##// END OF EJS Templates
py3: make stdout line-buffered if connected to a TTY...
py3: make stdout line-buffered if connected to a TTY Status messages that are to be shown on the terminal should be written to the file descriptor before anything further is done, to keep the user updated. One common way to achieve this is to make stdout line-buffered if it is connected to a TTY. This is done on Python 2 (except on Windows, where libc, which the CPython 2 streams depend on, does not properly support this). Python 3 rolls it own I/O streams. On Python 3, buffered binary streams can't be set line-buffered. The previous code (added in 227ba1afcb65) incorrectly assumed that on Python 3, pycompat.stdout (sys.stdout.buffer) is already line-buffered. However the interpreter initializes it with a block-buffered stream or an unbuffered stream (when the -u option or the PYTHONUNBUFFERED environment variable is set), never with a line-buffered stream. One example where the current behavior is unacceptable is when running `hg pull https://www.mercurial-scm.org/repo/hg` on Python 3, where the line "pulling from https://www.mercurial-scm.org/repo/hg" does not appear on the terminal before the hg process blocks while waiting for the server. Various approaches to fix this problem are possible, including: 1. Weaken the contract of procutil.stdout to not give any guarantees about buffering behavior. In this case, users of procutil.stdout need to be changed to do enough flushes. In particular, 1. either ui must insert enough flushes for ui.write() and friends, or 2. ui.write() and friends get split into flushing and fully buffered methods, or 3. users of ui.write() and friends must flush explicitly. 2. Make stdout unbuffered. 3. Make stdout line-buffered. Since Python 3 does not natively support that for binary streams, we must implement it ourselves. (2.) is problematic because using unbuffered I/O changes the performance characteristics significantly compared to line-buffered (which is used on Python 2) and this would be a regression. (1.2.) and (1.3) are a substantial amount of work. It’s unclear whether the added complexity would be justified, given that raw performance doesn’t matter that much when writing to a terminal much faster than the user could read it. (1.1.) pushes complexity into the ui class instead of separating the concern of how stdout is buffered. Other users of procutil.stdout would still need to take care of the flushes. This patch implements (3.). The general performance considerations are very similar to (1.1.). The extra method invocation and method forwarding add a little more overhead if the class is used. In exchange, it doesn’t add overhead if not used. For the benchmarks, I compared the previous implementation (incorrect on Python 3), (1.1.), (3.) and (2.). The command was chosen so that the streams were configured as if they were writing to a TTY, but actually write to a pager, which is also the default: HGRCPATH=/dev/null python3 ./hg --cwd ~/vcs/mozilla-central --time --pager yes --config pager.pager='cat > /dev/null' status --all previous: time: real 7.880 secs (user 7.290+0.050 sys 0.580+0.170) time: real 7.830 secs (user 7.220+0.070 sys 0.590+0.140) time: real 7.800 secs (user 7.210+0.050 sys 0.570+0.170) (1.1.) using Yuya Nishihara’s patch: time: real 9.860 secs (user 8.670+0.350 sys 1.160+0.830) time: real 9.540 secs (user 8.430+0.370 sys 1.100+0.770) time: real 9.830 secs (user 8.630+0.370 sys 1.180+0.840) (3.) using this patch: time: real 9.580 secs (user 8.480+0.350 sys 1.090+0.770) time: real 9.670 secs (user 8.480+0.330 sys 1.170+0.860) time: real 9.640 secs (user 8.500+0.350 sys 1.130+0.810) (2.) using a previous patch by me: time: real 10.480 secs (user 8.850+0.720 sys 1.590+1.500) time: real 10.490 secs (user 8.750+0.750 sys 1.710+1.470) time: real 10.240 secs (user 8.600+0.700 sys 1.590+1.510) As expected, there’s no difference on Python 2, as exactly the same code paths are used: previous: time: real 6.950 secs (user 5.870+0.330 sys 1.070+0.770) time: real 7.040 secs (user 6.040+0.360 sys 0.980+0.750) time: real 7.070 secs (user 5.950+0.360 sys 1.100+0.760) this patch: time: real 7.010 secs (user 5.900+0.390 sys 1.070+0.730) time: real 7.000 secs (user 5.850+0.350 sys 1.120+0.760) time: real 7.000 secs (user 5.790+0.380 sys 1.170+0.710)

File last commit:

r45022:07d9fd60 default
r45477:f9734b2d default
Show More
files.rs
384 lines | 11.2 KiB | application/rls-services+xml | RustLexer
// files.rs
//
// Copyright 2019
// Raphaël Gomès <rgomes@octobus.net>,
// Yuya Nishihara <yuya@tcha.org>
//
// This software may be used and distributed according to the terms of the
// GNU General Public License version 2 or any later version.
