diff --git a/pandas/core/format.py b/pandas/core/format.py
index c2f439877ca00..5a69d8f03f12b 100644
--- a/pandas/core/format.py
+++ b/pandas/core/format.py
@@ -26,6 +26,10 @@
from pandas.tseries.period import PeriodIndex, DatetimeIndex
+horiz_ellips = u('\u22ef') # ⋯
+vert_ellips = u('\u22ee') # ⋮
+diag_ellips = u('\u22f1') # ⋱
+
docstring_to_string = """
Parameters
----------
@@ -314,6 +318,10 @@ def __init__(self, frame, buf=None, columns=None, col_space=None,
self._chk_truncate()
def _chk_truncate(self):
+ '''
+ If truncation necessary, prepare data and additional
+ information.
+ '''
from pandas.tools.merge import concat
truncate_h = self.max_cols and (len(self.columns) > self.max_cols)
@@ -354,7 +362,6 @@ def _to_str_columns(self):
# may include levels names also
str_index = self._get_formatted_index(frame)
-
str_columns = self._get_formatted_column_labels(frame)
if self.header:
@@ -370,7 +377,6 @@ def _to_str_columns(self):
fmt_values = _make_fixed_width(fmt_values, self.justify,
minimum=max_colwidth)
-
max_len = max(np.max([_strlen(x) for x in fmt_values]),
max_colwidth)
if self.justify == 'left':
@@ -395,32 +401,26 @@ def _to_str_columns(self):
# Add ... to signal truncated
truncate_h = self.truncate_h
truncate_v = self.truncate_v
-
+ nlevels_row = frame.index.nlevels
+ nlevels_col = frame.columns.nlevels #len(str_index) - len(frame)
if truncate_h:
col_num = self.tr_col_num
col_width = len(strcols[col_num][0]) # infer from column header
- strcols.insert(col_num + 1, ['...'.center(col_width)] * (len(str_index)))
- if truncate_v:
- n_header_rows = len(str_index) - len(frame)
+ strcols.insert(col_num + 1, [ horiz_ellips.center(col_width) ] * (len(str_index)))
+ if truncate_v:
row_num = self.tr_row_num
for ix,col in enumerate(strcols):
- cwidth = len(strcols[ix][row_num]) # infer from above row
- is_dot_col = False
- if truncate_h:
- is_dot_col = ix == col_num + 1
- if cwidth > 3 or is_dot_col:
- my_str = '...'
- else:
- my_str = '..'
-
+ cwidth = max(len(ele) for ele in strcols[ix][:row_num + nlevels_col]) # infer from above rows
if ix == 0:
- dot_str = my_str.ljust(cwidth)
- elif is_dot_col:
- dot_str = my_str.center(cwidth)
+ cwidth_split = ( cwidth - ( nlevels_row - 1 ) ) // nlevels_row
+ dot_str = vert_ellips.center(cwidth_split)
+ for lvl in range(2,nlevels_row):
+ dot_str += ' ' + vert_ellips.center(cwidth_split)
else:
- dot_str = my_str.rjust(cwidth)
-
- strcols[ix].insert(row_num + n_header_rows, dot_str)
+ dot_str = vert_ellips.center(cwidth)
+ strcols[ix].insert(row_num + nlevels_row, dot_str)
+ if truncate_h and truncate_v:
+ strcols[col_num + 1][nlevels_col + row_num] = diag_ellips.center(col_width)
return strcols
@@ -428,7 +428,6 @@ def to_string(self):
"""
Render a DataFrame to a console-friendly tabular output.
"""
-
frame = self.frame
if len(frame.columns) == 0 or len(frame.index) == 0:
@@ -464,7 +463,7 @@ def _join_multiline(self, *strcols):
col_bins = _binify(col_widths, lwidth)
nbins = len(col_bins)
- if self.max_rows and len(self.frame) > self.max_rows:
+ if self.truncate_v:
nrows = self.max_rows + 1
else:
nrows = len(self.frame)
@@ -836,7 +835,7 @@ def _column_header():
elif tag + span > ins_col:
recs_new[tag] = span + 1
if lnum == inner_lvl:
- values = values[:ins_col] + (u('...'),) + \
+ values = values[:ins_col] + (horiz_ellips,) + \
values[ins_col:]
else: # sparse col headers do not receive a ...
