Merge branch 'dev' of github.com:hipstermojo/paperoni into dev

src/http.rs

@@ -72,6 +72,60 @@ pub async fn fetch_html(url: &str) -> Result<HTMLResource, PaperoniError> {
     })
 }
 
+type ImgItem<'a> = (&'a str, String, Option<String>);
+
+async fn process_img_response<'a>(
+    img_response: &mut surf::Response,
+    url: &'a str,
+) -> Result<ImgItem<'a>, ImgError> {
+    if !img_response.status().is_success() {
+        let kind = ErrorKind::HTTPError(format!(
+            "Non-success HTTP status code ({})",
+            img_response.status()
+        ));
+        return Err(ImgError::with_kind(kind));
+    }
+    let img_content: Vec<u8> = match img_response.body_bytes().await {
+        Ok(bytes) => bytes,
+        Err(e) => return Err(e.into()),
+    };
+    let img_mime = img_response
+        .content_type()
+        .map(|mime| mime.essence().to_string());
+    let img_ext = match img_response
+        .content_type()
+        .map(|mime| map_mime_subtype_to_ext(mime.subtype()).to_string())
+    {
+        Some(mime_str) => mime_str,
+        None => return Err(ErrorKind::HTTPError("Image has no Content-Type".to_owned()).into()),
+    };
+
+    let mut img_path = std::env::temp_dir();
+    img_path.push(format!("{}.{}", hash_url(url), &img_ext));
+    let mut img_file = match File::create(&img_path).await {
+        Ok(file) => file,
+        Err(e) => return Err(e.into()),
+    };
+    match img_file.write_all(&img_content).await {
+        Ok(_) => (),
+        Err(e) => return Err(e.into()),
+    }
+
+    Ok((
+        url,
+        img_path
+            .file_name()
+            .map(|os_str_name| {
+                os_str_name
+                    .to_str()
+                    .expect("Unable to get image file name")
+                    .to_string()
+            })
+            .unwrap(),
+        img_mime,
+    ))
+}
+
 pub async fn download_images(
     extractor: &mut Extractor,
     article_origin: &Url,
@@ -102,53 +156,9 @@ pub async fn download_images(
             bar.set_message(format!("Downloading images [{}/{}]", img_idx + 1, img_count).as_str());
             match req.await {
                 Ok(mut img_response) => {
-                    let process_response = async {
+                    let process_response =
-                        let img_content: Vec<u8> = match img_response.body_bytes().await {
+                        process_img_response(&mut img_response, url.as_ref()).await;
-                            Ok(bytes) => bytes,
+                    process_response.map_err(|mut e: ImgError| {
-                            Err(e) => return Err(e.into()),
-                        };
-                        let img_mime = img_response
-                            .content_type()
-                            .map(|mime| mime.essence().to_string());
-                        let img_ext = match img_response
-                            .content_type()
-                            .map(|mime| map_mime_subtype_to_ext(mime.subtype()).to_string())
-                        {
-                            Some(mime_str) => mime_str,
-                            None => {
-                                return Err(ErrorKind::HTTPError(
-                                    "Image has no Content-Type".to_owned(),
-                                )
-                                .into())
-                            }
-                        };
-
-                        let mut img_path = std::env::temp_dir();
-                        img_path.push(format!("{}.{}", hash_url(&url), &img_ext));
-                        let mut img_file = match File::create(&img_path).await {
-                            Ok(file) => file,
-                            Err(e) => return Err(e.into()),
-                        };
-                        match img_file.write_all(&img_content).await {
-                            Ok(_) => (),
-                            Err(e) => return Err(e.into()),
-                        }
-
-                        Ok((
-                            url,
-                            img_path
-                                .file_name()
-                                .map(|os_str_name| {
-                                    os_str_name
-                                        .to_str()
-                                        .expect("Unable to get image file name")
-                                        .to_string()
-                                })
-                                .unwrap(),
-                            img_mime,
-                        ))
-                    };
-                    process_response.await.map_err(|mut e: ImgError| {
                         e.set_url(url);
                         e
                     })
@@ -162,8 +172,7 @@ pub async fn download_images(
         });
 
     // A utility closure used when update the value of an image source after downloading is successful
-    let replace_existing_img_src =
+    let replace_existing_img_src = |img_item: ImgItem| -> (String, Option<String>) {
-        |img_item: (&String, String, Option<String>)| -> (String, Option<String>) {
         let (img_url, img_path, img_mime) = img_item;
         let img_ref = extractor
             .article()

src/moz_readability/mod.rs

@@ -1248,8 +1248,7 @@ impl Readability {
             let srcset = node_attr.get("srcset");
             let class = node_attr.get("class");
             if (src.is_some() || srcset.is_some())
-                && class.is_some()
+                && class.and_then(|classname| classname.find("lazy")).is_none()
-                && !class.unwrap().contains("lazy")
             {
                 continue;
             }
@@ -3160,6 +3159,7 @@ characters. For that reason, this <p> tag could not be a byline because it's too
                     <source media="(min-width:465px)" srcset="img_white_flower.jpg">
                     <img src="img_orange_flowers.jpg" alt="Flowers" style="width:auto;">
                 </picture>
+                <img id="no-lazy-class" src="https://image.url/" data-attrs="{&quot;src&quot;:&quot;https://other.url/1.png&quot;,&quot;alt&quot;:&quot;&quot;}"/>
             </body>
         </html>
         "#;
@@ -3189,6 +3189,13 @@ characters. For that reason, this <p> tag could not be a byline because it's too
             lazy_loaded_attrs.get("data-src"),
             lazy_loaded_attrs.get("src")
         );
+
+        let no_lazy_class = doc.root_node.select_first("#no-lazy-class").unwrap();
+        let no_lazy_class_attrs = no_lazy_class.attributes.borrow();
+        assert_eq!(
+            no_lazy_class_attrs.get("src").unwrap(),
+            "https://image.url/"
+        );
     }
 
     #[test]