Skip to content

cksum: accept non-UTF-8 filenames #6575

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 3 commits into from
Jul 30, 2024
Merged

cksum: accept non-UTF-8 filenames #6575

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
244634b
cksum: accept non-UTF-8 filenames
BenWiederhake Jul 15, 2024
6af9fd7
cksum: correctly output non-utf8 filename
BenWiederhake Jul 16, 2024
1fcff25
uucore: mark comment as rustdoc
BenWiederhake Jul 19, 2024
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
133 changes: 74 additions & 59 deletions src/uu/cksum/src/cksum.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,8 +4,9 @@
// file that was distributed with this source code.

// spell-checker:ignore (ToDO) fname, algo
use clap::builder::ValueParser;
use clap::{crate_version, value_parser, Arg, ArgAction, Command};
use std::ffi::OsStr;
use std::ffi::{OsStr, OsString};
use std::fs::File;
use std::io::{self, stdin, stdout, BufReader, Read, Write};
use std::iter;
Expand All @@ -18,7 +19,7 @@ use uucore::checksum::{
use uucore::{
encoding,
error::{FromIo, UResult, USimpleError},
format_usage, help_about, help_section, help_usage, show,
format_usage, help_about, help_section, help_usage, os_str_as_bytes, show,
sum::{div_ceil, Digest},
};

Expand Down Expand Up @@ -116,52 +117,64 @@ where
};
// The BSD checksum output is 5 digit integer
let bsd_width = 5;
match (options.algo_name, not_file) {
(ALGORITHM_OPTIONS_SYSV, true) => println!(
"{} {}",
sum.parse::<u16>().unwrap(),
div_ceil(sz, options.output_bits)
let (before_filename, should_print_filename, after_filename) = match options.algo_name {
ALGORITHM_OPTIONS_SYSV => (
format!(
"{} {}{}",
sum.parse::<u16>().unwrap(),
div_ceil(sz, options.output_bits),
if not_file { "" } else { " " }
),
!not_file,
String::new(),
),
(ALGORITHM_OPTIONS_SYSV, false) => println!(
"{} {} {}",
sum.parse::<u16>().unwrap(),
div_ceil(sz, options.output_bits),
filename.display()
ALGORITHM_OPTIONS_BSD => (
format!(
"{:0bsd_width$} {:bsd_width$}{}",
sum.parse::<u16>().unwrap(),
div_ceil(sz, options.output_bits),
if not_file { "" } else { " " }
),
!not_file,
String::new(),
),
(ALGORITHM_OPTIONS_BSD, true) => println!(
"{:0bsd_width$} {:bsd_width$}",
sum.parse::<u16>().unwrap(),
div_ceil(sz, options.output_bits)
ALGORITHM_OPTIONS_CRC => (
format!("{sum} {sz}{}", if not_file { "" } else { " " }),
!not_file,
String::new(),
),
(ALGORITHM_OPTIONS_BSD, false) => println!(
"{:0bsd_width$} {:bsd_width$} {}",
sum.parse::<u16>().unwrap(),
div_ceil(sz, options.output_bits),
filename.display()
),
(ALGORITHM_OPTIONS_CRC, true) => println!("{sum} {sz}"),
(ALGORITHM_OPTIONS_CRC, false) => println!("{sum} {sz} {}", filename.display()),
(ALGORITHM_OPTIONS_BLAKE2B, _) if options.tag => {
if let Some(length) = options.length {
// Multiply by 8 here, as we want to print the length in bits.
println!("BLAKE2b-{} ({}) = {sum}", length * 8, filename.display());
} else {
println!("BLAKE2b ({}) = {sum}", filename.display());
}
ALGORITHM_OPTIONS_BLAKE2B if options.tag => {
(
if let Some(length) = options.length {
// Multiply by 8 here, as we want to print the length in bits.
format!("BLAKE2b-{} (", length * 8)
} else {
"BLAKE2b (".to_owned()
},
true,
format!(") = {sum}"),
)
}
_ => {
if options.tag {
println!(
"{} ({}) = {sum}",
options.algo_name.to_ascii_uppercase(),
filename.display()
);
(
format!("{} (", options.algo_name.to_ascii_uppercase()),
true,
format!(") = {sum}"),
)
} else {
let prefix = if options.asterisk { "*" } else { " " };
println!("{sum} {prefix}{}", filename.display());
(format!("{sum} {prefix}"), true, String::new())
}
}
};
print!("{}", before_filename);
if should_print_filename {
// The filename might not be valid UTF-8, and filename.display() would mangle the names.
// Therefore, emit the bytes directly to stdout, without any attempt at encoding them.
let _dropped_result = stdout().write_all(os_str_as_bytes(filename.as_os_str())?);
}
println!("{}", after_filename);
}

Ok(())
Expand Down Expand Up @@ -209,7 +222,7 @@ fn prompt_asterisk(tag: bool, binary: bool, had_reset: bool) -> bool {
* Don't do it with clap because if it struggling with the --overrides_with
* marking the value as set even if not present
*/
fn had_reset(args: &[String]) -> bool {
fn had_reset(args: &[OsString]) -> bool {
// Indices where "--binary" or "-b", "--tag", and "--untagged" are found
let binary_index = args.iter().position(|x| x == "--binary" || x == "-b");
let tag_index = args.iter().position(|x| x == "--tag");
Expand All @@ -234,7 +247,7 @@ fn handle_tag_text_binary_flags(matches: &clap::ArgMatches) -> UResult<(bool, bo

let binary_flag: bool = matches.get_flag(options::BINARY);

let args: Vec<String> = std::env::args().collect();
let args: Vec<OsString> = std::env::args_os().collect();
let had_reset = had_reset(&args);

let asterisk: bool = prompt_asterisk(tag, binary_flag, had_reset);
Expand Down Expand Up @@ -298,7 +311,7 @@ pub fn uumain(args: impl uucore::Args) -> UResult<()> {

