CLN: Abolish stringify and _strify in favor of pprint_thing()

Author: y-p

pandas/core/common.py

@@ -721,18 +721,6 @@ def _index_labels_to_array(labels):
 
     return labels
 
-def _stringify(col, encoding='UTF8'):
-    # unicode workaround
-    try:
-        return unicode(col)
-    except UnicodeError:
-        try:
-            if isinstance(col, str):
-                return col.decode(encoding)
-        except UnicodeError:
-            pass
-        return console_encode(col)
-
 def _maybe_make_list(obj):
     if obj is not None and not isinstance(obj, (tuple, list)):
         return [obj]

pandas/core/format.py

@@ -8,7 +8,7 @@
 except:
     from io import StringIO
 
-from pandas.core.common import adjoin, isnull, notnull, _stringify
+from pandas.core.common import adjoin, isnull, notnull
 from pandas.core.index import MultiIndex, _ensure_index
 from pandas.util import py3compat
 
@@ -717,12 +717,7 @@ def _format_strings(self, use_unicode=False):
         else:
             float_format = self.float_format
 
-        if use_unicode:
-            def _strify(x):
-                return _stringify(x, print_config.encoding)
-            formatter = _strify if self.formatter is None else self.formatter
-        else:
-            formatter = str if self.formatter is None else self.formatter
+        formatter = com.pprint_thing if self.formatter is None else self.formatter
 
         def _format(x):
             if self.na_rep is not None and lib.checknull(x):

pandas/core/frame.py

@@ -24,8 +24,8 @@
 import numpy as np
 import numpy.ma as ma
 
-from pandas.core.common import (isnull, notnull, PandasError, _try_sort,
-                                _default_index,_stringify,_is_sequence)
+from pandas.core.common import (isnull, notnull, PandasError, _try_sort,\
+                                _default_index,_is_sequence)
 from pandas.core.generic import NDFrame
 from pandas.core.index import Index, MultiIndex, _ensure_index
 from pandas.core.indexing import _NDFrameIndexer, _maybe_droplevels
@@ -1368,7 +1368,7 @@ def info(self, verbose=True, buf=None):
         # hack
         if verbose and len(self.columns) < 100:
             lines.append('Data columns:')
-            space = max([len(_stringify(k)) for k in self.columns]) + 4
+            space = max([len(com.pprint_thing(k)) for k in self.columns]) + 4
             counts = self.count()
             assert(len(cols) == len(counts))
             for col, count in counts.iteritems():

pandas/core/index.py

@@ -1474,15 +1474,9 @@ def get_level_values(self, level):
 
     def format(self, space=2, sparsify=None, adjoin=True, names=False,
                na_rep='NaN'):
-        from pandas.core.common import _stringify
-        from pandas.core.format import print_config
-        def _strify(x):
-            return _stringify(x, print_config.encoding)
-
         if len(self) == 0:
             return []
 
-
         stringified_levels = []
         for lev, lab in zip(self.levels, self.labels):
             if len(lev) > 0:
@@ -1497,7 +1491,7 @@ def _strify(x):
             level = []
 
             if names:
-                level.append(_strify(name) if name is not None else '')
+                level.append(com.pprint_thing(name) if name is not None else '')
 
             level.extend(np.array(lev, dtype=object))
             result_levels.append(level)

pandas/tools/plotting.py

@@ -207,7 +207,7 @@ def normalize(series):
         line = ax.scatter(to_plot[class_][0],
                           to_plot[class_][1],
                           color=random_color(class_),
-                          label=com._stringify(class_), **kwds)
+                          label=com.pprint_thing(class_), **kwds)
     ax.legend()
 
     ax.add_patch(patches.Circle((0.0, 0.0), radius=1.0, facecolor='none'))
@@ -272,8 +272,8 @@ def random_color(column):
         f = function(row)
         y = [f(t) for t in x]
         label = None
-        if com._stringify(class_col[i]) not in used_legends:
-            label = com._stringify(class_col[i])
+        if com.pprint_thing(class_col[i]) not in used_legends:
+            label = com.pprint_thing(class_col[i])
             used_legends.add(label)
         ax.plot(x, y, color=random_color(class_col[i]), label=label)
     ax.legend(loc='upper right')
@@ -378,8 +378,8 @@ def random_color(column):
         y = row
         label = None
         kls = class_col.iget_value(i)
-        if com._stringify(kls) not in used_legends:
-            label = com._stringify(kls)
+        if com.pprint_thing(kls) not in used_legends:
+            label = com.pprint_thing(kls)
             used_legends.add(label)
         ax.plot(x, y, color=random_color(kls), label=label, **kwds)
 
@@ -671,7 +671,7 @@ def _adorn_subplots(self):
                 self.axes[0].set_title(self.title)
 
         if self._need_to_set_index:
-            labels = [_stringify(key) for key in self.data.index]
+            labels = [com.pprint_thing(key) for key in self.data.index]
             labels = dict(zip(range(len(self.data.index)), labels))
 
             for ax_ in self.axes:
@@ -685,10 +685,10 @@ def legend_title(self):
             if not isinstance(self.data.columns, MultiIndex):
                 name = self.data.columns.name
                 if name is not None:
-                    name = com._stringify(name)
+                    name = com.pprint_thing(name)
                 return name
             else:
-                stringified = map(com._stringify,
+                stringified = map(com.pprint_thing,
                                   self.data.columns.names)
                 return ','.join(stringified)
         else:
@@ -742,13 +742,13 @@ def _get_index_name(self):
         if isinstance(self.data.index, MultiIndex):
             name = self.data.index.names
             if any(x is not None for x in name):
-                name = ','.join([com._stringify(x) for x in name])
+                name = ','.join([com.pprint_thing(x) for x in name])
             else:
                 name = None
         else:
             name = self.data.index.name
             if name is not None:
-                name = com._stringify(name)
+                name = com.pprint_thing(name)
 
         return name
 
@@ -806,7 +806,7 @@ def _make_plot(self):
             ax = self._get_ax(i)
             style = self._get_style(i, label)
 
