RGBCube/uutils-coreutils
1
Fork 0
mirror of https://github.com/RGBCube/uutils-coreutils synced 2025-07-29 12:07:46 +00:00
Code Issues Activity

checksum: move regex detection to the line level

Browse source
This commit is contained in:
Dorian Péron 2024-12-05 15:57:01 +01:00 • committed by Dorian Péron
parent c8bcdb9267
commit 5cbe87620c
1 changed files with 36 additions and 73 deletions
Download patch file Download diff file Expand all files Collapse all files

109
src/uucore/src/lib/features/checksum.rs
View file

@ -8,7 +8,7 @@ use data_encoding::BASE64;
use os_display::Quotable; use os_display::Quotable;
use regex::bytes::{Captures, Regex}; use regex::bytes::{Captures, Regex};
use std::{ use std::{
ffi::{OsStr, OsString}, ffi::OsStr,
fmt::Display, fmt::Display,
fs::File, fs::File,
io::{self, stdin, BufReader, Read, Write}, io::{self, stdin, BufReader, Read, Write},
@ -130,9 +130,6 @@ enum FileCheckError {
ImproperlyFormatted, ImproperlyFormatted,
/// reading of the checksum file failed /// reading of the checksum file failed
CantOpenChecksumFile, CantOpenChecksumFile,
/// Algorithm detection was unsuccessful.
/// Either none is provided, or there is a conflict.
AlgoDetectionError,
} }
impl From<Box<dyn UError>> for FileCheckError { impl From<Box<dyn UError>> for FileCheckError {
@ -441,7 +438,7 @@ fn get_filename_for_output(filename: &OsStr, input_is_stdin: bool) -> String {
} }
/// Determines the appropriate regular expression to use based on the provided lines. /// Determines the appropriate regular expression to use based on the provided lines.
fn determine_regex(lines: &[OsString]) -> Option<(Regex, bool)> { fn determine_regex(line: impl AsRef<OsStr>) -> Option<(Regex, bool)> {
let regexes = [ let regexes = [
(Regex::new(ALGO_BASED_REGEX).unwrap(), true), (Regex::new(ALGO_BASED_REGEX).unwrap(), true),
(Regex::new(DOUBLE_SPACE_REGEX).unwrap(), false), (Regex::new(DOUBLE_SPACE_REGEX).unwrap(), false),
@ -449,12 +446,10 @@ fn determine_regex(lines: &[OsString]) -> Option<(Regex, bool)> {
(Regex::new(ALGO_BASED_REGEX_BASE64).unwrap(), true), (Regex::new(ALGO_BASED_REGEX_BASE64).unwrap(), true),
]; ];
for line in lines { let line_bytes = os_str_as_bytes(line.as_ref()).expect("UTF-8 decoding failed");
let line_bytes = os_str_as_bytes(line).expect("UTF-8 decoding failed"); for (regex, is_algo_based) in &regexes {
for (regex, is_algo_based) in &regexes { if regex.is_match(line_bytes) {
if regex.is_match(line_bytes) { return Some((regex.clone(), *is_algo_based));
return Some((regex.clone(), *is_algo_based));
}
} }
} }
@ -599,13 +594,20 @@ fn process_checksum_line(
filename_input: &OsStr, filename_input: &OsStr,
line: &OsStr, line: &OsStr,
i: usize, i: usize,
chosen_regex: &Regex,
is_algo_based_format: bool,
cli_algo_name: Option<&str>, cli_algo_name: Option<&str>,
cli_algo_length: Option<usize>, cli_algo_length: Option<usize>,
opts: ChecksumOptions, opts: ChecksumOptions,
) -> Result<(), LineCheckError> { ) -> Result<(), LineCheckError> {
let line_bytes = os_str_as_bytes(line)?; let line_bytes = os_str_as_bytes(line)?;
// early return on empty or commented lines.
