Add support for html underline and videos (#1955)

* fixed #1953 italic detection error fix: issue where when html2md parse <em> to "_" instead of "*" that won't be detected MarkdownToDelta converter feat(test): added test for DeltaX feat: added config classes to MarkdownToDelta and html2md that allow users configure their own styles * Added support for html underline and videos * removed print calls and fix no expect in test * removed useless element attr for underline * improved video url validator pattern * Added support for <video> tag * chore: dart fixes * fix: removed useless params * fix: imports issue --------- Co-authored-by: CatHood0 <[email protected]>
singerdmx · Jun 28, 2024 · 1f32e84 · 1f32e84
1 parent 83e2754
commit 1f32e84
Show file tree

Hide file tree

Showing 5 changed files with 136 additions and 83 deletions.
diff --git a/lib/src/models/documents/delta_x.dart b/lib/src/models/documents/delta_x.dart
@@ -3,6 +3,7 @@ import 'package:markdown/markdown.dart' as md;
 import 'package:meta/meta.dart';
 import '../../../markdown_quill.dart';
 import '../../../quill_delta.dart';
+import '../../utils/delta_x_utils.dart';
 
 @immutable
 @experimental
@@ -14,20 +15,12 @@ class DeltaX {
   /// This api is **experimental** and designed to be used **internally** and shouldn't
   /// used for **production applications**.
   @experimental
-  static Delta fromMarkdown(
-    String markdownText, {
-    Md2DeltaConfigs md2DeltaConfigs = const Md2DeltaConfigs(),
-  }) {
-    final mdDocument = md.Document(encodeHtml: false);
-    final mdToDelta = MarkdownToDelta(
-      markdownDocument: mdDocument,
-      customElementToBlockAttribute:
-          md2DeltaConfigs.customElementToBlockAttribute,
-      customElementToEmbeddable: md2DeltaConfigs.customElementToEmbeddable,
-      customElementToInlineAttribute:
-          md2DeltaConfigs.customElementToInlineAttribute,
-      softLineBreak: md2DeltaConfigs.softLineBreak,
+  static Delta fromMarkdown(String markdownText) {
+    final mdDocument = md.Document(
+      encodeHtml: false,
+      inlineSyntaxes: [UnderlineSyntax(), VideoSyntax()],
     );
+    final mdToDelta = MarkdownToDelta(markdownDocument: mdDocument);
     return mdToDelta.convert(markdownText);
   }
 
@@ -42,72 +35,15 @@ class DeltaX {
   /// used for **production applications**.
   ///
   @experimental
-  static Delta fromHtml(String htmlText, {Html2MdConfigs? configs}) {
-    final markdownText = html2md
-        .convert(
-          htmlText,
-          rules: configs?.customRules,
-          ignore: configs?.ignoreIf,
-          rootTag: configs?.rootTag,
-          imageBaseUrl: configs?.imageBaseUrl,
-          styleOptions: configs?.styleOptions,
-        )
-        .replaceAll(
-          'unsafe:',
-          '',
-        );
+  static Delta fromHtml(String htmlText) {
+    final markdownText = html2md.convert(
+      htmlText,
+      rules: [underlineRule, videoRule],
+      styleOptions: {'emDelimiter': '*'},
+    ).replaceAll(
+      'unsafe:',
+      '',
+    );
     return fromMarkdown(markdownText);
   }
 }
-
-@immutable
-@experimental
-class Md2DeltaConfigs {
-  const Md2DeltaConfigs({
-    this.customElementToInlineAttribute = const {},
-    this.customElementToBlockAttribute = const {},
-    this.customElementToEmbeddable = const {},
-    this.softLineBreak = false,
-  });
-  final Map<String, ElementToAttributeConvertor> customElementToInlineAttribute;
-  final Map<String, ElementToAttributeConvertor> customElementToBlockAttribute;
-  final Map<String, ElementToEmbeddableConvertor> customElementToEmbeddable;
-  final bool softLineBreak;
-}
-
-@immutable
-@experimental
-class Html2MdConfigs {
-  const Html2MdConfigs({
-    this.customRules,
-    this.ignoreIf,
-    this.rootTag,
-    this.imageBaseUrl,
-    this.styleOptions = const {'emDelimiter': '*'},
-    //emDelimiter set em to be "*" instead a "_"
-  });
-
-  /// The [rules] parameter can be used to customize element processing.
-  final List<html2md.Rule>? customRules;
-
-  /// Elements list in [ignore] would be ingored.
-  final List<String>? ignoreIf;
-
-  final String? rootTag;
-  final String? imageBaseUrl;
-
-  /// The default and available style options:
-  ///
-  /// | Name        | Default           | Options  |
-  /// | ------------- |:-------------:| -----:|
-  /// | headingStyle      | "setext" | "setext", "atx" |
-  /// | hr      | "* * *" | "* * *", "- - -", "_ _ _" |
-  /// | bulletListMarker      | "*" | "*", "-", "_" |
-  /// | codeBlockStyle      | "indented" | "indented", "fenced" |
-  /// | fence      | "\`\`\`" | "\`\`\`", "~~~" |
-  /// | emDelimiter      | "_" | "_", "*" |
-  /// | strongDelimiter      | "**" | "**", "__" |
-  /// | linkStyle      | "inlined" | "inlined", "referenced" |
-  /// | linkReferenceStyle      | "full" | "full", "collapsed", "shortcut" |
-  final Map<String, String>? styleOptions;
-}
diff --git a/lib/src/packages/quill_markdown/markdown_to_delta.dart b/lib/src/packages/quill_markdown/markdown_to_delta.dart
@@ -82,6 +82,7 @@ class MarkdownToDelta extends Converter<String, Delta>
 
