Vendor things
This commit is contained in:
parent
5deceec006
commit
977e3c17e5
19434 changed files with 10682014 additions and 0 deletions
201
third-party/vendor/data-url/src/forgiving_base64.rs
vendored
Normal file
201
third-party/vendor/data-url/src/forgiving_base64.rs
vendored
Normal file
|
|
@ -0,0 +1,201 @@
|
|||
//! <https://infra.spec.whatwg.org/#forgiving-base64-decode>
|
||||
|
||||
use alloc::vec::Vec;
|
||||
use core::fmt;
|
||||
|
||||
#[derive(Debug)]
|
||||
pub struct InvalidBase64(InvalidBase64Details);
|
||||
|
||||
impl fmt::Display for InvalidBase64 {
|
||||
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
|
||||
match self.0 {
|
||||
InvalidBase64Details::UnexpectedSymbol(code_point) => {
|
||||
write!(f, "symbol with codepoint {} not expected", code_point)
|
||||
}
|
||||
InvalidBase64Details::AlphabetSymbolAfterPadding => {
|
||||
write!(f, "alphabet symbol present after padding")
|
||||
}
|
||||
InvalidBase64Details::LoneAlphabetSymbol => write!(f, "lone alphabet symbol present"),
|
||||
InvalidBase64Details::Padding => write!(f, "incorrect padding"),
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
#[cfg(feature = "std")]
|
||||
impl std::error::Error for InvalidBase64 {}
|
||||
|
||||
#[derive(Debug)]
|
||||
enum InvalidBase64Details {
|
||||
UnexpectedSymbol(u8),
|
||||
AlphabetSymbolAfterPadding,
|
||||
LoneAlphabetSymbol,
|
||||
Padding,
|
||||
}
|
||||
|
||||
#[derive(Debug)]
|
||||
pub enum DecodeError<E> {
|
||||
InvalidBase64(InvalidBase64),
|
||||
WriteError(E),
|
||||
}
|
||||
|
||||
impl<E: fmt::Display> fmt::Display for DecodeError<E> {
|
||||
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
|
||||
match self {
|
||||
Self::InvalidBase64(inner) => write!(f, "base64 not valid: {}", inner),
|
||||
Self::WriteError(err) => write!(f, "write error: {}", err),
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
#[cfg(feature = "std")]
|
||||
impl<E: std::error::Error> std::error::Error for DecodeError<E> {}
|
||||
|
||||
impl<E> From<InvalidBase64Details> for DecodeError<E> {
|
||||
fn from(e: InvalidBase64Details) -> Self {
|
||||
DecodeError::InvalidBase64(InvalidBase64(e))
|
||||
}
|
||||
}
|
||||
|
||||
pub(crate) enum Impossible {}
|
||||
|
||||
impl From<DecodeError<Impossible>> for InvalidBase64 {
|
||||
fn from(e: DecodeError<Impossible>) -> Self {
|
||||
match e {
|
||||
DecodeError::InvalidBase64(e) => e,
|
||||
DecodeError::WriteError(e) => match e {},
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
/// `input` is assumed to be in an ASCII-compatible encoding
|
||||
pub fn decode_to_vec(input: &[u8]) -> Result<Vec<u8>, InvalidBase64> {
|
||||
let mut v = Vec::new();
|
||||
{
|
||||
let mut decoder = Decoder::new(|bytes| {
|
||||
v.extend_from_slice(bytes);
|
||||
Ok(())
|
||||
});
|
||||
decoder.feed(input)?;
|
||||
decoder.finish()?;
|
||||
}
|
||||
Ok(v)
|
||||
}
|
||||
|
||||
/// <https://infra.spec.whatwg.org/#forgiving-base64-decode>
|
||||
pub struct Decoder<F, E>
|
||||
where
|
||||
F: FnMut(&[u8]) -> Result<(), E>,
|
||||
{
|
||||
write_bytes: F,
|
||||
bit_buffer: u32,
|
||||
buffer_bit_length: u8,
|
||||
padding_symbols: u8,
|
||||
}
|
||||
|
||||
impl<F, E> Decoder<F, E>
|
||||
where
|
||||
F: FnMut(&[u8]) -> Result<(), E>,
|
||||
{
|
