mikros/rust
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

libregex: Remove all uses of ~str from libregex

Browse Source
This commit is contained in:
Patrick Walton 2014-05-12 21:12:50 -07:00
parent 62cf95f674
commit ec5911b2a6
5 changed files with 50 additions and 45 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
src/libregex/compile.rs
View File

@ -83,12 +83,12 @@ pub struct Program {
/// If the regular expression requires a literal prefix in order to have a
/// match, that prefix is stored here. (It's used in the VM to implement
/// an optimization.)
pub prefix: ~str,
pub prefix: StrBuf,
}
impl Program {
/// Compiles a Regex given its AST.
pub fn new(ast: parse::Ast) -> (Program, Vec<Option<~str>>) {
pub fn new(ast: parse::Ast) -> (Program, Vec<Option<StrBuf>>) {
let mut c = Compiler {
insts: Vec::with_capacity(100),
names: Vec::with_capacity(10),
@ -113,7 +113,7 @@ impl Program {
let Compiler { insts, names } = c;
let prog = Program {
insts: insts,
prefix: pre.into_owned(),
prefix: pre,
};
(prog, names)
}
@ -135,7 +135,7 @@ impl Program {
struct Compiler<'r> {
insts: Vec<Inst>,
names: Vec<Option<~str>>,
names: Vec<Option<StrBuf>>,
}
// The compiler implemented here is extremely simple. Most of the complexity

34
src/libregex/parse.rs
View File

@ -32,7 +32,7 @@ pub struct Error {
/// The *approximate* character index of where the error occurred.
pub pos: uint,
/// A message describing the error.
pub msg: ~str,
pub msg: StrBuf,
}
impl fmt::Show for Error {
@ -59,7 +59,7 @@ pub enum Ast {
Begin(Flags),
End(Flags),
WordBoundary(Flags),
Capture(uint, Option<~str>, Box<Ast>),
Capture(uint, Option<StrBuf>, Box<Ast>),
// Represent concatenation as a flat vector to avoid blowing the
// stack in the compiler.
Cat(Vec<Ast>),
@ -104,7 +104,7 @@ impl Greed {
#[deriving(Show)]
enum BuildAst {
Ast(Ast),
Paren(Flags, uint, ~str), // '('
Paren(Flags, uint, StrBuf), // '('
Bar, // '|'
}
@ -131,7 +131,7 @@ impl BuildAst {
}
}
fn capture_name(&self) -> Option<~str> {
fn capture_name(&self) -> Option<StrBuf> {
match *self {
Paren(_, 0, _) => None,
Paren(_, _, ref name) => {
@ -185,7 +185,7 @@ struct Parser<'a> {
// opening a capture group).
caps: uint,
// A set of all capture group names used only to detect duplicates.
names: Vec<~str>,
names: Vec<StrBuf>,
}
pub fn parse(s: &str) -> Result<Ast, Error> {
@ -222,7 +222,7 @@ impl<'a> Parser<'a> {
self.caps += 1;
self.stack.push(Paren(self.flags,
self.caps,
"".to_owned()))
"".to_strbuf()))
}
}
')' => {
@ -470,7 +470,7 @@ impl<'a> Parser<'a> {
FLAG_EMPTY
};
let name = self.slice(name_start, closer - 1);
match find_class(ASCII_CLASSES, name) {
match find_class(ASCII_CLASSES, name.as_slice()) {
None => None,
Some(ranges) => {
self.chari = closer;
@ -611,7 +611,7 @@ impl<'a> Parser<'a> {
// character).
fn parse_unicode_name(&mut self) -> Result<Ast, Error> {
let negated = if self.cur() == 'P' { FLAG_NEGATED } else { FLAG_EMPTY };
let mut name: ~str;
let mut name: StrBuf;
if self.peek_is(1, '{') {
try!(self.expect('{'))
let closer =
@ -633,7 +633,7 @@ impl<'a> Parser<'a> {
name = self.slice(self.chari + 1, self.chari + 2);
self.chari += 1;
}
match find_class(UNICODE_CLASSES, name) {
match find_class(UNICODE_CLASSES, name.as_slice()) {
None => return self.err(format!(
"Could not find Unicode class '{}'", name)),
Some(ranges) => {
@ -657,7 +657,7 @@ impl<'a> Parser<'a> {
}
}
let s = self.slice(start, end);
match num::from_str_radix::<u32>(s, 8) {
match num::from_str_radix::<u32>(s.as_slice(), 8) {
Some(n) => Ok(Literal(try!(self.char_from_u32(n)), FLAG_EMPTY)),
None => self.err(format!(
"Could not parse '{}' as octal number.", s)),
@ -679,7 +679,7 @@ impl<'a> Parser<'a> {
Some(i) => i,
};
self.chari = closer;
self.parse_hex_digits(self.slice(start, closer))
self.parse_hex_digits(self.slice(start, closer).as_slice())
}
// Parses a two-digit hex number.
@ -690,7 +690,7 @@ impl<'a> Parser<'a> {
let (start, end) = (self.chari, self.chari + 2);
let bad = self.slice(start - 2, self.chars.len());
try!(self.noteof(format!("Invalid hex escape sequence '{}'", bad)))
self.parse_hex_digits(self.slice(start, end))
self.parse_hex_digits(self.slice(start, end).as_slice())
}
// Parses `s` as a hexadecimal number.
@ -717,7 +717,7 @@ impl<'a> Parser<'a> {
return self.err("Capture names must have at least 1 character.")
}
let name = self.slice(self.chari, closer);
if !name.chars().all(is_valid_cap) {
if !name.as_slice().chars().all(is_valid_cap) {
return self.err(
"Capture names can only have underscores, letters and digits.")
}
@ -771,7 +771,7 @@ impl<'a> Parser<'a> {
}
if self.cur() == ':' {
// Save the old flags with the opening paren.
self.stack.push(Paren(self.flags, 0, "".to_owned()));
self.stack.push(Paren(self.flags, 0, "".to_strbuf()));
}
self.flags = flags;
return Ok(())
@ -892,7 +892,7 @@ impl<'a> Parser<'a> {
fn err<T>(&self, msg: &str) -> Result<T, Error> {
Err(Error {
pos: self.chari,
msg: msg.to_owned(),
msg: msg.to_strbuf(),
})
}
@ -911,8 +911,8 @@ impl<'a> Parser<'a> {
*self.chars.get(self.chari)
}
fn slice(&self, start: uint, end: uint) -> ~str {
str::from_chars(self.chars.as_slice().slice(start, end))
fn slice(&self, start: uint, end: uint) -> StrBuf {
str::from_chars(self.chars.as_slice().slice(start, end)).to_strbuf()
}
}