//! Functions for fiddling with files.
use crate::utils::{
hg_path::{path_to_hg_path_buf, HgPath, HgPathBuf, HgPathError},
path_auditor::PathAuditor,
replace_slice,
};
use lazy_static::lazy_static;
use same_file::is_same_file;
use std::borrow::ToOwned;
use std::fs::Metadata;
use std::iter::FusedIterator;
use std::ops::Deref;
use std::path::{Path, PathBuf};
pub fn get_path_from_bytes(bytes: &[u8]) -> &Path {
let os_str;
#[cfg(unix)]
{
use std::os::unix::ffi::OsStrExt;
os_str = std::ffi::OsStr::from_bytes(bytes);
}
// TODO Handle other platforms
// TODO: convert from WTF8 to Windows MBCS (ANSI encoding).
// Perhaps, the return type would have to be Result<PathBuf>.
Path::new(os_str)
}
// TODO: need to convert from WTF8 to MBCS bytes on Windows.
// that's why Vec<u8> is returned.
#[cfg(unix)]
pub fn get_bytes_from_path(path: impl AsRef<Path>) -> Vec<u8> {
use std::os::unix::ffi::OsStrExt;
path.as_ref().as_os_str().as_bytes().to_vec()
}
/// An iterator over repository path yielding itself and its ancestors.
#[derive(Copy, Clone, Debug)]
pub struct Ancestors<'a> {
next: Option<&'a HgPath>,
}
impl<'a> Iterator for Ancestors<'a> {
type Item = &'a HgPath;
fn next(&mut self) -> Option<Self::Item> {
let next = self.next;
self.next = match self.next {
Some(s) if s.is_empty() => None,
Some(s) => {
let p = s.bytes().rposition(|c| *c == b'/').unwrap_or(0);
Some(HgPath::new(&s.as_bytes()[..p]))
}
None => None,
};
next
}
}
impl<'a> FusedIterator for Ancestors<'a> {}
/// An iterator over repository path yielding itself and its ancestors.
#[derive(Copy, Clone, Debug)]
pub(crate) struct AncestorsWithBase<'a> {
next: Option<(&'a HgPath, &'a HgPath)>,
}
impl<'a> Iterator for AncestorsWithBase<'a> {
type Item = (&'a HgPath, &'a HgPath);
fn next(&mut self) -> Option<Self::Item> {
let next = self.next;
self.next = match self.next {
Some((s, _)) if s.is_empty() => None,
Some((s, _)) => Some(s.split_filename()),
None => None,
};
next
}
}
impl<'a> FusedIterator for AncestorsWithBase<'a> {}
/// Returns an iterator yielding ancestor directories of the given repository
/// path.
///
/// The path is separated by '/', and must not start with '/'.
///
/// The path itself isn't included unless it is b"" (meaning the root
/// directory.)
pub fn find_dirs<'a>(path: &'a HgPath) -> Ancestors<'a> {
let mut dirs = Ancestors { next: Some(path) };
if !path.is_empty() {
dirs.next(); // skip itself
}
dirs
}
/// Returns an iterator yielding ancestor directories of the given repository
/// path.
///
/// The path is separated by '/', and must not start with '/'.
