Data lost randomly from dictionary after creating the dictionary

[Y3KvBv]

BPO	39338
Nosy	@tim-one
Files	test.py: Example code that triggers dic data loss

^{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 2020-01-14.23:02:31.106>
created_at = <Date 2020-01-14.22:23:01.339>
labels = ['invalid', 'type-bug', '3.8']
title = 'Data lost randomly from dictionary after creating the dictionary'
updated_at = <Date 2020-01-15.00:40:44.636>
user = 'https://bugs.python.org/Y3KvBv'

bugs.python.org fields:

activity = <Date 2020-01-15.00:40:44.636>
actor = 'tim.peters'
assignee = 'none'
closed = True
closed_date = <Date 2020-01-14.23:02:31.106>
closer = 'tim.peters'
components = []
creation = <Date 2020-01-14.22:23:01.339>
creator = 'Y3Kv Bv'
dependencies = []
files = ['48841']
hgrepos = []
issue_num = 39338
keywords = []
message_count = 5.0
messages = ['360007', '360008', '360010', '360011', '360015']
nosy_count = 2.0
nosy_names = ['tim.peters', 'Y3Kv Bv']
pr_nums = []
priority = 'normal'
resolution = 'not a bug'
stage = 'resolved'
status = 'closed'
superseder = None
type = 'behavior'
url = 'https://bugs.python.org/issue39338'
versions = ['Python 3.8']

[Y3KvBv]

Windows 7 x64, Python 3.8.1

I've encountered a very weird issue where after creating a dictionary from a list the dictionary ends up being shorter/data is lost from it.
It's absolutely random when it loses, how many and which items are lost.

I've attached the example file with the code that always has a chance to trigger the issue for me.

I've managed to figure only this much that when "if useAmp" never triggers, data loss will never occur. I've added checkpoints to verify where the loss occurs and it's not caused by "if useAmp" block directly, data loss happens exactly after the dictionary is created.

[zware]

I suspect your useAmp branch is not doing what you think it's doing: it's effectively replacing a random number of entries in your crazyQuilt2 list with a duplicate entry (try printing the list every time around the main loop to see what's happening to it). When you later create a dict using the entries of the list as keys there are only so many unique keys, and thus the dict is not the same length as the list.

[tim-one]

What, exactly, in the output shows "the problem"? When I run it, the a == b part is always True, while len(x) and len(crazyQuilt2) are always 30. The other three (len(coordinates), len(x2), len(x3)) are always equal to each other, but are monotonically non-increasing across output lines. For example, a typical block of output lines:

(True, 30, 30, 30, 30, 30)
(True, 30, 30, 30, 30, 30)
(True, 30, 30, 30, 30, 30)
(True, 30, 30, 30, 30, 30)
(True, 19, 30, 30, 19, 19)
(True, 19, 30, 30, 19, 19)
(True, 19, 30, 30, 19, 19)
(True, 12, 30, 30, 12, 12)
(True, 12, 30, 30, 12, 12)
(True, 12, 30, 30, 12, 12)

None of that surprises me. Exactly what about it surprises you? Or do you get different kinds of output (and, if so, exactly what?)?

Here's my guess: you _intended_ these two lines:

      crazyQuilt2[coordinate2Index] = crazyQuilt2[index]
      crazyQuilt2[index] = crazyQuilt2[coordinate2Index]

to swap the values at indices index and coordinate2Index. But they don't. They copy the value originally at index into the coordinate2Index position, and leave the value originally at index untouched.

As more copies randomly build up, anything that builds a set or dict out of the list naturally gets smaller.