List inplace addition different from normal addition

[devkapilbansal]

BPO	45293
Nosy	@rhettinger, @ericvsmith, @tirkarthi, @devkapilbansal

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields:

assignee = None
closed_at = <Date 2021-09-26.20:58:22.124>
created_at = <Date 2021-09-26.12:13:59.531>
labels = ['3.8', 'type-bug', '3.7', '3.9', 'invalid']
title = 'List inplace addition different from normal addition'
updated_at = <Date 2021-09-26.20:58:22.119>
user = 'https://github.com/devkapilbansal'

bugs.python.org fields:

activity = <Date 2021-09-26.20:58:22.119>
actor = 'rhettinger'
assignee = 'none'
closed = True
closed_date = <Date 2021-09-26.20:58:22.124>
closer = 'rhettinger'
components = []
creation = <Date 2021-09-26.12:13:59.531>
creator = 'devkapilbansal'
dependencies = []
files = []
hgrepos = []
issue_num = 45293
keywords = []
message_count = 4.0
messages = ['402658', '402659', '402660', '402679']
nosy_count = 4.0
nosy_names = ['rhettinger', 'eric.smith', 'xtreak', 'devkapilbansal']
pr_nums = []
priority = 'normal'
resolution = 'not a bug'
stage = 'resolved'
status = 'closed'
superseder = None
type = 'behavior'
url = 'https://bugs.python.org/issue45293'
versions = ['Python 3.6', 'Python 3.7', 'Python 3.8', 'Python 3.9']

[devkapilbansal]

Hi,

I tried addition and in-place addition on a list.

>> l = ['a', 'b', 'c']

>> l = l + 'de' (raises an error)

>>> l += 'de'
>>> print(l)
['a', 'b' , 'c', 'd', 'e']

I want to ask why the behaviour of both these are different?? If it is done intentionally, then it should be there in the documentation but I am not able to find any reference.

[ericvsmith]

For those not in front of a computer, the error is:

>>> l = l + 'de'
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: can only concatenate list (not "str") to list

[tirkarthi]

https://docs.python.org/3/faq/programming.html#faq-augmented-assignment-tuple-error

for lists, __iadd__ is equivalent to calling extend on the list and returning the list. That’s why we say that for lists, += is a “shorthand” for list.extend

This example is calling extend on list of strings with another string as argument. Hence the target string is iterated and each character is added as an element. __add__ is different compared __iadd__. For += __iadd__ is called if defined and falls back to __add__

[rhettinger]

Kapil, this behavior was intentional (not a bug), but later regarded as a design mistake. GvR has said that if he had to do it over again list.__iadd__ would only accept other lists. The problem arises when people write "somelist += 'hello'" and expect it to add a single string with one word rather than five one letter strings. That said, the current behavior is guaranteed, people rely on it, and we cannot change it.

As Karthikeyan points out, this is documented. However since the docs have grown so voluminous, it is often difficult to find any one particular fact.