mirror of
https://github.com/simonask/libyaml-safer
synced 2024-11-25 21:00:25 +00:00
Use char rather than raw bytes where possible
This commit is contained in:
Simon Ask Ulsnes
parent
4d90f9a429
commit
5ef7d51da0
1 changed files with 52 additions and 38 deletions
|
|
@ -332,10 +332,9 @@ impl<'w> Emitter<'w> {
|
|||
}
|
||||
|
||||
/// Equivalent of the libyaml `PUT` macro.
|
||||
fn put(&mut self, value: u8) -> Result<(), WriterError> {
|
||||
fn put(&mut self, value: char) -> Result<(), WriterError> {
|
||||
self.flush_if_needed()?;
|
||||
let ch = char::from(value);
|
||||
self.buffer.push(ch);
|
||||
self.buffer.push(value);
|
||||
self.column += 1;
|
||||
Ok(())
|
||||
}
|
||||
|
|
@ -356,11 +355,23 @@ impl<'w> Emitter<'w> {
|
|||
}
|
||||
|
||||
/// Write UTF-8 charanters from `string` to `emitter` and increment
|
||||
/// `emitter.column` the appropriate number of times.
|
||||
/// `emitter.column` the appropriate number of times. It is assumed that the
|
||||
/// string does not contain line breaks!
|
||||
fn write_str(&mut self, string: &str) -> Result<(), WriterError> {
|
||||
for ch in string.chars() {
|
||||
self.write_char(ch)?;
|
||||
if self.buffer.len() + string.len() > OUTPUT_BUFFER_SIZE {
|
||||
self.flush()?;
|
||||
}
|
||||
|
||||
// Note: Reserves less than what is necessary if there are UTF-8
|
||||
// characters present.
|
||||
self.buffer.reserve(string.len());
|
||||
|
||||
self.column += string.chars().count() as i32;
|
||||
|
||||
// Note: This may cause the buffer to become slightly larger than
|
||||
// `OUTPUT_BUFFER_SIZE`, but not by much.
|
||||
self.buffer.push_str(string);
|
||||
|
||||
Ok(())
|
||||
}
|
||||
|
||||
|
|
@ -822,7 +833,7 @@ impl<'w> Emitter<'w> {
|
|||
) -> Result<(), EmitterError> {
|
||||
self.process_anchor(analysis)?;
|
||||
if self.simple_key_context {
|
||||
self.put(b' ')?;
|
||||
self.put(' ')?;
|
||||
}
|
||||
self.state = self.states.pop().unwrap();
|
||||
Ok(())
|
||||
|
|
@ -1390,7 +1401,7 @@ impl<'w> Emitter<'w> {
|
|||
self.put_break()?;
|
||||
}
|
||||
while self.column < indent {
|
||||
self.put(b' ')?;
|
||||
self.put(' ')?;
|
||||
}
|
||||
self.whitespace = true;
|
||||
self.indention = true;
|
||||
|
|
@ -1405,7 +1416,7 @@ impl<'w> Emitter<'w> {
|
|||
is_indention: bool,
|
||||
) -> Result<(), EmitterError> {
|
||||
if need_whitespace && !self.whitespace {
|
||||
self.put(b' ')?;
|
||||
self.put(' ')?;
|
||||
}
|
||||
self.write_str(indicator)?;
|
||||
self.whitespace = is_whitespace;
|
||||
|
|
@ -1422,7 +1433,7 @@ impl<'w> Emitter<'w> {
|
|||
|
||||
fn write_tag_handle(&mut self, value: &str) -> Result<(), EmitterError> {
|
||||
if !self.whitespace {
|
||||
self.put(b' ')?;
|
||||
self.put(' ')?;
|
||||
}
|
||||
self.write_str(value)?;
|
||||
self.whitespace = false;
|
||||
|
|
@ -1436,7 +1447,7 @@ impl<'w> Emitter<'w> {
|
|||
need_whitespace: bool,
|
||||
) -> Result<(), EmitterError> {
|
||||
if need_whitespace && !self.whitespace {
|
||||
self.put(b' ')?;
|
||||
self.put(' ')?;
|
||||
}
|
||||
|
||||
for ch in value.chars() {
|
||||
|
|
@ -1458,9 +1469,13 @@ impl<'w> Emitter<'w> {
|
|||
let mut encode_buffer = [0u8; 4];
|
||||
let encoded_char = ch.encode_utf8(&mut encode_buffer);
|
||||
for value in encoded_char.bytes() {
|
||||