35
src/libregex/re.rs
View File

@ -20,7 +20,7 @@ use vm::{CaptureLocs, MatchKind, Exists, Location, Submatches};
/// Escapes all regular expression meta characters in `text` so that it may be
/// safely used in a regular expression as a literal string.
pub fn quote(text: &str) -> ~str {
pub fn quote(text: &str) -> StrBuf {
let mut quoted = StrBuf::with_capacity(text.len());
for c in text.chars() {
if parse::is_punct(c) {
@ -28,7 +28,7 @@ pub fn quote(text: &str) -> ~str {
}
quoted.push_char(c);
}
quoted.into_owned()
quoted
}
/// Tests if the given regular expression matches somewhere in the text given.
@ -107,9 +107,9 @@ pub struct Regex {
/// See the comments for the `program` module in `lib.rs` for a more
/// detailed explanation for what `regex!` requires.
#[doc(hidden)]
pub original: ~str,
pub original: StrBuf,
#[doc(hidden)]
pub names: Vec<Option<~str>>,
pub names: Vec<Option<StrBuf>>,
#[doc(hidden)]
pub p: MaybeNative,
}
@ -146,7 +146,10 @@ impl Regex {
pub fn new(re: &str) -> Result<Regex, parse::Error> {
let ast = try!(parse::parse(re));
let (prog, names) = Program::new(ast);
Ok(Regex { original: re.to_owned(), names: names, p: Dynamic(prog) })
Ok(Regex {
original: re.to_strbuf(),
names: names, p: Dynamic(prog),
})
}
/// Returns true if and only if the regex matches the string given.
@ -404,7 +407,7 @@ impl Regex {
/// ```
///
/// But anything satisfying the `Replacer` trait will work. For example,
/// a closure of type `|&Captures| -> ~str` provides direct access to the
/// a closure of type `|&Captures| -> StrBuf` provides direct access to the
/// captures corresponding to a match. This allows one to access
/// submatches easily:
///
@ -414,7 +417,7 @@ impl Regex {
/// # use regex::Captures; fn main() {
/// let re = regex!(r"([^,\s]+),\s+(\S+)");
/// let result = re.replace("Springsteen, Bruce", |caps: &Captures| {
/// format!("{} {}", caps.at(2), caps.at(1))
/// format_strbuf!("{} {}", caps.at(2), caps.at(1))
/// });
/// assert_eq!(result.as_slice(), "Bruce Springsteen");
/// # }
@ -526,7 +529,7 @@ impl<'t> Replacer for &'t str {
}
}
impl<'a> Replacer for |&Captures|: 'a -> ~str {
impl<'a> Replacer for |&Captures|: 'a -> StrBuf {
fn reg_replace<'r>(&'r mut self, caps: &Captures) -> MaybeOwned<'r> {
Owned((*self)(caps).into_owned())
}
@ -605,7 +608,7 @@ impl<'r, 't> Iterator<&'t str> for RegexSplitsN<'r, 't> {
pub struct Captures<'t> {
text: &'t str,
locs: CaptureLocs,
named: Option<HashMap<~str, uint>>,
named: Option<HashMap<StrBuf, uint>>,
}
impl<'t> Captures<'t> {
@ -624,7 +627,7 @@ impl<'t> Captures<'t> {
match name {
&None => {},
&Some(ref name) => {
named.insert(name.to_owned(), i);
named.insert(name.to_strbuf(), i);
}
}
}
@ -707,12 +710,14 @@ impl<'t> Captures<'t> {
// How evil can you get?
// FIXME: Don't use regexes for this. It's completely unnecessary.
let re = Regex::new(r"(^|[^$]|\b)\$(\w+)").unwrap();
let text = re.replace_all(text, |refs: &Captures| -> ~str {
let text = re.replace_all(text, |refs: &Captures| -> StrBuf {
let (pre, name) = (refs.at(1), refs.at(2));
pre + match from_str::<uint>(name) {
None => self.name(name).to_owned(),
Some(i) => self.at(i).to_owned(),
}
format_strbuf!("{}{}",
pre,
match from_str::<uint>(name.as_slice()) {
None => self.name(name).to_strbuf(),
Some(i) => self.at(i).to_strbuf(),
})
});
let re = Regex::new(r"\$\$").unwrap();
re.replace_all(text.as_slice(), NoExpand("$"))

