diff options
author | Andrew Gallant <jamslam@gmail.com> | 2017-01-08 10:27:30 -0500 |
---|---|---|
committer | Andrew Gallant <jamslam@gmail.com> | 2017-01-09 16:12:08 -0500 |
commit | 461e0c4e33b38532545af4f97ec92bd367963017 (patch) | |
tree | 26bbdd3340c1d9b29f90ac7856b18d5d4bb3ebe0 /src/args.rs | |
parent | 82df3b768587e2839db849e6eed2db518a728778 (diff) |
Don't search stdout redirected file.
When running ripgrep like this:
rg foo > output
we must be careful not to search `output` since ripgrep is actively writing
to it. Searching it can cause massive blowups where the file grows without
bound.
While this is conceptually easy to fix (check the inode of the redirection
and the inode of the file you're about to search), there are a few problems
with it.
First, inodes are a Unix thing, so we need a Windows specific solution to
this as well. To resolve this concern, I created a new crate, `same-file`,
which provides a cross platform abstraction.
Second, stat'ing every file is costly. This is not avoidable on Windows,
but on Unix, we can get the inode number directly from directory traversal.
However, this information wasn't exposed, but now it is (through both the
ignore and walkdir crates).
Fixes #286
Diffstat (limited to 'src/args.rs')
-rw-r--r-- | src/args.rs | 36 |
1 files changed, 36 insertions, 0 deletions
diff --git a/src/args.rs b/src/args.rs index 6aec396b..29a7fa81 100644 --- a/src/args.rs +++ b/src/args.rs @@ -15,6 +15,7 @@ use grep::{Grep, GrepBuilder}; use log; use num_cpus; use regex; +use same_file; use termcolor; use app; @@ -65,6 +66,7 @@ pub struct Args { quiet_matched: QuietMatched, replace: Option<Vec<u8>>, sort_files: bool, + stdout_handle: Option<same_file::Handle>, text: bool, threads: usize, type_list: bool, @@ -182,6 +184,17 @@ impl Args { termcolor::Stdout::new(self.color_choice) } + /// Returns a handle to stdout for filtering search. + /// + /// A handle is returned if and only if ripgrep's stdout is being + /// redirected to a file. The handle returned corresponds to that file. + /// + /// This can be used to ensure that we do not attempt to search a file + /// that ripgrep is writing to. + pub fn stdout_handle(&self) -> Option<&same_file::Handle> { + self.stdout_handle.as_ref() + } + /// Create a new buffer writer for multi-threaded searching with color /// support. pub fn buffer_writer(&self) -> termcolor::BufferWriter { @@ -338,6 +351,7 @@ impl<'a> ArgMatches<'a> { quiet_matched: QuietMatched::new(quiet), replace: self.replace(), sort_files: self.is_present("sort-files"), + stdout_handle: self.stdout_handle(), text: self.text(), threads: try!(self.threads()), type_list: self.is_present("type-list"), @@ -518,6 +532,28 @@ impl<'a> ArgMatches<'a> { } } + /// Returns a handle to stdout for filtering search. + /// + /// A handle is returned if and only if ripgrep's stdout is being + /// redirected to a file. The handle returned corresponds to that file. + /// + /// This can be used to ensure that we do not attempt to search a file + /// that ripgrep is writing to. + fn stdout_handle(&self) -> Option<same_file::Handle> { + let h = match same_file::Handle::stdout() { + Err(_) => return None, + Ok(h) => h, + }; + let md = match h.as_file().metadata() { + Err(_) => return None, + Ok(md) => md, + }; + if !md.is_file() { + return None; + } + Some(h) + } + /// Returns true if and only if memory map searching should be tried. /// /// `paths` should be a slice of all top-level file paths that ripgrep |