Auto merge of #517 - servo:2.0, r=nox

2.0

<!-- Reviewable:start -->
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/servo/rust-url/517)
<!-- Reviewable:end -->

Author: bors-servo

.travis.yml

@@ -1,28 +1,12 @@
 language: rust
+script: cargo test --all-features --all
 
 jobs:
   include:
-  - rust: 1.17.0
-    install:
-    # --precise requires Cargo.lock to already exist
-    - cargo update
-    # getopts is only used in tests. Its versions 0.2.16+ don’t build on 1.17.0
-    - cargo update -p getopts --precise 0.2.15
-
-    - cargo update -p unicode-normalization --precise 0.1.5
-
-    # data-url uses pub(crate) which is unstable in 1.17
-    script: cargo test --all-features -p url -p idna -p percent-encoding -p url_serde
-
+  - rust: 1.30.0
   - rust: stable
-    script: cargo test --all-features --all
-
   - rust: beta
-    script: cargo test --all-features --all
-
   - rust: nightly
-    script: cargo test --all-features --all
-
   - rust: nightly
     env: TARGET=WASM32  # For job list UI
     install: rustup target add wasm32-unknown-unknown

Cargo.toml

@@ -2,7 +2,7 @@
 
 name = "url"
 # When updating version, also modify html_root_url in the lib.rs
-version = "1.7.2"
+version = "2.0.0"
 authors = ["The rust-url developers"]
 
 description = "URL library for Rust, based on the WHATWG URL Standard"
@@ -18,7 +18,7 @@ travis-ci = { repository = "servo/rust-url" }
 appveyor = { repository = "Manishearth/rust-url" }
 
 [workspace]
-members = [".", "idna", "percent_encoding", "url_serde", "data-url"]
+members = [".", "idna", "percent_encoding", "data-url"]
 
 [[test]]
 name = "unit"
@@ -32,27 +32,16 @@ test = false
 
 [dev-dependencies]
 rustc-test = "0.3"
-rustc-serialize = "0.3"
-serde_json = ">=0.6.1, <0.9"
+serde_json = "1.0"
 
 bencher = "0.1"
 
-[features]
-query_encoding = ["encoding"]
-heap_size = ["heapsize"]
-
 [dependencies]
-encoding = {version = "0.2", optional = true}
-heapsize = {version = ">=0.4.1, <0.5", optional = true}
-idna = { version = "0.1.0", path = "./idna" }
+idna = { version = "0.2.0", path = "./idna" }
 matches = "0.1"
 percent-encoding = { version = "1.0.0", path = "./percent_encoding" }
-rustc-serialize = {version = "0.3", optional = true}
-serde = {version = ">=0.6.1, <0.9", optional = true}
+serde = {version = "1.0", optional = true}
 
 [[bench]]
 name = "parse_url"
 harness = false
-
-[package.metadata.docs.rs]
-features = ["query_encoding"]

data-url/src/forgiving_base64.rs

@@ -29,7 +29,7 @@ impl From<DecodeError<Impossible>> for InvalidBase64 {
     fn from(e: DecodeError<Impossible>) -> Self {
         match e {
             DecodeError::InvalidBase64(e) => e,
-            DecodeError::WriteError(e) => match e {}
+            DecodeError::WriteError(e) => match e {},
         }
     }
 }
@@ -46,14 +46,20 @@ pub fn decode_to_vec(input: &[u8]) -> Result<Vec<u8>, InvalidBase64> {
 }
 
 /// <https://infra.spec.whatwg.org/#forgiving-base64-decode>
-pub struct Decoder<F, E> where F: FnMut(&[u8]) -> Result<(), E> {
+pub struct Decoder<F, E>
+where
+    F: FnMut(&[u8]) -> Result<(), E>,
+{
     write_bytes: F,
     bit_buffer: u32,
     buffer_bit_length: u8,
     padding_symbols: u8,
 }
 
-impl<F, E> Decoder<F, E> where F: FnMut(&[u8]) -> Result<(), E> {
+impl<F, E> Decoder<F, E>
+where
+    F: FnMut(&[u8]) -> Result<(), E>,
+{
     pub fn new(write_bytes: F) -> Self {
         Self {
             write_bytes,
@@ -72,12 +78,12 @@ impl<F, E> Decoder<F, E> where F: FnMut(&[u8]) -> Result<(), E> {
 
                 // Remove ASCII whitespace
                 if matches!(byte, b' ' | b'\t' | b'\n' | b'\r' | b'\x0C') {
-                    continue
+                    continue;
                 }
 
                 if byte == b'=' {
                     self.padding_symbols = self.padding_symbols.saturating_add(1);
-                    continue
+                    continue;
                 }
 