6
src/libregex/test/bench.rs
View File

@ -140,7 +140,7 @@ macro_rules! throughput(
fn $name(b: &mut Bencher) {
let text = gen_text($size);
b.bytes = $size;
b.iter(|| if $regex.is_match(text) { fail!("match") });
b.iter(|| if $regex.is_match(text.as_slice()) { fail!("match") });
}
);
)
@ -151,7 +151,7 @@ fn medium() -> Regex { regex!("[XYZ]ABCDEFGHIJKLMNOPQRSTUVWXYZ$") }
fn hard() -> Regex { regex!("[ -~]*ABCDEFGHIJKLMNOPQRSTUVWXYZ$") }
#[allow(deprecated_owned_vector)]
fn gen_text(n: uint) -> ~str {
fn gen_text(n: uint) -> StrBuf {
let mut rng = task_rng();
let mut bytes = rng.gen_ascii_str(n).into_bytes();
for (i, b) in bytes.mut_iter().enumerate() {
@ -159,7 +159,7 @@ fn gen_text(n: uint) -> ~str {
*b = '\n' as u8
}
}
str::from_utf8(bytes).unwrap().to_owned()
str::from_utf8(bytes).unwrap().to_strbuf()
}
throughput!(easy0_32, easy0(), 32)

12
src/libregex_macros/lib.rs
View File

@ -105,8 +105,8 @@ struct NfaGen<'a> {
cx: &'a ExtCtxt<'a>,
sp: codemap::Span,
prog: Program,
names: Vec<Option<~str>>,
original: ~str,
names: Vec<Option<StrBuf>>,
original: StrBuf,
}
impl<'a> NfaGen<'a> {
@ -119,7 +119,7 @@ impl<'a> NfaGen<'a> {
|cx, name| match *name {
Some(ref name) => {
let name = name.as_slice();
quote_expr!(cx, Some($name.to_owned()))
quote_expr!(cx, Some($name.to_strbuf()))
}
None => cx.expr_none(self.sp),
}
@ -311,7 +311,7 @@ fn exec<'t>(which: ::regex::native::MatchKind, input: &'t str,
}
::regex::Regex {
original: $regex.to_owned(),
original: $regex.to_strbuf(),
names: vec!$cap_names,
p: ::regex::native::Native(exec),
}
@ -601,14 +601,14 @@ fn exec<'t>(which: ::regex::native::MatchKind, input: &'t str,
/// Looks for a single string literal and returns it.
/// Otherwise, logs an error with cx.span_err and returns None.
fn parse(cx: &mut ExtCtxt, tts: &[ast::TokenTree]) -> Option<~str> {
fn parse(cx: &mut ExtCtxt, tts: &[ast::TokenTree]) -> Option<StrBuf> {
let mut parser = parse::new_parser_from_tts(cx.parse_sess(), cx.cfg(),
Vec::from_slice(tts));
let entry = cx.expand_expr(parser.parse_expr());
let regex = match entry.node {
ast::ExprLit(lit) => {
match lit.node {
ast::LitStr(ref s, _) => s.to_str(),
ast::LitStr(ref s, _) => s.to_str().to_strbuf(),
_ => {
cx.span_err(entry.span, format!(
"expected string literal but got `{}`",