A cross platform Audiobookshelf client for Android and iOS
Dart
import 'package:flutter/gestures.dart';
import 'package:flutter/material.dart';
import 'package:url_launcher/url_launcher.dart';
import '../l10n/app_localizations.dart';
/// Lightweight HTML-to-RichText widget with clickable links.
///
/// Handles common podcast/book description HTML:
/// - `<a href>` → tappable links
/// - `<b>`, `<strong>` → bold
/// - `<i>`, `<em>` → italic
/// - `<br>` → line break, `<p>` → paragraph break (blank line between)
/// - Bare URLs → tappable links
/// - HTML entities → decoded
///
/// With [maxLines] set it truncates and offers a "Show more / Show less"
/// toggle; null shows the whole thing with no toggle.
class HtmlDescription extends StatefulWidget {
final String html;
final int? maxLines;
final TextStyle? style;
final Color? linkColor;
const HtmlDescription({
super.key,
required this.html,
this.maxLines = 4,
this.style,
this.linkColor,
});
@override
State<HtmlDescription> createState() => _HtmlDescriptionState();
}
class _HtmlDescriptionState extends State<HtmlDescription> {
bool _expanded = false;
@override
Widget build(BuildContext context) {
final cs = Theme.of(context).colorScheme;
final l = AppLocalizations.of(context)!;
final baseStyle = widget.style ?? Theme.of(context).textTheme.bodySmall?.copyWith(
color: cs.onSurface.withValues(alpha: 0.7), height: 1.5);
final linkColor = widget.linkColor ?? cs.primary;
final spans = _parseHtml(widget.html, baseStyle ?? const TextStyle(), linkColor);
final textSpan = TextSpan(children: spans);
final maxLines = widget.maxLines;
if (maxLines == null) return Text.rich(textSpan);
return LayoutBuilder(builder: (context, constraints) {
// Measure whether the text actually overflows at maxLines
final tp = TextPainter(
text: textSpan,
maxLines: maxLines,
textDirection: TextDirection.ltr,
)..layout(maxWidth: constraints.maxWidth);
final isTruncated = tp.didExceedMaxLines;
tp.dispose();
return Column(
crossAxisAlignment: CrossAxisAlignment.start,
children: [
Text.rich(
textSpan,
maxLines: _expanded ? null : maxLines,
overflow: _expanded ? TextOverflow.clip : TextOverflow.ellipsis,
),
if (isTruncated)
Padding(
padding: const EdgeInsets.only(top: 4),
child: GestureDetector(
onTap: () => setState(() => _expanded = !_expanded),
child: Text(
_expanded ? l.showLess : l.showMore,
style: TextStyle(fontSize: 12, color: linkColor, fontWeight: FontWeight.w500),
),
),
),
],
);
});
}
}
// ─── HTML Parsing ──────────────────────────────────────────
/// Regex for bare URLs not already inside an <a> tag.
final _bareUrlRegex = RegExp(
r'https?://[^\s<>\])"]+',
caseSensitive: false,
);
/// Parse HTML string into a list of styled TextSpans.
List<InlineSpan> _parseHtml(String html, TextStyle baseStyle, Color linkColor) {
// Pre-process: convert block elements to newlines. A closing paragraph
// becomes a blank line, the way a browser spaces <p> blocks - one newline
// ran every paragraph straight into the next. CR/LF from descriptions
// typed on Windows is normalised first so it counts as a plain newline.
var text = html
.replaceAll('\r\n', '\n')
.replaceAll('\r', '\n')
.replaceAll(RegExp(r'<br\s*/?>'), '\n')
.replaceAll(RegExp(r'</p>'), '\n\n')
.replaceAll(RegExp(r'<p[^>]*>'), '')
.replaceAll(RegExp(r'</div>'), '\n')
.replaceAll(RegExp(r'<div[^>]*>'), '')
.replaceAll(RegExp(r'</?ul[^>]*>'), '\n')
.replaceAll(RegExp(r'</?ol[^>]*>'), '\n')
.replaceAll(RegExp(r'<li[^>]*>'), ' \u2022 ')
.replaceAll(RegExp(r'</li>'), '\n')
.replaceAll(RegExp(r'\n{3,}'), '\n\n'); // collapse excess newlines
final spans = <InlineSpan>[];
final tagPattern = RegExp(
r'<(/?)(a|b|strong|i|em)(\s[^>]*)?>',
caseSensitive: false,
);
var bold = false;
var italic = false;
String? linkUrl;
var pos = 0;
for (final match in tagPattern.allMatches(text)) {
// Add text before this tag
if (match.start > pos) {
final segment = text.substring(pos, match.start);
spans.addAll(_textWithLinks(
_decodeEntities(segment),
_buildStyle(baseStyle, bold, italic, linkUrl != null, linkColor),
linkColor,
linkUrl,
));
}
pos = match.end;
final isClosing = match.group(1) == '/';
final tag = match.group(2)!.toLowerCase();
switch (tag) {
case 'a':
if (isClosing) {
linkUrl = null;
} else {
final attrs = match.group(3) ?? '';
final hrefMatch = RegExp(r'href\s*=\s*"([^"]*)"').firstMatch(attrs) ??