                 Err(InvalidBase64Details::UnexpectedSymbol(byte))?
@@ -115,39 +121,30 @@ impl<F, E> Decoder<F, E> where F: FnMut(&[u8]) -> Result<(), E> {
             (12, 2) | (12, 0) => {
                 // A multiple of four of alphabet symbols, followed by two more symbols,
                 // optionally followed by two padding characters (which make a total multiple of four).
-                let byte_buffer = [
-                    (self.bit_buffer >> 4) as u8,
-                ];
+                let byte_buffer = [(self.bit_buffer >> 4) as u8];
                 (self.write_bytes)(&byte_buffer).map_err(DecodeError::WriteError)?;
             }
             (18, 1) | (18, 0) => {
                 // A multiple of four of alphabet symbols, followed by three more symbols,
                 // optionally followed by one padding character (which make a total multiple of four).
-                let byte_buffer = [
-                    (self.bit_buffer >> 10) as u8,
-                    (self.bit_buffer >> 2) as u8,
-                ];
+                let byte_buffer = [(self.bit_buffer >> 10) as u8, (self.bit_buffer >> 2) as u8];
                 (self.write_bytes)(&byte_buffer).map_err(DecodeError::WriteError)?;
             }
-            (6, _) => {
-                Err(InvalidBase64Details::LoneAlphabetSymbol)?
-            }
-            _ => {
-                Err(InvalidBase64Details::Padding)?
-            }
+            (6, _) => Err(InvalidBase64Details::LoneAlphabetSymbol)?,
+            _ => Err(InvalidBase64Details::Padding)?,
         }
         Ok(())
     }
 }
 
-
 /// Generated by `make_base64_decode_table.py` based on "Table 1: The Base 64 Alphabet"
 /// at <https://tools.ietf.org/html/rfc4648#section-4>
 ///
 /// Array indices are the byte value of symbols.
 /// Array values are their positions in the base64 alphabet,
 /// or -1 for symbols not in the alphabet.
 /// The position contributes 6 bits to the decoded bytes.
+#[rustfmt::skip]
 const BASE64_DECODE_TABLE: [i8; 256] = [
     -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
     -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,

data-url/src/lib.rs

@@ -15,14 +15,15 @@
 //! assert!(fragment.is_none());
 //! ```
 
-#[macro_use] extern crate matches;
+#[macro_use]
+extern crate matches;
 
 macro_rules! require {
     ($condition: expr) => {
         if !$condition {
-            return None
+            return None;
         }
-    }
+    };
 }
 
 pub mod forgiving_base64;
@@ -53,7 +54,11 @@ impl<'a> DataUrl<'a> {
 
         let (mime_type, base64) = parse_header(from_colon_to_comma);
 
-        Ok(DataUrl { mime_type, base64, encoded_body_plus_fragment })
+        Ok(DataUrl {
+            mime_type,
+            base64,
+            encoded_body_plus_fragment,
+        })
     }
 
     pub fn mime_type(&self) -> &mime::Mime {
@@ -62,9 +67,12 @@ impl<'a> DataUrl<'a> {
 
     /// Streaming-decode the data URL’s body to `write_body_bytes`,
     /// and return the URL’s fragment identifier if it has one.
-    pub fn decode<F, E>(&self, write_body_bytes: F)
-                        -> Result<Option<FragmentIdentifier<'a>>, forgiving_base64::DecodeError<E>>
-        where F: FnMut(&[u8]) -> Result<(), E>
+    pub fn decode<F, E>(
+        &self,
+        write_body_bytes: F,
+    ) -> Result<Option<FragmentIdentifier<'a>>, forgiving_base64::DecodeError<E>>
+    where
+        F: FnMut(&[u8]) -> Result<(), E>,
     {
         if self.base64 {
             decode_with_base64(self.encoded_body_plus_fragment, write_body_bytes)
@@ -75,9 +83,9 @@ impl<'a> DataUrl<'a> {
     }
 
     /// Return the decoded body, and the URL’s fragment identifier if it has one.
-    pub fn decode_to_vec(&self)
-        -> Result<(Vec<u8>, Option<FragmentIdentifier<'a>>), forgiving_base64::InvalidBase64>
-    {
+    pub fn decode_to_vec(
+        &self,
+    ) -> Result<(Vec<u8>, Option<FragmentIdentifier<'a>>), forgiving_base64::InvalidBase64> {
         let mut body = Vec::new();
         let fragment = self.decode(|bytes| Ok(body.extend_from_slice(bytes)))?;
         Ok((body, fragment))
@@ -100,7 +108,7 @@ impl<'a> FragmentIdentifier<'a> {
                     percent_encode(byte, &mut string)
                 }
                 // Printable ASCII
-                _ => string.push(byte as char)
+                _ => string.push(byte as char),
             }
         }
         string
@@ -125,7 +133,9 @@ fn pretend_parse_data_url(input: &str) -> Option<&str> {
     let mut bytes = left_trimmed.bytes();
     {
         // Ignore ASCII tabs or newlines like the URL parser would
-        let mut iter = bytes.by_ref().filter(|&byte| !matches!(byte, b'\t' | b'\n' | b'\r'));
+        let mut iter = bytes
+            .by_ref()
+            .filter(|&byte| !matches!(byte, b'\t' | b'\n' | b'\r'));
         require!(iter.next()?.to_ascii_lowercase() == b'd');
         require!(iter.next()?.to_ascii_lowercase() == b'a');
         require!(iter.next()?.to_ascii_lowercase() == b't');
@@ -142,10 +152,10 @@ fn pretend_parse_data_url(input: &str) -> Option<&str> {
 fn find_comma_before_fragment(after_colon: &str) -> Option<(&str, &str)> {
     for (i, byte) in after_colon.bytes().enumerate() {
         if byte == b',' {
-            return Some((&after_colon[..i], &after_colon[i + 1..]))
+            return Some((&after_colon[..i], &after_colon[i + 1..]));
         }
         if byte == b'#' {
-            break
+            break;
         }
     }
     None
@@ -187,18 +197,16 @@ fn parse_header(from_colon_to_comma: &str) -> (mime::Mime, bool) {
             }
 