||||
pub fn new(write_bytes: F) -> Self {
|
||||
Self {
|
||||
write_bytes,
|
||||
bit_buffer: 0,
|
||||
buffer_bit_length: 0,
|
||||
padding_symbols: 0,
|
||||
}
|
||||
}
|
||||
|
||||
/// Feed to the decoder partial input in an ASCII-compatible encoding
|
||||
pub fn feed(&mut self, input: &[u8]) -> Result<(), DecodeError<E>> {
|
||||
for &byte in input.iter() {
|
||||
let value = BASE64_DECODE_TABLE[byte as usize];
|
||||
if value < 0 {
|
||||
// A character that’s not part of the alphabet
|
||||
|
||||
// Remove ASCII whitespace
|
||||
if matches!(byte, b' ' | b'\t' | b'\n' | b'\r' | b'\x0C') {
|
||||
continue;
|
||||
}
|
||||
|
||||
if byte == b'=' {
|
||||
self.padding_symbols = self.padding_symbols.saturating_add(1);
|
||||
continue;
|
||||
}
|
||||
|
||||
return Err(InvalidBase64Details::UnexpectedSymbol(byte).into());
|
||||
}
|
||||
if self.padding_symbols > 0 {
|
||||
return Err(InvalidBase64Details::AlphabetSymbolAfterPadding.into());
|
||||
}
|
||||
self.bit_buffer <<= 6;
|
||||
self.bit_buffer |= value as u32;
|
||||
// 18 before incrementing means we’ve just reached 24
|
||||
if self.buffer_bit_length < 18 {
|
||||
self.buffer_bit_length += 6;
|
||||
} else {
|
||||
// We’ve accumulated four times 6 bits, which equals three times 8 bits.
|
||||
let byte_buffer = [
|
||||
(self.bit_buffer >> 16) as u8,
|
||||
(self.bit_buffer >> 8) as u8,
|
||||
self.bit_buffer as u8,
|
||||
];
|
||||
(self.write_bytes)(&byte_buffer).map_err(DecodeError::WriteError)?;
|
||||
self.buffer_bit_length = 0;
|
||||
// No need to reset bit_buffer,
|
||||
// since next time we’re only gonna read relevant bits.
|
||||
}
|
||||
}
|
||||
Ok(())
|
||||
}
|
||||
|
||||
/// Call this to signal the end of the input
|
||||
pub fn finish(mut self) -> Result<(), DecodeError<E>> {
|
||||
match (self.buffer_bit_length, self.padding_symbols) {
|
||||
(0, 0) => {
|
||||
// A multiple of four of alphabet symbols, and nothing else.
|
||||
}
|
||||
(12, 2) | (12, 0) => {
|
||||
// A multiple of four of alphabet symbols, followed by two more symbols,
|
||||
// optionally followed by two padding characters (which make a total multiple of four).
|
||||
let byte_buffer = [(self.bit_buffer >> 4) as u8];
|
||||
(self.write_bytes)(&byte_buffer).map_err(DecodeError::WriteError)?;
|
||||
}
|
||||
(18, 1) | (18, 0) => {
|
||||
// A multiple of four of alphabet symbols, followed by three more symbols,
|
||||
// optionally followed by one padding character (which make a total multiple of four).
|
||||
let byte_buffer = [(self.bit_buffer >> 10) as u8, (self.bit_buffer >> 2) as u8];
|
||||
(self.write_bytes)(&byte_buffer).map_err(DecodeError::WriteError)?;
|
||||
}
|
||||
(6, _) => return Err(InvalidBase64Details::LoneAlphabetSymbol.into()),
|
||||
_ => return Err(InvalidBase64Details::Padding.into()),
|
||||
}
|
||||
Ok(())
|
||||
}
|
||||
}
|
||||
|
||||
/// Generated by `make_base64_decode_table.py` based on "Table 1: The Base 64 Alphabet"
|
||||
/// at <https://tools.ietf.org/html/rfc4648#section-4>
|
||||
///
|
||||
/// Array indices are the byte value of symbols.
|
||||
/// Array values are their positions in the base64 alphabet,
|
||||
/// or -1 for symbols not in the alphabet.
|
||||
/// The position contributes 6 bits to the decoded bytes.
