight',
+    'matplotlib.tests.test_category',
     'matplotlib.tests.test_cbook',
     'matplotlib.tests.test_coding_standards',
     'matplotlib.tests.test_collections',

diff --git a/lib/matplotlib/axes/_axes.py b/lib/matplotlib/axes/_axes.py
@@ -21,6 +21,7 @@
 import matplotlib.collections as mcoll
 import matplotlib.colors as mcolors
 import matplotlib.contour as mcontour
+import matplotlib.category as _  # <-registers a date unit converter
 import matplotlib.dates as _  # <-registers a date unit converter
 from matplotlib import docstring
 import matplotlib.image as mimage

diff --git a/lib/matplotlib/axis.py b/lib/matplotlib/axis.py
@@ -662,6 +662,7 @@ def __init__(self, axes, pickradius=15):
         self.offsetText = self._get_offset_text()
         self.majorTicks = []
         self.minorTicks = []
+        self.unit_data = []
         self.pickradius = pickradius
 
         # Initialize here for testing; later add API
@@ -712,6 +713,17 @@ def _set_scale(self, value, **kwargs):
     def limit_range_for_scale(self, vmin, vmax):
         return self._scale.limit_range_for_scale(vmin, vmax, self.get_minpos())
 
+    @property
+    def unit_data(self):
+        """Holds data that a ConversionInterface subclass relys on
+        to convert between labels and indexes
+        """
+        return self._unit_data
+
+    @unit_data.setter
+    def unit_data(self, data):
+        self._unit_data = data
+
     def get_children(self):
         children = [self.label, self.offsetText]
         majorticks = self.get_major_ticks()

diff --git a/lib/matplotlib/category.py b/lib/matplotlib/category.py
@@ -0,0 +1,117 @@
+"""
+catch all for categorical functions
+"""
+from __future__ import (absolute_import, division, print_function,
+                        unicode_literals)
+
+import six
+import numpy as np
+
+import matplotlib.units as units
+import matplotlib.ticker as ticker
+
+
+class StrCategoryConverter(units.ConversionInterface):
+    @staticmethod
+    def convert(value, unit, axis):
+        """Uses axis.unit_data map to encode
+        data as integers
+        """
+
+        if isinstance(value, six.string_types):
+            return dict(axis.unit_data)[value]
+
+        vals = np.asarray(value, dtype='str')
+        for label, loc in axis.unit_data:
+            vals[vals == label] = loc
+        return vals.astype('float')
+
+    @staticmethod
+    def axisinfo(unit, axis):
+        seq, locs = zip(*axis.unit_data)
+        majloc = StrCategoryLocator(locs)
+        majfmt = StrCategoryFormatter(seq)
+        return units.AxisInfo(majloc=majloc, majfmt=majfmt)
+
+    @staticmethod
+    def default_units(data, axis):
+        # the conversion call stack is:
+        # default_units->axis_info->convert
+        axis.unit_data = map_categories(data, axis.unit_data)
+        return None
+
+
+def map_categories(data, old_map=[], sort=True):
+    """Create mapping between unique categorical
+    values and numerical identifier.
+
+    Paramters
+    ---------
+    data: iterable
+        sequence of values
+    old_map: list of tuple, optional
+        if not `None`, than old_mapping will be updated with new values and
+        previous mappings will remain unchanged)
+    sort: bool, optional
+        sort keys by ASCII value
+
+    Returns
+    -------
+    list of tuple
+        [(label, ticklocation),...]
+
+    """
+
+    # code typical missing data in the negative range because
+    # everything else will always have positive encoding
+    # question able if it even makes sense
+    spdict = {'nan': -1, 'inf': -2, '-inf': -3}
+
+    # cast all data to str
+    strdata = [str(d) for d in data]
+
+    uniq = set(strdata)
+
+    category_map = old_map.copy()
+
+    if old_map:
+        olabs, okeys = zip(*old_map)
+        olabs, okeys = set(olabs), set(okeys)
+        svalue = max(okeys) + 1
+    else:
+        olabs, okeys = set(), set()
+        svalue = 0
+
+    new_labs = (uniq - olabs)
+
+    missing = (new_labs & set(spdict.keys()))
+    category_map.extend([(m, spdict[m]) for m in missing])
+
+    new_labs = (new_labs - missing)
+    if sort:
+        new_labs = list(new_labs)
+        new_labs.sort()
+
+    new_locs = range(svalue, svalue + len(new_labs))
+    category_map.extend(list(zip(new_labs, new_locs)))
+    return category_map
+
+
+class StrCategoryLocator(ticker.FixedLocator):
+    def __init__(self, locs):
+        super(StrCategoryLocator, self).__init__(locs, None)
+
+
+class StrCategoryFormatter(ticker.FixedFormatter):
+    def __init__(self, seq):
+        super(StrCategoryFormatter, self).__init__(seq)
+
+
+# Connects the convertor to matplotlib
+units.registry[bytearray] = StrCategoryConverter()
+units.registry[str] = StrCategoryConverter()
+
+if six.PY3:
+    units.registry[bytes] = StrCategoryConverter()
+elif six.PY2:
+    units.registry[unicode] = StrCategoryConverter()
diff --git a/lib/matplotlib/tests/baseline_images/test_axes/units_strings.pdf b/lib/matplotlib/tests/baseline_images/test_axes/units_strings.pdf
diff --git a/lib/matplotlib/tests/baseline_images/test_axes/units_strings.png b/lib/matplotlib/tests/baseline_images/test_axes/units_strings.png