|
| 1 | +# CWE-230: Improper Handling of Missing Values |
| 2 | + |
| 3 | +The `NaN` value should be stripped before as they can cause surprising or undefined behaviours in the statistics functions that sort or count occurrences [[2024 doc.python.org]](https://docs.python.org/3/library/statistics.html). |
| 4 | +In python, some datasets use `NaN` (not-a-number) to represent the missing data. This can be problematic as the `NaN` values are unordered. Any ordered comparison of a number to a not-a-number value are `False`. A counter-intuitive implication is that `not-a-number` values are not equal to themselves. |
| 5 | + |
| 6 | +This behavior is compliant with IEEE 754[[2024 Wikipedia]](https://en.wikipedia.org/wiki/IEEE_754) a hardware induced compromise. |
| 7 | +The [example01.py](example01.py) code demonstrates various comparisons of `float('NaN')` all resulting in `False`. |
| 8 | + |
| 9 | +```python |
| 10 | +# SPDX-FileCopyrightText: OpenSSF project contributors |
| 11 | +# SPDX-License-Identifier: MIT |
| 12 | +""" Code Example """ |
| 13 | + |
| 14 | +foo = float('NaN') |
| 15 | +print(f"foo={foo} type = {type(foo)}") |
| 16 | + |
| 17 | + |
| 18 | +print(foo == float("NaN") or |
| 19 | + foo is float("NaN") or |
| 20 | + foo < 3 or |
| 21 | + foo == foo or |
| 22 | + foo is None |
| 23 | + ) |
| 24 | + |
| 25 | +``` |
| 26 | + |
| 27 | +## Non-Compliant Code Example |
| 28 | + |
| 29 | +This noncompliant code example [[2024 docs.python.org]](https://docs.python.org/3/reference/expressions.html#value-comparisons) attempts a direct comparison with `NaN` in `_value == float("NaN")`. |
| 30 | +*[noncompliant01.py](noncompliant01.py):* |
| 31 | + |
| 32 | +```python |
| 33 | +# SPDX-FileCopyrightText: OpenSSF project contributors |
| 34 | +# SPDX-License-Identifier: MIT |
| 35 | +""" Non-compliant Code Example """ |
| 36 | + |
| 37 | + |
| 38 | +def balance_is_positive(value: str) -> bool: |
| 39 | + """Returns True if there is still enough value for a transaction""" |
| 40 | + _value = float(value) |
| 41 | + if _value == float("NaN") or _value is float("NaN") or _value is None: |
| 42 | + raise ValueError("Expected a float") |
| 43 | + if _value <= 0: |
| 44 | + return False |
| 45 | + else: |
| 46 | + return True |
| 47 | + |
| 48 | + |
| 49 | +##################### |
| 50 | +# attempting to exploit above code example |
| 51 | +##################### |
| 52 | +print(balance_is_positive("0.01")) |
| 53 | +print(balance_is_positive("0.001")) |
| 54 | +print(balance_is_positive("NaN")) |
| 55 | + |
| 56 | +``` |
| 57 | + |
| 58 | +## Compliant Solution |
| 59 | + |
| 60 | +In the `compliant01.py` code example, the method `Decimal.quantize` is used to gain control over known rounding errors in floating point values. |
| 61 | + |
| 62 | +The decision by the `balance_is_positive` method is to `ROUND_DOWN` instead of the default `ROUND_HALF_EVEN`. |
| 63 | + |
| 64 | +*[compliant01.py](compliant01.py):* |
| 65 | + |
| 66 | +```python |
| 67 | +# SPDX-FileCopyrightText: OpenSSF project contributors |
| 68 | +# SPDX-License-Identifier: MIT |
| 69 | +""" Compliant Code Example """ |
| 70 | + |
| 71 | +from decimal import ROUND_DOWN, Decimal |
| 72 | + |
| 73 | + |
| 74 | +def balance_is_positive(value: str) -> bool: |
| 75 | + """Returns True if there is still enough value for a transaction""" |
| 76 | + # TODO: additional input sanitation for expected type |
| 77 | + _value = Decimal(value) |
| 78 | + # TODO: exception handling |
| 79 | + return _value.quantize(Decimal(".01"), rounding=ROUND_DOWN) > Decimal("0.00") |