///
/// The path itself isn't included unless it is b"" (meaning the root
/// directory.)
pub(crate) fn find_dirs_with_base<'a>(
path: &'a HgPath,
) -> AncestorsWithBase<'a> {
let mut dirs = AncestorsWithBase {
next: Some((path, HgPath::new(b""))),
};
if !path.is_empty() {
dirs.next(); // skip itself
}
dirs
}
/// TODO more than ASCII?
pub fn normalize_case(path: &HgPath) -> HgPathBuf {
#[cfg(windows)] // NTFS compares via upper()
return path.to_ascii_uppercase();
#[cfg(unix)]
path.to_ascii_lowercase()
}
lazy_static! {
static ref IGNORED_CHARS: Vec<Vec<u8>> = {
[
0x200c, 0x200d, 0x200e, 0x200f, 0x202a, 0x202b, 0x202c, 0x202d,
0x202e, 0x206a, 0x206b, 0x206c, 0x206d, 0x206e, 0x206f, 0xfeff,
]
.iter()
.map(|code| {
std::char::from_u32(*code)
.unwrap()
.encode_utf8(&mut [0; 3])
.bytes()
.collect()
})
.collect()
};
}
fn hfs_ignore_clean(bytes: &[u8]) -> Vec<u8> {
let mut buf = bytes.to_owned();
let needs_escaping = bytes.iter().any(|b| *b == b'\xe2' || *b == b'\xef');
if needs_escaping {
for forbidden in IGNORED_CHARS.iter() {
replace_slice(&mut buf, forbidden, &[])
}
buf
} else {
buf
}
}
pub fn lower_clean(bytes: &[u8]) -> Vec<u8> {
hfs_ignore_clean(&bytes.to_ascii_lowercase())
}
#[derive(Eq, PartialEq, Ord, PartialOrd, Copy, Clone)]
pub struct HgMetadata {
pub st_dev: u64,
pub st_mode: u32,
pub st_nlink: u64,
pub st_size: u64,
pub st_mtime: i64,
pub st_ctime: i64,
}
// TODO support other plaforms
#[cfg(unix)]
impl HgMetadata {
pub fn from_metadata(metadata: Metadata) -> Self {
use std::os::unix::fs::MetadataExt;
Self {
st_dev: metadata.dev(),
st_mode: metadata.mode(),
st_nlink: metadata.nlink(),
st_size: metadata.size(),
st_mtime: metadata.mtime(),
st_ctime: metadata.ctime(),
}
}
}
/// Returns the canonical path of `name`, given `cwd` and `root`
pub fn canonical_path(
root: impl AsRef<Path>,
cwd: impl AsRef<Path>,
name: impl AsRef<Path>,
) -> Result<PathBuf, HgPathError> {
// TODO add missing normalization for other platforms
let root = root.as_ref();
let cwd = cwd.as_ref();
let name = name.as_ref();
let name = if !name.is_absolute() {
root.join(&cwd).join(&name)
} else {
name.to_owned()
};
let auditor = PathAuditor::new(&root);
if name != root && name.starts_with(&root) {
let name = name.strip_prefix(&root).unwrap();
auditor.audit_path(path_to_hg_path_buf(name)?)?;
return Ok(name.to_owned());
} else if name == root {
return Ok("".into());
} else {
// Determine whether `name' is in the hierarchy at or beneath `root',
// by iterating name=name.parent() until it returns `None` (can't
// check name == '/', because that doesn't work on windows).