   final _elementToInlineAttr = <String, ElementToAttributeConvertor>{
     'em': (_) => [Attribute.italic],
+    'u': (_) => [Attribute.underline],
     'strong': (_) => [Attribute.bold],
     'del': (_) => [Attribute.strikeThrough],
     'a': (element) => [LinkAttribute(element.attributes['href'])],
@@ -91,6 +92,7 @@ class MarkdownToDelta extends Converter<String, Delta>
   final _elementToEmbed = <String, ElementToEmbeddableConvertor>{
     'hr': (_) => horizontalRule,
     'img': (elAttrs) => BlockEmbed.image(elAttrs['src'] ?? ''),
+    'video': (elAttrs) => BlockEmbed.video(elAttrs['src'] ?? '')
   };
 
   var _delta = Delta();

diff --git a/lib/src/utils/delta_x_utils.dart b/lib/src/utils/delta_x_utils.dart
@@ -0,0 +1,80 @@
+import 'package:html2md/html2md.dart' as hmd;
+import 'package:markdown/markdown.dart' as md;
+
+// [ character
+const int _$lbracket = 0x5B;
+final RegExp _youtubeVideoUrlValidator = RegExp(
+    r'^((?:https?:)?\/\/)?((?:www|m)\.)?((?:youtube\.com|youtu.be))(\/(?:[\w\-]+\?v=|embed\/|v\/)?)([\w\-]+)(\S+)?$');
+
+///Local syntax implementation for underline
+class UnderlineSyntax extends md.DelimiterSyntax {
+  UnderlineSyntax()
+      : super(
+          '<und>',
+          requiresDelimiterRun: true,
+          allowIntraWord: true,
+          tags: [md.DelimiterTag('u', 5)],
+        );
+}
+
+class VideoSyntax extends md.LinkSyntax {
+  VideoSyntax({super.linkResolver})
+      : super(
+          pattern: r'\[',
+          startCharacter: _$lbracket,
+        );
+
+  @override
+  md.Element createNode(
+    String destination,
+    String? title, {
+    required List<md.Node> Function() getChildren,
+  }) {
+    final element = md.Element.empty('video');
+    element.attributes['src'] = destination;
+    if (title != null && title.isNotEmpty) {
+      element.attributes['title'] = title;
+    }
+    return element;
+  }
+}
+
+///This rule avoid the default converter from html2md ignore underline tag for <u> or <ins>
+final underlineRule =
+    hmd.Rule('underline', filters: ['u', 'ins'], replacement: (content, node) {
+  //Is used a local underline implemenation since markdown just use underline with html tags
+  return '<und>$content<und>';
+});
+final videoRule = hmd.Rule('video', filters: ['iframe', 'video'],
+    replacement: (content, node) {
+  //This need to be verified by a different way of iframes, since video tag can have <source> children
+  if (node.nodeName == 'video') {
+    //if has children then just will be taked as different part of code
+    if (node.childNum > 0) {
+      var child = node.firstChild!;
+      var src = child.getAttribute('src');
+      if (src == null) {
+        child = node.childNodes().last;
+        src = child.getAttribute('src');
+      }
+      if (!_youtubeVideoUrlValidator.hasMatch(src ?? '')) {
+        return '<video>${child.outerHTML}</video>';
+      }
+      return '[$content]($src)';
+    }
+    final src = node.getAttribute('src');
+    if (src == null || !_youtubeVideoUrlValidator.hasMatch(src)) {
+      return node.outerHTML;
+    }
+    return '[$content]($src)';
+  }
+  //by now, we can only access to src
+  final src = node.getAttribute('src');
+  //if the source is null or is not valid youtube url, then just return the html instead remove it
+  //by now is only available validation for youtube videos
+  if (src == null || !_youtubeVideoUrlValidator.hasMatch(src)) {
+    return node.outerHTML;
+  }
+  final title = node.getAttribute('title');
+  return '[$title]($src)';
+});
diff --git a/pubspec.yaml b/pubspec.yaml
@@ -75,6 +75,7 @@ dev_dependencies:
   yaml: ^3.1.2
   http: ^1.2.1
 
