Merge pull request #780 from Mingun/end-attributes

Allow attributes in the `Event::End` and fix `.error_position()`

Author: Mingun

Changelog.md

@@ -19,9 +19,15 @@
 
 - [#781]: Fix conditions to start CDATA section. Only uppercase `<![CDATA[` can start it.
   Previously any case was allowed.
+- [#780]: Fixed incorrect `.error_position()` when encountering syntax error for open or self-closed tag.
 
 ### Misc Changes
 
+- [#780]: `reader::Parser`, `reader::ElementParser` and `reader::PiParser` moved to the new module `parser`.
+- [#776]: Allow to have attributes in the end tag for compatibility reasons with Adobe Flash XML parser.
+
+[#776]: https://github.com/tafia/quick-xml/issues/776
+[#780]: https://github.com/tafia/quick-xml/pull/780
 [#781]: https://github.com/tafia/quick-xml/pull/781
 
 

README.md

@@ -40,7 +40,7 @@ loop {
     // when the input is a &str or a &[u8], we don't actually need to use another
     // buffer, we could directly call `reader.read_event()`
     match reader.read_event_into(&mut buf) {
-        Err(e) => panic!("Error at position {}: {:?}", reader.buffer_position(), e),
+        Err(e) => panic!("Error at position {}: {:?}", reader.error_position(), e),
         // exits the loop when reaching end of file
         Ok(Event::Eof) => break,
 
@@ -98,7 +98,7 @@ loop {
         Ok(Event::Eof) => break,
         // we can either move or borrow the event to write, depending on your use-case
         Ok(e) => assert!(writer.write_event(e).is_ok()),
-        Err(e) => panic!("Error at position {}: {:?}", reader.buffer_position(), e),
+        Err(e) => panic!("Error at position {}: {:?}", reader.error_position(), e),
     }
 }
 

examples/custom_entities.rs

@@ -68,7 +68,7 @@ fn main() -> Result<(), Box<dyn std::error::Error>> {
                 );
             }
             Ok(Event::Eof) => break,
-            Err(e) => panic!("Error at position {}: {:?}", reader.buffer_position(), e),
+            Err(e) => panic!("Error at position {}: {:?}", reader.error_position(), e),
             _ => (),
         }
     }

examples/read_buffered.rs

@@ -23,7 +23,7 @@ fn main() -> Result<(), quick_xml::Error> {
                 count += 1;
             }
             Ok(Event::Eof) => break, // exits the loop when reaching end of file
-            Err(e) => panic!("Error at position {}: {:?}", reader.buffer_position(), e),
+            Err(e) => panic!("Error at position {}: {:?}", reader.error_position(), e),
             _ => (), // There are several other `Event`s we do not consider here
         }
     }

examples/read_texts.rs

@@ -18,7 +18,7 @@ fn main() {
                 println!("{:?}", txt);
             }
             Ok(Event::Eof) => break, // exits the loop when reaching end of file
-            Err(e) => panic!("Error at position {}: {:?}", reader.buffer_position(), e),
+            Err(e) => panic!("Error at position {}: {:?}", reader.error_position(), e),
             _ => (), // There are several other `Event`s we do not consider here
         }
     }

src/lib.rs

@@ -59,6 +59,7 @@ pub mod errors;
 pub mod escape;
 pub mod events;
 pub mod name;
+pub mod parser;
 pub mod reader;
 #[cfg(feature = "serialize")]
 pub mod se;

src/reader/element.rs renamed to src/parser/element.rs

@@ -1,7 +1,7 @@
 //! Contains a parser for an XML element.
 
 use crate::errors::SyntaxError;
-use crate::reader::Parser;
+use crate::parser::Parser;
 
 /// A parser that search a `>` symbol in the slice outside of quoted regions.
 ///
@@ -25,7 +25,7 @@ use crate::reader::Parser;
 ///
 /// ```
 /// # use pretty_assertions::assert_eq;
-/// use quick_xml::reader::{ElementParser, Parser};
+/// use quick_xml::parser::{ElementParser, Parser};
 ///
 /// let mut parser = ElementParser::default();
 ///

src/parser/mod.rs

@@ -0,0 +1,29 @@
+//! Contains low-level parsers of different XML pieces.
+
+use crate::errors::SyntaxError;
+
+mod element;
+mod pi;
+
+pub use element::ElementParser;
+pub use pi::PiParser;
+
+/// Used to decouple reading of data from data source and parsing XML structure from it.
+/// This is a state preserved between getting chunks of bytes from the reader.
+///
+/// This trait is implemented for every parser that processes piece of XML grammar.
+pub trait Parser {
+    /// Process new data and try to determine end of the parsed thing.
+    ///
+    /// Returns position of the end of thing in `bytes` in case of successful search
+    /// and `None` otherwise.
+    ///
+    /// # Parameters
+    /// - `bytes`: a slice to find the end of a thing.
+    ///   Should contain text in ASCII-compatible encoding
+    fn feed(&mut self, bytes: &[u8]) -> Option<usize>;
+
+    /// Returns parse error produced by this parser in case of reaching end of
+    /// input without finding the end of a parsed thing.
+    fn eof_error() -> SyntaxError;
+}

src/reader/pi.rs renamed to src/parser/pi.rs

@@ -1,7 +1,7 @@
 //! Contains a parser for an XML processing instruction.
 