|
||||
#[rustfmt::skip]
|
||||
const BASE64_DECODE_TABLE: [i8; 256] = [
|
||||
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
|
||||
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
|
||||
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, 62, -1, -1, -1, 63,
|
||||
52, 53, 54, 55, 56, 57, 58, 59, 60, 61, -1, -1, -1, -1, -1, -1,
|
||||
-1, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,
|
||||
15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, -1, -1, -1, -1, -1,
|
||||
-1, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40,
|
||||
41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, -1, -1, -1, -1, -1,
|
||||
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
|
||||
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
|
||||
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
|
||||
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
|
||||
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
|
||||
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
|
||||
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
|
||||
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
|
||||
];
|
||||
347
third-party/vendor/data-url/src/lib.rs
vendored
Normal file
347
third-party/vendor/data-url/src/lib.rs
vendored
Normal file
|
|
@ -0,0 +1,347 @@
|
|||
//! Processing of `data:` URLs according to the Fetch Standard:
|
||||
//! <https://fetch.spec.whatwg.org/#data-urls>
|
||||
//! but starting from a string rather than a parsed URL to avoid extra copies.
|
||||
//!
|
||||
//! ```rust
|
||||
//! use data_url::{DataUrl, mime};
|
||||
//!
|
||||
//! let url = DataUrl::process("data:,Hello%20World!").unwrap();
|
||||
//! let (body, fragment) = url.decode_to_vec().unwrap();
|
||||
//!
|
||||
//! assert_eq!(url.mime_type().type_, "text");
|
||||
//! assert_eq!(url.mime_type().subtype, "plain");
|
||||
//! assert_eq!(url.mime_type().get_parameter("charset"), Some("US-ASCII"));
|
||||
//! assert_eq!(body, b"Hello World!");
|
||||
//! assert!(fragment.is_none());
|
||||
//! ```
|
||||
#![no_std]
|
||||
|
||||
// For forwards compatibility
|
||||
#[cfg(feature = "std")]
|
||||
extern crate std;
|
||||
|
||||
#[macro_use]
|
||||
extern crate alloc;
|
||||
|
||||
#[cfg(not(feature = "alloc"))]
|
||||
compile_error!("the `alloc` feature must be enabled");
|
||||
|
||||
use alloc::{string::String, vec::Vec};
|
||||
use core::fmt;
|
||||
|
||||
macro_rules! require {
|
||||
($condition: expr) => {
|
||||
if !$condition {
|
||||
return None;
|
||||
}
|
||||
};
|
||||
}
|
||||
|
||||
pub mod forgiving_base64;
|
||||
pub mod mime;
|
||||
|
||||
pub struct DataUrl<'a> {
|
||||
mime_type: mime::Mime,
|
||||
base64: bool,
|
||||
encoded_body_plus_fragment: &'a str,
|
||||
}
|
||||
|
||||
#[derive(Debug)]
|
||||
pub enum DataUrlError {
|
||||
NotADataUrl,
|
||||
NoComma,
|
||||
}
|
||||
|
||||
impl fmt::Display for DataUrlError {
|
||||
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
|
||||
match self {
|
||||
Self::NotADataUrl => write!(f, "not a valid data url"),
|
||||
Self::NoComma => write!(
|
||||
f,
|
||||
"data url is missing comma delimiting attributes and body"
|
||||
),
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
#[cfg(feature = "std")]
|
||||
impl std::error::Error for DataUrlError {}
|
||||
|
||||
impl<'a> DataUrl<'a> {
|
||||
/// <https://fetch.spec.whatwg.org/#data-url-processor>
|
||||
/// but starting from a string rather than a parsed `Url`, to avoid extra string copies.
|
||||
pub fn process(input: &'a str) -> Result<Self, DataUrlError> {
|
||||
use crate::DataUrlError::*;
|
||||
|
||||
let after_colon = pretend_parse_data_url(input).ok_or(NotADataUrl)?;
|
||||
|
||||
let (from_colon_to_comma, encoded_body_plus_fragment) =
|
||||
find_comma_before_fragment(after_colon).ok_or(NoComma)?;
|
||||
|
||||
let (mime_type, base64) = parse_header(from_colon_to_comma);
|
||||
|
||||
Ok(DataUrl {
|
||||
mime_type,
|
||||
base64,
|
||||
encoded_body_plus_fragment,
|
||||
})
|
||||
}
|
||||
|
||||
pub fn mime_type(&self) -> &mime::Mime {
|
||||
&self.mime_type
|
||||
}
|
||||
|
||||
/// Streaming-decode the data URL’s body to `write_body_bytes`,
|
||||
/// and return the URL’s fragment identifier if it has one.