values = values[:ins_col] + \
@@ -846,7 +845,7 @@ def _column_header():
# if ins_col lies between tags, all col headers get ...
if tag + span == ins_col:
recs_new[ins_col] = 1
- values = values[:ins_col] + (u('...'),) + \
+ values = values[:ins_col] + (horiz_ellips,) + \
values[ins_col:]
records = recs_new
inner_lvl = len(level_lengths) - 1
@@ -861,8 +860,7 @@ def _column_header():
recs_new[tag] = span
recs_new[ins_col] = 1
records = recs_new
- values = values[:ins_col] + [u('...')] + values[ins_col:]
-
+ values = values[:ins_col] + [horiz_ellips] + values[ins_col:]
name = self.columns.names[lnum]
row = [''] * (row_levels - 1) + ['' if name is None
else com.pprint_thing(name)]
@@ -876,6 +874,7 @@ def _column_header():
continue
j += 1
row.append(v)
+
self.write_tr(row, indent, self.indent_delta, tags=tags,
header=True)
else:
@@ -884,8 +883,7 @@ def _column_header():
if truncate_h:
ins_col = row_levels + self.fmt.tr_col_num
- col_row.insert(ins_col, '...')
-
+ col_row.insert(ins_col, horiz_ellips)
self.write_tr(col_row, indent, self.indent_delta, header=True,
align=align)
@@ -931,6 +929,8 @@ def _write_body(self, indent):
def _write_regular_rows(self, fmt_values, indent):
truncate_h = self.fmt.truncate_h
truncate_v = self.fmt.truncate_v
+ if truncate_h:
+ dot_col_ix = self.fmt.tr_col_num + 1
ncols = len(self.fmt.tr_frame.columns)
nrows = len(self.fmt.tr_frame)
@@ -941,19 +941,21 @@ def _write_regular_rows(self, fmt_values, indent):
index_values = self.fmt.tr_frame.index.format()
for i in range(nrows):
-
if truncate_v and i == (self.fmt.tr_row_num):
- str_sep_row = [ '...' for ele in row ]
+ str_sep_row = [ vert_ellips for ele in row ]
+ if truncate_h:
+ str_sep_row = str_sep_row[:dot_col_ix] + \
+ [diag_ellips] + \
+ str_sep_row[dot_col_ix+1:]
self.write_tr(str_sep_row, indent, self.indent_delta, tags=None,
nindex_levels=1)
-
row = []
row.append(index_values[i])
row.extend(fmt_values[j][i] for j in range(ncols))
if truncate_h:
- dot_col_ix = self.fmt.tr_col_num + 1
- row.insert(dot_col_ix, '...')
+ row.insert(dot_col_ix, horiz_ellips)
+
self.write_tr(row, indent, self.indent_delta, tags=None,
nindex_levels=1)
@@ -962,6 +964,8 @@ def _write_hierarchical_rows(self, fmt_values, indent):
truncate_h = self.fmt.truncate_h
truncate_v = self.fmt.truncate_v
+ if truncate_h:
+ dot_col_ix = self.fmt.tr_col_num + 1
frame = self.fmt.tr_frame
ncols = len(frame.columns)
nrows = len(frame)
@@ -991,7 +995,7 @@ def _write_hierarchical_rows(self, fmt_values, indent):
elif tag + span > ins_row:
rec_new[tag] = span + 1
dot_row = list(idx_values[ins_row - 1])
- dot_row[-1] = u('...')
+ dot_row[-1] = vert_ellips
idx_values.insert(ins_row,tuple(dot_row))
else:
rec_new[tag] = span
@@ -999,12 +1003,12 @@ def _write_hierarchical_rows(self, fmt_values, indent):
if tag + span == ins_row:
rec_new[ins_row] = 1
if lnum == 0:
- idx_values.insert(ins_row,tuple([u('...')]*len(level_lengths)))
+ idx_values.insert(ins_row,tuple([vert_ellips]*len(level_lengths)))
level_lengths[lnum] = rec_new
level_lengths[inner_lvl][ins_row] = 1
for ix_col in range(len(fmt_values)):
- fmt_values[ix_col].insert(ins_row,'...')
+ fmt_values[ix_col].insert(ins_row,vert_ellips)
nrows += 1
for i in range(nrows):
@@ -1026,19 +1030,33 @@ def _write_hierarchical_rows(self, fmt_values, indent):
row.extend(fmt_values[j][i] for j in range(ncols))
if truncate_h:
- row.insert(row_levels - sparse_offset + self.fmt.tr_col_num, '...')
