added exclude flag for excluding extensions

CHANGELOG.md

@@ -2,6 +2,10 @@
 Dates are given in YYYY-MM-DD format.
 
 ## v0.2
+### v0.2.14 (2021-xx-yy)
+#### Features
+- Added `-x`/`--exclude` flag for excluding file extensions (overrides `-e` or `-E`)
+
 ### v0.2.13 (2021-04-26)
 #### Features
 - Added `-v`/`--verbose` flag for setting verbosity without using `RUST_LOG`

Cargo.lock

@@ -313,9 +313,9 @@ dependencies = [
 
 [[package]]
 name = "libc"
-version = "0.2.93"
+version = "0.2.94"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9385f66bf6105b241aa65a61cb923ef20efc665cb9f9bb50ac2f0c4b7f378d41"
+checksum = "18794a8ad5b29321f790b55d93dfba91e125cb1a9edbd4f8e3150acc771c1a5e"
 
 [[package]]
 name = "log"

Cargo.toml

@@ -10,7 +10,7 @@ repository = "https://gitlab.com/Lynnesbian/fif"
 readme = "README.md"
 keywords = ["mime", "mimetype", "utilities", "tools"]
 categories = ["command-line-utilities"]
-exclude = [".idea/", "Cross.toml", "*.sh", "*.py", ".drone.yml", "pkg/"]
+exclude = [".idea/", "*.toml", "!Cargo.toml", "*.sh", "*.py", "*.yml", "*.md", ".mailmap", "pkg/"]
 #resolver = "2"
 #license-file = "LICENSE"
 
@@ -66,8 +66,9 @@ fastrand = "1.4.1"
 [profile.release]
 lto = "thin"
 
+# perform some simple optimisations when testing
 [profile.test]
-opt-level = 0
+opt-level = 1
 
 # optimise dependencies, even when producing debug and test builds
 [profile.dev.package."*"]

clippy.sh

@@ -1,6 +1,6 @@
 #!/bin/bash
 fd -e rs -x touch {}
-cargo clippy --all-features -- \
+cargo clippy --all-features --tests -- \
  -W clippy::nursery \
  -W clippy::perf \
  -W clippy::pedantic \

src/main.rs

@@ -78,10 +78,17 @@ fn main() {
 	debug!("Iterating directory: {:?}", args.dirs);
 
 	let extensions = args.extensions();
+	let excludes = args.excluded_extensions();
 
+	if let Some(extensions) = &extensions {
 		debug!("Checking files with extensions: {:?}", extensions);
+	} else if let Some(excludes) = &excludes {
+		debug!("Skipping files with extensions: {:?}", excludes);
+	} else {
+		debug!("Checking files regardless of extensions");
+	}
 
-	let entries = scan_directory(&args.dirs, extensions.as_ref(), &args.get_scan_opts());
+	let entries = scan_directory(&args.dirs, extensions.as_ref(), excludes.as_ref(), &args.get_scan_opts());
 
 	if entries.is_none() {
 		// no need to log anything for fatal errors - fif will already have printed something obvious like
@@ -172,7 +179,7 @@ cfg_if! {
 
 /// Returns `true` if a file matches the given criteria. This means checking whether the file's extension appears in
 /// `exts` (if specified), potentially skipping over hidden files, and so on.
-fn wanted_file(entry: &DirEntry, exts: Option<&Vec<&str>>, scan_opts: &ScanOpts) -> bool {
+fn wanted_file(entry: &DirEntry, exts: Option<&Vec<&str>>, exclude: Option<&Vec<&str>>, scan_opts: &ScanOpts) -> bool {
 	if entry.depth() == 0 {
 		// the root directory should always be scanned.
 		return true;
@@ -188,19 +195,21 @@ fn wanted_file(entry: &DirEntry, exts: Option<&Vec<&str>>, scan_opts: &ScanOpts)
 		return true;
 	}
 