|
||||
pub fn decode<F, E>(
|
||||
&self,
|
||||
write_body_bytes: F,
|
||||
) -> Result<Option<FragmentIdentifier<'a>>, forgiving_base64::DecodeError<E>>
|
||||
where
|
||||
F: FnMut(&[u8]) -> Result<(), E>,
|
||||
{
|
||||
if self.base64 {
|
||||
decode_with_base64(self.encoded_body_plus_fragment, write_body_bytes)
|
||||
} else {
|
||||
decode_without_base64(self.encoded_body_plus_fragment, write_body_bytes)
|
||||
.map_err(forgiving_base64::DecodeError::WriteError)
|
||||
}
|
||||
}
|
||||
|
||||
/// Return the decoded body, and the URL’s fragment identifier if it has one.
|
||||
pub fn decode_to_vec(
|
||||
&self,
|
||||
) -> Result<(Vec<u8>, Option<FragmentIdentifier<'a>>), forgiving_base64::InvalidBase64> {
|
||||
let mut body = Vec::new();
|
||||
let fragment = self.decode(|bytes| {
|
||||
body.extend_from_slice(bytes);
|
||||
Ok(())
|
||||
})?;
|
||||
Ok((body, fragment))
|
||||
}
|
||||
}
|
||||
|
||||
/// The URL’s fragment identifier (after `#`)
|
||||
pub struct FragmentIdentifier<'a>(&'a str);
|
||||
|
||||
impl<'a> FragmentIdentifier<'a> {
|
||||
/// Like in a parsed URL
|
||||
pub fn to_percent_encoded(&self) -> String {
|
||||
let mut string = String::new();
|
||||
for byte in self.0.bytes() {
|
||||
match byte {
|
||||
// Ignore ASCII tabs or newlines like the URL parser would
|
||||
b'\t' | b'\n' | b'\r' => continue,
|
||||
// https://url.spec.whatwg.org/#fragment-percent-encode-set
|
||||
b'\0'..=b' ' | b'"' | b'<' | b'>' | b'`' | b'\x7F'..=b'\xFF' => {
|
||||
percent_encode(byte, &mut string)
|
||||
}
|
||||
// Printable ASCII
|
||||
_ => string.push(byte as char),
|
||||
}
|
||||
}
|
||||
string
|
||||
}
|
||||
}
|
||||
|
||||
/// Similar to <https://url.spec.whatwg.org/#concept-basic-url-parser>
|
||||
/// followed by <https://url.spec.whatwg.org/#concept-url-serializer>
|
||||
///
|
||||
/// * `None`: not a data URL.
|
||||
///
|
||||
/// * `Some(s)`: sort of the result of serialization, except:
|
||||
///
|
||||
/// - `data:` prefix removed
|
||||
/// - The fragment is included
|
||||
/// - Other components are **not** UTF-8 percent-encoded
|
||||
/// - ASCII tabs and newlines in the middle are **not** removed
|
||||
fn pretend_parse_data_url(input: &str) -> Option<&str> {
|
||||
// Trim C0 control or space
|
||||
let left_trimmed = input.trim_start_matches(|ch| ch <= ' ');
|
||||
|
||||
let mut bytes = left_trimmed.bytes();
|
||||
{
|
||||
// Ignore ASCII tabs or newlines like the URL parser would
|
||||
let mut iter = bytes
|
||||
.by_ref()
|
||||
.filter(|&byte| !matches!(byte, b'\t' | b'\n' | b'\r'));
|
||||
require!(iter.next()?.to_ascii_lowercase() == b'd');
|
||||
require!(iter.next()?.to_ascii_lowercase() == b'a');
|
||||
require!(iter.next()?.to_ascii_lowercase() == b't');
|
||||
require!(iter.next()?.to_ascii_lowercase() == b'a');
|
||||
require!(iter.next()? == b':');
|
||||
}
|
||||
let bytes_consumed = left_trimmed.len() - bytes.len();
|
||||
let after_colon = &left_trimmed[bytes_consumed..];
|
||||
|
||||
// Trim C0 control or space
|
||||
Some(after_colon.trim_end_matches(|ch| ch <= ' '))
|
||||
}
|
||||
|
||||
fn find_comma_before_fragment(after_colon: &str) -> Option<(&str, &str)> {
|
||||
for (i, byte) in after_colon.bytes().enumerate() {
|
||||
if byte == b',' {
|
||||
return Some((&after_colon[..i], &after_colon[i + 1..]));
|
||||
}
|
||||
if byte == b'#' {
|
||||
break;
|
||||
}
|
||||
}
|
||||
None
|
||||
}
|
||||
|
||||
fn parse_header(from_colon_to_comma: &str) -> (mime::Mime, bool) {
