RS-R1000 · Function with cyclomatic complexity higher than threshold

Function with cyclomatic complexity higher than threshold RS-R1000

Anti-pattern

Minor

6 months ago — 6 months old

fn call has a cyclomatic complexity of 79 with "critical" risk

141        self.api_impl.poll_ready(cx)
142    }
143
144    fn call(&mut self, req: (Request<Body>, C)) -> Self::Future {145        async fn run<T, C>(
146            mut api_impl: T,
147            req: (Request<Body>, C),

fn load_section_data_with_skel_value has a cyclomatic complexity of 31 with "very-high" risk

bpf-loader-rs/bpf-loader-lib/src/skeleton/preload/section_loader.rs

 43        vec}
 44    };
 45}
 46pub(crate) fn load_section_data_with_skel_value( 47    btf: &Btf,
 48    section: &DataSectionMeta,
 49    buffer: &mut [u8],

fn load_and_attach has a cyclomatic complexity of 27 with "very-high" risk

bpf-loader-rs/bpf-loader-lib/src/skeleton/preload/mod.rs

 52    /// load and attach the ebpf program to the kernel to run the ebpf program
 53    /// if the ebpf program has maps to export to user space, you need to call
 54    /// the wait and export.
 55    pub fn load_and_attach(mut self) -> Result<BpfSkeleton> { 56        // This function differs from the C++ version `int bpf_skeleton::load_and_attach_prog(void)`
 57        // Because we put the call to bpf_object_open in `BpfSkeletonBuilder::build`
 58        // So Here are just the calls to load and attach

fn wait_and_poll_to_handler_with_multiple_exporter has a cyclomatic complexity of 27 with "very-high" risk

bpf-loader-rs/bpf-loader-lib/src/skeleton/mod.rs

211    }
212    /// Start poll with each map corresponding to a different exporter
213    /// The function `exporter_provider` should return the ExportFormatType, EventHandler, and UserContext(if applies) for the given map name (If you want to set the exporter)
214    pub fn wait_and_poll_to_handler_with_multiple_exporter(215        &self,
216        exporter_provider: impl Fn(
217            &str,

Description

A function with high cyclomatic complexity can be hard to understand and maintain. Cyclomatic complexity is a software metric that measures the number of independent paths through a function. A higher cyclomatic complexity indicates that the function has more decision points and is more complex.

Functions with high cyclomatic complexity are more likely to have bugs and be harder to test. They may lead to reduced code maintainability and increased development time.

To reduce the cyclomatic complexity of a function, you can:

Break the function into smaller, more manageable functions.
Refactor complex logic into separate functions or classes.
Avoid multiple return paths and deeply nested control expressions.

Bad practice

use anyhow::{anyhow, Context, Result};
use std::io::{Lines, StdinLock};

trait InputHandle {
    type Input<'a>
    where
        Self: 'a;
    fn get_input<'a>(&self) -> Self::Input<'a>;
    fn process_input<'a, T>(&self, f: impl FnOnce(Self::Input<'a>) -> Result<T>) -> Result<T>
    where
        Self: 'a;
}

struct StdioInput;
impl InputHandle for StdioInput {
    type Input<'a> = Lines<StdinLock<'a>>;

    fn get_input<'a>(&self) -> Self::Input<'a> {
        std::io::stdin().lines()
    }

    fn process_input<'a, T>(&self, f: impl FnOnce(Self::Input<'a>) -> Result<T>) -> Result<T>
    where
        Self: 'a,
    {
        f(self.get_input())
    }
}

fn main() -> Result<()> {
    // cc = 1 (default)
    let index: usize = StdioInput.process_input(|mut input| {
        // cc = 3 (?, ?)
        str::parse(input.next().context("no input from stdin")??.as_str()).context("parse failed")
    })?; // cc = 4 (?)

    // cc = 5 (for)
    for i in 0..index {
        let fizzbuzz = match i % 15 {
            // cc = 6 (match_arm 0)
            0 => "fizzbuzz".into(),
            // cc = 9 (match_arm 3 | 6 | 9)
            3 | 6 | 9 | 12 => "fizz".into(),
            // cc = 11 (match_arm 5 | 10)
            5 | 10 => "buzz".into(),
            x => format!("{x}"),
        };
        println!("{i}: {fizzbuzz}");
    }

    Ok(())
}

use anyhow::{anyhow, Context, Result};
use std::io::{Lines, StdinLock};

trait InputHandle {
    type Input<'a>
    where
        Self: 'a;
    fn get_input<'a>(&self) -> Self::Input<'a>;
    fn process_input<'a, T>(&self, f: impl FnOnce(Self::Input<'a>) -> Result<T>) -> Result<T>
    where
        Self: 'a;
}

struct StdioInput;
impl InputHandle for StdioInput {
    type Input<'a> = Lines<StdinLock<'a>>;

    fn get_input<'a>(&self) -> Self::Input<'a> {
        std::io::stdin().lines()
    }

    fn process_input<'a, T>(&self, f: impl FnOnce(Self::Input<'a>) -> Result<T>) -> Result<T>
    where
        Self: 'a,
    {
        f(self.get_input())
    }
}

fn fizzbuzz(u: usize) -> String {
    // cc = 1 (default)
    match u % 15 {
        // cc = 2 (match_arm 0)
        0 => "fizzbuzz".into(),
        // cc = 9 (match_arm 3 | 6 | 9)
        3 | 6 | 9 | 12 => "fizz".into(),
        // cc = 11 (match_arm 5 | 10)
        5 | 10 => "buzz".into(),
        x => format!("{x}"),
    }
} // total cc =

fn main() -> Result<()> {
    // cc = 1 (default)
    let index: usize = StdioInput.process_input(|mut input| {
        // cc = 3 (?, ?)
        str::parse(input.next().context("no input from stdin")??.as_str()).context("parse failed")
    })?; // cc = 4 (?)

    // cc = 5 (for)
    for i in 0..index {
        let val = fizzbuzz(i);
        println!("{i}: {val}");
    }

    Ok(())
}

Issue configuration

Cyclomatic complexity threshold can be configured using the cyclomatic_complexity_threshold (docs) in the .deepsource.toml config file.

Configuring this is optional. If you don't provide a value, the Analyzer will raise issues for functions with complexity higher than the default threshold, which is high(only raise issues for >25) for the Rust Analyzer.

Here's the mapping of the risk category to the cyclomatic complexity score to help you configure this better:

Risk category	Cyclomatic complexity range	Recommended action
low	1-5	No action needed.
medium	6-15	Review and monitor.
high	16-25	Review and refactor. Recommended to add comments if the function is absolutely needed to be kept as it is.
very-high.	26-50	Refactor to reduce the complexity.
critical	>50	Must refactor this. This can make the code untestable and very difficult to understand.

eunomia-bpf / eunomia-bpf

Bad practice

Recommended

Issue configuration