let mut name = name.deref();
let original_name = name.to_owned();
loop {
let same = is_same_file(&name, &root).unwrap_or(false);
if same {
if name == original_name {
// `name` was actually the same as root (maybe a symlink)
return Ok("".into());
}
// `name` is a symlink to root, so `original_name` is under
// root
let rel_path = original_name.strip_prefix(&name).unwrap();
auditor.audit_path(path_to_hg_path_buf(&rel_path)?)?;
return Ok(rel_path.to_owned());
}
name = match name.parent() {
None => break,
Some(p) => p,
};
}
// TODO hint to the user about using --cwd
// Bubble up the responsibility to Python for now
Err(HgPathError::NotUnderRoot {
path: original_name.to_owned(),
root: root.to_owned(),
})
}
}
#[cfg(test)]
mod tests {
use super::*;
use pretty_assertions::assert_eq;
#[test]
fn find_dirs_some() {
let mut dirs = super::find_dirs(HgPath::new(b"foo/bar/baz"));
assert_eq!(dirs.next(), Some(HgPath::new(b"foo/bar")));
assert_eq!(dirs.next(), Some(HgPath::new(b"foo")));
assert_eq!(dirs.next(), Some(HgPath::new(b"")));
assert_eq!(dirs.next(), None);
assert_eq!(dirs.next(), None);
}
#[test]
fn find_dirs_empty() {
// looks weird, but mercurial.pathutil.finddirs(b"") yields b""
let mut dirs = super::find_dirs(HgPath::new(b""));
assert_eq!(dirs.next(), Some(HgPath::new(b"")));
assert_eq!(dirs.next(), None);
assert_eq!(dirs.next(), None);
}
#[test]
fn test_find_dirs_with_base_some() {
let mut dirs = super::find_dirs_with_base(HgPath::new(b"foo/bar/baz"));
assert_eq!(
dirs.next(),
Some((HgPath::new(b"foo/bar"), HgPath::new(b"baz")))
);
assert_eq!(
dirs.next(),
Some((HgPath::new(b"foo"), HgPath::new(b"bar")))
);
assert_eq!(dirs.next(), Some((HgPath::new(b""), HgPath::new(b"foo"))));
assert_eq!(dirs.next(), None);
assert_eq!(dirs.next(), None);
}
#[test]
fn test_find_dirs_with_base_empty() {
let mut dirs = super::find_dirs_with_base(HgPath::new(b""));
assert_eq!(dirs.next(), Some((HgPath::new(b""), HgPath::new(b""))));
assert_eq!(dirs.next(), None);
assert_eq!(dirs.next(), None);
}
#[test]
fn test_canonical_path() {
let root = Path::new("/repo");
let cwd = Path::new("/dir");
let name = Path::new("filename");
assert_eq!(
canonical_path(root, cwd, name),
Err(HgPathError::NotUnderRoot {
path: PathBuf::from("/dir/filename"),
root: root.to_path_buf()
})
);
let root = Path::new("/repo");
let cwd = Path::new("/");
let name = Path::new("filename");
assert_eq!(
canonical_path(root, cwd, name),
Err(HgPathError::NotUnderRoot {
path: PathBuf::from("/filename"),
root: root.to_path_buf()
})
);
let root = Path::new("/repo");
let cwd = Path::new("/");
let name = Path::new("repo/filename");
assert_eq!(
canonical_path(root, cwd, name),
Ok(PathBuf::from("filename"))
);
let root = Path::new("/repo");
let cwd = Path::new("/repo");
let name = Path::new("filename");
assert_eq!(
canonical_path(root, cwd, name),
Ok(PathBuf::from("filename"))
);
let root = Path::new("/repo");
let cwd = Path::new("/repo/subdir");
let name = Path::new("filename");
assert_eq!(
canonical_path(root, cwd, name),
Ok(PathBuf::from("subdir/filename"))
);
}
#[test]
fn test_canonical_path_not_rooted() {
use std::fs::create_dir;
use tempfile::tempdir;
let base_dir = tempdir().unwrap();
let base_dir_path = base_dir.path();
let beneath_repo = base_dir_path.join("a");
let root = base_dir_path.join("a/b");
let out_of_repo = base_dir_path.join("c");
let under_repo_symlink = out_of_repo.join("d");
create_dir(&beneath_repo).unwrap();
create_dir(&root).unwrap();
// TODO make portable
std::os::unix::fs::symlink(&root, &out_of_repo).unwrap();
assert_eq!(
canonical_path(&root, Path::new(""), out_of_repo),
Ok(PathBuf::from(""))
);
assert_eq!(
canonical_path(&root, Path::new(""), &beneath_repo),
Err(HgPathError::NotUnderRoot {
path: beneath_repo.to_owned(),
root: root.to_owned()
})
);
assert_eq!(
canonical_path(&root, Path::new(""), &under_repo_symlink),
Ok(PathBuf::from("d"))
);
}
}