Introduce Config struct that holds parser configuration and implement #513

Cargo.toml

@@ -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

@@ -10,9 +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().<...>`.
 
 ### Bug Fixes
 
@@ -30,6 +37,7 @@
 
 [#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

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

@@ -30,7 +30,7 @@ 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() {
@@ -49,7 +49,9 @@ 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() {
@@ -74,7 +76,7 @@ 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() {
@@ -93,7 +95,9 @@ 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() {
@@ -120,7 +124,9 @@ 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),
@@ -135,7 +141,9 @@ 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),
@@ -150,7 +158,9 @@ 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),
@@ -168,7 +178,7 @@ 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() {
@@ -189,7 +199,7 @@ 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() {
@@ -210,7 +220,7 @@ 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

@@ -105,14 +105,16 @@ fn main() -> Result<(), AppError> {
     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 +17,10 @@ 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);
+    config.trim_text_end = true;
     loop {
         let event_result = reader.read_event_into(&mut buf);
         if let Ok(ref event) = event_result {

fuzz/fuzz_targets/structured_roundtrip.rs

@@ -84,12 +84,13 @@ fn fuzz_round_trip(driver: Driver) -> quick_xml::Result<()> {
     // 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));
+    let config = reader.config_mut();
+    config.check_comments = *config_iter.next().unwrap_or(&false);
+    config.check_end_names = *config_iter.next().unwrap_or(&false);
+    config.expand_empty_elements = *config_iter.next().unwrap_or(&false);
+    config.trim_markup_names_in_closing_tags = *config_iter.next().unwrap_or(&false);
+    config.trim_text(*config_iter.next().unwrap_or(&false));
+    config.trim_text_end = *config_iter.next().unwrap_or(&false);
 
     loop {
         let event = black_box(reader.read_event()?);
@@ -99,12 +100,13 @@ 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));
+    let config = reader.config_mut();
+    config.check_comments = *config_iter.next().unwrap_or(&false);
+    config.check_end_names = *config_iter.next().unwrap_or(&false);
+    config.expand_empty_elements = *config_iter.next().unwrap_or(&false);
+    config.trim_markup_names_in_closing_tags = *config_iter.next().unwrap_or(&false);
+    config.trim_text(*config_iter.next().unwrap_or(&false));
+    config.trim_text_end = *config_iter.next().unwrap_or(&false);
 
     loop {
         let event = black_box(reader.read_event()?);

src/de/mod.rs

@@ -2476,7 +2476,7 @@ where
     ///
     /// [`deserialize_seq`]: serde::Deserializer::deserialize_seq
     /// [DoS]: https://en.wikipedia.org/wiki/Denial-of-service_attack
-    /// [auto-expanding feature]: Reader::expand_empty_elements
+    /// [auto-expanding feature]: crate::reader::Config::expand_empty_elements
     #[cfg(feature = "overlapped-lists")]
     pub fn event_buffer_size(&mut self, limit: Option<NonZeroUsize>) -> &mut Self {
         self.limit = limit;
@@ -2761,7 +2761,8 @@ where
     /// and use specified entity resolver.
     pub fn from_str_with_resolver(source: &'de str, entity_resolver: E) -> Self {
         let mut reader = Reader::from_str(source);
-        reader.expand_empty_elements(true);
+        let config = reader.config_mut();
+        config.expand_empty_elements = true;
 
         Self::new(
             SliceReader {
@@ -2803,7 +2804,8 @@ where
     /// UTF-8, you can decode it first before using [`from_str`].
     pub fn with_resolver(reader: R, entity_resolver: E) -> Self {
         let mut reader = Reader::from_reader(reader);
-        reader.expand_empty_elements(true);
+        let config = reader.config_mut();
+        config.expand_empty_elements = true;
 
         Self::new(
             IoReader {
@@ -3709,7 +3711,8 @@ mod tests {
             start_trimmer: StartTrimmer::default(),
         };
 
