aboutsummaryrefslogtreecommitdiff
path: root/src/utf8
diff options
context:
space:
mode:
authorJoe Wilm <joe@jwilm.com>2016-09-17 15:51:45 -0700
committerJoe Wilm <joe@jwilm.com>2016-09-17 17:03:20 -0700
commitcffdb6de59ceb3fd9983a1c19476e5109da8db97 (patch)
tree26603abf607d21eefd3b9a6ac79a36dab63b5781 /src/utf8
parent930f8cc30a5bc4943c1b56e18cf1a3f8bb00bc2a (diff)
downloadr-alacritty-vte-cffdb6de59ceb3fd9983a1c19476e5109da8db97.tar.gz
r-alacritty-vte-cffdb6de59ceb3fd9983a1c19476e5109da8db97.tar.bz2
r-alacritty-vte-cffdb6de59ceb3fd9983a1c19476e5109da8db97.zip
Add support for UTF-8
This adds a table-driven UTF-8 parser which only has a single branch for the entire parser. UTF-8 support is essentially bolted onto the VTE parser. Not the most elegant, but it does prevent the transition tables from blowing up. Instead of refactoring the syntax extension to handle both table definitions, I've opted to copy/paste now for both simplicities sake and because I can't see a clear path to a minimal shared solution.
Diffstat (limited to 'src/utf8')
-rw-r--r--src/utf8/mod.rs91
-rw-r--r--src/utf8/table.rs184
-rw-r--r--src/utf8/table.rs.in60
-rw-r--r--src/utf8/types.rs77
4 files changed, 412 insertions, 0 deletions
diff --git a/src/utf8/mod.rs b/src/utf8/mod.rs
new file mode 100644
index 0000000..3d099b1
--- /dev/null
+++ b/src/utf8/mod.rs
@@ -0,0 +1,91 @@
+//! A table-driven UTF-8 Parser
+//!
+//! This module implements a table-driven UTF-8 parser which should
+//! theoretically contain the minimal number of branches (1). The only branch is
+//! on the `Action` returned from unpacking a transition.
+use std::char;
+
+mod types;
+use self::types::{State, Action, unpack};
+
+mod table;
+use self::table::TRANSITIONS;
+
+/// Handles codepoint and invalid sequence events from the parser.
+pub trait Receiver {
+ /// Code point parsed
+ ///
+ /// Called with the codepoint
+ fn codepoint(&mut self, char);
+
+ /// Invalid sequence encountered
+ fn invalid_sequence(&mut self);
+}
+
+/// A parser for Utf8 Characters
+///
+/// Repeatedly call `advance` with bytes to emit Utf8 characters
+pub struct Parser {
+ point: u32,
+ state: State,
+}
+
+/// Continuation bytes are masked with this value.
+const CONTINUATION_MASK: u8 = 0b0011_1111;
+
+impl Parser {
+ /// Create a new Parser
+ pub fn new() -> Parser {
+ Parser {
+ point: 0,
+ state: State::Ground,
+ }
+ }
+
+ pub fn advance<R>(&mut self, receiver: &mut R, byte: u8)
+ where R: Receiver
+ {
+ let cur = self.state as usize;
+ let change = TRANSITIONS[cur][byte as usize];
+ let (state, action) = unsafe { unpack(change) };
+
+ self.perform_action(receiver, byte, action);
+ self.state = state;
+ }
+
+ fn perform_action<R>(&mut self, receiver: &mut R, byte: u8, action: Action)
+ where R: Receiver
+ {
+ match action {
+ Action::InvalidSequence => {
+ self.point = 0;
+ receiver.invalid_sequence();
+ },
+ Action::EmitByte => {
+ receiver.codepoint(byte as char);
+ },
+ Action::SetByte1 => {
+ let point = self.point | ((byte & CONTINUATION_MASK) as u32);
+ let c = unsafe { char::from_u32_unchecked(point) };
+ self.point = 0;
+
+ receiver.codepoint(c);
+ },
+ Action::SetByte2 => {
+ self.point |= ((byte & CONTINUATION_MASK) as u32) << 6;
+ },
+ Action::SetByte2Top => {
+ self.point |= ((byte & 0b0001_1111) as u32) << 6;
+ },
+ Action::SetByte3 => {
+ self.point |= ((byte & CONTINUATION_MASK) as u32) << 12;
+ },
+ Action::SetByte3Top => {
+ self.point |= ((byte & 0b0000_1111) as u32) << 12;
+ },
+ Action::SetByte4 => {
+ self.point |= ((byte & 0b0000_0111) as u32) << 18;
+ },
+ }
+ }
+}
diff --git a/src/utf8/table.rs b/src/utf8/table.rs
new file mode 100644
index 0000000..5a1292b
--- /dev/null
+++ b/src/utf8/table.rs
@@ -0,0 +1,184 @@
+//! UTF-8 Parse Transition Table
+
+/// Transition table for parsing UTF-8. This is built from the grammar described
+/// at https://tools.ietf.org/html/rfc3629#section-4 which I have copied and
+/// formatted below.
+///
+/// # UTF-8 Grammar
+///
+/// ```ignore
+/// UTF8-octets = *( UTF8-char )
+/// UTF8-char = UTF8-1 / UTF8-2 / UTF8-3 / UTF8-4
+/// UTF8-1 = %x00-7F
+/// UTF8-2 = %xC2-DF UTF8-tail
+/// UTF8-3 = %xE0 %xA0-BF UTF8-tail /
+/// %xE1-EC 2( UTF8-tail ) /
+/// %xED %x80-9F UTF8-tail /
+/// %xEE-EF 2( UTF8-tail )
+/// UTF8-4 = %xF0 %x90-BF 2( UTF8-tail ) /
+/// %xF1-F3 3( UTF8-tail ) /
+/// %xF4 %x80-8F 2( UTF8-tail )
+/// UTF8-tail = %x80-BF
+/// ```
+///
+/// Not specifying an action in this table is equivalent to specifying
+/// Action::InvalidSequence. Not specifying a state is equivalent to specifying
+/// state::ground.
+pub static TRANSITIONS: [[u8; 256]; 8] =
+ [[16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8,
+ 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8,
+ 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8,
+ 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8,
+ 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8,
+ 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8,
+ 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8,
+ 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8,
+ 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8,
+ 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8,
+ 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 16u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 67u8, 67u8, 67u8, 67u8, 67u8, 67u8, 67u8, 67u8,
+ 67u8, 67u8, 67u8, 67u8, 67u8, 67u8, 67u8, 67u8, 67u8, 67u8, 67u8, 67u8,
+ 67u8, 67u8, 67u8, 67u8, 67u8, 67u8, 67u8, 67u8, 67u8, 67u8, 100u8, 98u8,
+ 98u8, 98u8, 98u8, 98u8, 98u8, 98u8, 98u8, 98u8, 98u8, 98u8, 98u8, 101u8,
+ 98u8, 98u8, 118u8, 113u8, 113u8, 113u8, 119u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8],
+ [0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8,
+ 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8,
+ 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8,
+ 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8,
+ 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8,
+ 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8],
+ [0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8,
+ 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8,
+ 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8,
+ 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8,
+ 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8,
+ 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8],
+ [0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8,
+ 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8,
+ 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8,
+ 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8,
+ 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 32u8,
+ 32u8, 32u8, 32u8, 32u8, 32u8, 32u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8],
+ [0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8,
+ 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8,
+ 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8,
+ 51u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8],
+ [0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8,
+ 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8,
+ 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 51u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8],
+ [0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8,
+ 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8,
+ 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8,
+ 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8,
+ 82u8, 82u8, 82u8, 82u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8],
+ [0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 82u8,
+ 82u8, 82u8, 82u8, 82u8, 82u8, 82u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8,
+ 0u8, 0u8, 0u8, 0u8, 0u8, 0u8, 0u8]];
diff --git a/src/utf8/table.rs.in b/src/utf8/table.rs.in
new file mode 100644
index 0000000..2acafe7
--- /dev/null
+++ b/src/utf8/table.rs.in
@@ -0,0 +1,60 @@
+//! UTF-8 Parse Transition Table
+
+/// Transition table for parsing UTF-8. This is built from the grammar described
+/// at https://tools.ietf.org/html/rfc3629#section-4 which I have copied and
+/// formatted below.
+///
+/// # UTF-8 Grammar
+///
+/// ```ignore
+/// UTF8-octets = *( UTF8-char )
+/// UTF8-char = UTF8-1 / UTF8-2 / UTF8-3 / UTF8-4
+/// UTF8-1 = %x00-7F
+/// UTF8-2 = %xC2-DF UTF8-tail
+/// UTF8-3 = %xE0 %xA0-BF UTF8-tail /
+/// %xE1-EC 2( UTF8-tail ) /
+/// %xED %x80-9F UTF8-tail /
+/// %xEE-EF 2( UTF8-tail )
+/// UTF8-4 = %xF0 %x90-BF 2( UTF8-tail ) /
+/// %xF1-F3 3( UTF8-tail ) /
+/// %xF4 %x80-8F 2( UTF8-tail )
+/// UTF8-tail = %x80-BF
+/// ```
+///
+/// Not specifying an action in this table is equivalent to specifying
+/// Action::InvalidSequence. Not specifying a state is equivalent to specifying
+/// state::ground.
+pub static TRANSITIONS: [[u8; 256]; 8] = utf8_state_table! {
+ State::Ground => {
+ 0x00...0x7f => (State::Ground, Action::EmitByte),
+ 0xc2...0xdf => (State::Tail1, Action::SetByte2Top),
+ 0xe0 => (State::U3_2_e0, Action::SetByte3Top),
+ 0xe1...0xec => (State::Tail2, Action::SetByte3Top),
+ 0xed => (State::U3_2_ed, Action::SetByte3Top),
+ 0xee...0xef => (State::Tail2, Action::SetByte3Top),
+ 0xf0 => (State::Utf8_4_3_f0, Action::SetByte4),
+ 0xf1...0xf3 => (State::Tail3, Action::SetByte4),
+ 0xf4 => (State::Utf8_4_3_f4, Action::SetByte4),
+ },
+ State::U3_2_e0 => {
+ 0xa0...0xbf => (State::Tail1, Action::SetByte2),
+ },
+ State::U3_2_ed => {
+ 0x80...0x9f => (State::Tail1, Action::SetByte2),
+ },
+ State::Utf8_4_3_f0 => {
+ 0x90...0xbf => (State::Tail2, Action::SetByte3),
+ },
+ State::Utf8_4_3_f4 => {
+ 0x80...0x8f => (State::Tail2, Action::SetByte3),
+ },
+ State::Tail3 => {
+ 0x80...0xbf => (State::Tail2, Action::SetByte3),
+ },
+ State::Tail2 => {
+ 0x80...0xbf => (State::Tail1, Action::SetByte2),
+ },
+ State::Tail1 => {
+ 0x80...0xbf => (State::Ground, Action::SetByte1),
+ },
+};
diff --git a/src/utf8/types.rs b/src/utf8/types.rs
new file mode 100644
index 0000000..4c604f4
--- /dev/null
+++ b/src/utf8/types.rs
@@ -0,0 +1,77 @@
+//! Types supporting the UTF-8 parser
+#![allow(non_camel_case_types)]
+use std::mem;
+
+/// States the parser can be in.
+///
+/// There is a state for each initial input of the 3 and 4 byte sequences since
+/// the following bytes are subject to different conditions than a tail byte.
+#[allow(dead_code)]
+#[derive(Debug, Copy, Clone)]
+pub enum State {
+ /// Ground state; expect anything
+ Ground = 0,
+ /// 3 tail bytes
+ Tail3 = 1,
+ /// 2 tail bytes
+ Tail2 = 2,
+ /// 1 tail byte
+ Tail1 = 3,
+ /// UTF8-3 starting with E0
+ U3_2_e0 = 4,
+ /// UTF8-3 starting with ED
+ U3_2_ed = 5,
+ /// UTF8-4 starting with F0
+ Utf8_4_3_f0 = 6,
+ /// UTF8-4 starting with F4
+ Utf8_4_3_f4 = 7,
+}
+
+/// Action to take when receiving a byte
+#[allow(dead_code)]
+#[derive(Debug, Copy, Clone)]
+pub enum Action {
+ /// Unexpected byte; sequence is invalid
+ InvalidSequence = 0,
+ /// Received valid 7-bit ASCII byte which can be directly emitted.
+ EmitByte = 1,
+ /// Set the bottom continuation byte
+ SetByte1 = 2,
+ /// Set the 2nd-from-last continuation byte
+ SetByte2 = 3,
+ /// Set the 2nd-from-last byte which is part of a two byte sequence
+ SetByte2Top = 4,
+ /// Set the 3rd-from-last continuation byte
+ SetByte3 = 5,
+ /// Set the 3rd-from-last byte which is part of a three byte sequence
+ SetByte3Top = 6,
+ /// Set the top byte of a four byte sequence.
+ SetByte4 = 7,
+}
+
+/// Convert a state and action to a u8
+///
+/// State will be the bottom 4 bits and action the top 4
+#[inline]
+#[allow(dead_code)]
+pub fn pack(state: State, action: Action) -> u8 {
+ ((action as u8) << 4) | (state as u8)
+}
+
+/// Convert a u8 to a state and action
+///
+/// # Unsafety
+///
+/// If this function is called with a byte that wasn't encoded with the `pack`
+/// function in this module, there is no guarantee that a valid state and action
+/// can be produced.
+#[inline]
+pub unsafe fn unpack(val: u8) -> (State, Action) {
+ (
+ // State is stored in bottom 4 bits
+ mem::transmute(val & 0x0f),
+
+ // Action is stored in top 4 bits
+ mem::transmute(val >> 4),
+ )
+}