+ if i == self.fmt.tr_row_num:
+ row.insert(row_levels - sparse_offset + self.fmt.tr_col_num,
+ diag_ellips)
+ else:
+ row.insert(row_levels - sparse_offset + self.fmt.tr_col_num,
+ horiz_ellips)
+
self.write_tr(row, indent, self.indent_delta, tags=tags,
- nindex_levels=len(levels) - sparse_offset)
+ nindex_levels=len(levels) - sparse_offset)
else:
for i in range(len(frame)):
idx_values = list(zip(*frame.index.format(sparsify=False,
adjoin=False,
names=False)))
+ if truncate_v and i == self.fmt.tr_row_num:
+ str_sep_row = vert_ellips * row_levels + \
+ vert_ellips * ( dot_col_ix - 1 ) + \
+ diag_ellips + \
+ vert_ellips * (dot_col_ix - 1)
+ self.write_tr(str_sep_row, indent, self.indent_delta,
+ tags=None, align='center')
row = []
row.extend(idx_values[i])
row.extend(fmt_values[j][i] for j in range(ncols))
if truncate_h:
- row.insert(row_levels + self.fmt.tr_col_num, '...')
+ row.insert(row_levels + self.fmt.tr_col_num,
+ horiz_ellips)
self.write_tr(row, indent, self.indent_delta, tags=None,
nindex_levels=frame.index.nlevels)
diff --git a/pandas/core/series.py b/pandas/core/series.py
index ea3656662ab06..ea1e519db5141 100644
--- a/pandas/core/series.py
+++ b/pandas/core/series.py
@@ -847,6 +847,7 @@ def _tidy_repr(self, max_vals=20):
Internal function, should always return unicode string
"""
+ vert_ellips = fmt.vert_ellips
if max_vals > 1:
num = max_vals // 2
else:
@@ -858,7 +859,10 @@ def _tidy_repr(self, max_vals=20):
length=False,
name=False,
dtype=False)
- result = head + '\n...\n' + tail
+ wdth = len(head.split(u('\n'))[0])
+ # from IPython.core.debugger import Tracer
+ # Tracer()()
+ result = head + u('\n') + vert_ellips.center(wdth) + u('\n') + tail
result = '%s\n%s' % (result, self._repr_footer())
return compat.text_type(result)
diff --git a/pandas/tests/test_format.py b/pandas/tests/test_format.py
index 8e405dc98f3da..83f0cd6a4d37a 100644
--- a/pandas/tests/test_format.py
+++ b/pandas/tests/test_format.py
@@ -29,6 +29,10 @@
_frame = DataFrame(tm.getSeriesData())
+# Truncation symbols
+horiz_ellips = u('\u22ef') # ⋯
+vert_ellips = u('\u22ee') # ⋮
+diag_ellips = u('\u22f1') # ⋱
def curpath():
pth, _ = os.path.split(os.path.abspath(__file__))
@@ -48,22 +52,24 @@ def has_non_verbose_info_repr(df):
def has_horizontally_truncated_repr(df):
try: # Check header row
- fst_line = np.array(repr(df).splitlines()[0].split())
- cand_col = np.where(fst_line=='...')[0][0]
+ fst_line = np.array(df.__unicode__().splitlines()[0].split())
+ cand_col = np.where(fst_line==horiz_ellips)[0][0]
except:
return False
# Make sure each row has this ... in the same place
- r = repr(df)
+ r = df.__unicode__()
for ix,l in enumerate(r.splitlines()):
- if not r.split()[cand_col] == '...':
+ ele = r.split()[cand_col]
+ if not (ele == horiz_ellips or ele == diag_ellips):
return False
return True
def has_vertically_truncated_repr(df):
- r = repr(df)
+ regex = u('^[') + vert_ellips + diag_ellips + u('\s]+$')
+ r = df.__unicode__()
only_dot_row = False
- for row in r.splitlines():
- if re.match('^[\.\ ]+$',row):
+ for ix_row,row in enumerate(r.splitlines()):
+ if re.match(regex,row):
