Description
Feature Type
-
Adding new functionality to pandas
-
Changing existing functionality in pandas
-
Removing existing functionality in pandas
Problem Description
I use pandas inside ipython. Sometimes I want all float items are printted with 3 digits. But pd.DataFrame({"seq": [[pi] * 5], "foo": [pi]}) got displayed like this:
seq foo
0 [3.141592653589793, 3.141592653589793, 3.14159... 3.142
i.e. "foo" in 3 digits but items of "seq" lots digits.
Feature Description
I want pd.DataFrame({"seq": [[pi] * 5], "foo": [pi]}) as
seq foo
0 [3.142, 3.142, 3.142, 3.142, 3.142] 3.142
in one line and all 5 items shown.
Alternative Solutions
I tried to define a str method for that sequence, but in like this issue customized representation method is not called. So I proposed to patch pprint_thing like this:
diff -u /home/renlifeng/ve3/lib/python3.12/site-packages/pandas/io/formats/printing.py.\~orig\~ /home/renlifeng/ve3/lib/python3.12/site-packages/pandas/io/formats/printing.py
--- /home/renlifeng/ve3/lib/python3.12/site-packages/pandas/io/formats/printing.py.~orig~ 2024-10-06 06:11:00.000000000 +0800
+++ /home/renlifeng/ve3/lib/python3.12/site-packages/pandas/io/formats/printing.py 2024-12-04 15:47:37.481159765 +0800
@@ -8,6 +8,10 @@
Mapping,
Sequence,
)
+from numbers import (
+ Integral,
+ Real,
+)
import sys
from typing import (
Any,
@@ -206,7 +210,10 @@
else:
escape_chars = escape_chars or ()
- result = str(thing)
+ if isinstance(thing, Real) and not isinstance(thing, Integral):
+ result = f"%.{get_option('display.precision')}f" % thing
+ else:
+ result = str(thing)
for c in escape_chars:
result = result.replace(c, translate[c])
return result
Diff finished. Wed Dec 4 15:48:27 2024
Additional Context
I use pandas inside ipython. Sometimes I want all float items are printted with 3 digits.