mirror of
https://github.com/servo/servo.git
synced 2025-07-22 23:03:42 +01:00
Move ParserContext to script::parse
This commit is contained in:
Anthony Ramine
parent
881f7f4de7
commit
02162a8bda
3 changed files with 174 additions and 173 deletions
|
|
@ -5,44 +5,29 @@
|
|||
//! The bulk of the HTML parser integration is in `script::parse::html`.
|
||||
//! This module is mostly about its interaction with DOM memory management.
|
||||
|
||||
use document_loader::LoadType;
|
||||
use dom::bindings::cell::DOMRefCell;
|
||||
use dom::bindings::codegen::Bindings::DocumentBinding::DocumentMethods;
|
||||
use dom::bindings::codegen::Bindings::HTMLImageElementBinding::HTMLImageElementMethods;
|
||||
use dom::bindings::codegen::Bindings::NodeBinding::NodeMethods;
|
||||
use dom::bindings::codegen::Bindings::ServoHTMLParserBinding;
|
||||
use dom::bindings::inheritance::Castable;
|
||||
use dom::bindings::js::{JS, Root};
|
||||
use dom::bindings::refcounted::Trusted;
|
||||
use dom::bindings::reflector::reflect_dom_object;
|
||||
use dom::bindings::str::DOMString;
|
||||
use dom::bindings::trace::JSTraceable;
|
||||
use dom::document::Document;
|
||||
use dom::globalscope::GlobalScope;
|
||||
use dom::htmlimageelement::HTMLImageElement;
|
||||
use dom::node::Node;
|
||||
use dom::servoparser::ServoParser;
|
||||
use dom::window::Window;
|
||||
use encoding::all::UTF_8;
|
||||
use encoding::types::{DecoderTrap, Encoding};
|
||||
use html5ever::tokenizer;
|
||||
use html5ever::tree_builder;
|
||||
use html5ever::tree_builder::{TreeBuilder, TreeBuilderOpts};
|
||||
use hyper::header::ContentType;
|
||||
use hyper::mime::{Mime, SubLevel, TopLevel};
|
||||
use hyper_serde::Serde;
|
||||
use js::jsapi::JSTracer;
|
||||
use msg::constellation_msg::PipelineId;
|
||||
use net_traits::{AsyncResponseListener, Metadata, NetworkError};
|
||||
use network_listener::PreInvoke;
|
||||
use parse::{Parser, ParserRef, TrustedParser};
|
||||
use parse::{Parser, ParserRef};
|
||||
use profile_traits::time::{TimerMetadata, TimerMetadataFrameType, TimerMetadataReflowType, profile};
|
||||
use profile_traits::time::ProfilerCategory;
|
||||
use script_thread::ScriptThread;
|
||||
use std::cell::Cell;
|
||||
use std::default::Default;
|
||||
use url::Url;
|
||||
use util::resource_files::read_resource_file;
|
||||
|
||||
#[must_root]
|
||||
#[derive(JSTraceable, HeapSizeOf)]
|
||||
|
|
@ -61,161 +46,6 @@ pub struct FragmentContext<'a> {
|
|||
|
||||
pub type Tokenizer = tokenizer::Tokenizer<TreeBuilder<JS<Node>, Sink>>;
|
||||
|
||||
/// The context required for asynchronously fetching a document and parsing it progressively.
|
||||
pub struct ParserContext {
|
||||
/// The parser that initiated the request.
|
||||
parser: Option<TrustedParser>,
|
||||
/// Is this a synthesized document
|
||||
is_synthesized_document: bool,
|
||||
/// The pipeline associated with this document.
|
||||
id: PipelineId,
|
||||
/// The URL for this document.
