Fix query-string: accept array/bracketed keys and trailing &

[Copilot]

from_query_string rejected valid query strings using array-style (a[]=1&a[]=2) or bracketed (user[name]=joe) keys because the old character-whitelist regex excluded [ and ]. Trailing & was also silently broken by the new validation.

Changes

• benedict/serializers/query_string.py: Replace opaque whitelist regex with structural pair validation — split on &, assert each segment matches ^[^\s=&]+=[^\s&]*$. Strip exactly one trailing empty segment so "a=1&" remains accepted (backward-compatible).
• tests/dicts/io/test_io_dict_query_string.py: Add test_from_query_string_with_array_style_keys and test_from_query_string_with_bracketed_keys, each asserting both the static method and constructor (IODict(s, format="query_string")) entry points.

from benedict import benedict

# Previously raised ValueError, now work correctly:
benedict.from_query_string("a[]=1&a[]=2")        # {"a[]": "1"}
benedict.from_query_string("user[name]=joe&user[age]=42")  # {"user[name]": "joe", "user[age]": "42"}
benedict.from_query_string("a=1&")               # {"a": "1"}

Checklist before requesting a review

• I have performed a self-review of my code.
• I have added tests for the proposed changes.
• I have run the tests and there are not errors.

[fabiocaccamo]

These are the right tests that the implementation must satisfy:

def test_from_query_string_with_array_style_keys(self) -> None:
    # array-style keys (PHP / HTML form syntax) are valid query strings
    s = "a[]=1&a[]=2"
    r = {"a": ["1", "2"]}
    d = IODict.from_query_string(s)
    self.assertTrue(isinstance(d, dict))
    self.assertEqual(d, r)
    d = IODict(s, format="query_string")
    self.assertTrue(isinstance(d, dict))
    self.assertEqual(d, r)

def test_from_query_string_with_bracketed_keys(self) -> None:
    s = "user[name]=joe&user[age]=42"
    r = {"user": {"name": "joe", "age": "42"}}
    d = IODict.from_query_string(s)
    self.assertTrue(isinstance(d, dict))
    self.assertEqual(d, r)

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 98.38%. Comparing base (c7c934f) to head (9e42364).
⚠️ Report is 3 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #586      +/-   ##
==========================================
+ Coverage   98.33%   98.38%   +0.05%     
==========================================
  Files          64       64              
  Lines        2400     2417      +17     
==========================================
+ Hits         2360     2378      +18     
+ Misses         40       39       -1     

Flag	Coverage Δ
unittests	98.38% <100.00%> (+0.05%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[Copilot]

These are the right tests that the implementation must satisfy:

Done in 3d5c0ba. Updated the implementation and tests to match your specifications:

• a[]=1&a[]=2 → {"a": ["1", "2"]} (array keys collected into a list under the base name)
• user[name]=joe&user[age]=42 → {"user": {"name": "joe", "age": "42"}} (bracket notation expanded into a nested dict)

The _parse_bracket_notation helper handles both patterns while regular keys continue to honour the flat parameter as before.

[fabiocaccamo]

Should run pre-commit.

[Copilot]

Should run pre-commit.

Done in 29c14b3. Pre-commit found two mypy type-arg errors (bare dict annotations in _parse_bracket_notation) — fixed both to dict[str, Any]. All hooks now pass.