-            label = com._stringify(label)
+            label = com.pprint_thing(label)
 
             gkde = gaussian_kde(y)
             sample_range = max(y) - min(y)
@@ -902,7 +902,7 @@ def _maybe_add_color(kwargs, style, i):
 
                 _maybe_add_color(kwds, style, i)
 
-                label = _stringify(label)
+                label = com.pprint_thing(label).encode('utf-8')
 
                 mask = com.isnull(y)
                 if mask.any():
@@ -947,7 +947,7 @@ def to_leg_label(label, i):
         if isinstance(data, Series):
             ax = self._get_ax(0) #self.axes[0]
             style = self.style or ''
-            label = com._stringify(self.label)
+            label = com.pprint_thing(self.label)
             kwds = kwargs.copy()
             _maybe_add_color(kwds, style, 0)
 
@@ -959,7 +959,7 @@ def to_leg_label(label, i):
             labels.append(leg_label)
         else:
             for i, col in enumerate(data.columns):
-                label = com._stringify(col)
+                label = com.pprint_thing(col)
                 ax = self._get_ax(i)
                 style = self._get_style(i, col)
                 kwds = kwargs.copy()
@@ -1097,7 +1097,7 @@ def _make_plot(self):
         K = self.nseries
 
         for i, (label, y) in enumerate(self._iter_data()):
-            label = com._stringify(label)
+            label = com.pprint_thing(label)
             kwds = self.kwds.copy()
             kwds['color'] = colors[i % len(colors)]
 
@@ -1125,7 +1125,7 @@ def _make_plot(self):
 
     def _post_plot_logic(self):
         for ax in self.axes:
-            str_index = [_stringify(key) for key in self.data.index]
+            str_index = [com.pprint_thing(key) for key in self.data.index]
 
             name = self._get_index_name()
             if self.kind == 'bar':
@@ -1359,7 +1359,7 @@ def boxplot(data, column=None, by=None, ax=None, fontsize=None,
 
     def plot_group(grouped, ax):
         keys, values = zip(*grouped)
-        keys = [_stringify(x) for x in keys]
+        keys = [com.pprint_thing(x) for x in keys]
         values = [remove_na(v) for v in values]
         ax.boxplot(values, **kwds)
         if kwds.get('vert', 1):
@@ -1394,7 +1394,7 @@ def plot_group(grouped, ax):
             cols = columns
         else:
             cols = data.columns
-        keys = [_stringify(x) for x in cols]
+        keys = [com.pprint_thing(x) for x in cols]
 
         # Return boxplot dict in single plot case
 
@@ -1411,14 +1411,6 @@ def plot_group(grouped, ax):
     fig.subplots_adjust(bottom=0.15, top=0.9, left=0.1, right=0.9, wspace=0.2)
     return ret
 
-
-def _stringify(x):
-    if isinstance(x, tuple):
-        return '|'.join(com._stringify(y) for y in x)
-    else:
-        return com._stringify(x)
-
-
 def format_date_labels(ax, rot):
     # mini version of autofmt_xdate
     try:
@@ -1455,8 +1447,8 @@ def plot_group(group, ax):
         else:
             fig = ax.get_figure()
         plot_group(data, ax)
-        ax.set_ylabel(com._stringify(y))
-        ax.set_xlabel(com._stringify(x))
+        ax.set_ylabel(com.pprint_thing(y))
+        ax.set_xlabel(com.pprint_thing(x))
 
         ax.grid(grid)
 
@@ -1620,7 +1612,7 @@ def boxplot_frame_groupby(grouped, subplots=True, column=None, fontsize=None,
         for (key, group), ax in zip(grouped, axes):
             d = group.boxplot(ax=ax, column=column, fontsize=fontsize,
                               rot=rot, grid=grid, figsize=figsize, **kwds)
-            ax.set_title(_stringify(key))
+            ax.set_title(com.pprint_thing(key))
             ret[key] = d
     else:
         from pandas.tools.merge import concat
@@ -1676,7 +1668,7 @@ def _grouped_plot(plotf, data, column=None, by=None, numeric_only=True,
         if numeric_only and isinstance(group, DataFrame):
             group = group._get_numeric_data()
         plotf(group, ax)
-        ax.set_title(com._stringify(key))
+        ax.set_title(com.pprint_thing(key))
 
     return fig, axes
 
@@ -1710,7 +1702,7 @@ def _grouped_plot_by_column(plotf, data, columns=None, by=None,
         gp_col = grouped[col]
         plotf(gp_col, ax)
         ax.set_title(col)
-        ax.set_xlabel(com._stringify(by))
+        ax.set_xlabel(com.pprint_thing(by))
         ax.grid(grid)
 
     byline = by[0] if len(by) == 1 else by

pandas/tseries/plotting.py

@@ -158,7 +158,7 @@ def _replot_ax(ax, freq, plotf, kwargs):
             ax._plot_data.append(series)
             args = _maybe_mask(series)
             lines.append(plotf(ax, *args, **kwds)[0])
-            labels.append(com._stringify(series.name))
+            labels.append(com.pprint_thing(series.name))
 
     return lines, labels