RegExp(r"href\s*=\s*'([^']*)'").firstMatch(attrs);
linkUrl = hrefMatch?.group(1);
}
break;
case 'b':
case 'strong':
bold = !isClosing;
break;
case 'i':
case 'em':
italic = !isClosing;
break;
}
}
// Add remaining text after last tag
if (pos < text.length) {
final segment = text.substring(pos);
// Strip any remaining unknown tags
final cleaned = segment.replaceAll(RegExp(r'<[^>]+>'), '');
spans.addAll(_textWithLinks(
_decodeEntities(cleaned),
_buildStyle(baseStyle, bold, italic, linkUrl != null, linkColor),
linkColor,
linkUrl,
));
}
// Trim leading newlines (an empty opening paragraph would otherwise start
// the text with a blank line)
while (spans.isNotEmpty) {
final first = spans.first;
if (first is TextSpan && first.text != null) {
final trimmed = first.text!.replaceAll(RegExp(r'^\n+'), '');
if (trimmed.isEmpty) {
spans.removeAt(0);
} else {
spans[0] = TextSpan(
text: trimmed,
style: first.style,
recognizer: first.recognizer,
);
break;
}
} else {
break;
}
}
// Trim trailing newlines
while (spans.isNotEmpty) {
final last = spans.last;
if (last is TextSpan && last.text != null) {
final trimmed = last.text!.replaceAll(RegExp(r'\n+$'), '');
if (trimmed.isEmpty) {
spans.removeLast();
} else {
spans[spans.length - 1] = TextSpan(
text: trimmed,
style: last.style,
recognizer: last.recognizer,
);
break;
}
} else {
break;
}
}
return spans;
}
TextStyle _buildStyle(TextStyle base, bool bold, bool italic, bool isLink, Color linkColor) {
var style = base;
if (bold) style = style.copyWith(fontWeight: FontWeight.w600);
if (italic) style = style.copyWith(fontStyle: FontStyle.italic);
if (isLink) style = style.copyWith(color: linkColor, decoration: TextDecoration.underline, decorationColor: linkColor.withValues(alpha: 0.4));
return style;
}
/// Split text into plain segments and bare-URL segments, making URLs tappable.
List<InlineSpan> _textWithLinks(String text, TextStyle style, Color linkColor, String? explicitUrl) {
if (explicitUrl != null) {
// Entire segment is inside an <a> tag
return [_tappableSpan(text, style, explicitUrl)];
}
// Detect bare URLs
final spans = <InlineSpan>[];
var pos = 0;
for (final match in _bareUrlRegex.allMatches(text)) {
if (match.start > pos) {
spans.add(TextSpan(text: text.substring(pos, match.start), style: style));
}
final url = match.group(0)!;
final linkStyle = style.copyWith(
color: linkColor,
decoration: TextDecoration.underline,
decorationColor: linkColor.withValues(alpha: 0.4),
);
spans.add(_tappableSpan(url, linkStyle, url));
pos = match.end;
}
if (pos < text.length) {
spans.add(TextSpan(text: text.substring(pos), style: style));
}
return spans;
}
TextSpan _tappableSpan(String text, TextStyle style, String url) {
return TextSpan(
text: text,
style: style,
recognizer: TapGestureRecognizer()
..onTap = () => launchUrl(Uri.parse(url), mode: LaunchMode.externalApplication),
);
}
String _decodeEntities(String text) {
return text
.replaceAll(' ', ' ')
.replaceAll('&', '&')
.replaceAll('<', '<')
.replaceAll('>', '>')
.replaceAll('"', '"')
.replaceAll(''', "'")
.replaceAll(''', "'")
.replaceAll(''', "'")
.replaceAll('—', '\u2014')
.replaceAll('–', '\u2013')
.replaceAll('…', '\u2026')
.replaceAll('«', '\u00AB')
.replaceAll('»', '\u00BB');
}