|
||||
url: Url,
|
||||
}
|
||||
|
||||
impl ParserContext {
|
||||
pub fn new(id: PipelineId, url: Url) -> ParserContext {
|
||||
ParserContext {
|
||||
parser: None,
|
||||
is_synthesized_document: false,
|
||||
id: id,
|
||||
url: url,
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
impl AsyncResponseListener for ParserContext {
|
||||
fn headers_available(&mut self, meta_result: Result<Metadata, NetworkError>) {
|
||||
let mut ssl_error = None;
|
||||
let metadata = match meta_result {
|
||||
Ok(meta) => Some(meta),
|
||||
Err(NetworkError::SslValidation(url, reason)) => {
|
||||
ssl_error = Some(reason);
|
||||
let mut meta = Metadata::default(url);
|
||||
let mime: Option<Mime> = "text/html".parse().ok();
|
||||
meta.set_content_type(mime.as_ref());
|
||||
Some(meta)
|
||||
},
|
||||
Err(_) => None,
|
||||
};
|
||||
let content_type =
|
||||
metadata.clone().and_then(|meta| meta.content_type).map(Serde::into_inner);
|
||||
let parser = match ScriptThread::page_headers_available(&self.id,
|
||||
metadata) {
|
||||
Some(parser) => parser,
|
||||
None => return,
|
||||
};
|
||||
|
||||
let parser = parser.r();
|
||||
let servo_parser = parser.as_servo_parser();
|
||||
self.parser = Some(match parser {
|
||||
ParserRef::HTML(parser) => TrustedParser::HTML(
|
||||
Trusted::new(parser)),
|
||||
ParserRef::XML(parser) => TrustedParser::XML(
|
||||
Trusted::new(parser)),
|
||||
});
|
||||
|
||||
match content_type {
|
||||
Some(ContentType(Mime(TopLevel::Image, _, _))) => {
|
||||
self.is_synthesized_document = true;
|
||||
let page = "<html><body></body></html>".into();
|
||||
servo_parser.push_input_chunk(page);
|
||||
parser.parse_sync();
|
||||
|
||||
let doc = servo_parser.document();
|
||||
let doc_body = Root::upcast::<Node>(doc.GetBody().unwrap());
|
||||
let img = HTMLImageElement::new(atom!("img"), None, doc);
|
||||
img.SetSrc(DOMString::from(self.url.to_string()));
|
||||
doc_body.AppendChild(&Root::upcast::<Node>(img)).expect("Appending failed");
|
||||
|
||||
},
|
||||
Some(ContentType(Mime(TopLevel::Text, SubLevel::Plain, _))) => {
|
||||
// https://html.spec.whatwg.org/multipage/#read-text
|
||||
let page = "<pre>\n".into();
|
||||
servo_parser.push_input_chunk(page);
|
||||
parser.parse_sync();
|
||||
parser.set_plaintext_state();
|
||||
},
|
||||
Some(ContentType(Mime(TopLevel::Text, SubLevel::Html, _))) => { // Handle text/html
|
||||
if let Some(reason) = ssl_error {
|
||||
self.is_synthesized_document = true;
|
||||
let page_bytes = read_resource_file("badcert.html").unwrap();
|
||||
let page = String::from_utf8(page_bytes).unwrap();
|
||||
let page = page.replace("${reason}", &reason);
|
||||
servo_parser.push_input_chunk(page);
|
||||
parser.parse_sync();
|
||||
}
|
||||
},
|
||||
Some(ContentType(Mime(TopLevel::Text, SubLevel::Xml, _))) => {}, // Handle text/xml
|
||||
Some(ContentType(Mime(toplevel, sublevel, _))) => {
|
||||
if toplevel.as_str() == "application" && sublevel.as_str() == "xhtml+xml" {
|
||||
// Handle xhtml (application/xhtml+xml).
|
||||
return;
|
||||
}
|
||||
|
||||
// Show warning page for unknown mime types.
|
||||
let page = format!("<html><body><p>Unknown content type ({}/{}).</p></body></html>",
|
||||
toplevel.as_str(), sublevel.as_str());
|
||||
self.is_synthesized_document = true;
|
||||
servo_parser.push_input_chunk(page);
|
||||
parser.parse_sync();
|
||||
},
|
||||
None => {
|
||||
// No content-type header.
|
||||
// Merge with #4212 when fixed.