|
||||
// "Strip leading and trailing ASCII whitespace"
|
||||
// \t, \n, and \r would have been filtered by the URL parser
|
||||
// \f percent-encoded by the URL parser
|
||||
// space is the only remaining ASCII whitespace
|
||||
let trimmed = from_colon_to_comma.trim_matches(|c| matches!(c, ' ' | '\t' | '\n' | '\r'));
|
||||
|
||||
let without_base64_suffix = remove_base64_suffix(trimmed);
|
||||
let base64 = without_base64_suffix.is_some();
|
||||
let mime_type = without_base64_suffix.unwrap_or(trimmed);
|
||||
|
||||
let mut string = String::new();
|
||||
if mime_type.starts_with(';') {
|
||||
string.push_str("text/plain")
|
||||
}
|
||||
let mut in_query = false;
|
||||
for byte in mime_type.bytes() {
|
||||
match byte {
|
||||
// Ignore ASCII tabs or newlines like the URL parser would
|
||||
b'\t' | b'\n' | b'\r' => continue,
|
||||
|
||||
// https://url.spec.whatwg.org/#c0-control-percent-encode-set
|
||||
b'\0'..=b'\x1F' | b'\x7F'..=b'\xFF' => percent_encode(byte, &mut string),
|
||||
|
||||
// Bytes other than the C0 percent-encode set that are percent-encoded
|
||||
// by the URL parser in the query state.
|
||||
// '#' is also in that list but cannot occur here
|
||||
// since it indicates the start of the URL’s fragment.
|
||||
b' ' | b'"' | b'<' | b'>' if in_query => percent_encode(byte, &mut string),
|
||||
|
||||
b'?' => {
|
||||
in_query = true;
|
||||
string.push('?')
|
||||
}
|
||||
|
||||
// Printable ASCII
|
||||
_ => string.push(byte as char),
|
||||
}
|
||||
}
|
||||
|
||||
// FIXME: does Mime::from_str match the MIME Sniffing Standard’s parsing algorithm?
|
||||
// <https://mimesniff.spec.whatwg.org/#parse-a-mime-type>
|
||||
let mime_type = string.parse().unwrap_or_else(|_| mime::Mime {
|
||||
type_: String::from("text"),
|
||||
subtype: String::from("plain"),
|
||||
parameters: vec![(String::from("charset"), String::from("US-ASCII"))],
|
||||
});
|
||||
|
||||
(mime_type, base64)
|
||||
}
|
||||
|
||||
/// None: no base64 suffix
|
||||
#[allow(clippy::skip_while_next)]
|
||||
fn remove_base64_suffix(s: &str) -> Option<&str> {
|
||||
let mut bytes = s.bytes();
|
||||
{
|
||||
// Ignore ASCII tabs or newlines like the URL parser would
|
||||
let iter = bytes
|
||||
.by_ref()
|
||||
.filter(|&byte| !matches!(byte, b'\t' | b'\n' | b'\r'));
|
||||
|
||||
// Search from the end
|
||||
let mut iter = iter.rev();
|
||||
|
||||
require!(iter.next()? == b'4');
|
||||
require!(iter.next()? == b'6');
|
||||
require!(iter.next()?.to_ascii_lowercase() == b'e');
|
||||
require!(iter.next()?.to_ascii_lowercase() == b's');
|
||||
require!(iter.next()?.to_ascii_lowercase() == b'a');
|
||||
require!(iter.next()?.to_ascii_lowercase() == b'b');
|
||||
require!(iter.skip_while(|&byte| byte == b' ').next()? == b';');
|
||||
}
|
||||
Some(&s[..bytes.len()])
|
||||
}
|
||||
|
||||
fn percent_encode(byte: u8, string: &mut String) {
|
||||
const HEX_UPPER: [u8; 16] = *b"0123456789ABCDEF";
|
||||
string.push('%');
|
||||
string.push(HEX_UPPER[(byte >> 4) as usize] as char);
|
||||
string.push(HEX_UPPER[(byte & 0x0f) as usize] as char);
|
||||
}
|
||||
|
||||
/// This is <https://url.spec.whatwg.org/#string-percent-decode> while also:
|
||||
///
|
||||
/// * Ignoring ASCII tab or newlines
|
||||
/// * Stopping at the first '#' (which indicates the start of the fragment)
|
||||
///
|
||||
/// Anything that would have been UTF-8 percent-encoded by the URL parser
|
||||
/// would be percent-decoded here.