 use crate::errors::SyntaxError;
-use crate::reader::Parser;
+use crate::parser::Parser;
 
 /// A parser that search a `?>` sequence in the slice.
 ///
@@ -19,7 +19,7 @@ use crate::reader::Parser;
 ///
 /// ```
 /// # use pretty_assertions::assert_eq;
-/// use quick_xml::reader::{Parser, PiParser};
+/// use quick_xml::parser::{Parser, PiParser};
 ///
 /// let mut parser = PiParser::default();
 ///

src/reader/async_tokio.rs

@@ -7,10 +7,9 @@ use tokio::io::{self, AsyncBufRead, AsyncBufReadExt};
 use crate::errors::{Error, Result, SyntaxError};
 use crate::events::Event;
 use crate::name::{QName, ResolveResult};
+use crate::parser::{ElementParser, Parser, PiParser};
 use crate::reader::buffered_reader::impl_buffered_source;
-use crate::reader::{
-    BangType, ElementParser, NsReader, ParseState, Parser, PiParser, ReadTextResult, Reader, Span,
-};
+use crate::reader::{BangType, NsReader, ParseState, ReadTextResult, Reader, Span};
 use crate::utils::is_whitespace;
 
 /// A struct for read XML asynchronously from an [`AsyncBufRead`].
@@ -59,7 +58,7 @@ impl<R: AsyncBufRead + Unpin> Reader<R> {
     ///     match reader.read_event_into_async(&mut buf).await {
     ///         Ok(Event::Start(_)) => count += 1,
     ///         Ok(Event::Text(e)) => txt.push(e.unescape().unwrap().into_owned()),
-    ///         Err(e) => panic!("Error at position {}: {:?}", reader.buffer_position(), e),
+    ///         Err(e) => panic!("Error at position {}: {:?}", reader.error_position(), e),
     ///         Ok(Event::Eof) => break,
     ///         _ => (),
     ///     }

src/reader/buffered_reader.rs

@@ -8,7 +8,8 @@ use std::path::Path;
 use crate::errors::{Error, Result};
 use crate::events::Event;
 use crate::name::QName;
-use crate::reader::{BangType, Parser, ReadTextResult, Reader, Span, XmlSource};
+use crate::parser::Parser;
+use crate::reader::{BangType, ReadTextResult, Reader, Span, XmlSource};
 use crate::utils::is_whitespace;
 
 macro_rules! impl_buffered_source {
@@ -100,54 +101,6 @@ macro_rules! impl_buffered_source {
             ReadTextResult::UpToEof(&buf[start..])
         }
 
-        #[inline]
-        $($async)? fn read_bytes_until $(<$lf>)? (
-            &mut self,
-            byte: u8,
-            buf: &'b mut Vec<u8>,
-            position: &mut u64,
-        ) -> io::Result<(&'b [u8], bool)> {
-            // search byte must be within the ascii range
-            debug_assert!(byte.is_ascii());
-
-            let mut read = 0;
-            let start = buf.len();
-            loop {
-                let available = match self $(.$reader)? .fill_buf() $(.$await)? {
-                    Ok(n) if n.is_empty() => break,
-                    Ok(n) => n,
-                    Err(ref e) if e.kind() == io::ErrorKind::Interrupted => continue,
-                    Err(e) => {
-                        *position += read;
-                        return Err(e);
-                    }
-                };
-
-                match memchr::memchr(byte, available) {
-                    Some(i) => {
-                        buf.extend_from_slice(&available[..i]);
-
-                        let used = i + 1;
-                        self $(.$reader)? .consume(used);
-                        read += used as u64;
-
-                        *position += read;
-                        return Ok((&buf[start..], true));
-                    }
-                    None => {
-                        buf.extend_from_slice(available);
-
-                        let used = available.len();
-                        self $(.$reader)? .consume(used);
-                        read += used as u64;
-                    }
-                }
-            }
-
-            *position += read;
-            Ok((&buf[start..], false))
-        }
-
         #[inline]
         $($async)? fn read_with<$($lf,)? P: Parser>(
             &mut self,
@@ -327,7 +280,7 @@ impl<R: BufRead> Reader<R> {
     ///     match reader.read_event_into(&mut buf) {
     ///         Ok(Event::Start(_)) => count += 1,
     ///         Ok(Event::Text(e)) => txt.push(e.unescape().unwrap().into_owned()),
-    ///         Err(e) => panic!("Error at position {}: {:?}", reader.buffer_position(), e),
+    ///         Err(e) => panic!("Error at position {}: {:?}", reader.error_position(), e),
     ///         Ok(Event::Eof) => break,
     ///         _ => (),
     ///     }