| 80 | + |
| 81 | + |
| 82 | +##################### |
| 83 | +# attempting to exploit above code example |
| 84 | +##################### |
| 85 | +print(balance_is_positive("0.01")) |
| 86 | +print(balance_is_positive("0.001")) |
| 87 | +print(balance_is_positive("NaN")) |
| 88 | + |
| 89 | +``` |
| 90 | + |
| 91 | +`Decimal` throws a `decimal.InvalidOperation` for `NaN` values, the controlled rounding causes only `"0.01"` to return `True`. |
| 92 | + |
| 93 | +In `compliant02.py` we use the `math.isnan` to verify if the value passed is a valid `float` value. |
| 94 | + |
| 95 | +*[compliant02.py](compliant02.py):* |
| 96 | + |
| 97 | +```python |
| 98 | +# SPDX-FileCopyrightText: OpenSSF project contributors |
| 99 | +# SPDX-License-Identifier: MIT |
| 100 | +""" Compliant Code Example """ |
| 101 | + |
| 102 | +import math |
| 103 | + |
| 104 | + |
| 105 | +def balance_is_positive(value: str) -> bool: |
| 106 | + """Returns True if there is still enough value for a transaction""" |
| 107 | + _value = float(value) |
| 108 | + if math.isnan(_value) or _value is None: |
| 109 | + raise ValueError("Expected a float") |
| 110 | + if _value < 0.01: |
| 111 | + return False |
| 112 | + else: |
| 113 | + return True |
| 114 | + |
| 115 | + |
| 116 | +##################### |
| 117 | +# attempting to exploit above code example |
| 118 | +##################### |
| 119 | +print(balance_is_positive("0.01")) |
| 120 | +print(balance_is_positive("0.001")) |
| 121 | +print(balance_is_positive("NaN")) |
| 122 | + |
| 123 | +``` |
| 124 | + |
| 125 | +The `balance_is_poitive` method will raise an `ValueError` for `NaN` values. |
| 126 | + |
| 127 | +## Automated Detection |
| 128 | + |
| 129 | +|Tool|Version|Checker|Description| |
| 130 | +|:----|:----|:----|:----| |
| 131 | +|Bandit|1.7.4 on Python 3.10.4|Not Available|| |
| 132 | +|flake8|flake8-4.0.1 on python 3.10.4||Not Available| |
| 133 | + |
| 134 | +## Related Guidelines |
| 135 | + |
| 136 | +||| |
| 137 | +|:---|:---| |
| 138 | +|[SEI CERT Coding Standard for Java](https://wiki.sei.cmu.edu/confluence/display/java/SEI+CERT+Oracle+Coding+Standard+for+Java)|[NUM07-J. Do not attempt comparisons with NaN](https://wiki.sei.cmu.edu/confluence/display/java/NUM07-J.+Do+not+attempt+comparisons+with+NaN)| |
| 139 | +|[ISO/IEC TR 24772:2013](https://wiki.sei.cmu.edu/confluence/display/java/Rule+AA.+References#RuleAA.References-ISO/IECTR24772-2013)|Injection RST| |
| 140 | +|[MITRE CWE Pillar](http://cwe.mitre.org/)|[CWE-703: Improper Check or Handling of Exceptional Conditions (mitre.org)](https://cwe.mitre.org/data/definitions/703.html)| |
| 141 | +|[MITRE CWE Pillar](http://cwe.mitre.org/)|[CWE-230: Improper Handling of Missing Values](https://cwe.mitre.org/data/definitions/230.html)| |
| 142 | + |
| 143 | +## Bibliography |
| 144 | + |
| 145 | +||| |
| 146 | +|:---|:---| |
| 147 | +|[[Python 3.10.4 docs]](https://docs.python.org/3/library/string.html#formatstrings)|Format String Syntax. Available from: <https://docs.python.org/3/library/string.html#formatstrings> \[Accessed 22 July 2025]| |
| 148 | +|[Python docs](https://docs.python.org/3/)|<https://docs.python.org/3/library/math.html#math.nan> \[Accessed 22 July 2025]| |
| 149 | +|[Python docs](https://docs.python.org/3/)|Python Value comparisons<https://docs.python.org/3/reference/expressions.html#value-comparisons> \[Accessed 22 July 2025]| |
| 150 | +|[[Wikipedia 2024]](https://realpython.com/python-string-formatting/)|IEEE 754: <https://en.wikipedia.org/wiki/IEEE_754> \[Accessed 22 July 2025]| |
0 commit comments