|
||||
/// We skip that round-trip and pass it through unchanged.
|
||||
fn decode_without_base64<F, E>(
|
||||
encoded_body_plus_fragment: &str,
|
||||
mut write_bytes: F,
|
||||
) -> Result<Option<FragmentIdentifier<'_>>, E>
|
||||
where
|
||||
F: FnMut(&[u8]) -> Result<(), E>,
|
||||
{
|
||||
let bytes = encoded_body_plus_fragment.as_bytes();
|
||||
let mut slice_start = 0;
|
||||
for (i, &byte) in bytes.iter().enumerate() {
|
||||
// We only need to look for 5 different "special" byte values.
|
||||
// For everything else we make slices as large as possible, borrowing the input,
|
||||
// in order to make fewer write_all() calls.
|
||||
if matches!(byte, b'%' | b'#' | b'\t' | b'\n' | b'\r') {
|
||||
// Write everything (if anything) "non-special" we’ve accumulated
|
||||
// before this special byte
|
||||
if i > slice_start {
|
||||
write_bytes(&bytes[slice_start..i])?;
|
||||
}
|
||||
// Then deal with the special byte.
|
||||
match byte {
|
||||
b'%' => {
|
||||
let l = bytes.get(i + 2).and_then(|&b| (b as char).to_digit(16));
|
||||
let h = bytes.get(i + 1).and_then(|&b| (b as char).to_digit(16));
|
||||
if let (Some(h), Some(l)) = (h, l) {
|
||||
// '%' followed by two ASCII hex digits
|
||||
let one_byte = h as u8 * 0x10 + l as u8;
|
||||
write_bytes(&[one_byte])?;
|
||||
slice_start = i + 3;
|
||||
} else {
|
||||
// Do nothing. Leave slice_start unchanged.
|
||||
// The % sign will be part of the next slice.
|
||||
}
|
||||
}
|
||||
|
||||
b'#' => {
|
||||
let fragment_start = i + 1;
|
||||
let fragment = &encoded_body_plus_fragment[fragment_start..];
|
||||
return Ok(Some(FragmentIdentifier(fragment)));
|
||||
}
|
||||
|
||||
// Ignore over '\t' | '\n' | '\r'
|
||||
_ => slice_start = i + 1,
|
||||
}
|
||||
}
|
||||
}
|
||||
write_bytes(&bytes[slice_start..])?;
|
||||
Ok(None)
|
||||
}
|
||||
|
||||
/// `decode_without_base64()` composed with
|
||||
/// <https://infra.spec.whatwg.org/#isomorphic-decode> composed with
|
||||
/// <https://infra.spec.whatwg.org/#forgiving-base64-decode>.
|
||||
fn decode_with_base64<F, E>(
|
||||
encoded_body_plus_fragment: &str,
|
||||
write_bytes: F,
|
||||
) -> Result<Option<FragmentIdentifier<'_>>, forgiving_base64::DecodeError<E>>
|
||||
where
|
||||
F: FnMut(&[u8]) -> Result<(), E>,
|
||||
{
|
||||
let mut decoder = forgiving_base64::Decoder::new(write_bytes);
|
||||
let fragment = decode_without_base64(encoded_body_plus_fragment, |bytes| decoder.feed(bytes))?;
|
||||
decoder.finish()?;
|
||||
Ok(fragment)
|
||||
}
|
||||
19
third-party/vendor/data-url/src/make_base64_decode_table.py
vendored
Normal file
19
third-party/vendor/data-url/src/make_base64_decode_table.py
vendored
Normal file
|
|
@ -0,0 +1,19 @@
|
|||
"""
|
||||
Generate the BASE64_DECODE_TABLE constant. See its doc-comment.
