Cargo.toml

@@ -16,7 +16,7 @@ include = ["src/*", "LICENSE-MIT.md", "README.md"]
 [dependencies]
 document-features = { version = "0.2", optional = true }
 encoding_rs = { version = "0.8", optional = true }
-serde = { version = ">=1.0.100", optional = true }
+serde = { version = ">=1.0.139", optional = true }
 tokio = { version = "1.10", optional = true, default-features = false, features = ["io-util"] }
 memchr = "2.1"
 arbitrary = { version = "1", features = ["derive"], optional = true }
@@ -85,7 +85,7 @@ async-tokio = ["tokio"]
 ## # }
 ## let xml = to_utf16le_with_bom(r#"<?xml encoding='UTF-16'><element/>"#);
 ## let mut reader = Reader::from_reader(xml.as_ref());
-## reader.trim_text(true);
+## reader.config_mut().trim_text(true);
 ##
 ## let mut buf = Vec::new();
 ## let mut unsupported = false;

Changelog.md

@@ -10,8 +10,16 @@
 
 ## Unreleased
 
+The way to configure parser is changed. Now all configuration is contained in the
+`Config` struct and can be applied at once. When `serde-types` feature is enabled,
+configuration is serializable.
+
 ### New Features
 
+- [#513]: Allow to continue parsing after getting new `Error::IllFormed`.
+- [#677]: Added methods `config()` and `config_mut()` to inspect and change the parser
+  configuration. Previous builder methods on `Reader` / `NsReader` was replaced by
+  direct access to fields of config using `reader.config_mut().<...>`.
 - [#379]: Improved compliance with the XML attribute value normalization process by
   adding `Attribute::normalized_value()` and `Attribute::normalized_value_with()`,
   which ought to be used in place of `Attribute::unescape_value()` and
@@ -21,9 +29,21 @@
 
 ### Misc Changes
 
+- [#675]: Minimum supported version of serde raised to 1.0.139
+- [#675]: Rework the `quick_xml::Error` type to provide more accurate information:
+  - `Error::EndEventMismatch` replaced by `IllFormedError::MismatchedEnd` in some cases
+  - `Error::EndEventMismatch` replaced by `IllFormedError::UnmatchedEnd` in some cases
+  - `Error::TextNotFound` was removed because not used
+  - `Error::UnexpectedBang` replaced by `SyntaxError`
+  - `Error::UnexpectedEof` replaced by `SyntaxError` in some cases
+  - `Error::UnexpectedEof` replaced by `IllFormedError` in some cases
+  - `Error::UnexpectedToken` replaced by `IllFormedError::DoubleHyphenInComment`
 - [#379]: Added tests for attribute value normalization
 
 [#379]: https://github.com/tafia/quick-xml/pull/379
+[#513]: https://github.com/tafia/quick-xml/issues/513
+[#675]: https://github.com/tafia/quick-xml/pull/675
+[#677]: https://github.com/tafia/quick-xml/pull/677
 
 ## 0.31.0 -- 2023-10-22
 
@@ -91,14 +111,14 @@ serde >= 1.0.181
 
 ### Bug Fixes
 
-- [#618]: Avoid crashing on wrong comments like `<!-->` when using `read_event_into*` functions.
+- [#604]: Avoid crashing on wrong comments like `<!-->` when using `read_event_into*` functions.
 
 ### Misc Changes
 
+[#604]: https://github.com/tafia/quick-xml/issue/604
 [#609]: https://github.com/tafia/quick-xml/pull/609
 [#615]: https://github.com/tafia/quick-xml/pull/615
 [#617]: https://github.com/tafia/quick-xml/pull/617
-[#618]: https://github.com/tafia/quick-xml/pull/618
 