             // Printable ASCII
-            _ => string.push(byte as char)
+            _ => string.push(byte as char),
         }
     }
 
     // FIXME: does Mime::from_str match the MIME Sniffing Standard’s parsing algorithm?
     // <https://mimesniff.spec.whatwg.org/#parse-a-mime-type>
-    let mime_type = string.parse().unwrap_or_else(|_| {
-        mime::Mime {
-            type_: String::from("text"),
-            subtype: String::from("plain"),
-            parameters: vec![(String::from("charset"), String::from("US-ASCII"))],
-        }
+    let mime_type = string.parse().unwrap_or_else(|_| mime::Mime {
+        type_: String::from("text"),
+        subtype: String::from("plain"),
+        parameters: vec![(String::from("charset"), String::from("US-ASCII"))],
     });
 
     (mime_type, base64)
@@ -209,7 +217,9 @@ fn remove_base64_suffix(s: &str) -> Option<&str> {
     let mut bytes = s.bytes();
     {
         // Ignore ASCII tabs or newlines like the URL parser would
-        let iter = bytes.by_ref().filter(|&byte| !matches!(byte, b'\t' | b'\n' | b'\r'));
+        let iter = bytes
+            .by_ref()
+            .filter(|&byte| !matches!(byte, b'\t' | b'\n' | b'\r'));
 
         // Search from the end
         let mut iter = iter.rev();
@@ -240,9 +250,12 @@ fn percent_encode(byte: u8, string: &mut String) {
 /// Anything that would have been UTF-8 percent-encoded by the URL parser
 /// would be percent-decoded here.
 /// We skip that round-trip and pass it through unchanged.
-fn decode_without_base64<F, E>(encoded_body_plus_fragment: &str, mut write_bytes: F)
-                               -> Result<Option<FragmentIdentifier>, E>
-    where F: FnMut(&[u8]) -> Result<(), E>
+fn decode_without_base64<F, E>(
+    encoded_body_plus_fragment: &str,
+    mut write_bytes: F,
+) -> Result<Option<FragmentIdentifier>, E>
+where
+    F: FnMut(&[u8]) -> Result<(), E>,
 {
     let bytes = encoded_body_plus_fragment.as_bytes();
     let mut slice_start = 0;
@@ -275,11 +288,11 @@ fn decode_without_base64<F, E>(encoded_body_plus_fragment: &str, mut write_bytes
                 b'#' => {
                     let fragment_start = i + 1;
                     let fragment = &encoded_body_plus_fragment[fragment_start..];
-                    return Ok(Some(FragmentIdentifier(fragment)))
+                    return Ok(Some(FragmentIdentifier(fragment)));
                 }
 
                 // Ignore over '\t' | '\n' | '\r'
-                _ => slice_start = i + 1
+                _ => slice_start = i + 1,
             }
         }
     }
@@ -290,9 +303,12 @@ fn decode_without_base64<F, E>(encoded_body_plus_fragment: &str, mut write_bytes
 /// `decode_without_base64()` composed with
 /// <https://infra.spec.whatwg.org/#isomorphic-decode> composed with
 /// <https://infra.spec.whatwg.org/#forgiving-base64-decode>.
-fn decode_with_base64<F, E>(encoded_body_plus_fragment: &str, write_bytes: F)
-                            -> Result<Option<FragmentIdentifier>, forgiving_base64::DecodeError<E>>
-    where F: FnMut(&[u8]) -> Result<(), E>
+fn decode_with_base64<F, E>(
+    encoded_body_plus_fragment: &str,
+    write_bytes: F,
+) -> Result<Option<FragmentIdentifier>, forgiving_base64::DecodeError<E>>
+where
+    F: FnMut(&[u8]) -> Result<(), E>,
 {
     let mut decoder = forgiving_base64::Decoder::new(write_bytes);
     let fragment = decode_without_base64(encoded_body_plus_fragment, |bytes| decoder.feed(bytes))?;