// Execute the checksum validation based on the presence of files or the use of stdin

let files = matches.get_many::<String>(options::FILE).map_or_else(
let files = matches.get_many::<OsString>(options::FILE).map_or_else(
|| iter::once(OsStr::new("-")).collect::<Vec<_>>(),
|files| files.map(OsStr::new).collect::<Vec<_>>(),
);
Expand Down Expand Up @@ -337,7 +350,7 @@ pub fn uumain(args: impl uucore::Args) -> UResult<()> {
asterisk,
};

match matches.get_many::<String>(options::FILE) {
match matches.get_many::<OsString>(options::FILE) {
Some(files) => cksum(opts, files.map(OsStr::new))?,
None => cksum(opts, iter::once(OsStr::new("-")))?,
};
Expand All @@ -356,6 +369,7 @@ pub fn uu_app() -> Command {
Arg::new(options::FILE)
.hide(true)
.action(clap::ArgAction::Append)
.value_parser(ValueParser::os_string())
.value_hint(clap::ValueHint::FilePath),
)
.arg(
Expand Down Expand Up @@ -469,61 +483,62 @@ mod tests {
use super::had_reset;
use crate::calculate_blake2b_length;
use crate::prompt_asterisk;
use std::ffi::OsString;

#[test]
fn test_had_reset() {
let args = ["--binary", "--tag", "--untagged"]
.iter()
.map(|&s| s.to_string())
.collect::<Vec<String>>();
.map(|&s| s.into())
.collect::<Vec<OsString>>();
assert!(had_reset(&args));

let args = ["-b", "--tag", "--untagged"]
.iter()
.map(|&s| s.to_string())
.collect::<Vec<String>>();
.map(|&s| s.into())
.collect::<Vec<OsString>>();
assert!(had_reset(&args));

let args = ["-b", "--binary", "--tag", "--untagged"]
.iter()
.map(|&s| s.to_string())
.collect::<Vec<String>>();
.map(|&s| s.into())
.collect::<Vec<OsString>>();
assert!(had_reset(&args));

let args = ["--untagged", "--tag", "--binary"]
.iter()
.map(|&s| s.to_string())
.collect::<Vec<String>>();
.map(|&s| s.into())
.collect::<Vec<OsString>>();
assert!(!had_reset(&args));

let args = ["--untagged", "--tag", "-b"]
.iter()
.map(|&s| s.to_string())
.collect::<Vec<String>>();
.map(|&s| s.into())
.collect::<Vec<OsString>>();
assert!(!had_reset(&args));

let args = ["--binary", "--tag"]
.iter()
.map(|&s| s.to_string())
.collect::<Vec<String>>();
.map(|&s| s.into())
.collect::<Vec<OsString>>();
assert!(!had_reset(&args));

let args = ["--tag", "--untagged"]
.iter()
.map(|&s| s.to_string())
.collect::<Vec<String>>();
.map(|&s| s.into())
.collect::<Vec<OsString>>();
assert!(!had_reset(&args));

let args = ["--text", "--untagged"]
.iter()
.map(|&s| s.to_string())
.collect::<Vec<String>>();
.map(|&s| s.into())
.collect::<Vec<OsString>>();
assert!(!had_reset(&args));

let args = ["--binary", "--untagged"]
.iter()
.map(|&s| s.to_string())
.collect::<Vec<String>>();
.map(|&s| s.into())
.collect::<Vec<OsString>>();
assert!(!had_reset(&args));
}

Expand Down
6 changes: 3 additions & 3 deletions src/uucore/src/lib/lib.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -222,9 +222,9 @@ pub fn read_yes() -> bool {
}
}

// Helper function for processing delimiter values (which could be non UTF-8)
// It converts OsString to &[u8] for unix targets only
// On non-unix (i.e. Windows) it will just return an error if delimiter value is not UTF-8
/// Helper function for processing delimiter values (which could be non UTF-8)
/// It converts OsString to &[u8] for unix targets only
/// On non-unix (i.e. Windows) it will just return an error if delimiter value is not UTF-8
pub fn os_str_as_bytes(os_string: &OsStr) -> mods::error::UResult<&[u8]> {
#[cfg(unix)]
let bytes = os_string.as_bytes();
Expand Down
29 changes: 28 additions & 1 deletion tests/by-util/test_cksum.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@
//
// For the full copyright and license information, please view the LICENSE
// file that was distributed with this source code.
// spell-checker:ignore (words) asdf algo algos mgmt
// spell-checker:ignore (words) asdf algo algos asha mgmt xffname

use crate::common::util::TestScenario;

Expand Down Expand Up @@ -1250,3 +1250,30 @@ fn test_several_files_error_mgmt() {
.stderr_contains("empty: no properly ")
.stderr_contains("incorrect: no properly ");
}

#[cfg(target_os = "linux")]
#[test]
fn test_non_utf8_filename() {
use std::ffi::OsString;
use std::os::unix::ffi::OsStringExt;

let scene = TestScenario::new(util_name!());
let at = &scene.fixtures;
let filename: OsString = OsStringExt::from_vec(b"funky\xffname".to_vec());

at.touch(&filename);

scene
.ucmd()
.arg(&filename)
.succeeds()
.stdout_is_bytes(b"4294967295 0 funky\xffname\n")
.no_stderr();
scene
.ucmd()
.arg("-asha256")
.arg(filename)
.succeeds()
.stdout_is_bytes(b"SHA256 (funky\xffname) = e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855\n")
.no_stderr();
}
Loading