twin/ripgrep
1
0
Fork 0
mirror of https://github.com/BurntSushi/ripgrep.git synced 2025-08-14 19:55:47 -07:00
Code Issues Packages Projects Releases Wiki Activity

deps: update to pcre2 0.2.4

Browse Source 
0.2.4 updates to PCRE2 10.42 and has a few other nice changes. For
example, when `utf` is enabled, the crate will always set the
PCRE2_MATCH_INVALID_UTF option. That means we no longer need to do
transcoding or UTF-8 validity checks.

Because of this, we actually get to remove one of the two uses of
`unsafe` in ripgrep's `main` program.

(This also updates a couple other dependencies for convenience.)
This commit is contained in:
Andrew Gallant
2023-06-18 13:07:48 -04:00
parent a775b493fd
commit 4b8aa91ae5
4 changed files with 23 additions and 57 deletions
Download Patch File Download Diff File Expand all files Collapse all files

33
crates/core/args.rs
View File

@@ -472,24 +472,6 @@ enum EncodingMode {
Disabled,
}
impl EncodingMode {
/// Checks if an explicit encoding has been set. Returns false for
/// automatic BOM sniffing and no sniffing.
///
/// This is only used to determine whether PCRE2 needs to have its own
/// UTF-8 checking enabled. If we have an explicit encoding set, then
/// we're always guaranteed to get UTF-8, so we can disable PCRE2's check.
/// Otherwise, we have no such guarantee, and must enable PCRE2' UTF-8
/// check.
#[cfg(feature = "pcre2")]
fn has_explicit_encoding(&self) -> bool {
match self {
EncodingMode::Some(_) => true,
_ => false,
}
}
}
impl ArgMatches {
/// Create an ArgMatches from clap's parse result.
fn new(clap_matches: clap::ArgMatches<'static>) -> ArgMatches {
@@ -732,14 +714,6 @@ impl ArgMatches {
}
if self.unicode() {
builder.utf(true).ucp(true);
if self.encoding()?.has_explicit_encoding() {
// SAFETY: If an encoding was specified, then we're guaranteed
// to get valid UTF-8, so we can disable PCRE2's UTF checking.
// (Feeding invalid UTF-8 to PCRE2 is undefined behavior.)
unsafe {
builder.disable_utf_check();
}
}
}
if self.is_present("multiline") {
builder.dotall(self.is_present("multiline-dotall"));
@@ -1080,7 +1054,6 @@ impl ArgMatches {
}
let label = match self.value_of_lossy("encoding") {
None if self.pcre2_unicode() => "utf-8".to_string(),
None => return Ok(EncodingMode::Auto),
Some(label) => label,
};
@@ -1641,12 +1614,6 @@ impl ArgMatches {
!(self.is_present("no-unicode") || self.is_present("no-pcre2-unicode"))
}
/// Returns true if and only if PCRE2 is enabled and its Unicode mode is
/// enabled.
fn pcre2_unicode(&self) -> bool {
self.is_present("pcre2") && self.unicode()
}
/// Returns true if and only if file names containing each match should
/// be emitted.
fn with_filename(&self, paths: &[PathBuf]) -> bool {