Mirrors/servo
1
0
Fork 0
mirror of https://github.com/servo/servo.git synced 2025-08-03 04:30:10 +01:00
Code Issues Projects Releases Packages Wiki Activity

Resolves #4183 - Implemementing context-based MIME type sniffing

Browse source 
The version of the standard is not finalized at the time of this writing.
Specifications may be found here: https://mimesniff.spec.whatwg.org/#context-specific-sniffing .
This commit is contained in:
David Rajchenbach-Teller 2015-09-30 12:22:00 +02:00 • committed by Josh Matthews
parent 66c8aa8cda
commit 1e81b8c133
18 changed files with 226 additions and 93 deletions
Download patch file Download diff file Expand all files Collapse all files

127
components/net/mime_classifier.rs
View file

@ -2,6 +2,7 @@
* 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 net_traits::LoadContext;
use std::borrow::ToOwned;
pub struct MIMEClassifier {
@ -11,7 +12,8 @@ pub struct MIMEClassifier {
plaintext_classifier: GroupedClassifier,
archive_classifier: GroupedClassifier,
binary_or_plaintext: BinaryOrPlaintextClassifier,
feeds_classifier: FeedsClassifier
feeds_classifier: FeedsClassifier,
font_classifier: GroupedClassifier,
}
pub enum MediaType {
@ -33,35 +35,105 @@ pub enum NoSniffFlag {
}
impl MIMEClassifier {
//Performs MIME Type Sniffing Algorithm (section 7)
//Performs MIME Type Sniffing Algorithm (sections 7 and 8)
pub fn classify(&self,
context: LoadContext,
no_sniff_flag: NoSniffFlag,
apache_bug_flag: ApacheBugFlag,
supplied_type: &Option<(String, String)>,
data: &[u8]) -> (String, String) {
match *supplied_type {
None => self.sniff_unknown_type(no_sniff_flag, data),
Some(ref supplied_type) => {
let &(ref media_type, ref media_subtype) = supplied_type;
if MIMEClassifier::is_explicit_unknown(media_type, media_subtype) {
self.sniff_unknown_type(no_sniff_flag, data)
} else {
match no_sniff_flag {
NoSniffFlag::ON => supplied_type.clone(),
NoSniffFlag::OFF => match apache_bug_flag {
ApacheBugFlag::ON => self.sniff_text_or_data(data),
ApacheBugFlag::OFF => match MIMEClassifier::get_media_type(media_type,
media_subtype) {
Some(MediaType::Xml) => None,
Some(MediaType::Html) => self.feeds_classifier.classify(data),
Some(MediaType::Image) => self.image_classifier.classify(data),
Some(MediaType::AudioVideo) => self.audio_video_classifier.classify(data),
None => None
}.unwrap_or(supplied_type.clone())
let supplied_type_or_octet_stream = supplied_type.clone()
.unwrap_or(("application".to_owned(),
"octet-stream".to_owned()));
match context {
LoadContext::Browsing => match *supplied_type {
None => self.sniff_unknown_type(no_sniff_flag, data),
Some(ref supplied_type) => {
let &(ref media_type, ref media_subtype) = supplied_type;
if MIMEClassifier::is_explicit_unknown(media_type, media_subtype) {
self.sniff_unknown_type(no_sniff_flag, data)
} else {
match no_sniff_flag {
NoSniffFlag::ON => supplied_type.clone(),
NoSniffFlag::OFF => match apache_bug_flag {
ApacheBugFlag::ON => self.sniff_text_or_data(data),
ApacheBugFlag::OFF => match MIMEClassifier::get_media_type(media_type,
media_subtype) {
Some(MediaType::Html) => self.feeds_classifier.classify(data),
Some(MediaType::Image) => self.image_classifier.classify(data),
Some(MediaType::AudioVideo) => self.audio_video_classifier.classify(data),
Some(MediaType::Xml) | None => None,
}.unwrap_or(supplied_type.clone())
}
}
}
}
}
},
LoadContext::Image => {
// Section 8.2 Sniffing an image context
match MIMEClassifier::maybe_get_media_type(supplied_type) {
Some(MediaType::Xml) => None,
_ => self.image_classifier.classify(data),
}.unwrap_or(supplied_type_or_octet_stream)
},
LoadContext::AudioVideo => {
// Section 8.3 Sniffing an image context
match MIMEClassifier::maybe_get_media_type(supplied_type) {
Some(MediaType::Xml) => None,
_ => self.audio_video_classifier.classify(data),
}.unwrap_or(supplied_type_or_octet_stream)
},
LoadContext::Plugin => {
// 8.4 Sniffing in a plugin context
//
// This section was *not* finalized in the specs at the time
// of this implementation.
match *supplied_type {
None => ("application".to_owned(), "octet-stream".to_owned()),
_ => supplied_type_or_octet_stream,
}
},
LoadContext::Style => {
// 8.5 Sniffing in a style context
//
// This section was *not* finalized in the specs at the time
// of this implementation.
match *supplied_type {
None => ("text".to_owned(), "css".to_owned()),
_ => supplied_type_or_octet_stream,
}
},
LoadContext::Script => {
// 8.6 Sniffing in a script context
//
// This section was *not* finalized in the specs at the time
// of this implementation.
match *supplied_type {
None => ("text".to_owned(), "javascript".to_owned()),
_ => supplied_type_or_octet_stream,
}
},
LoadContext::Font => {
// 8.7 Sniffing in a font context
match MIMEClassifier::maybe_get_media_type(supplied_type) {
Some(MediaType::Xml) => None,
_ => self.font_classifier.classify(data),
}.unwrap_or(supplied_type_or_octet_stream)
},
LoadContext::TextTrack => {
// 8.8 Sniffing in a text track context
//
// This section was *not* finalized in the specs at the time
// of this implementation.
("text".to_owned(), "vtt".to_owned())
},
LoadContext::CacheManifest => {
// 8.9 Sniffing in a cache manifest context
//
// This section was *not* finalized in the specs at the time
// of this implementation.
("text".to_owned(), "cache-manifest".to_owned())
},
}
}
@ -73,7 +145,8 @@ impl MIMEClassifier {
plaintext_classifier: GroupedClassifier::plaintext_classifier(),
archive_classifier: GroupedClassifier::archive_classifier(),
binary_or_plaintext: BinaryOrPlaintextClassifier,
feeds_classifier: FeedsClassifier
feeds_classifier: FeedsClassifier,
font_classifier: GroupedClassifier::font_classifier()
}
}
@ -143,6 +216,12 @@ impl MIMEClassifier {
None
}
}
fn maybe_get_media_type(supplied_type: &Option<(String, String)>) -> Option<MediaType> {
supplied_type.as_ref().and_then(|&(ref media_type, ref media_subtype)| {
MIMEClassifier::get_media_type(media_type, media_subtype)
})
}
}
pub fn as_string_option(tup: Option<(&'static str, &'static str)>) -> Option<(String, String)> {
@ -375,8 +454,6 @@ impl GroupedClassifier {
}
}
// TODO: Use this in font context classifier
#[allow(dead_code)]
fn font_classifier() -> GroupedClassifier {
GroupedClassifier {
byte_matchers: vec![