-	let ext = extension_from_path(entry.path());
-
-	if ext.is_none() && !scan_opts.extensionless {
-		// don't scan files without extensions.
-		return false;
-	}
+	if let Some(ext) = extension_from_path(entry.path()) {
+		// file has extension - discard invalid UTF-8 and normalise it to lowercase.
+		let ext = ext.to_string_lossy().to_lowercase();
+		let ext = ext.as_str();
 
 		if let Some(exts) = exts {
 			// only scan if the file has one of the specified extensions.
-		exts.contains(&ext.unwrap().to_string_lossy().to_lowercase().as_str())
+			exts.contains(&ext)
 		} else {
-		// no extensions specified - no reason not to scan this file.
-		true
+			// no extensions specified - the file should be scanned unless its extension is on the exclude list.
+			exclude.map_or(true, |exclude| !exclude.contains(&ext))
+		}
+	} else {
+		// no file extension
+		scan_opts.extensionless
 	}
 }
 
@@ -276,11 +285,11 @@ fn scan_from_walkdir(entries: &[DirEntry]) -> Vec<Result<Findings, ScanError>> {
 
 /// Scans a given directory with [`WalkDir`], filters with [`wanted_file`], checks for errors, and returns a vector of
 /// [DirEntry]s.
-fn scan_directory(dirs: &Path, exts: Option<&Vec<&str>>, scan_opts: &ScanOpts) -> Option<Vec<DirEntry>> {
+fn scan_directory(dirs: &Path, exts: Option<&Vec<&str>>, exclude: Option<&Vec<&str>>, scan_opts: &ScanOpts) -> Option<Vec<DirEntry>> {
 	let stepper = WalkDir::new(dirs).follow_links(scan_opts.follow_symlinks).into_iter();
 	let mut probably_fatal_error = false;
 	let entries: Vec<DirEntry> = stepper
-		.filter_entry(|e| wanted_file(e, exts, scan_opts)) // filter out unwanted files
+		.filter_entry(|e| wanted_file(e, exts, exclude, scan_opts)) // filter out unwanted files
 		.filter_map(|e| {
 			if let Err(err) = &e {
 				debug!("uh oh spaghettio!! {:#?}", e);

src/parameters.rs

@@ -40,7 +40,7 @@ pub enum OutputFormat {
 	setting(AppSettings::ColoredHelp)
 )]
 pub struct Parameters {
-	/// Only examine files with these extensions (Comma-separated list).
+	/// Only examine files with these extensions (comma-separated list).
 	/// This argument conflicts with `-E`.
 	#[clap(short, long, use_delimiter = true, require_delimiter = true, group = "extensions")]
 	pub exts: Option<Vec<StringType>>,
@@ -50,6 +50,16 @@ pub struct Parameters {
 	#[clap(short = 'E', long, arg_enum, group = "extensions")]
 	pub ext_set: Option<ExtensionSet>,
 
+	/// Don't scan files with these extensions (comma-separated list).
+	/// This option takes preference over files specified with -e or -E.
+	#[clap(
+		short = 'x',
+		long,
+		use_delimiter = true,
+		require_delimiter = true,
+	)]
+	pub exclude: Option<Vec<StringType>>,
+
 	/// Don't skip hidden files and directories.
 	/// Even if this flag is not present, fif will still recurse into a hidden root directory - for example, `fif
 	/// ~/.hidden` will recurse into `~/.hidden` regardless of whether or not -s was passed as an argument.
@@ -94,19 +104,41 @@ pub struct ScanOpts {
 }
 
 impl Parameters {
+	/// Returns an optional vec of the extensions to be scanned - i.e., extensions specified via the `-e` or `-E` flag,
+	/// minus the extensions excluded with the `-x` flag.
 	pub fn extensions(&self) -> Option<Vec<&str>> {
+		let empty_vec = vec![];
+		let exclude = &self.excluded_extensions().unwrap_or(empty_vec);
+
+		// TODO: bleugh
 		if let Some(exts) = &self.exts {
 			// extensions supplied like "-e png,jpg,jpeg"
-			Some(exts.iter().map(|s| s.as_str()).collect())
+			Some(
+				exts
+					.iter()
+					.map(|ext| ext.as_str())
+					.filter(|ext| !exclude.contains(ext))
+					.collect(),
+			)
 		} else if let Some(exts) = &self.ext_set {
 			// extensions supplied like "-E images"
-			Some(exts.extensions())
+			Some(
+				exts
+					.extensions()
+					.into_iter()
+					.filter(|ext| !exclude.contains(ext))
+					.collect(),
+			)
 		} else {
 			// neither -E nor -e was passed
 			None
 		}
 	}
 
