Skip to content

annotate test files for secrets #676

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 5 commits into from
Apr 7, 2025
Merged

annotate test files for secrets #676

Show file tree
Hide file tree
Changes from 3 commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
59d5467
address comments
juli1 Apr 3, 2025
9180536
Merge branch 'main' into julien/secrets-test-file
juli1 Apr 5, 2025
69f5b1d
fix linting
juli1 Apr 5, 2025
42a4ade
address feedback
juli1 Apr 7, 2025
7736712
style
juli1 Apr 7, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
121 changes: 85 additions & 36 deletions crates/bins/src/bin/datadog-static-analyzer.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,13 +3,7 @@ use getopts::Options;
use indicatif::ProgressBar;
use itertools::Itertools;
use rayon::prelude::*;
use std::collections::HashMap;
use std::io::prelude::*;
use std::path::PathBuf;
use std::process::exit;
use std::sync::Arc;
use std::time::{Duration, Instant, SystemTime};
use std::{env, fs};

use cli::config_file::get_config;
use cli::constants::{
Expand All @@ -25,7 +19,7 @@ use cli::datadog_utils::{
};
use cli::file_utils::{
are_subdirectories_safe, filter_files_by_diff_aware_info, filter_files_by_size,
filter_files_for_language, get_files, read_files_from_gitignore,
filter_files_for_language, get_files, get_language_for_file, read_files_from_gitignore,
};
use cli::model::cli_configuration::CliConfiguration;
use cli::model::datadog_api::DiffAwareData;
Expand Down Expand Up @@ -54,6 +48,14 @@ use secrets::model::secret_result::{SecretResult, SecretValidationStatus};
use secrets::scanner::{build_sds_scanner, find_secrets};
use secrets::secret_files::should_ignore_file_for_secret;
use std::cell::Cell;
use std::collections::hash_map::Entry;
use std::collections::HashMap;
use std::io::prelude::*;
use std::path::Path;
use std::process::exit;
use std::sync::Arc;
use std::time::{Duration, Instant, SystemTime};
use std::{env, fs};

fn print_usage(program: &str, opts: Options) {
let brief = format!("Usage: {} FILE [options]", program);
Expand Down Expand Up @@ -751,7 +753,9 @@ fn main() -> Result<()> {
// Secrets detection
let mut secrets_results: Vec<SecretResult> = vec![];
if secrets_enabled {
let path_metadata;
let secrets_start = Instant::now();
let all_path_metadata_secrets = HashMap::<String, Option<ArtifactClassification>>::new();

let secrets_files: Vec<PathBuf> = files_to_analyze
.into_iter()
Expand All @@ -768,36 +772,72 @@ fn main() -> Result<()> {
let nb_secrets_rules: usize = secrets_rules.len();
let nb_secrets_files = secrets_files.len();

secrets_results = secrets_files
.par_iter()
.flat_map(|path| {
let relative_path = path
.strip_prefix(directory_path)
.unwrap()
.to_str()
.expect("path contains non-Unicode characters");
let res = if let Ok(file_content) = read_file(path) {
let file_content = Arc::from(file_content);
find_secrets(
&sds_scanner,
&secrets_rules,
relative_path,
&file_content,
&analysis_options,
)
} else {
// this is generally because the file is binary.
if use_debug {
eprintln!("error when getting content of path {}", &path.display());
(secrets_results, path_metadata) = secrets_files
.into_par_iter()
.fold(
|| (Vec::new(), HashMap::new()),
|(_, mut path_metadata), path| {
let relative_path = path
.strip_prefix(directory_path)
.expect("cannot strip prefix from path")
.to_str()
.expect("path contains non-Unicode characters");
let res = if let Ok(file_content) = read_file(&path) {
let file_content = Arc::from(file_content);
let secrets = find_secrets(
&sds_scanner,
&secrets_rules,
relative_path,
&file_content,
&analysis_options,
);

if !secrets.is_empty()
&& !all_path_metadata_secrets.contains_key(relative_path)
{
let cloned_path_str = relative_path.to_string();
let language_opt = get_language_for_file(&path);

