Mirrors/servo
1
0
Fork 0
mirror of https://github.com/servo/servo.git synced 2025-08-03 20:50:07 +01:00
Code Issues Projects Releases Packages Wiki Activity

Hack the HTML lexer to skip comments

Browse source
This commit is contained in:
Brian Anderson 2012-08-21 16:01:55 -07:00
parent e15f2d50cb
commit 05efc6a612
1 changed files with 52 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

59
src/servo/parser/html_lexer.rs
View file

@ -33,6 +33,7 @@ trait HtmlLexerMethods {
fn parse_html() -> Token; fn parse_html() -> Token;
fn parse_in_normal_state(c: u8) -> Token; fn parse_in_normal_state(c: u8) -> Token;
fn parse_in_tag_state(c: u8) -> Token; fn parse_in_tag_state(c: u8) -> Token;
fn eat_until_end_of_comment();
} }
impl HtmlLexer : HtmlLexerMethods { impl HtmlLexer : HtmlLexerMethods {
@ -53,6 +54,7 @@ impl HtmlLexer : HtmlLexerMethods {
fn parse_in_normal_state(c: u8) -> Token { fn parse_in_normal_state(c: u8) -> Token {
let mut ch = c; let mut ch = c;
if ch == ('<' as u8) { if ch == ('<' as u8) {
match self.input_state.get() { match self.input_state.get() {
CoeChar(c) => { ch = c; } CoeChar(c) => { ch = c; }
@ -60,13 +62,30 @@ impl HtmlLexer : HtmlLexerMethods {
} }
if ch == ('!' as u8) { if ch == ('!' as u8) {
self.input_state.eat_whitespace(); let ch = self.input_state.get();
self.input_state.expect_ident(~"DOCTYPE"); // FIXME: This comment parsing is very hacky
self.input_state.eat_whitespace(); if ch == CoeChar('-' as u8) {
self.input_state.expect_ident(~"html"); self.eat_until_end_of_comment();
self.input_state.eat_whitespace(); return match self.input_state.get() {
self.input_state.expect('>' as u8); CoeChar(c) => self.parse_in_normal_state(c),
return Doctype; CoeEof => self.input_state.parse_err(~"FIXME")
}
} else if ch == CoeChar('D' as u8) {
self.input_state.expect_ident(~"OCTYPE");
self.input_state.eat_whitespace();
self.input_state.expect_ident(~"html");
self.input_state.eat_whitespace();
self.input_state.expect('>' as u8);
return Doctype;
} else {
self.input_state.eat_whitespace();
self.input_state.expect_ident(~"DOCTYPE");
self.input_state.eat_whitespace();
self.input_state.expect_ident(~"html");
self.input_state.eat_whitespace();
self.input_state.expect('>' as u8);
return Doctype;
}
} }
if ch == ('/' as u8) { if ch == ('/' as u8) {
@ -100,6 +119,32 @@ impl HtmlLexer : HtmlLexerMethods {
} }
} }
} }
fn eat_until_end_of_comment() {
let mut state = none;
loop {
match self.input_state.get() {
CoeChar(c) => {
match c {
'-' as u8 if state == none => {
state = some(~"-")
}
'-' as u8 if state == some(~"-") => {
state = some(~"--")
}
'>' as u8 if state == some(~"--") => {
return
}
_ => {
state = none
}
}
}
CoeEof => return
}
}
}
fn parse_in_tag_state(c: u8) -> Token { fn parse_in_tag_state(c: u8) -> Token {
let mut ch = c; let mut ch = c;