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
fn data_available(&mut self, payload: Vec<u8>) {
|
||||
if !self.is_synthesized_document {
|
||||
// FIXME: use Vec<u8> (html5ever #34)
|
||||
let data = UTF_8.decode(&payload, DecoderTrap::Replace).unwrap();
|
||||
let parser = match self.parser.as_ref() {
|
||||
Some(parser) => parser.root(),
|
||||
None => return,
|
||||
};
|
||||
parser.r().parse_chunk(data);
|
||||
}
|
||||
}
|
||||
|
||||
fn response_complete(&mut self, status: Result<(), NetworkError>) {
|
||||
let parser = match self.parser.as_ref() {
|
||||
Some(parser) => parser.root(),
|
||||
None => return,
|
||||
};
|
||||
|
||||
if let Err(NetworkError::Internal(ref reason)) = status {
|
||||
// Show an error page for network errors,
|
||||
// certificate errors are handled earlier.
|
||||
self.is_synthesized_document = true;
|
||||
let parser = parser.r();
|
||||
let page_bytes = read_resource_file("neterror.html").unwrap();
|
||||
let page = String::from_utf8(page_bytes).unwrap();
|
||||
let page = page.replace("${reason}", reason);
|
||||
parser.as_servo_parser().push_input_chunk(page);
|
||||
parser.parse_sync();
|
||||
} else if let Err(err) = status {
|
||||
// TODO(Savago): we should send a notification to callers #5463.
|
||||
debug!("Failed to load page URL {}, error: {:?}", self.url, err);
|
||||
}
|
||||
|
||||
let parser = parser.r();
|
||||
let servo_parser = parser.as_servo_parser();
|
||||
|
||||
servo_parser.document()
|
||||
.finish_load(LoadType::PageSource(self.url.clone()));
|
||||
|
||||
servo_parser.mark_last_chunk_received();
|
||||
if !parser.is_suspended() {
|
||||
parser.parse_sync();
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
impl PreInvoke for ParserContext {
|
||||
}
|
||||
|
||||
#[dom_struct]
|
||||
pub struct ServoHTMLParser {
|
||||
servoparser: ServoParser,
|
||||
|
|
|
|||
|
|
@ -2,15 +2,33 @@
|
|||
* License, v. 2.0. If a copy of the MPL was not distributed with this
|
||||
* file, You can obtain one at http://mozilla.org/MPL/2.0/. */
|
||||
|
||||
use document_loader::LoadType;
|
||||
use dom::bindings::codegen::Bindings::DocumentBinding::DocumentMethods;
|
||||
use dom::bindings::codegen::Bindings::HTMLImageElementBinding::HTMLImageElementMethods;
|
||||
use dom::bindings::codegen::Bindings::NodeBinding::NodeMethods;
|
||||
use dom::bindings::inheritance::Castable;
|
||||
use dom::bindings::js::{JS, Root};
|
||||
use dom::bindings::refcounted::Trusted;
|
||||
use dom::bindings::str::DOMString;
|
||||
use dom::htmlimageelement::HTMLImageElement;
|
||||
use dom::node::Node;
|
||||
use dom::servohtmlparser::ServoHTMLParser;
|
||||
use dom::servoparser::ServoParser;
|
||||
use dom::servoxmlparser::ServoXMLParser;
|
||||
use dom::window::Window;
|
||||
use encoding::all::UTF_8;
|
||||
use encoding::types::{DecoderTrap, Encoding};
|
||||
use hyper::header::ContentType;
|
||||
use hyper::mime::{Mime, SubLevel, TopLevel};
|
||||
use hyper_serde::Serde;
|
||||
use msg::constellation_msg::PipelineId;
|
||||
use net_traits::{AsyncResponseListener, Metadata, NetworkError};
|
||||
use network_listener::PreInvoke;
|
||||
use script_thread::ScriptThread;
|
||||
use std::cell::UnsafeCell;
|
||||
use std::ptr;
|
||||
use url::Url;
|
||||
use util::resource_files::read_resource_file;
|
||||
|
||||
pub mod html;
|
||||
pub mod xml;
|
||||
|
|
@ -160,3 +178,157 @@ impl<'a> ParserRef<'a> {
|
|||
}
|
||||
}
|
||||
|
||||
/// The context required for asynchronously fetching a document
|
||||
/// and parsing it progressively.
