From 9dbaf467fe01686b00ca5d2ef2b7d24287f23a0a Mon Sep 17 00:00:00 2001
From: Bert Palm <bert.palm@ufz.de>
Date: Fri, 13 Mar 2020 11:01:18 +0100
Subject: [PATCH] added length in print, some bugfixes
---
dios/dios.py | 51 +++++++++++++++++++++++++++++++++++++--------------
dios/lib.py | 12 ++++++------
2 files changed, 43 insertions(+), 20 deletions(-)
diff --git a/dios/dios.py b/dios/dios.py
index 8117f3c..5ce12b8 100644
--- a/dios/dios.py
+++ b/dios/dios.py
@@ -98,7 +98,7 @@ class DictOfSeries:
# we are called internally
if fastpath:
- self._itype = itype or MixedItype
+ self._itype = itype or ObjItype
self._policy = cast_policy
if data is not None:
@@ -112,7 +112,7 @@ class DictOfSeries:
# itype=None means infer the itype by the data, so we first set to the highest
# possible itype, then insert data, then find the best-fitting itype.
if itype is None:
- self._itype = MixedItype
+ self._itype = ObjItype
else:
self._itype = get_itype(itype)
@@ -216,7 +216,7 @@ class DictOfSeries:
if method in ['union', 'all']:
res = ftools.reduce(pd.Index.union, indexes)
- elif method == ['intersection', 'shared']:
+ elif method in ['intersection', 'shared']:
res = ftools.reduce(pd.Index.intersection, indexes)
elif method in ['uniques', 'non-uniques']:
res = ftools.reduce(pd.Index.append, indexes)
@@ -441,10 +441,11 @@ class DictOfSeries:
def __repr__(self):
repr = dios_options[OptsFields.dios_repr]
- return self.to_string(method=repr)
+ showdim = self.lengths.max() > dios_options[OptsFields.disp_max_rows]
+ return self.to_string(method=repr, show_dimensions=showdim)
def to_string(self, max_rows=None, min_rows=None, max_cols=None,
- na_rep=np.nan,
+ na_rep=np.nan, show_dimensions=False,
method=Opts.repr_indexed,
no_value=' ', empty_series_rep='no data',
col_delim=' | ', header_delim='=', col_space=None, ):
@@ -487,7 +488,7 @@ class DictOfSeries:
min_rows = min_rows or dios_options[OptsFields.disp_min_rows] or 100
kwargs = dict(max_rows=max_rows, min_rows=min_rows, max_cols=max_cols,
- na_rep=na_rep, col_space=col_space)
+ na_rep=na_rep, col_space=col_space, show_dimensions=show_dimensions)
if method == Opts.repr_aligned:
return _to_aligned_df(self, no_value=no_value).to_string(**kwargs)
@@ -508,7 +509,7 @@ class DictOfSeries:
where all columns of dtype obj/string.
DO NOT USE FOR CALCULATIONS (!)"""
- return _to_aligned_df(self)
+ return self.to_df()
def apply(self, func, axis=0, raw=False, args=(), **kwds):
if axis in [1, 'columns']:
@@ -660,12 +661,12 @@ def _empty_repr(di):
def pprint_dios(dios,
max_rows=None, min_rows=None, max_cols=None,
na_rep="NaN", empty_series_rep='no data',
- col_space=None,
+ col_space=None, show_dimensions=True,
col_delim=' | ', header_delim='='
):
+
na_rep = str(na_rep)
empty_series_rep = str(empty_series_rep)
-
col_delim = col_delim or ' '
min_rows = min(max_rows, min_rows)
@@ -673,6 +674,7 @@ def pprint_dios(dios,
if dios.empty:
return _empty_repr(dios)
+ maxlen = dios.lengths.max()