if let Some(language) = language_opt {
let metadata = if is_test_file(
language,
file_content.as_ref(),
Path::new(&cloned_path_str),
None,
) {
Some(ArtifactClassification { is_test_file: true })
} else {
None
};
path_metadata.insert(cloned_path_str.clone(), metadata);
}
}

secrets
} else {
// this is generally because the file is binary.
if use_debug {
eprintln!("error when getting content of path {}", path.display());
}
vec![]
};
if let Some(pb) = &progress_bar {
pb.inc(1);
}
vec![]
};
if let Some(pb) = &progress_bar {
pb.inc(1);
}
res
})
.collect();
(res, path_metadata)
},
)
.reduce(
|| (Vec::new(), HashMap::new()),
|mut base, other| {
let (other_results, other_classifications) = other;
base.0.extend(other_results);
for (k, v) in other_classifications {
base.1.insert(k, v);
}
base
},
);

let nb_secrets_found: u32 = secrets_results.iter().map(|x| x.matches.len() as u32).sum();
let nb_secrets_validated: u32 = secrets_results
Expand All @@ -811,6 +851,15 @@ fn main() -> Result<()> {
})
.sum();

// adding metadata from secrets
for (k, v) in path_metadata {
if let Entry::Vacant(e) = all_path_metadata.entry(k) {
if let Some(artifact_classification) = v {
e.insert(artifact_classification);
}
}
}

if let Some(pb) = &progress_bar {
pb.finish();
}
Expand Down
65 changes: 65 additions & 0 deletions crates/cli/src/file_utils.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -78,6 +78,47 @@ fn get_prefix_for_language(language: &Language) -> Option<Vec<String>> {
None
}

/// Find the language for a given file.
pub fn get_language_for_file(path: &Path) -> Option<Language> {
// match for extensions (myfile.c, myfile.php, etc).
for (language, extensions) in FILE_EXTENSIONS_PER_LANGUAGE_LIST {
let extensions_string = extensions
.to_vec()
.iter()
.map(|x| x.to_string())
.collect::<Vec<String>>();
if match_extension(path, extensions_string.as_slice()) {
return Some(*language);
}
}

// match for exact match (e.g. BUILD.bazel, Dockerfile, etc)
for (language, filenames) in FILE_EXACT_MATCH_PER_LANGUAGE_LIST {
let filename_strings = filenames
.to_vec()
.iter()
.map(|x| x.to_string())
.collect::<Vec<String>>();
if match_exact_filename(path, filename_strings.as_slice()) {
return Some(*language);
}
}

// match for prefix (e.g. Dockerfile.something)
for (language, prefixes) in FILE_PREFIX_PER_LANGUAGE_LIST {
let prefix_string = prefixes
.to_vec()
.iter()
.map(|x| x.to_string())
.collect::<Vec<String>>();
if match_prefix_filename(path, prefix_string.as_slice()) {
return Some(*language);
}
}

None
}

// Read the .gitignore file in a directory and return the lines that are not commented
// or empty.
// We ignore pattern that start with # (comments) or contains ! (cause repositories
Expand Down Expand Up @@ -862,4 +903,28 @@ mod tests {
.len()
);
}

#[test]
fn test_get_language_for_file() {
// extension
assert_eq!(
get_language_for_file(&PathBuf::from("path/to/foo.java")),
Some(Language::Java)
);

// exact filename
assert_eq!(
get_language_for_file(&PathBuf::from("BUILD.bazel")),
Some(Language::Starlark)
);

// prefix
assert_eq!(
get_language_for_file(&PathBuf::from("Dockerfile.foobar")),
Some(Language::Dockerfile)
);

// none
assert_eq!(get_language_for_file(&PathBuf::from("wefwefwef")), None);
}
}
6 changes: 4 additions & 2 deletions misc/integration-test-secrets.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,8 +22,10 @@ RES=`jq '.runs[0].results | length ' "${REPO_DIR}/results1.json"`

echo "Found $RES errors on first run"

if [ "$RES" -ne "3" ]; then
echo "incorrect number of errors found"
EXPECTING=5

if [ "$RES" -ne "$EXPECTING" ]; then
echo "incorrect number of errors found, found $RES, expected $EXPECTING"
exit 1
fi

Expand Down
Loading