let upper = (value >> 4) + if (value >> 4) < 10 { b'0' } else { b'A' - 10 };
|
||||
let lower = (value & 0x0F) + if (value & 0x0F) < 10 { b'0' } else { b'A' - 10 };
|
||||
self.put(b'%')?;
|
||||
let upper = char::from_digit(value as u32 >> 4, 16)
|
||||
.expect("invalid digit")
|
||||
.to_ascii_uppercase();
|
||||
let lower = char::from_digit(value as u32 & 0x0F, 16)
|
||||
.expect("invalid digit")
|
||||
.to_ascii_uppercase();
|
||||
self.put('%')?;
|
||||
self.put(upper)?;
|
||||
self.put(lower)?;
|
||||
}
|
||||
|
|
@ -1475,7 +1490,7 @@ impl<'w> Emitter<'w> {
|
|||
let mut spaces = false;
|
||||
let mut breaks = false;
|
||||
if !self.whitespace && (!value.is_empty() || self.flow_level != 0) {
|
||||
self.put(b' ')?;
|
||||
self.put(' ')?;
|
||||
}
|
||||
|
||||
let mut chars = value.chars();
|
||||
|
|
@ -1550,7 +1565,7 @@ impl<'w> Emitter<'w> {
|
|||
self.write_indent()?;
|
||||
}
|
||||
if ch == '\'' {
|
||||
self.put(b'\'')?;
|
||||
self.put('\'')?;
|
||||
}
|
||||
self.write_char(ch)?;
|
||||
self.indention = false;
|
||||
|
|
@ -1586,61 +1601,61 @@ impl<'w> Emitter<'w> {
|
|||
|| ch == '"'
|
||||
|| ch == '\\'
|
||||
{
|
||||
self.put(b'\\')?;
|
||||
self.put('\\')?;
|
||||
match ch {
|
||||
// TODO: Double check these character mappings.
|
||||
'\0' => {
|
||||
self.put(b'0')?;
|
||||
self.put('0')?;
|
||||
}
|
||||
'\x07' => {
|
||||
self.put(b'a')?;
|
||||
self.put('a')?;
|
||||
}
|
||||
'\x08' => {
|
||||
self.put(b'b')?;
|
||||
self.put('b')?;
|
||||
}
|
||||
'\x09' => {
|
||||
self.put(b't')?;
|
||||
self.put('t')?;
|
||||
}
|
||||
'\x0A' => {
|
||||
self.put(b'n')?;
|
||||
self.put('n')?;
|
||||
}
|
||||
'\x0B' => {
|
||||
self.put(b'v')?;
|
||||
self.put('v')?;
|
||||
}
|
||||
'\x0C' => {
|
||||
self.put(b'f')?;
|
||||
self.put('f')?;
|
||||
}
|
||||
'\x0D' => {
|
||||
self.put(b'r')?;
|
||||
self.put('r')?;
|
||||
}
|
||||
'\x1B' => {
|
||||
self.put(b'e')?;
|
||||
self.put('e')?;
|
||||
}
|
||||
'\x22' => {
|
||||
self.put(b'"')?;
|
||||
self.put('"')?;
|
||||
}
|
||||
'\x5C' => {
|
||||
self.put(b'\\')?;
|
||||
self.put('\\')?;
|
||||
}
|
||||
'\u{0085}' => {
|
||||
self.put(b'N')?;
|
||||
self.put('N')?;
|
||||
}
|
||||
'\u{00A0}' => {
|
||||
self.put(b'_')?;
|
||||
self.put('_')?;
|
||||
}
|
||||
'\u{2028}' => {
|
||||
self.put(b'L')?;
|
||||
self.put('L')?;
|
||||
}
|
||||
'\u{2029}' => {
|
||||
self.put(b'P')?;
|
||||
self.put('P')?;
|
||||
}
|
||||
_ => {
|
||||
let (prefix, width) = if ch <= '\u{00ff}' {
|
||||
(b'x', 2)
|
||||
('x', 2)
|
||||
} else if ch <= '\u{ffff}' {
|
||||
(b'u', 4)
|
||||
('u', 4)
|
||||
} else {
|
||||
(b'U', 8)
|
||||
('U', 8)
|
||||
};
|
||||
self.put(prefix)?;
|
||||
let mut k = (width - 1) * 4;
|
||||
|
|
@ -1652,8 +1667,7 @@ impl<'w> Emitter<'w> {
|
|||
};
|
||||
// The libyaml emitter encodes unicode sequences as uppercase hex.
|
||||
let digit_char = digit_char.to_ascii_uppercase();
|
||||
let digit_byte = digit_char as u8;
|
||||
self.put(digit_byte)?;
|
||||
self.put(digit_char)?;
|
||||
k -= 4;
|
||||
}
|
||||
}
|
||||
|
|
@ -1668,7 +1682,7 @@ impl<'w> Emitter<'w> {
|
|||
{
|
||||
self.write_indent()?;
|
||||
if is_space(chars.clone().next()) {
|
||||
self.put(b'\\')?;
|
||||
self.put('\\')?;
|
||||
}
|
||||
} else {
|
||||
self.write_char(ch)?;
|
||||
|
|
|
|||
Loading…
Reference in a new issue