+	pub fn excluded_extensions(&self) -> Option<Vec<&str>> {
+		self.exclude.as_ref().map(|exclude| exclude.iter().map(|ext| ext.as_str()).collect())
+	}
+
 	pub const fn get_scan_opts(&self) -> ScanOpts {
 		ScanOpts {
 			hidden: self.scan_hidden,

src/tests/mod.rs

@@ -113,7 +113,7 @@ fn simple_directory() {
 		follow_symlinks: false,
 	};
 
-	let entries = scan_directory(&dir.path().to_path_buf(), None, &scan_opts).expect("Directory scan failed.");
+	let entries = scan_directory(&dir.path().to_path_buf(), None, None, &scan_opts).expect("Directory scan failed.");
 
 	assert_eq!(entries.len(), files.len());
 
@@ -168,7 +168,6 @@ fn simple_directory() {
 /// Ensure that command line argument parsing works correctly - flags are interpreted, booleans are set, and so on.
 fn argument_parsing() {
 	use crate::parameters::{Parameters, ScanOpts};
-
 	use clap::Clap;
 
 	// pass `-f`, which enables following symlinks, and `-E images`, which scans files with image extensions
@@ -177,8 +176,10 @@ fn argument_parsing() {
 	// check if "jpg" is in the list of extensions to be scanned
 	assert!(args
 		.extensions()
-		.expect("args.extensions() should contain the `images` set!")
-		.contains(&"jpg"));
+		.expect("args.extensions() should be Some(_)!")
+		.contains(&"jpg"),
+		"args.extensions() should contain the `images` set!"
+	);
 
 	// make sure "scan_hidden" is false
 	assert!(!args.scan_hidden);
@@ -186,6 +187,9 @@ fn argument_parsing() {
 	// exts should be none
 	assert!(args.exts.is_none());
 
+	// there shouldn't be any excluded extensions
+	assert!(args.excluded_extensions().is_none());
+
 	// get the ScanOpts, and make sure they match expectations
 	assert_eq!(
 		args.get_scan_opts(),
@@ -198,6 +202,36 @@ fn argument_parsing() {
 	)
 }
 
+#[test]
+/// Ensure exclude overrides `-e` and `-E`.
+fn exclude_overrides() {
+	use crate::parameters::{Parameters};
+	use clap::Clap;
+
+	// pass `-E images`, which includes many image extensions, and `-x jpg,png`, which should remove "jpg" and "png" from
+	// the extensions list
+	let args: Parameters = Parameters::parse_from(vec!["fif", "-x", "jpg,png", "-E", "images"]);
+	let extensions = args.extensions();
+	assert!(extensions.is_some(), "Extensions should contain the `images` set!");
+	let extensions = extensions.unwrap();
+
+	assert!(!extensions.contains(&"jpg"), "\"jpg\" should be excluded!");
+	assert!(!extensions.contains(&"png"), "\"png\" should be excluded!");
+	assert!(extensions.contains(&"jpeg"), "\"jpeg\" should be included!");
+
+	// pass `-e abc,def,ghi,jkl` and `-x abc,def` -- extensions() should only contain "ghi" and "jkl"
+	let args: Parameters = Parameters::parse_from(vec!["fif", "-e", "abc,def,ghi,jkl", "-x", "abc,def"]);
+	let extensions = args.extensions();
+	assert!(extensions.is_some(), "Extensions should be set!");
+	let extensions = extensions.unwrap();
+
+	assert!(!extensions.contains(&"abc"));
+	assert!(!extensions.contains(&"def"));
+	assert!(extensions.contains(&"ghi"));
+	assert!(extensions.contains(&"jkl"));
+}
+
+
 #[test]
 /// Ensure that badly formed command line arguments are rejected.
 fn rejects_bad_args() {