Python 3.6 f-Strings
02 November 2018
f-strings are TOTALLY AWESOME!¶
In [1]:
website = 'Braithwaite I/O'
f'Welcome to {website}!'
I found out about them at PyCon Canada 2017 at Mariatta Wijaya keynote.¶
In [15]:
from IPython.display import HTML
HTML('<iframe width="560" height="315" src="https://www.youtube-nocookie.com/embed/kGd1TVfP_Ds" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>')
Out[15]:
They also support expressions inside the { }¶
In [3]:
f'Welcome to {website.upper()}!'
Even dict¶
In [4]:
portfolio = {'AAPL': 222.77, 'RHT': 122.31, 'GOOGL': 1111.77}
In [5]:
f'${portfolio['GOOGL']}USD'
While that's a little embarrassing...¶
In [6]:
f"${portfolio['GOOGL']}USD"
Out[6]:
But look what else you can do...¶
In [7]:
f"${portfolio['GOOGL']:,}USD"
Out[7]:
Hey I heard you like dates and times...¶
In [8]:
import datetime
NOW = datetime.datetime.now()
f'This notebook was run on {NOW:%d %B %Y} at {NOW:%H:%M}!'
Out[8]:
You can use them in class...¶
In [9]:
class Drink:
def __init__(self, name, caffine):
self.name = name
self.caffine = caffine
def __str__(self):
return f'{self.name}'
def caffine_for_quantity(self, size=100000):
return (
f'{self.name} has {self.caffine*size:.0f} '
f'mg for {size/1000:.0f} grams.'
)
In [10]:
coffee = Drink('Coffee', .0004)
coffee.caffine_for_quantity()
Out[10]:
In [11]:
f'{coffee.caffine_for_quantity(459732)}'
Out[11]:
In [12]:
import timeit
format_funcs = {
'f-strings': """
def format(superhero, rank):
return f'{superhero} has a rank of {rank}!'
""",
'%-formatting': """
def format(superhero, rank):
return '%s has a rank of %s!' % (superhero, str(rank))
""",
'.format()': """
def format(superhero, rank):
return '{} has a rank of {}!'.format(superhero, str(rank))
""",
'concatenation +': """
def format(superhero, rank):
return superhero + ' has a rank of ' + str(rank) + '!'
""",
'concatenation ()': """
def format(superhero, rank):
return superhero, ' has a rank of ', str(rank), '!'
"""
}
test_func = """def test_format():
for superhero in ('Wonder Woman', 'Supergirl', 'Batman', 'Robin'):
for rank in range (1, 101):
format(superhero, rank)
"""
data = []
for key, func in format_funcs.items():
data.append({
'method': key,
'time': float(timeit.timeit('test_format()', func + test_func, number=10000))
})
In [13]:
%matplotlib inline
import pandas as pd
df = pd.DataFrame([
{'method': 'f-strings', 'time': 1.2825216680000153},
{'method': '%-formatting', 'time': 2.282235770999989},
{'method': '.format()', 'time': 2.983771015000002},
{'method': 'concatenation +', 'time': 1.9718771689999812},
{'method': 'concatenation ()', 'time': 1.402194948999977}
])
and it's fast...¶
In [14]:
df.sort_values('time').plot(
kind='barh',
x='method',
y='time',
figsize=(12, 6),
fontsize=14,
legend=False
)
Out[14]:
