Skip to content

Commit

Permalink
pyDoc2GitHub Adding documentation for CWE-197 01 as part of #531 to G…
Browse files Browse the repository at this point in the history
…itHub (#649)

* Adding documentation for CWE-197 01 as part of #531 to GitHub
Signed-off-by: Helge Wehder <[email protected]>

* Fixed formatting according to comments

Signed-off-by: myteron <[email protected]>

* Update README.md

Adding link/reference to 8 rounding modes

Signed-off-by: myteron <[email protected]>

---------

Signed-off-by: myteron <[email protected]>
  • Loading branch information
myteron authored Oct 17, 2024
1 parent 6fc707f commit e247db1
Show file tree
Hide file tree
Showing 6 changed files with 142 additions and 7 deletions.
104 changes: 104 additions & 0 deletions docs/Secure-Coding-Guide-for-Python/CWE-664/CWE-197/01/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,104 @@
# CWE-197: Control rounding when converting to less precise numbers

While defensive coding requires enforcing types, it is important to make conscious design decisions on how conversions are rounded.

The `example01.py` code demonstrates how `int()` behaves differently to `round()`.

[*example01.py:*](example01.py)

```py
""" Code Example """

print(int(0.5)) # prints 0
print(int(1.5)) # prints 1
print(int(1.45)) # prints 1
print(int(1.51)) # prints 1
print(int(-1.5)) # prints -1

print(round(0.5)) # prints 0
print(round(1.5)) # prints 2
print(round(1.45)) # prints 1
print(round(1.51)) # prints 2
print(round(-1.5)) # prints -2

print(type(round(0.5))) # prints <class 'int'>

```

The build in `round()` does not allow to specify the type of rounding in use [[python round( ) 2024]](https://docs.python.org/3/library/functions.html#round). In Python 3 the `round()` function uses "bankers' rounding" (rounds to the nearest even number in case of ties). This is different to Python 2 which always rounds away from zero. Rounding provided by the `decimal` module allows a choice between 8 rounding modes [[python decimal 2024]](https://docs.python.org/3/library/decimal.html#rounding-modes). Rounding in mathematics and science is not discussed here as it requires a deeper knowledge of computer floating-point arithmetic's.

## Non-Compliant Code Example (float to int)

In `noncompliant01.py` there is no conscious choice of rounding mode.

[*noncompliant01.py:*](noncompliant01.py)

```py
""" Non-compliant Code Example """

print(int(0.5)) # prints 0
print(int(1.5)) # prints 1
print(round(0.5)) # prints 0
print(round(1.5)) # prints 2
```

## Compliant Solution (float to int)

Using the `Decimal` class from the `decimal` module allows more control over rounding by choosing one of the `8` rounding modes [[python decimal 2024]](https://docs.python.org/3/library/decimal.html#rounding-modes).

[*compliant01.py:*](compliant01.py)

```py
""" Compliant Code Example """
from decimal import Decimal, ROUND_HALF_UP, ROUND_HALF_DOWN

print(Decimal("0.5").quantize(Decimal("1"), rounding=ROUND_HALF_UP)) # prints 1
print(Decimal("1.5").quantize(Decimal("1"), rounding=ROUND_HALF_UP)) # prints 2
print(Decimal("0.5").quantize(Decimal("1"), rounding=ROUND_HALF_DOWN)) # prints 0
print(Decimal("1.5").quantize(Decimal("1"), rounding=ROUND_HALF_DOWN)) # prints 1
```

The `.quantize(Decimal("1")`, determines the precision to be `integer` and `rounding=ROUND_HALF_UP` determines the type of rounding applied. Specifying numbers as strings avoids issues such as floating-point representations in binary.

That `Decimal` can have unexpected results when operated without `Decimal.quantize()` on floating point numbers is demonstrated in `example02.py`.

[*example02.py:*](example02.py)

```py
# SPDX-FileCopyrightText: OpenSSF project contributors
# SPDX-License-Identifier: MIT
""" Code Example """
from decimal import ROUND_HALF_UP, Decimal

print(Decimal("0.10")) # prints 0.10
print(Decimal(0.10)) # prints 0.1000000000000000055511151231257827021181583404541015625
print(Decimal("0.10").quantize(Decimal("0.10"), rounding=ROUND_HALF_UP)) # prints 0.10
print(Decimal(0.10).quantize(Decimal("0.10"), rounding=ROUND_HALF_UP)) # prints 0.10
```

Initializing `Decimal` with an actual `float`, such as `0.10`, and without rounding creates an unprecise number `0.1000000000000000055511151231257827021181583404541015625` in `Python 3.9.2`.

## Automated Detection

|Tool|Version|Checker|Description|
|:---|:---|:---|:---|
|Bandit|1.7.4 on Python 3.10.4|Not Available||
|Flake8|8-4.0.1 on Python 3.10.4|Not Available||

## Related Guidelines

|||
|:---|:---|
|[MITRE CWE](http://cwe.mitre.org/)|Pillar [CWE-682, Incorrect Conversion between Numeric Types (mitre.org)](http://cwe.mitre.org/data/definitions/682.html)|
|[MITRE CWE](http://cwe.mitre.org/)|Class [CWE-197, Numeric Truncation Error](https://cwe.mitre.org/data/definitions/197.html)|
|[SEI CERT C Coding Standard](https://wiki.sei.cmu.edu/confluence/display/c/SEI+CERT+C+Coding+Standard)|[INT31-C. Ensure that integer conversions do not result in lost or misinterpreted data](https://wiki.sei.cmu.edu/confluence/display/c/INT31-C.+Ensure+that+integer+conversions+do+not+result+in+lost+or+misinterpreted+data)|
|[SEI CERT C Coding Standard](https://wiki.sei.cmu.edu/confluence/display/c/SEI+CERT+C+Coding+Standard)|[FLP34-C. Ensure that floating-point conversions are within range of the new type](https://wiki.sei.cmu.edu/confluence/display/c/FLP34-C.+Ensure+that+floating-point+conversions+are+within+range+of+the+new+type)|
|[ISO/IEC TR 24772:2019](https://www.iso.org/standard/71091.html)|Programming languages — Guidance to avoiding vulnerabilities in programming languages, available from [https://www.iso.org/standard/71091.html](https://www.iso.org/standard/71091.html)|
|[SEI CERT Coding Standard for Java](https://wiki.sei.cmu.edu/confluence/display/java/SEI+CERT+Oracle+Coding+Standard+for+Java)|[NUM12-J. Ensure conversions of numeric types to narrower types do not result in lost or misinterpreted data](https://wiki.sei.cmu.edu/confluence/display/java/NUM12-J.+Ensure+conversions+of+numeric+types+to+narrower+types+do+not+result+in+lost+or+misinterpreted+data)|

## Biblography

|||
|:---|:---|
|[python round( ) 2024](https://docs.python.org/3/library/functions.html#round)|python round( ), available from: [https://docs.python.org/3/library/functions.html#round](https://docs.python.org/3/library/functions.html#round), [Last accessed June 2024] |
|[python decimal 2024](https://docs.python.org/3/library/decimal.html#rounding-modes)|Python decimal module, available from: [https://docs.python.org/3/library/decimal.html#rounding-modes](https://docs.python.org/3/library/decimal.html#rounding-modes)|
Original file line number Diff line number Diff line change
@@ -1,6 +1,9 @@
# SPDX-FileCopyrightText: OpenSSF project contributors
# SPDX-License-Identifier: MIT
""" Compliant Code Example """
foo = int(round(0.9))
type(foo) # class int
print(foo) # prints 1
from decimal import Decimal, ROUND_HALF_UP, ROUND_HALF_DOWN

print(Decimal("0.5").quantize(Decimal("1"), rounding=ROUND_HALF_UP)) # prints 1
print(Decimal("1.5").quantize(Decimal("1"), rounding=ROUND_HALF_UP)) # prints 2
print(Decimal("0.5").quantize(Decimal("1"), rounding=ROUND_HALF_DOWN)) # prints 0
print(Decimal("1.5").quantize(Decimal("1"), rounding=ROUND_HALF_DOWN)) # prints 1
Original file line number Diff line number Diff line change
@@ -0,0 +1,17 @@
# SPDX-FileCopyrightText: OpenSSF project contributors
# SPDX-License-Identifier: MIT
""" Code Example """

print(int(0.5)) # prints 0
print(int(1.5)) # prints 1
print(int(1.45)) # prints 1
print(int(1.51)) # prints 1
print(int(-1.5)) # prints -1

print(round(0.5)) # prints 0
print(round(1.5)) # prints 2
print(round(1.45)) # prints 1
print(round(1.51)) # prints 2
print(round(-1.5)) # prints -2

print(type(round(0.5))) # prints <class 'int'>
Original file line number Diff line number Diff line change
@@ -0,0 +1,9 @@
# SPDX-FileCopyrightText: OpenSSF project contributors
# SPDX-License-Identifier: MIT
""" Code Example """
from decimal import ROUND_HALF_UP, Decimal

print(Decimal("0.10")) # prints 0.10
print(Decimal(0.10)) # prints 0.1000000000000000055511151231257827021181583404541015625
print(Decimal("0.10").quantize(Decimal("0.10"), rounding=ROUND_HALF_UP)) # prints 0.10
print(Decimal(0.10).quantize(Decimal("0.10"), rounding=ROUND_HALF_UP)) # prints 0.10
Original file line number Diff line number Diff line change
@@ -1,6 +1,8 @@
# SPDX-FileCopyrightText: OpenSSF project contributors
# SPDX-License-Identifier: MIT
""" Non-compliant Code Example """
foo = int(0.9)
type(foo) # class int
print(foo) # prints 0

print(int(0.5)) # prints 0
print(int(1.5)) # prints 1
print(round(0.5)) # prints 0
print(round(1.5)) # prints 2
2 changes: 1 addition & 1 deletion docs/Secure-Coding-Guide-for-Python/readme.md
Original file line number Diff line number Diff line change
Expand Up @@ -42,7 +42,7 @@ It is **not production code** and requires code-style or python best practices t
|:-----------------------------------------------------------------------------------------------------------------------------------------------|:----|
|[CWE-134: Use of Externally-Controlled Format String](CWE-664/CWE-134/README.md)|[CVE-2022-27177](https://www.cvedetails.com/cve/CVE-2022-27177/),<br/>CVSSv3.1: **9.8**,<br/>EPSS: **00.37** (01.12.2023)|
|[CWE-197: Numeric Truncation Error](CWE-664/CWE-197/README.md)||
|[CWE-197: Control rounding when converting to less precise numbers](CWE-664/CWE-197/01/.)||
|[CWE-197: Control rounding when converting to less precise numbers](CWE-664/CWE-197/01/README.md)||
|[CWE-400: Uncontrolled Resource Consumption](CWE-664/CWE-400/README.md)||
|[CWE-409: Improper Handling of Highly Compressed Data (Data Amplification)](CWE-664/CWE-409/.)||
|[CWE-410: Insufficient Resource Pool](CWE-664/CWE-410/README.md)||
Expand Down

0 comments on commit e247db1

Please sign in to comment.