Skip to content

Commit

Permalink
std: Respect formatting flags for str-like OsStr
Browse files Browse the repository at this point in the history
Historically many `Display` and `Debug` implementations for `OsStr`-like
abstractions have gone through `String::from_utf8_lossy`, but this was updated
in #42613 to use an internal `Utf8Lossy` abstraction instead. This had the
unfortunate side effect of causing a regression (#43765) in code which relied on
these `fmt` trait implementations respecting the various formatting flags
specified.

This commit opportunistically adds back interpretation of formatting trait flags
in the "common case" where where `OsStr`-like "thing" is all valid utf-8 and can
delegate to the formatting implementation for `str`. This doesn't entirely solve
the regression as non-utf8 paths will format differently than they did before
still (in that they will not respect formatting flags), but this should solve
the regression for all "real world" use cases of paths and such. The door's also
still open for handling these flags in the future!

Closes #43765
  • Loading branch information
alexcrichton committed Aug 14, 2017
1 parent f3cf206 commit 742ca0c
Show file tree
Hide file tree
Showing 4 changed files with 29 additions and 6 deletions.
6 changes: 6 additions & 0 deletions src/libstd/path.rs
Original file line number Diff line number Diff line change
Expand Up @@ -3953,4 +3953,10 @@ mod tests {
assert_eq!(path, path_buf);
assert!(path_buf.into_os_string().capacity() >= 15);
}

#[test]
fn display_format_flags() {
assert_eq!(format!("a{:#<5}b", Path::new("").display()), "a#####b");
assert_eq!(format!("a{:#<5}b", Path::new("a").display()), "aa####b");
}
}
10 changes: 7 additions & 3 deletions src/libstd/sys_common/wtf8.rs
Original file line number Diff line number Diff line change
Expand Up @@ -452,10 +452,14 @@ impl fmt::Display for Wtf8 {
pos = surrogate_pos + 3;
},
None => {
formatter.write_str(unsafe {
let s = unsafe {
str::from_utf8_unchecked(&wtf8_bytes[pos..])
})?;
return Ok(());
};
if pos == 0 {
return s.fmt(formatter)
} else {
return formatter.write_str(s)
}
}
}
}
Expand Down
1 change: 0 additions & 1 deletion src/libstd_unicode/lib.rs
Original file line number Diff line number Diff line change
Expand Up @@ -34,7 +34,6 @@

#![feature(core_char_ext)]
#![feature(str_internals)]
#![feature(core_intrinsics)]
#![feature(decode_utf8)]
#![feature(fused)]
#![feature(fn_traits)]
Expand Down
18 changes: 16 additions & 2 deletions src/libstd_unicode/lossy.rs
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@ use core::str as core_str;
use core::fmt;
use core::fmt::Write;
use char;
use core::intrinsics;
use core::mem;


/// Lossy UTF-8 string.
Expand All @@ -27,7 +27,7 @@ impl Utf8Lossy {
}

pub fn from_bytes(bytes: &[u8]) -> &Utf8Lossy {
unsafe { intrinsics::transmute(bytes) }
unsafe { mem::transmute(bytes) }
}

pub fn chunks(&self) -> Utf8LossyChunksIter {
Expand Down Expand Up @@ -153,7 +153,21 @@ impl<'a> Iterator for Utf8LossyChunksIter<'a> {

impl fmt::Display for Utf8Lossy {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
// If we're the empty string then our iterator won't actually yield
// anything, so perform the formatting manually
if self.bytes.len() == 0 {
return "".fmt(f)
}

for Utf8LossyChunk { valid, broken } in self.chunks() {
// If we successfully decoded the whole chunk as a valid string then
// we can return a direct formatting of the string which will also
// respect various formatting flags if possible.
if valid.len() == self.bytes.len() {
assert!(broken.is_empty());
return valid.fmt(f)
}

f.write_str(valid)?;
if !broken.is_empty() {
f.write_char(char::REPLACEMENT_CHARACTER)?;
Expand Down

0 comments on commit 742ca0c

Please sign in to comment.