|
||||
pub struct ParserContext {
|
||||
/// The parser that initiated the request.
|
||||
parser: Option<TrustedParser>,
|
||||
/// Is this a synthesized document
|
||||
is_synthesized_document: bool,
|
||||
/// The pipeline associated with this document.
|
||||
id: PipelineId,
|
||||
/// The URL for this document.
|
||||
url: Url,
|
||||
}
|
||||
|
||||
impl ParserContext {
|
||||
pub fn new(id: PipelineId, url: Url) -> ParserContext {
|
||||
ParserContext {
|
||||
parser: None,
|
||||
is_synthesized_document: false,
|
||||
id: id,
|
||||
url: url,
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
impl AsyncResponseListener for ParserContext {
|
||||
fn headers_available(&mut self, meta_result: Result<Metadata, NetworkError>) {
|
||||
let mut ssl_error = None;
|
||||
let metadata = match meta_result {
|
||||
Ok(meta) => Some(meta),
|
||||
Err(NetworkError::SslValidation(url, reason)) => {
|
||||
ssl_error = Some(reason);
|
||||
let mut meta = Metadata::default(url);
|
||||
let mime: Option<Mime> = "text/html".parse().ok();
|
||||
meta.set_content_type(mime.as_ref());
|
||||
Some(meta)
|
||||
},
|
||||
Err(_) => None,
|
||||
};
|
||||
let content_type =
|
||||
metadata.clone().and_then(|meta| meta.content_type).map(Serde::into_inner);
|
||||
let parser = match ScriptThread::page_headers_available(&self.id,
|
||||
metadata) {
|
||||
Some(parser) => parser,
|
||||
None => return,
|
||||
};
|
||||
|
||||
let parser = parser.r();
|
||||
let servo_parser = parser.as_servo_parser();
|
||||
self.parser = Some(match parser {
|
||||
ParserRef::HTML(parser) => TrustedParser::HTML(
|
||||
Trusted::new(parser)),
|
||||
ParserRef::XML(parser) => TrustedParser::XML(
|
||||
Trusted::new(parser)),
|
||||
});
|
||||
|
||||
match content_type {
|
||||
Some(ContentType(Mime(TopLevel::Image, _, _))) => {
|
||||
self.is_synthesized_document = true;
|
||||
let page = "<html><body></body></html>".into();
|
||||
servo_parser.push_input_chunk(page);
|
||||
parser.parse_sync();
|
||||
|
||||
let doc = servo_parser.document();
|
||||
let doc_body = Root::upcast::<Node>(doc.GetBody().unwrap());
|
||||
let img = HTMLImageElement::new(atom!("img"), None, doc);
|
||||
img.SetSrc(DOMString::from(self.url.to_string()));
|
||||
doc_body.AppendChild(&Root::upcast::<Node>(img)).expect("Appending failed");
|
||||
|
||||
},
|
||||
Some(ContentType(Mime(TopLevel::Text, SubLevel::Plain, _))) => {
|
||||
// https://html.spec.whatwg.org/multipage/#read-text
|
||||
let page = "<pre>\n".into();
|
||||
servo_parser.push_input_chunk(page);
|
||||
parser.parse_sync();
|
||||
parser.set_plaintext_state();
|
||||
},
|
||||
Some(ContentType(Mime(TopLevel::Text, SubLevel::Html, _))) => { // Handle text/html
|
||||
if let Some(reason) = ssl_error {
|
||||
self.is_synthesized_document = true;
|
||||
let page_bytes = read_resource_file("badcert.html").unwrap();
|
||||
let page = String::from_utf8(page_bytes).unwrap();
|
||||
let page = page.replace("${reason}", &reason);
|
||||
servo_parser.push_input_chunk(page);
|
||||
parser.parse_sync();
|
||||
}
|
||||
},
|
||||
Some(ContentType(Mime(TopLevel::Text, SubLevel::Xml, _))) => {}, // Handle text/xml
|
||||
Some(ContentType(Mime(toplevel, sublevel, _))) => {
|
||||
if toplevel.as_str() == "application" && sublevel.as_str() == "xhtml+xml" {
|
||||
// Handle xhtml (application/xhtml+xml).
