deepdiff
deepdiff copied to clipboard
iterable_compare_func not working properly
Please checkout the F.A.Q page before creating a bug ticket to make sure it is not already addressed.
Describe the bug
the issue is related to the use of the ability to compare iterable element with a customized function.
with an iterable_compare_func defined as in the documentation : to compare iterables only if they have the same id, the result of deepdiff is wrong as it is missing the differences.
To Reproduce Exactly with the same examples on documentation.
from deepdiff import DeepDiff
from deepdiff.helper import CannotCompare
t1 = [
{
'id': 1,
'value': [1]
},
{
'id': 2,
'value': [7, 8, 1]
},
{
'id': 3,
'value': [7, 8],
},
]
t2 = [
{
'id': 2,
'value': [7, 8]
},
{
'id': 3,
'value': [7, 8, 1],
},
{
'id': 1,
'value': [1]
},
]
def compare_func(x, y, level=None):
try:
return x['id'] == y['id']
except Exception:
raise CannotCompare() from None
without iterable function - result is as expected
>>>DeepDiff(t1, t2)
{'values_changed': {"root[0]['id']": {'new_value': 2, 'old_value': 1}, "root[0]['value'][0]": {'new_value': 7, 'old_value': 1}, "root[1]['id']": {'new_value': 3, 'old_value': 2}, "root[2]['id']": {'new_value': 1, 'old_value': 3}, "root[2]['value'][0]": {'new_value': 1, 'old_value': 7}}, 'iterable_item_added': {"root[0]['value'][1]": 8}, 'iterable_item_removed': {"root[2]['value'][1]": 8}}
without iterable function - result is wrong
>>> DeepDiff(t1, t2, iterable_compare_func=compare_func)
{}
>>> DeepDiff(t1, t2, iterable_compare_func=compare_func, verbose_level=2)
{'iterable_item_moved': {'root[0]': {'new_path': 'root[2]', 'value': {'id': 1, 'value': [1]}}, 'root[1]': {'new_path': 'root[0]', 'value': {'id': 2, 'value': [7, 8]}}, 'root[2]': {'new_path': 'root[1]', 'value': {'id': 3, 'value': [7, 8, 1]}}}}
>>> DeepDiff(t1, t2, iterable_compare_func=compare_func, cutoff_distance_for_pairs=1, cutoff_intersection_for_pairs=1)
{}
as you can see on verbosity_level=2 the element is marked as moved even if it is not even the same value ? how this could happen ?
Expected behavior Expected behavior to list compare with ids so the change will be.
>>>DeepDiff(t1, t2, iterable_compare_func=compare_func)
{'iterable_item_added': {"root[2]['value'][2]": 1}, 'iterable_item_removed': {"root[1]['value'][2]": 1}}
OS, DeepDiff version and Python version (please complete the following information): The tests has been performed on several environments.
- OS: Debian, SilverBlue (Fedora), Container (Alpine)
- Python Version 3.10, 3.11.4, 3.10 respectively
- DeepDiff Version : 6.3.1
Hello. Just to add a tracking for the behavior on several deepdiff versions.
The test are done inside a container with python 3.10.5 and pip 22.1.1.
-
with deepdiff 6.4.0 - Another Bug like mentioned on an other issue
>>> from deepdiff import DeepDiff Traceback (most recent call last): File "<stdin>", line 1, in <module> File "/usr/lib/python3.10/site-packages/deepdiff/__init__.py", line 10, in <module> from .diff import DeepDiff File "/usr/lib/python3.10/site-packages/deepdiff/diff.py", line 19, in <module> from deepdiff.helper import (strings, bytes_type, numbers, uuids, datetimes, ListItemRemovedOrAdded, notpresent, File "/usr/lib/python3.10/site-packages/deepdiff/helper.py", line 158, in <module> if get_semvar_as_integer(np.__version__) < 1019000: AttributeError: 'NoneType' object has no attribute '__version__' >>> from deepdiff.helper import CannotCompare Traceback (most recent call last): File "<stdin>", line 1, in <module> File "/usr/lib/python3.10/site-packages/deepdiff/__init__.py", line 10, in <module> from .diff import DeepDiff File "/usr/lib/python3.10/site-packages/deepdiff/diff.py", line 19, in <module> from deepdiff.helper import (strings, bytes_type, numbers, uuids, datetimes, ListItemRemovedOrAdded, notpresent, File "/usr/lib/python3.10/site-packages/deepdiff/helper.py", line 158, in <module> if get_semvar_as_integer(np.__version__) < 1019000: AttributeError: 'NoneType' object has no attribute '__version__' -
with deepdiff 6.3.1, 6.3.0, 6.2.3, 6.2.2, 6.2.1
>>> DeepDiff(t1, t2, iterable_compare_func=compare_func) {} -
with deepdiff 5.8.1 - working
>>> DeepDiff(t1, t2, iterable_compare_func=compare_func) {'iterable_item_added': {"root[2]['value'][2]": 1}, 'iterable_item_removed': {"root[1]['value'][2]": 1}}
Hello @seperman
Is there any plan regarding a fix.
I want to mention that even with the previous working version (v5.8.1) there some cases where the iterable_compare_function is not working properly.
For example, when an iterable item is deleted on t1[index_K] and another iterable item is added on t2[index_K] even in the case that these two items have different id, DeepDiff reports this as a value change.
+1 I am also affected.
+1 same
@seperman Is there any info on this bug?
Hello, thanks for reminding me. I have not had a chance to take a look at it yet. I will keep you posted once I fix it. PRs are always very welcome!
I have opened a PR that fixes the issue. With the PR here are the outputs for the initial test case:
>>> DeepDiff(t1, t2, iterable_compare_func=compare_func)
{'iterable_item_added': {"root[1]['value'][2]": 1}, 'iterable_item_removed': {"root[0]['value'][2]": 1}}
>>> DeepDiff(t1, t2, iterable_compare_func=compare_func, verbose_level=2)
{'iterable_item_added': {"root[1]['value'][2]": 1}, 'iterable_item_removed': {"root[0]['value'][2]": 1}, 'iterable_item_moved': {'root[0]': {'new_path': 'root[2]', 'value': {'id': 1, 'value': [1]}}, 'root[1]': {'new_path': 'root[0]', 'value': {'id': 2, 'value': [7, 8]}}, 'root[2]': {'new_path': 'root[1]', 'value': {'id': 3, 'value': [7, 8, 1]}}}}
Note: I believe 6.1.0 is the latest version that will not have this bug.