if line.is_empty() || line_bytes.starts_with(b"#") {
return Err(LineCheckError::Skipped);
}
let (chosen_regex, is_algo_based_format) =
determine_regex(line).ok_or(LineCheckError::ImproperlyFormatted)?;
if let Some(caps) = chosen_regex.captures(line_bytes) { if let Some(caps) = chosen_regex.captures(line_bytes) {
let mut filename_to_check = caps.name("filename").unwrap().as_bytes(); let mut filename_to_check = caps.name("filename").unwrap().as_bytes();
@ -617,7 +619,7 @@ fn process_checksum_line(
filename_to_check = &filename_to_check[1..]; filename_to_check = &filename_to_check[1..];
} }
let expected_checksum = get_expected_digest_as_hex_string(&caps, chosen_regex) let expected_checksum = get_expected_digest_as_hex_string(&caps, &chosen_regex)
.ok_or(LineCheckError::ImproperlyFormatted)?; .ok_or(LineCheckError::ImproperlyFormatted)?;
// If the algo_name is provided, we use it, otherwise we try to detect it // If the algo_name is provided, we use it, otherwise we try to detect it
@ -672,10 +674,6 @@ fn process_checksum_line(
Err(LineCheckError::DigestMismatch) Err(LineCheckError::DigestMismatch)
} }
} else { } else {
if line.is_empty() || line_bytes.starts_with(b"#") {
// Don't show any warning for empty or commented lines.
return Err(LineCheckError::Skipped);
}
if opts.warn { if opts.warn {
let algo = if let Some(algo_name_input) = cli_algo_name { let algo = if let Some(algo_name_input) = cli_algo_name {
algo_name_input.to_uppercase() algo_name_input.to_uppercase()
@ -723,19 +721,11 @@ fn process_checksum_file(
let reader = BufReader::new(file); let reader = BufReader::new(file);
let lines = read_os_string_lines(reader).collect::<Vec<_>>(); let lines = read_os_string_lines(reader).collect::<Vec<_>>();
let Some((chosen_regex, is_algo_based_format)) = determine_regex(&lines) else {
log_no_properly_formatted(get_filename_for_output(filename_input, input_is_stdin));
set_exit_code(1);
return Err(FileCheckError::AlgoDetectionError);
};
for (i, line) in lines.iter().enumerate() { for (i, line) in lines.iter().enumerate() {
let line_result = process_checksum_line( let line_result = process_checksum_line(
filename_input, filename_input,
line, line,
i, i,
&chosen_regex,
is_algo_based_format,
cli_algo_name, cli_algo_name,
cli_algo_length, cli_algo_length,
opts, opts,
@ -816,8 +806,7 @@ where
use FileCheckError::*; use FileCheckError::*;
match process_checksum_file(filename_input, algo_name_input, length_input, opts) { match process_checksum_file(filename_input, algo_name_input, length_input, opts) {
Err(UError(e)) => return Err(e), Err(UError(e)) => return Err(e),
Err(ImproperlyFormatted) => break, Err(CantOpenChecksumFile | ImproperlyFormatted) | Ok(_) => continue,
Err(CantOpenChecksumFile | AlgoDetectionError) | Ok(_) => continue,
} }
} }
@ -926,6 +915,7 @@ pub fn escape_filename(filename: &Path) -> (String, &'static str) {
#[cfg(test)] #[cfg(test)]
mod tests { mod tests {
use super::*; use super::*;
use std::ffi::OsString;
#[test] #[test]
fn test_unescape_filename() { fn test_unescape_filename() {
@ -1161,66 +1151,39 @@ mod tests {
#[test] #[test]
fn test_determine_regex() { fn test_determine_regex() {
// Test algo-based regex // Test algo-based regex
let lines_algo_based = ["MD5 (example.txt) = d41d8cd98f00b204e9800998ecf8427e"] let line_algo_based =
.iter() OsString::from("MD5 (example.txt) = d41d8cd98f00b204e9800998ecf8427e");
.map(|s| OsString::from(s.to_string())) let (regex, algo_based) = determine_regex(&line_algo_based).unwrap();
.collect::<Vec<_>>();
let (regex, algo_based) = determine_regex(&lines_algo_based).unwrap();
assert!(algo_based); assert!(algo_based);
assert!(regex.is_match(os_str_as_bytes(&lines_algo_based[0]).unwrap())); assert!(regex.is_match(os_str_as_bytes(&line_algo_based).unwrap()));
// Test double-space regex // Test double-space regex
let lines_double_space = ["d41d8cd98f00b204e9800998ecf8427e example.txt"] let line_double_space = OsString::from("d41d8cd98f00b204e9800998ecf8427e example.txt");
.iter() let (regex, algo_based) = determine_regex(&line_double_space).unwrap();
.map(|s| OsString::from(s.to_string()))
.collect::<Vec<_>>();
let (regex, algo_based) = determine_regex(&lines_double_space).unwrap();
assert!(!algo_based); assert!(!algo_based);
assert!(regex.is_match(os_str_as_bytes(&lines_double_space[0]).unwrap())); assert!(regex.is_match(os_str_as_bytes(&line_double_space).unwrap()));
// Test single-space regex // Test single-space regex
let lines_single_space = ["d41d8cd98f00b204e9800998ecf8427e example.txt"] let line_single_space = OsString::from("d41d8cd98f00b204e9800998ecf8427e example.txt");
.iter() let (regex, algo_based) = determine_regex(&line_single_space).unwrap();
.map(|s| OsString::from(s.to_string()))
.collect::<Vec<_>>();
let (regex, algo_based) = determine_regex(&lines_single_space).unwrap();
assert!(!algo_based); assert!(!algo_based);
assert!(regex.is_match(os_str_as_bytes(&lines_single_space[0]).unwrap())); assert!(regex.is_match(os_str_as_bytes(&line_single_space).unwrap()));
// Test double-space regex start with invalid
let lines_double_space = ["ERR", "d41d8cd98f00b204e9800998ecf8427e example.txt"]
.iter()
.map(|s| OsString::from(s.to_string()))
.collect::<Vec<_>>();
let (regex, algo_based) = determine_regex(&lines_double_space).unwrap();
assert!(!algo_based);
assert!(!regex.is_match(os_str_as_bytes(&lines_double_space[0]).unwrap()));
assert!(regex.is_match(os_str_as_bytes(&lines_double_space[1]).unwrap()));
// Test invalid checksum line // Test invalid checksum line
let lines_invalid = ["invalid checksum line"] let line_invalid = OsString::from("invalid checksum line");
.iter() assert!(determine_regex(&line_invalid).is_none());
.map(|s| OsString::from(s.to_string()))
.collect::<Vec<_>>();
assert!(determine_regex(&lines_invalid).is_none());
// Test leading space before checksum line // Test leading space before checksum line
let lines_algo_based_leading_space = let line_algo_based_leading_space =
[" MD5 (example.txt) = d41d8cd98f00b204e9800998ecf8427e"] OsString::from(" MD5 (example.txt) = d41d8cd98f00b204e9800998ecf8427e");
.iter() let res = determine_regex(&line_algo_based_leading_space);
.map(|s| OsString::from(s.to_string()))
.collect::<Vec<_>>();
let res = determine_regex(&lines_algo_based_leading_space);
assert!(res.is_some()); assert!(res.is_some());
assert_eq!(res.unwrap().0.as_str(), ALGO_BASED_REGEX); assert_eq!(res.unwrap().0.as_str(), ALGO_BASED_REGEX);
// Test trailing space after checksum line (should fail) // Test trailing space after checksum line (should fail)
let lines_algo_based_leading_space = let line_algo_based_leading_space =
["MD5 (example.txt) = d41d8cd98f00b204e9800998ecf8427e "] OsString::from("MD5 (example.txt) = d41d8cd98f00b204e9800998ecf8427e ");
.iter() let res = determine_regex(&line_algo_based_leading_space);
.map(|s| OsString::from(s.to_string()))
.collect::<Vec<_>>();
let res = determine_regex(&lines_algo_based_leading_space);
assert!(res.is_none()); assert!(res.is_none());
} }