|
||||
return;
|
||||
}
|
||||
|
||||
// Show warning page for unknown mime types.
|
||||
let page = format!("<html><body><p>Unknown content type ({}/{}).</p></body></html>",
|
||||
toplevel.as_str(), sublevel.as_str());
|
||||
self.is_synthesized_document = true;
|
||||
servo_parser.push_input_chunk(page);
|
||||
parser.parse_sync();
|
||||
},
|
||||
None => {
|
||||
// No content-type header.
|
||||
// Merge with #4212 when fixed.
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
fn data_available(&mut self, payload: Vec<u8>) {
|
||||
if !self.is_synthesized_document {
|
||||
// FIXME: use Vec<u8> (html5ever #34)
|
||||
let data = UTF_8.decode(&payload, DecoderTrap::Replace).unwrap();
|
||||
let parser = match self.parser.as_ref() {
|
||||
Some(parser) => parser.root(),
|
||||
None => return,
|
||||
};
|
||||
parser.r().parse_chunk(data);
|
||||
}
|
||||
}
|
||||
|
||||
fn response_complete(&mut self, status: Result<(), NetworkError>) {
|
||||
let parser = match self.parser.as_ref() {
|
||||
Some(parser) => parser.root(),
|
||||
None => return,
|
||||
};
|
||||
|
||||
if let Err(NetworkError::Internal(ref reason)) = status {
|
||||
// Show an error page for network errors,
|
||||
// certificate errors are handled earlier.
|
||||
self.is_synthesized_document = true;
|
||||
let parser = parser.r();
|
||||
let page_bytes = read_resource_file("neterror.html").unwrap();
|
||||
let page = String::from_utf8(page_bytes).unwrap();
|
||||
let page = page.replace("${reason}", reason);
|
||||
parser.as_servo_parser().push_input_chunk(page);
|
||||
parser.parse_sync();
|
||||
} else if let Err(err) = status {
|
||||
// TODO(Savago): we should send a notification to callers #5463.
|
||||
debug!("Failed to load page URL {}, error: {:?}", self.url, err);
|
||||
}
|
||||
|
||||
let parser = parser.r();
|
||||
let servo_parser = parser.as_servo_parser();
|
||||
|
||||
servo_parser.document()
|
||||
.finish_load(LoadType::PageSource(self.url.clone()));
|
||||
|
||||
servo_parser.mark_last_chunk_received();
|
||||
if !parser.is_suspended() {
|
||||
parser.parse_sync();
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
impl PreInvoke for ParserContext {}
|
||||
|
|
|
|||
|
|
@ -44,7 +44,6 @@ use dom::htmlanchorelement::HTMLAnchorElement;
|
|||
use dom::node::{Node, NodeDamage, window_from_node};
|
||||
use dom::serviceworker::TrustedServiceWorkerAddress;
|
||||
use dom::serviceworkerregistration::ServiceWorkerRegistration;
|
||||
use dom::servohtmlparser::ParserContext;
|
||||
use dom::uievent::UIEvent;
|
||||
use dom::window::{ReflowReason, Window};
|
||||
use dom::worker::TrustedWorkerAddress;
|
||||
|
|
@ -71,7 +70,7 @@ use net_traits::{IpcSend, LoadData as NetLoadData};
|
|||
use net_traits::bluetooth_thread::BluetoothMethodMsg;
|
||||
use net_traits::image_cache_thread::{ImageCacheChan, ImageCacheResult, ImageCacheThread};
|
||||
use network_listener::NetworkListener;
|
||||
use parse::ParserRoot;
|
||||
use parse::{ParserContext, ParserRoot};
|
||||
use parse::html::{ParseContext, parse_html};
|
||||
use parse::xml::{self, parse_xml};
|
||||
use profile_traits::mem::{self, OpaqueSender, Report, ReportKind, ReportsChan};
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue