1 files changed, 244 insertions, 0 deletions
diff --git a/src/tokenizer.rs b/src/tokenizer.rs
new file mode 100644
index 0000000..d7e60ac
--- /dev/null
+++ b/src/tokenizer.rs
@@ -0,0 +1,244 @@
+use crate::machine;
+use crate::utils::{control_pat, noncharacter_pat, surrogate_pat, ControlToken, State};
+use crate::{DefaultEmitter, Emitter, Error, Never, Readable, Reader};
+
+// this is a stack that can hold 0 to 2 Ts
+#[derive(Debug, Default)]
+struct Stack2<T: Copy>(Option<(T, Option<T>)>);
+
+impl<T: Copy> Stack2<T> {
+    #[inline]
+    fn push(&mut self, c: T) {
+        self.0 = match self.0 {
+            None => Some((c, None)),
+            Some((c1, None)) => Some((c1, Some(c))),
+            Some((_c1, Some(_c2))) => panic!("stack full!"),
+        }
+    }
+
+    #[inline]
+    fn pop(&mut self) -> Option<T> {
+        let (new_self, rv) = match self.0 {
+            Some((c1, Some(c2))) => (Some((c1, None)), Some(c2)),
+            Some((c1, None)) => (None, Some(c1)),
+            None => (None, None),
+        };
+        self.0 = new_self;
+        rv
+    }
+
+    #[inline]
+    fn is_empty(&self) -> bool {
+        matches!(self.0, None)
+    }
+}
+
+/// A HTML tokenizer. See crate-level docs for basic usage.
+pub struct Tokenizer<R: Reader, E: Emitter = DefaultEmitter> {
+    eof: bool,
+    pub(crate) state: State,
+    pub(crate) emitter: E,
+    pub(crate) temporary_buffer: String,
+    reader: R,
+    to_reconsume: Stack2<Option<char>>,
+    pub(crate) character_reference_code: u32,
+    pub(crate) return_state: Option<State>,
+}
+
+impl<R: Reader> Tokenizer<R> {
+    /// Create a new tokenizer from some input.
+    ///
+    /// `input` can be `&String` or `&str` at the moment, as those are the types for which
+    /// [`crate::Readable`] is implemented, but you can implement that trait on your own types.
+    ///
+    /// Patches are welcome for providing an efficient implementation over async streams,
+    /// iterators, files, etc, as long as any dependencies come behind featureflags.
+    pub fn new<'a, S: Readable<'a, Reader = R>>(input: S) -> Self {
+        Tokenizer::<S::Reader>::new_with_emitter(input, DefaultEmitter::default())
+    }
+}
+
+impl<R: Reader, E: Emitter> Tokenizer<R, E> {
+    /// Construct a new tokenizer from some input and a custom emitter.
+    ///
+    /// Use this method over [`Tokenizer::new`] when you want to have more control over string allocation for
+    /// tokens.
+    pub fn new_with_emitter<'a, S: Readable<'a, Reader = R>>(input: S, emitter: E) -> Self {
+        Tokenizer {
+            eof: false,
+            state: State::Data,
+            emitter,
+            temporary_buffer: String::new(),
+            to_reconsume: Stack2::default(),
+            reader: input.to_reader(),
+            character_reference_code: 0,
+            return_state: None,
+        }
+    }
+
+    /// Test-internal function to override internal state.
+    ///
+    /// Only available with the `integration-tests` feature which is not public API.
+    #[cfg(feature = "integration-tests")]
+    pub fn set_state(&mut self, state: State) {
+        self.state = state;
+    }
+
+    /// Set the statemachine to start/continue in [plaintext
+    /// state](https://html.spec.whatwg.org/#plaintext-state).
+    ///
+    /// This tokenizer never gets into that state naturally.
+    pub fn set_plaintext_state(&mut self) {
+        self.state = State::PlainText;
+    }
+
+    /// Test-internal function to override internal state.
+    ///
+    /// Only available with the `integration-tests` feature which is not public API.
+    #[cfg(feature = "integration-tests")]
+    pub fn set_last_start_tag(&mut self, last_start_tag: Option<&str>) {
+        self.emitter.set_last_start_tag(last_start_tag);
+    }
+
+    #[inline]
+    pub(crate) fn unread_char(&mut self, c: Option<char>) {
+        self.to_reconsume.push(c);
+    }
+
+    #[inline]
+    fn validate_char(&mut self, c: char) {
+        match c as u32 {
+            surrogate_pat!() => {
+                self.emitter.emit_error(Error::SurrogateInInputStream);
+            }
+            noncharacter_pat!() => {
+                self.emitter.emit_error(Error::NoncharacterInInputStream);
+            }
+            // control without whitespace or nul
+            x @ control_pat!()
+                if !matches!(x, 0x0000 | 0x0009 | 0x000a | 0x000c | 0x000d | 0x0020) =>
+            {
+                self.emitter
+                    .emit_error(Error::ControlCharacterInInputStream);
+            }
+            _ => (),
+        }
+    }
+
+    pub(crate) fn read_char(&mut self) -> Result<Option<char>, R::Error> {
+        let (c_res, reconsumed) = match self.to_reconsume.pop() {
+            Some(c) => (Ok(c), true),
+            None => (self.reader.read_char(), false),
+        };
+
+        let mut c = match c_res {
+            Ok(Some(c)) => c,
+            res => return res,
+        };
+
+        if c == '\r' {
+            c = '\n';
+            let c2 = self.reader.read_char()?;
+            if c2 != Some('\n') {
+                self.unread_char(c2);
+            }
+        }
+
+        if !reconsumed {
+            self.validate_char(c);
+        }
+
+        Ok(Some(c))
+    }
+
+    #[inline]
+    pub(crate) fn try_read_string(
+        &mut self,
+        s: &str,
+        case_sensitive: bool,
+    ) -> Result<bool, R::Error> {
+        debug_assert!(!s.is_empty());
+        debug_assert!(self.to_reconsume.is_empty());
+        self.reader.try_read_string(s, case_sensitive)
+    }
+
+    pub(crate) fn is_consumed_as_part_of_an_attribute(&self) -> bool {
+        matches!(
+            self.return_state,
+            Some(
+                State::AttributeValueDoubleQuoted
+                    | State::AttributeValueSingleQuoted
+                    | State::AttributeValueUnquoted
+            )
+        )
+    }
+
+    pub(crate) fn flush_code_points_consumed_as_character_reference(&mut self) {
+        if self.is_consumed_as_part_of_an_attribute() {
+            self.emitter.push_attribute_value(&self.temporary_buffer);
+            self.temporary_buffer.clear();
+        } else {
+            self.flush_buffer_characters();
+        }
+    }
+
+    pub(crate) fn next_input_character(&mut self) -> Result<Option<char>, R::Error> {
+        let rv = self.read_char()?;
+        self.unread_char(rv);
+        Ok(rv)
+    }
+
+    pub(crate) fn flush_buffer_characters(&mut self) {
+        self.emitter.emit_string(&self.temporary_buffer);
+        self.temporary_buffer.clear();
+    }
+}
+
+impl<R: Reader, E: Emitter> Iterator for Tokenizer<R, E> {
+    type Item = Result<E::Token, R::Error>;
+
+    fn next(&mut self) -> Option<Self::Item> {
+        loop {
+            if let Some(token) = self.emitter.pop_token() {
+                break Some(Ok(token));
+            } else if !self.eof {
+                match machine::consume(self) {
+                    Ok(ControlToken::Continue) => (),
+                    Ok(ControlToken::Eof) => {
+                        self.eof = true;
+                        self.emitter.emit_eof();
+                    }
+                    Err(e) => break Some(Err(e)),
+                }
+            } else {
+                break None;
+            }
+        }
+    }
+}
+
+/// A kind of tokenizer that directly yields tokens when used as an iterator, so `Token` instead of
+/// `Result<Token, _>`.
+///
+/// This is the return value of [`Tokenizer::infallible`].
+pub struct InfallibleTokenizer<R: Reader<Error = Never>, E: Emitter>(Tokenizer<R, E>);
+
+impl<R: Reader<Error = Never>, E: Emitter> Tokenizer<R, E> {
+    /// Statically assert that this iterator is infallible.
+    ///
+    /// Call this to get rid of error handling when parsing HTML from strings.
+    pub fn infallible(self) -> InfallibleTokenizer<R, E> {
+        InfallibleTokenizer(self)
+    }
+}
+
+impl<R: Reader<Error = Never>, E: Emitter> Iterator for InfallibleTokenizer<R, E> {
+    type Item = E::Token;
+
+    fn next(&mut self) -> Option<Self::Item> {
+        match self.0.next()? {
+            Ok(token) => Some(token),
+            Err(e) => match e {},
+        }
+    }
+}