|
||||
"""
|
||||
|
||||
import string
|
||||
|
||||
# https://tools.ietf.org/html/rfc4648#section-4
|
||||
alphabet = string.ascii_uppercase + string.ascii_lowercase + string.digits + "+/"
|
||||
assert len(alphabet) == 64
|
||||
|
||||
reverse_table = [-1] * 256
|
||||
for i, symbol in enumerate(alphabet):
|
||||
reverse_table[ord(symbol)] = i
|
||||
|
||||
print("[")
|
||||
per_line = 16
|
||||
for line in range(0, 256, per_line):
|
||||
print(" " + "".join(" %2s," % value for value in reverse_table[line:][:per_line]))
|
||||
print("]")
|
||||
207
third-party/vendor/data-url/src/mime.rs
vendored
Normal file
207
third-party/vendor/data-url/src/mime.rs
vendored
Normal file
|
|
@ -0,0 +1,207 @@
|
|||
use alloc::{borrow::ToOwned, string::String, vec::Vec};
|
||||
use core::fmt::{self, Write};
|
||||
use core::str::FromStr;
|
||||
|
||||
/// <https://mimesniff.spec.whatwg.org/#mime-type-representation>
|
||||
#[derive(Debug, PartialEq, Eq)]
|
||||
pub struct Mime {
|
||||
pub type_: String,
|
||||
pub subtype: String,
|
||||
/// (name, value)
|
||||
pub parameters: Vec<(String, String)>,
|
||||
}
|
||||
|
||||
impl Mime {
|
||||
pub fn get_parameter<P>(&self, name: &P) -> Option<&str>
|
||||
where
|
||||
P: ?Sized + PartialEq<str>,
|
||||
{
|
||||
self.parameters
|
||||
.iter()
|
||||
.find(|&(n, _)| name == &**n)
|
||||
.map(|(_, v)| &**v)
|
||||
}
|
||||
}
|
||||
|
||||
#[derive(Debug)]
|
||||
pub struct MimeParsingError(());
|
||||
|
||||
impl fmt::Display for MimeParsingError {
|
||||
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
|
||||
write!(f, "invalid mime type")
|
||||
}
|
||||
}
|
||||
|
||||
#[cfg(feature = "std")]
|
||||
impl std::error::Error for MimeParsingError {}
|
||||
|
||||
/// <https://mimesniff.spec.whatwg.org/#parsing-a-mime-type>
|
||||
impl FromStr for Mime {
|
||||
type Err = MimeParsingError;
|
||||
|
||||
fn from_str(s: &str) -> Result<Self, Self::Err> {
|
||||
parse(s).ok_or(MimeParsingError(()))
|
||||
}
|
||||
}
|
||||
|
||||
fn parse(s: &str) -> Option<Mime> {
|
||||
let trimmed = s.trim_matches(http_whitespace);
|
||||
|
||||
let (type_, rest) = split2(trimmed, '/');
|
||||
require!(only_http_token_code_points(type_) && !type_.is_empty());
|
||||
|
||||
let (subtype, rest) = split2(rest?, ';');
|
||||
let subtype = subtype.trim_end_matches(http_whitespace);
|
||||
require!(only_http_token_code_points(subtype) && !subtype.is_empty());
|
||||
|
||||
let mut parameters = Vec::new();
|
||||
if let Some(rest) = rest {
|
||||
parse_parameters(rest, &mut parameters)
|
||||
}
|
||||
|
||||
Some(Mime {
|
||||
type_: type_.to_ascii_lowercase(),
|
||||
subtype: subtype.to_ascii_lowercase(),
|
||||
parameters,
|
||||
})
|
||||
}
|
||||
|
||||
fn split2(s: &str, separator: char) -> (&str, Option<&str>) {
|
||||
let mut iter = s.splitn(2, separator);
|
||||
let first = iter.next().unwrap();
|
||||
(first, iter.next())
|
||||
}
|
||||
|
||||
fn parse_parameters(s: &str, parameters: &mut Vec<(String, String)>) {
|
||||
let mut semicolon_separated = s.split(';');
|
||||
|
||||
while let Some(piece) = semicolon_separated.next() {
|
||||
let piece = piece.trim_start_matches(http_whitespace);
|
||||
let (name, value) = split2(piece, '=');
|
||||
// We can not early return on an invalid name here, because the value
|
||||
// parsing later may consume more semicolon seperated pieces.