 
 ## 0.29.0 -- 2023-06-13

README.md

@@ -28,7 +28,7 @@ let xml = r#"<tag1 att1 = "test">
                 <tag2>Test 2</tag2>
              </tag1>"#;
 let mut reader = Reader::from_str(xml);
-reader.trim_text(true);
+reader.config_mut().trim_text(true);
 
 let mut count = 0;
 let mut txt = Vec::new();
@@ -73,7 +73,7 @@ use std::io::Cursor;
 
 let xml = r#"<this_tag k1="v1" k2="v2"><child>text</child></this_tag>"#;
 let mut reader = Reader::from_str(xml);
-reader.trim_text(true);
+reader.config_mut().trim_text(true);
 let mut writer = Writer::new(Cursor::new(Vec::new()));
 loop {
     match reader.read_event() {

benches/microbenches.rs

@@ -33,7 +33,7 @@ fn read_event(c: &mut Criterion) {
     group.bench_function("trim_text = false", |b| {
         b.iter(|| {
             let mut r = Reader::from_str(SAMPLE);
-            r.check_end_names(false);
+            r.config_mut().check_end_names = false;
             let mut count = criterion::black_box(0);
             loop {
                 match r.read_event() {
@@ -52,7 +52,9 @@ fn read_event(c: &mut Criterion) {
     group.bench_function("trim_text = true", |b| {
         b.iter(|| {
             let mut r = Reader::from_str(SAMPLE);
-            r.trim_text(true).check_end_names(false);
+            let config = r.config_mut();
+            config.trim_text(true);
+            config.check_end_names = false;
             let mut count = criterion::black_box(0);
             loop {
                 match r.read_event() {
@@ -77,7 +79,7 @@ fn read_resolved_event_into(c: &mut Criterion) {
     group.bench_function("trim_text = false", |b| {
         b.iter(|| {
             let mut r = NsReader::from_str(SAMPLE);
-            r.check_end_names(false);
+            r.config_mut().check_end_names = false;
             let mut count = criterion::black_box(0);
             loop {
                 match r.read_resolved_event() {
@@ -96,7 +98,9 @@ fn read_resolved_event_into(c: &mut Criterion) {
     group.bench_function("trim_text = true", |b| {
         b.iter(|| {
             let mut r = NsReader::from_str(SAMPLE);
-            r.trim_text(true).check_end_names(false);
+            let config = r.config_mut();
+            config.trim_text(true);
+            config.check_end_names = false;
             let mut count = criterion::black_box(0);
             loop {
                 match r.read_resolved_event() {
@@ -123,7 +127,9 @@ fn one_event(c: &mut Criterion) {
         b.iter(|| {
             let mut r = Reader::from_str(&src);
             let mut nbtxt = criterion::black_box(0);
-            r.trim_text(true).check_end_names(false);
+            let config = r.config_mut();
+            config.trim_text(true);
+            config.check_end_names = false;
             match r.read_event() {
                 Ok(Event::Start(ref e)) => nbtxt += e.len(),
                 something_else => panic!("Did not expect {:?}", something_else),
@@ -138,7 +144,9 @@ fn one_event(c: &mut Criterion) {
         b.iter(|| {
             let mut r = Reader::from_str(&src);
             let mut nbtxt = criterion::black_box(0);
-            r.trim_text(true).check_end_names(false);
+            let config = r.config_mut();
+            config.trim_text(true);
+            config.check_end_names = false;
             match r.read_event() {
                 Ok(Event::Comment(e)) => nbtxt += e.unescape().unwrap().len(),
                 something_else => panic!("Did not expect {:?}", something_else),
@@ -153,7 +161,9 @@ fn one_event(c: &mut Criterion) {
         b.iter(|| {
             let mut r = Reader::from_str(&src);
             let mut nbtxt = criterion::black_box(0);
-            r.trim_text(true).check_end_names(false);
+            let config = r.config_mut();
+            config.trim_text(true);
+            config.check_end_names = false;
             match r.read_event() {
                 Ok(Event::CData(ref e)) => nbtxt += e.len(),
                 something_else => panic!("Did not expect {:?}", something_else),
@@ -171,7 +181,7 @@ fn attributes(c: &mut Criterion) {
     group.bench_function("with_checks = true", |b| {
         b.iter(|| {
             let mut r = Reader::from_str(PLAYERS);
-            r.check_end_names(false);
+            r.config_mut().check_end_names = false;
             let mut count = criterion::black_box(0);
             loop {
                 match r.read_event() {
@@ -192,7 +202,7 @@ fn attributes(c: &mut Criterion) {
     group.bench_function("with_checks = false", |b| {
         b.iter(|| {
             let mut r = Reader::from_str(PLAYERS);
-            r.check_end_names(false);
+            r.config_mut().check_end_names = false;
             let mut count = criterion::black_box(0);
             loop {
                 match r.read_event() {
@@ -213,7 +223,7 @@ fn attributes(c: &mut Criterion) {
     group.bench_function("try_get_attribute", |b| {
         b.iter(|| {
             let mut r = Reader::from_str(PLAYERS);
-            r.check_end_names(false);
+            r.config_mut().check_end_names = false;
             let mut count = criterion::black_box(0);
             loop {
                 match r.read_event() {