-        reader.reader.expand_empty_elements(true);
+        let config = reader.reader.config_mut();
+        config.expand_empty_elements = true;
 
         let mut events = Vec::new();
 

src/errors.rs

@@ -96,7 +96,7 @@ pub enum IllFormedError {
     /// mostly artificial, but you can enable it in the [configuration].
     ///
     /// [specification]: https://www.w3.org/TR/xml11/#sec-comments
-    /// [configuration]: crate::reader::Reader::check_comments
+    /// [configuration]: crate::reader::Config::check_comments
     DoubleHyphenInComment,
 }
 

src/reader/async_tokio.rs

@@ -49,7 +49,7 @@ impl<R: AsyncBufRead + Unpin> Reader<R> {
     ///        <tag2>Test 2</tag2>
     ///     </tag1>
     /// "#.as_bytes());
-    /// reader.trim_text(true);
+    /// reader.config_mut().trim_text(true);
     ///
     /// let mut count = 0;
     /// let mut buf = Vec::new();
@@ -111,7 +111,7 @@ impl<R: AsyncBufRead + Unpin> Reader<R> {
     ///         </inner>
     ///     </outer>
     /// "#.as_bytes());
-    /// reader.trim_text(true);
+    /// reader.config_mut().trim_text(true);
     /// let mut buf = Vec::new();
     ///
     /// let start = BytesStart::new("outer");
@@ -186,7 +186,7 @@ impl<R: AsyncBufRead + Unpin> NsReader<R> {
     ///        <y:tag2>Test 2</y:tag2>
     ///     </x:tag1>
     /// "#.as_bytes());
-    /// reader.trim_text(true);
+    /// reader.config_mut().trim_text(true);
     ///
     /// let mut count = 0;
     /// let mut buf = Vec::new();
@@ -257,7 +257,7 @@ impl<R: AsyncBufRead + Unpin> NsReader<R> {
     ///         </inner>
     ///     </outer>
     /// "#.as_bytes());
-    /// reader.trim_text(true);
+    /// reader.config_mut().trim_text(true);
     /// let mut buf = Vec::new();
     ///
     /// let ns = Namespace(b"namespace 1");
@@ -292,7 +292,7 @@ impl<R: AsyncBufRead + Unpin> NsReader<R> {
         buf: &mut Vec<u8>,
     ) -> Result<Span> {
         // According to the https://www.w3.org/TR/xml11/#dt-etag, end name should
-        // match literally the start name. See `Reader::check_end_names` documentation
+        // match literally the start name. See `Config::check_end_names` documentation
         self.reader.read_to_end_into_async(end, buf).await
     }
 
@@ -321,7 +321,7 @@ impl<R: AsyncBufRead + Unpin> NsReader<R> {
     ///        <y:tag2>Test 2</y:tag2>
     ///     </x:tag1>
     /// "#.as_bytes());
-    /// reader.trim_text(true);
+    /// reader.config_mut().trim_text(true);
     ///
     /// let mut count = 0;
     /// let mut buf = Vec::new();

src/reader/buffered_reader.rs

@@ -283,7 +283,7 @@ impl<R: BufRead> Reader<R> {
     ///                 <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 buf = Vec::new();
     /// let mut txt = Vec::new();
@@ -368,7 +368,7 @@ impl<R: BufRead> Reader<R> {
     ///         </inner>
     ///     </outer>
     /// "#);
-    /// reader.trim_text(true);
+    /// reader.config_mut().trim_text(true);
     /// let mut buf = Vec::new();
     ///
     /// let start = BytesStart::new("outer");
@@ -390,8 +390,8 @@ impl<R: BufRead> Reader<R> {
     /// [`End`]: Event::End
     /// [`BytesStart::to_end()`]: crate::events::BytesStart::to_end
     /// [`read_to_end()`]: Self::read_to_end
-    /// [`expand_empty_elements`]: Self::expand_empty_elements
-    /// [`check_end_names`]: Self::check_end_names
+    /// [`expand_empty_elements`]: crate::reader::Config::expand_empty_elements
+    /// [`check_end_names`]: crate::reader::Config::check_end_names
     /// [the specification]: https://www.w3.org/TR/xml11/#dt-etag
     pub fn read_to_end_into(&mut self, end: QName, buf: &mut Vec<u8>) -> Result<Span> {
         Ok(read_to_end!(self, end, buf, read_event_impl, {