+  path: any
 flutter:
   uses-material-design: true
   generate: true
diff --git a/test/utils/delta_x_test.dart b/test/utils/delta_x_test.dart
@@ -5,18 +5,52 @@ import 'package:test/test.dart';
 void main() {
   const htmlWithEmp =
       '<p>This is a normal sentence, and this section has greater emp<em>hasis.</em></p>';
+
+  const htmlWithUnderline =
+      '<p>This is a normal sentence, and this section has greater <u>underline</u>';
+
+  const htmlWithIframeVideo =
+      '<iframe src="https://www.youtube.com/embed/dQw4w9WgXcQ" title="YouTube video player"></iframe>';
+
+  const htmlWithVideoTag =
+      '''<video src="https://www.youtube.com/embed/dQw4w9WgXcQ">Your browser does not support the video tag.</video> 
+''';
   final expectedDeltaEmp = Delta.fromOperations([
     Operation.insert(
         'This is a normal sentence, and this section has greater emp'),
     Operation.insert('hasis.', {'italic': true}),
     Operation.insert('\n'),
   ]);
 
+  final expectedDeltaUnderline = Delta.fromOperations([
+    Operation.insert(
+        'This is a normal sentence, and this section has greater '),
+    Operation.insert('underline', {'underline': true}),
+    Operation.insert('\n'),
+  ]);
+
+  final expectedDeltaVideo = Delta.fromOperations([
+    Operation.insert({'video': 'https://www.youtube.com/embed/dQw4w9WgXcQ'}),
+    Operation.insert('\n'),
+  ]);
+
   test('should detect emphasis and parse correctly', () {
-    final delta = DeltaX.fromHtml(
-      htmlWithEmp,
-      configs: const Html2MdConfigs(),
-    );
+    final delta = DeltaX.fromHtml(htmlWithEmp);
     expect(delta, expectedDeltaEmp);
   });
+
+  test('should detect underline and parse correctly', () {
+    final delta = DeltaX.fromHtml(htmlWithUnderline);
+    expect(delta, expectedDeltaUnderline);
+  });
+
+  test('should detect iframe and parse correctly', () {
+    final delta = DeltaX.fromHtml(htmlWithIframeVideo);
+    expect(delta, expectedDeltaVideo);
+  });
+
+  test('should detect video and parse correctly', () {
+    final delta = DeltaX.fromHtml(htmlWithVideoTag);
+    expect(delta, expectedDeltaVideo);
+  });
 }