compare/benches/bench.rs

@@ -59,7 +59,7 @@ fn low_level_comparison(c: &mut Criterion) {
             |b, input| {
                 b.iter(|| {
                     let mut r = Reader::from_reader(input.as_bytes());
-                    r.check_end_names(false);
+                    r.config_mut().check_end_names = false;
                     let mut count = criterion::black_box(0);
                     let mut buf = Vec::new();
                     loop {

examples/custom_entities.rs

@@ -25,7 +25,7 @@ const DATA: &str = r#"
 
 fn main() -> Result<(), Box<dyn std::error::Error>> {
     let mut reader = Reader::from_str(DATA);
-    reader.trim_text(true);
+    reader.config_mut().trim_text(true);
 
     let mut custom_entities: HashMap<String, String> = HashMap::new();
     let entity_re = Regex::new(r#"<!ENTITY\s+([^ \t\r\n]+)\s+"([^"]*)"\s*>"#)?;

examples/read_buffered.rs

@@ -8,7 +8,7 @@ fn main() -> Result<(), quick_xml::Error> {
     use quick_xml::reader::Reader;
 
     let mut reader = Reader::from_file("tests/documents/document.xml")?;
-    reader.trim_text(true);
+    reader.config_mut().trim_text(true);
 
     let mut buf = Vec::new();
 

examples/read_nodes.rs

@@ -2,6 +2,7 @@
 // Note: for this specific data set using serde feature would simplify
 //       this simple data is purely to make it easier to understand the code
 
+use quick_xml::events::attributes::AttrError;
 use quick_xml::events::{BytesStart, Event};
 use quick_xml::name::QName;
 use quick_xml::reader::Reader;
@@ -29,6 +30,26 @@ const XML: &str = r#"
   </Localization>
 "#;
 
+#[derive(Debug)]
+enum AppError {
+    /// XML parsing error
+    Xml(quick_xml::Error),
+    /// The `Translation/Text` node is missed
+    NoText(String),
+}
+
+impl From<quick_xml::Error> for AppError {
+    fn from(error: quick_xml::Error) -> Self {
+        Self::Xml(error)
+    }
+}
+
+impl From<AttrError> for AppError {
+    fn from(error: AttrError) -> Self {
+        Self::Xml(quick_xml::Error::InvalidAttr(error))
+    }
+}
+
 #[derive(Debug)]
 struct Translation {
     tag: String,
@@ -40,7 +61,7 @@ impl Translation {
     fn new_from_element(
         reader: &mut Reader<&[u8]>,
         element: BytesStart,
-    ) -> Result<Translation, quick_xml::Error> {
+    ) -> Result<Translation, AppError> {
         let mut tag = Cow::Borrowed("");
         let mut lang = Cow::Borrowed("");
 
@@ -68,30 +89,32 @@ impl Translation {
             } else {
                 dbg!("Expected Event::Start for Text, got: {:?}", &event);
                 let name_string = reader.decoder().decode(name.as_ref())?;
-                Err(quick_xml::Error::UnexpectedToken(name_string.into()))
+                Err(AppError::NoText(name_string.into()))
             }
         } else {
             let event_string = format!("{:?}", event);
-            Err(quick_xml::Error::UnexpectedToken(event_string))
+            Err(AppError::NoText(event_string))
         }
     }
 }
 
-fn main() -> Result<(), quick_xml::Error> {
+fn main() -> Result<(), AppError> {
     // In a real-world use case, Settings would likely be a struct
     // HashMap here is just to make the sample code short
     let mut settings: HashMap<String, String>;
     let mut translations: Vec<Translation> = Vec::new();
 
     let mut reader = Reader::from_str(XML);
-    reader.trim_text(true);
+    let config = reader.config_mut();
 