data = dios._data
trunc_cols = len(data) > max_cols
@@ -683,7 +685,8 @@ def pprint_dios(dios,
# now data only contains series that we want to print.
# if any series exceed max_rows we trim all series to min_rows
- series_maxlen = max(data.apply(len))
+ series_lengths = data.apply(len)
+ series_maxlen = max(series_lengths)
trunc_rows = series_maxlen > max_rows
max_rows = min_rows if trunc_rows else series_maxlen
@@ -693,12 +696,14 @@ def pprint_dios(dios,
outer = []
for colname in data.index:
s = data.at[colname]
- idx = True
- if s.empty:
+
+ isempty = s.empty
+ if isempty:
s = pd.Series(empty_series_rep)
idx = False
cspace = col_space
else:
+ idx = True
cspace = col_space // 2 if col_space else col_space
sstr = s.to_frame().to_string(col_space=cspace,
@@ -709,9 +714,10 @@ def pprint_dios(dios,
min_rows=min_rows,
)
li = sstr.split('\n')
+
# HACK: empty series produce a unnecessary space,
# because index is omitted
- if not idx:
+ if isempty:
cstr, vstr = li
if len(cstr.lstrip()) < len(vstr) and (cspace or 0) < len(vstr):
li = [cstr[1:], vstr[1:]]
@@ -733,8 +739,8 @@ def pprint_dios(dios,
colheader = (header_delim * valstr_len[i])[:valstr_len[i]]
c.insert(1, colheader)
+ delim = ' ... '
if trunc_cols:
- delim = ' ... '
outer.insert(max_cols // 2, [delim] * rows)
valstr_len.insert(max_cols // 2, len(delim))
@@ -748,6 +754,23 @@ def pprint_dios(dios,
txt += vstr + col_delim
txt += '\n'
+ # add footer
+ if show_dimensions:
+ # correction because the collections `series_lengths`
+ # and `valstr_len` differ by 1 in length. very hacky!
+ # todo: use a dict/df or dios(?!) to store the str-data
+ corr = 0
+ for i, c in enumerate(outer):
+ # ignore the dot-column
+ if trunc_cols and i == max_cols//2:
+ txt += delim + ' ' * len(col_delim)
+ corr = 1
+ else:
+ txt += f"[{series_lengths.iat[i-corr]}]".ljust(valstr_len[i] + len(col_delim))
+
+ txt += f'\n\nmax: [{maxlen} rows x {len(dios.columns)} columns]'
+ txt += '\n'
+
return txt
diff --git a/dios/lib.py b/dios/lib.py
index df152ae..c901478 100644
--- a/dios/lib.py
+++ b/dios/lib.py
@@ -54,10 +54,10 @@ class NumItype(__Itype):
min_pdindex = pd.Float64Index([])
-class MixedItype(__Itype):
- name = "mixed"
+class ObjItype(__Itype):
+ name = "object"
unique = False
- _subitypes = (DtItype, IntItype, FloatItype, NumItype)
+ _subitypes = (DtItype, IntItype, FloatItype, NumItype, str)
_otheritypes = (pd.CategoricalIndex, pd.IntervalIndex, pd.PeriodIndex, pd.TimedeltaIndex, pd.Index)
subtypes = (_subitypes + _otheritypes + DtItype.subtypes + NumItype.subtypes)
min_pdindex = pd.Index([])
@@ -111,7 +111,7 @@ def get_itype(obj):
return obj
# check if it is the actual type, not a subtype
- types = [DtItype, IntItype, FloatItype, NumItype, MixedItype]
+ types = [DtItype, IntItype, FloatItype, NumItype, ObjItype]
for t in types:
if is_itype(obj, t):
return t
@@ -138,7 +138,7 @@ def _itype_le(a, b):
def _find_least_common_itype(iterable_of_series):
itypes = [NumItype, FloatItype, IntItype, DtItype]
tlist = [get_itype(s.index) for s in iterable_of_series]
- found = MixedItype
+ found = ObjItype
if tlist:
for itype in itypes:
for t in tlist:
@@ -252,7 +252,7 @@ class OptsFields:
``__repr__`` or ``__str__`` like e.g. ``print(dios)`` do.
"""
disp_max_rows = "disp_max_rows "
- disp_min_rows = "disp_max_rows "
+ disp_min_rows = "disp_min_rows "
disp_max_cols = "disp_max_vars"
"""
--
GitLab