only_dot_row = True
return only_dot_row
@@ -800,7 +806,7 @@ def test_to_html_truncate(self):
|
0 |
1 |
- ... |
+ ⋯ |
18 |
19 |
@@ -810,7 +816,7 @@ def test_to_html_truncate(self):
2001-01-01 |
NaN |
NaN |
- ... |
+ ⋯ |
NaN |
NaN |
@@ -818,7 +824,7 @@ def test_to_html_truncate(self):
2001-01-02 |
NaN |
NaN |
- ... |
+ ⋯ |
NaN |
NaN |
@@ -826,7 +832,7 @@ def test_to_html_truncate(self):
2001-01-03 |
NaN |
NaN |
- ... |
+ ⋯ |
NaN |
NaN |
@@ -834,23 +840,23 @@ def test_to_html_truncate(self):
2001-01-04 |
NaN |
NaN |
- ... |
+ ⋯ |
NaN |
NaN |
- | ... |
- ... |
- ... |
- ... |
- ... |
- ... |
+ ⋮ |
+ ⋮ |
+ ⋮ |
+ ⋱ |
+ ⋮ |
+ ⋮ |
| 2001-01-17 |
NaN |
NaN |
- ... |
+ ⋯ |
NaN |
NaN |
@@ -858,7 +864,7 @@ def test_to_html_truncate(self):
2001-01-18 |
NaN |
NaN |
- ... |
+ ⋯ |
NaN |
NaN |
@@ -866,7 +872,7 @@ def test_to_html_truncate(self):
2001-01-19 |
NaN |
NaN |
- ... |
+ ⋯ |
NaN |
NaN |
@@ -874,7 +880,7 @@ def test_to_html_truncate(self):
2001-01-20 |
NaN |
NaN |
- ... |
+ ⋯ |
NaN |
NaN |
@@ -902,7 +908,7 @@ def test_to_html_truncate_multi_index(self):
|
bar |
baz |
- ... |
+ ⋯ |
foo |
qux |
@@ -912,7 +918,7 @@ def test_to_html_truncate_multi_index(self):
one |
two |
one |
- ... |
+ ⋯ |
two |
one |
two |
@@ -925,7 +931,7 @@ def test_to_html_truncate_multi_index(self):
NaN |
NaN |
NaN |
- ... |
+ ⋯ |
NaN |
NaN |
NaN |
@@ -935,7 +941,7 @@ def test_to_html_truncate_multi_index(self):
NaN |
NaN |
NaN |
- ... |
+ ⋯ |
NaN |
NaN |
NaN |
@@ -946,21 +952,21 @@ def test_to_html_truncate_multi_index(self):
NaN |
NaN |
NaN |
- ... |
+ ⋯ |
NaN |
NaN |
NaN |
- | ... |
- ... |
- ... |
- ... |
- ... |
- ... |
- ... |
- ... |
- ... |
+ ⋮ |
+ ⋮ |
+ ⋮ |
+ ⋮ |
+ ⋮ |
+ ⋱ |
+ ⋮ |
+ ⋮ |
+ ⋮ |
| foo |
@@ -968,7 +974,7 @@ def test_to_html_truncate_multi_index(self):
NaN |
NaN |
NaN |
- ... |
+ ⋯ |
NaN |
NaN |
NaN |
@@ -979,7 +985,7 @@ def test_to_html_truncate_multi_index(self):
NaN |
NaN |
NaN |
- ... |
+ ⋯ |
NaN |
NaN |
NaN |
@@ -989,7 +995,7 @@ def test_to_html_truncate_multi_index(self):
NaN |
NaN |
NaN |
- ... |
+ ⋯ |
NaN |
NaN |
NaN |
@@ -1006,7 +1012,7 @@ def test_to_html_truncate_multi_index_sparse_off(self):
arrays = [['bar', 'bar', 'baz', 'baz', 'foo', 'foo', 'qux', 'qux'],
['one', 'two', 'one', 'two', 'one', 'two', 'one', 'two']]
df = pd.DataFrame(index=arrays,columns=arrays)
- fmt.set_option('display.max_rows',7)
+ fmt.set_option('display.max_rows',4)
fmt.set_option('display.max_columns',7)
fmt.set_option('display.multi_sparse',False)
result = df._repr_html_()
@@ -1020,7 +1026,7 @@ def test_to_html_truncate_multi_index_sparse_off(self):
bar |
bar |
baz |
- ... |
+ ⋯ |
foo |
qux |
qux |
@@ -1031,7 +1037,7 @@ def test_to_html_truncate_multi_index_sparse_off(self):
one |
two |
one |
- ... |
+ ⋯ |
two |
one |
two |
@@ -1044,7 +1050,7 @@ def test_to_html_truncate_multi_index_sparse_off(self):
NaN |
NaN |
NaN |
- ... |
+ ⋯ |
NaN |
NaN |
NaN |
@@ -1055,32 +1061,21 @@ def test_to_html_truncate_multi_index_sparse_off(self):
NaN |
NaN |
NaN |
- ... |
- NaN |
- NaN |
- NaN |
-
-
- | baz |
- one |
- NaN |
- NaN |
- NaN |
- ... |
+ ⋯ |
NaN |
NaN |
NaN |
-
- | foo |
- two |
- NaN |
- NaN |
- NaN |