|
||||
let name_valid =
|
||||
!name.is_empty() && only_http_token_code_points(name) && !contains(parameters, name);
|
||||
if let Some(value) = value {
|
||||
let value = if let Some(stripped) = value.strip_prefix('"') {
|
||||
let max_len = stripped.len().saturating_sub(1); // without end quote
|
||||
let mut unescaped_value = String::with_capacity(max_len);
|
||||
let mut chars = stripped.chars();
|
||||
'until_closing_quote: loop {
|
||||
while let Some(c) = chars.next() {
|
||||
match c {
|
||||
'"' => break 'until_closing_quote,
|
||||
'\\' => unescaped_value.push(chars.next().unwrap_or_else(|| {
|
||||
semicolon_separated
|
||||
.next()
|
||||
.map(|piece| {
|
||||
// A semicolon inside a quoted value is not a separator
|
||||
// for the next parameter, but part of the value.
|
||||
chars = piece.chars();
|
||||
';'
|
||||
})
|
||||
.unwrap_or('\\')
|
||||
})),
|
||||
_ => unescaped_value.push(c),
|
||||
}
|
||||
}
|
||||
if let Some(piece) = semicolon_separated.next() {
|
||||
// A semicolon inside a quoted value is not a separator
|
||||
// for the next parameter, but part of the value.
|
||||
unescaped_value.push(';');
|
||||
chars = piece.chars()
|
||||
} else {
|
||||
break;
|
||||
}
|
||||
}
|
||||
if !name_valid || !valid_value(value) {
|
||||
continue;
|
||||
}
|
||||
unescaped_value
|
||||
} else {
|
||||
let value = value.trim_end_matches(http_whitespace);
|
||||
if value.is_empty() {
|
||||
continue;
|
||||
}
|
||||
if !name_valid || !valid_value(value) {
|
||||
continue;
|
||||
}
|
||||
value.to_owned()
|
||||
};
|
||||
parameters.push((name.to_ascii_lowercase(), value))
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
fn contains(parameters: &[(String, String)], name: &str) -> bool {
|
||||
parameters.iter().any(|(n, _)| n == name)
|
||||
}
|
||||
|
||||
fn valid_value(s: &str) -> bool {
|
||||
s.chars().all(|c| {
|
||||
// <https://mimesniff.spec.whatwg.org/#http-quoted-string-token-code-point>
|
||||
matches!(c, '\t' | ' '..='~' | '\u{80}'..='\u{FF}')
|
||||
})
|
||||
}
|
||||
|
||||
/// <https://mimesniff.spec.whatwg.org/#serializing-a-mime-type>
|
||||
impl fmt::Display for Mime {
|
||||
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
|
||||
f.write_str(&self.type_)?;
|
||||
f.write_str("/")?;
|
||||
f.write_str(&self.subtype)?;
|
||||
for (name, value) in &self.parameters {
|
||||
f.write_str(";")?;
|
||||
f.write_str(name)?;
|
||||
f.write_str("=")?;
|
||||
if only_http_token_code_points(value) && !value.is_empty() {
|
||||
f.write_str(value)?
|
||||
} else {
|
||||
f.write_str("\"")?;
|
||||
for c in value.chars() {
|
||||
if c == '"' || c == '\\' {
|
||||
f.write_str("\\")?
|
||||
}
|
||||
f.write_char(c)?
|
||||
}
|
||||
f.write_str("\"")?
|
||||
}
|
||||
}
|
||||
Ok(())
|
||||
}
|
||||
}
|
||||
|
||||
fn http_whitespace(c: char) -> bool {
|
||||
matches!(c, ' ' | '\t' | '\n' | '\r')
|
||||
}
|
||||
|
||||
fn only_http_token_code_points(s: &str) -> bool {
|
||||
s.bytes().all(|byte| IS_HTTP_TOKEN[byte as usize])
|
||||
}
|
||||
|
||||
macro_rules! byte_map {
|
||||
($($flag:expr,)*) => ([
|
||||
$($flag != 0,)*
|
||||
])
|
||||
}
|
||||
|
||||
// Copied from https://github.com/hyperium/mime/blob/v0.3.5/src/parse.rs#L293
|
||||
#[rustfmt::skip]
|
||||
static IS_HTTP_TOKEN: [bool; 256] = byte_map![
|
||||
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
|
||||
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
|
||||
0, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 1, 0, 1, 1, 0,
|
||||
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0,
|
||||
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
|
||||
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1,
|
||||
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
|
||||
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 1, 0,
|
||||
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
|
||||
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
|
||||
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
|
||||
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
|
||||
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
|
||||
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
|
||||
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
|
||||
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
|
||||
];
|
||||
Loading…
Add table
Add a link
Reference in a new issue