From 26a029d407be480d791972afb5975cf62c9360a6 Mon Sep 17 00:00:00 2001 From: Daniel Baumann Date: Fri, 19 Apr 2024 02:47:55 +0200 Subject: Adding upstream version 124.0.1. Signed-off-by: Daniel Baumann --- servo/components/style/str.rs | 181 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 181 insertions(+) create mode 100644 servo/components/style/str.rs (limited to 'servo/components/style/str.rs') diff --git a/servo/components/style/str.rs b/servo/components/style/str.rs new file mode 100644 index 0000000000..9badcdf413 --- /dev/null +++ b/servo/components/style/str.rs @@ -0,0 +1,181 @@ +/* This Source Code Form is subject to the terms of the Mozilla Public + * License, v. 2.0. If a copy of the MPL was not distributed with this + * file, You can obtain one at https://mozilla.org/MPL/2.0/. */ + +//! String utils for attributes and similar stuff. + +#![deny(missing_docs)] + +use num_traits::ToPrimitive; +use std::borrow::Cow; +use std::convert::AsRef; +use std::iter::{Filter, Peekable}; +use std::str::Split; + +/// A static slice of characters. +pub type StaticCharVec = &'static [char]; + +/// A static slice of `str`s. +pub type StaticStringVec = &'static [&'static str]; + +/// A "space character" according to: +/// +/// +pub static HTML_SPACE_CHARACTERS: StaticCharVec = + &['\u{0020}', '\u{0009}', '\u{000a}', '\u{000c}', '\u{000d}']; + +/// Whether a character is a HTML whitespace character. +#[inline] +pub fn char_is_whitespace(c: char) -> bool { + HTML_SPACE_CHARACTERS.contains(&c) +} + +/// Whether all the string is HTML whitespace. +#[inline] +pub fn is_whitespace(s: &str) -> bool { + s.chars().all(char_is_whitespace) +} + +#[inline] +fn not_empty(&split: &&str) -> bool { + !split.is_empty() +} + +/// Split a string on HTML whitespace. +#[inline] +pub fn split_html_space_chars<'a>( + s: &'a str, +) -> Filter, fn(&&str) -> bool> { + s.split(HTML_SPACE_CHARACTERS) + .filter(not_empty as fn(&&str) -> bool) +} + +/// Split a string on commas. +#[inline] +pub fn split_commas<'a>(s: &'a str) -> Filter, fn(&&str) -> bool> { + s.split(',').filter(not_empty as fn(&&str) -> bool) +} + +/// Character is ascii digit +pub fn is_ascii_digit(c: &char) -> bool { + match *c { + '0'..='9' => true, + _ => false, + } +} + +fn is_decimal_point(c: char) -> bool { + c == '.' +} + +fn is_exponent_char(c: char) -> bool { + match c { + 'e' | 'E' => true, + _ => false, + } +} + +/// Read a set of ascii digits and read them into a number. +pub fn read_numbers>(mut iter: Peekable) -> (Option, usize) { + match iter.peek() { + Some(c) if is_ascii_digit(c) => (), + _ => return (None, 0), + } + + iter.take_while(is_ascii_digit) + .map(|d| d as i64 - '0' as i64) + .fold((Some(0i64), 0), |accumulator, d| { + let digits = accumulator + .0 + .and_then(|accumulator| accumulator.checked_mul(10)) + .and_then(|accumulator| accumulator.checked_add(d)); + (digits, accumulator.1 + 1) + }) +} + +/// Read a decimal fraction. +pub fn read_fraction>( + mut iter: Peekable, + mut divisor: f64, + value: f64, +) -> (f64, usize) { + match iter.peek() { + Some(c) if is_decimal_point(*c) => (), + _ => return (value, 0), + } + iter.next(); + + iter.take_while(is_ascii_digit) + .map(|d| d as i64 - '0' as i64) + .fold((value, 1), |accumulator, d| { + divisor *= 10f64; + (accumulator.0 + d as f64 / divisor, accumulator.1 + 1) + }) +} + +/// Reads an exponent from an iterator over chars, for example `e100`. +pub fn read_exponent>(mut iter: Peekable) -> Option { + match iter.peek() { + Some(c) if is_exponent_char(*c) => (), + _ => return None, + } + iter.next(); + + match iter.peek() { + None => None, + Some(&'-') => { + iter.next(); + read_numbers(iter).0.map(|exp| -exp.to_i32().unwrap_or(0)) + }, + Some(&'+') => { + iter.next(); + read_numbers(iter).0.map(|exp| exp.to_i32().unwrap_or(0)) + }, + Some(_) => read_numbers(iter).0.map(|exp| exp.to_i32().unwrap_or(0)), + } +} + +/// Join a set of strings with a given delimiter `join`. +pub fn str_join(strs: I, join: &str) -> String +where + I: IntoIterator, + T: AsRef, +{ + strs.into_iter() + .enumerate() + .fold(String::new(), |mut acc, (i, s)| { + if i > 0 { + acc.push_str(join); + } + acc.push_str(s.as_ref()); + acc + }) +} + +/// Returns true if a given string has a given prefix with case-insensitive match. +pub fn starts_with_ignore_ascii_case(string: &str, prefix: &str) -> bool { + string.len() >= prefix.len() && + string.as_bytes()[0..prefix.len()].eq_ignore_ascii_case(prefix.as_bytes()) +} + +/// Returns an ascii lowercase version of a string, only allocating if needed. +pub fn string_as_ascii_lowercase<'a>(input: &'a str) -> Cow<'a, str> { + if input.bytes().any(|c| matches!(c, b'A'..=b'Z')) { + input.to_ascii_lowercase().into() + } else { + // Already ascii lowercase. + Cow::Borrowed(input) + } +} + +/// To avoid accidentally instantiating multiple monomorphizations of large +/// serialization routines, we define explicit concrete types and require +/// them in those routines. This avoids accidental mixing of String and +/// nsACString arguments in Gecko, which would cause code size to blow up. +#[cfg(feature = "gecko")] +pub type CssStringWriter = ::nsstring::nsACString; + +/// String type that coerces to CssStringWriter, used when serialization code +/// needs to allocate a temporary string. +#[cfg(feature = "gecko")] +pub type CssString = ::nsstring::nsCString; -- cgit v1.2.3