Numbers written in scientific notation don't work with itertools.islice.

Check this script: 

# a usual function works
## def works as well
f = lambda x : x 
f(1e+6)
# 1000000.0
import itertools
# islice without scientific notation works
itertools.islice([], 1000000)
# <itertools.islice object at 0x7fc7ce55be90>
itertools.islice([], 1e+6)
# Traceback (most recent call last):
#   File "<stdin>", line 1, in <module>
# ValueError: Stop argument for islice() must be None or an integer: 0 <= x <= sys.maxsize.

All this works well in Python 2.7.17, but scientific notation fails in Python 3.7.5. I use Fedora Core 29.

Hello Yaroslav.  What you've observed was a intentional change to make islice() behave more like regular slices.  Neither of those accept floats which are precarious to use for indexing.

A workaround is to replace 1e+6 with 10**6.

Hello Raymond. Many thanks for your explanation.

In this case I suggest any of the following:

1) distinguish between integer and floating numbers in scientific notation. Definitely, 1e+6 is an integer. I can't see where else numbers in scientific notation are defined as only floats.

2) to write explicitly in the documentation that scientific notation is always float. I searched documentation of 'scientific notation', but didn't found that (https://docs.python.org/3/search.html?q=scientific+notation&check_keywords=yes&area=default)

3) to provide a better exception message in islice if this is a known issue. 

Should I change parameters of my report or create a new one? I'm afraid this message may get lost if closed.

The behavior is documented here: https://docs.python.org/3.9/library/stdtypes.html#numeric-types-int-float-complex

> Numeric literals containing a decimal point or an exponent sign yield floating point numbers.