+    config.trim_text(true);
     // == Handling empty elements ==
     // To simply our processing code
     // we want the same events for empty elements, like:
     //   <DefaultSettings Language="es" Greeting="HELLO"/>
     //   <Text/>
-    reader.expand_empty_elements(true);
+    config.expand_empty_elements = true;
+
     let mut buf = Vec::new();
 
     loop {

examples/read_texts.rs

@@ -6,7 +6,7 @@ fn main() {
                <tag1>text3</tag1><tag1><tag2>text4</tag2></tag1>";
 
     let mut reader = Reader::from_str(xml);
-    reader.trim_text(true);
+    reader.config_mut().trim_text(true);
 
     loop {
         match reader.read_event() {

fuzz/fuzz_targets/fuzz_target_1.rs

@@ -17,10 +17,9 @@ where
 {
     let mut writer = Writer::new(Cursor::new(Vec::new()));
     let mut buf = vec![];
-    let reader = reader
-        .expand_empty_elements(true)
-        .trim_text(true)
-        .trim_text_end(true);
+    let config = reader.config_mut();
+    config.expand_empty_elements = true;
+    config.trim_text(true);
     loop {
         let event_result = reader.read_event_into(&mut buf);
         if let Ok(ref event) = event_result {

fuzz/fuzz_targets/structured_roundtrip.rs

@@ -3,7 +3,7 @@
 use arbitrary::{Arbitrary, Unstructured};
 use libfuzzer_sys::fuzz_target;
 use quick_xml::events::{BytesCData, BytesText, Event};
-use quick_xml::reader::{NsReader, Reader};
+use quick_xml::reader::{Config, NsReader, Reader};
 use quick_xml::writer::Writer;
 use std::{hint::black_box, io::Cursor};
 
@@ -41,7 +41,7 @@ enum WriterFunc<'a> {
 #[derive(Debug, arbitrary::Arbitrary)]
 struct Driver<'a> {
     writer_funcs: Vec<WriterFunc<'a>>,
-    reader_config: Vec<bool>,
+    reader_config: Config,
 }
 
 fn fuzz_round_trip(driver: Driver) -> quick_xml::Result<()> {
@@ -83,13 +83,7 @@ fn fuzz_round_trip(driver: Driver) -> quick_xml::Result<()> {
     let xml = writer.into_inner().into_inner();
     // The str should be valid as we just generated it, unwrapping **should** be safe.
     let mut reader = Reader::from_str(std::str::from_utf8(&xml).unwrap());
-    let mut config_iter = driver.reader_config.iter();
-    reader.check_comments(*config_iter.next().unwrap_or(&false));
-    reader.check_end_names(*config_iter.next().unwrap_or(&false));
-    reader.expand_empty_elements(*config_iter.next().unwrap_or(&false));
-    reader.trim_markup_names_in_closing_tags(*config_iter.next().unwrap_or(&false));
-    reader.trim_text(*config_iter.next().unwrap_or(&false));
-    reader.trim_text_end(*config_iter.next().unwrap_or(&false));
+    *reader.config_mut() = driver.reader_config.clone();
 
     loop {
         let event = black_box(reader.read_event()?);
@@ -99,12 +93,7 @@ fn fuzz_round_trip(driver: Driver) -> quick_xml::Result<()> {
     }
 
     let mut reader = NsReader::from_reader(&xml[..]);
-    reader.check_comments(*config_iter.next().unwrap_or(&false));
-    reader.check_end_names(*config_iter.next().unwrap_or(&false));
-    reader.expand_empty_elements(*config_iter.next().unwrap_or(&false));
-    reader.trim_markup_names_in_closing_tags(*config_iter.next().unwrap_or(&false));
-    reader.trim_text(*config_iter.next().unwrap_or(&false));
-    reader.trim_text_end(*config_iter.next().unwrap_or(&false));
+    *reader.config_mut() = driver.reader_config;
 
     loop {
         let event = black_box(reader.read_event()?);