- ... |
- NaN |
- NaN |
- NaN |
+
+ | ⋮ |
+ ⋮ |
+ ⋮ |
+ ⋮ |
+ ⋮ |
+ ⋱ |
+ ⋮ |
+ ⋮ |
+ ⋮ |
| qux |
@@ -1088,7 +1083,7 @@ def test_to_html_truncate_multi_index_sparse_off(self):
NaN |
NaN |
NaN |
- ... |
+ ⋯ |
NaN |
NaN |
NaN |
@@ -1099,7 +1094,7 @@ def test_to_html_truncate_multi_index_sparse_off(self):
NaN |
NaN |
NaN |
- ... |
+ ⋯ |
NaN |
NaN |
NaN |
@@ -1856,11 +1851,11 @@ def test_repr_html_wide(self):
max_cols = get_option('display.max_columns')
df = DataFrame([row(max_cols-1, 25) for _ in range(10)])
reg_repr = df._repr_html_()
- assert "..." not in reg_repr
+ assert horiz_ellips not in reg_repr
wide_df = DataFrame([row(max_cols+1, 25) for _ in range(10)])
wide_repr = wide_df._repr_html_()
- assert "..." in wide_repr
+ assert horiz_ellips in wide_repr
def test_repr_html_wide_multiindex_cols(self):
row = lambda l, k: [tm.rands(k) for _ in range(l)]
@@ -1877,20 +1872,20 @@ def test_repr_html_wide_multiindex_cols(self):
mcols = pandas.MultiIndex.from_tuples(tuples, names=['first', 'second'])
df = DataFrame([row(len(mcols), 25) for _ in range(10)], columns=mcols)
wide_repr = df._repr_html_()
- assert '...' in wide_repr
+ assert horiz_ellips in wide_repr
def test_repr_html_long(self):
max_rows = get_option('display.max_rows')
h = max_rows - 1
df = pandas.DataFrame({'A':np.arange(1,1+h), 'B':np.arange(41, 41+h)})
reg_repr = df._repr_html_()
- assert '..' not in reg_repr
+ assert vert_ellips not in reg_repr
assert str(41 + max_rows // 2) in reg_repr
h = max_rows + 1
df = pandas.DataFrame({'A':np.arange(1,1+h), 'B':np.arange(41, 41+h)})
long_repr = df._repr_html_()
- assert '..' in long_repr
+ assert vert_ellips in long_repr
assert str(41 + max_rows // 2) not in long_repr
assert u('%d rows ') % h in long_repr
assert u('2 columns') in long_repr
@@ -1900,13 +1895,13 @@ def test_repr_html_float(self):
h = max_rows - 1
df = pandas.DataFrame({'idx':np.linspace(-10,10,h), 'A':np.arange(1,1+h), 'B': np.arange(41, 41+h) }).set_index('idx')
reg_repr = df._repr_html_()
- assert '..' not in reg_repr
+ assert vert_ellips not in reg_repr
assert str(40 + h) in reg_repr
h = max_rows + 1
df = pandas.DataFrame({'idx':np.linspace(-10,10,h), 'A':np.arange(1,1+h), 'B': np.arange(41, 41+h) }).set_index('idx')
long_repr = df._repr_html_()
- assert '..' in long_repr
+ assert vert_ellips in long_repr
assert '31' not in long_repr
assert u('%d rows ') % h in long_repr
assert u('2 columns') in long_repr
@@ -1927,7 +1922,7 @@ def test_repr_html_long_multiindex(self):
df = DataFrame(np.random.randn((max_L1+1)*2, 2), index=idx,
columns=['A', 'B'])
long_repr = df._repr_html_()
- assert '...' in long_repr
+ assert vert_ellips in long_repr
def test_repr_html_long_and_wide(self):
max_cols = get_option('display.max_columns')
@@ -1939,7 +1934,8 @@ def test_repr_html_long_and_wide(self):
h, w = max_rows+1, max_cols+1
df = pandas.DataFrame(dict((k,np.arange(1,1+h)) for k in np.arange(w)))
- assert '...' in df._repr_html_()
+ df_html = df._repr_html_()
+ assert horiz_ellips in df_html and vert_ellips in df_html and diag_ellips in df_html
def test_info_repr(self):
max_rows = get_option('display.max_rows')