736 lines
22 KiB
Dart
736 lines
22 KiB
Dart
/// ============================================================
|
||
/// 闲言APP — RSS订阅服务
|
||
/// 创建时间: 2026-05-30
|
||
/// 更新时间: 2026-05-30
|
||
/// 作用: RSS/Atom订阅源管理+文章解析+持久化存储
|
||
/// 上次更新: 新增OPML导入导出功能
|
||
/// ============================================================
|
||
|
||
import 'dart:convert';
|
||
|
||
import 'package:dio/dio.dart';
|
||
import 'package:hive_ce/hive.dart';
|
||
import 'package:rss_dart/dart_rss.dart';
|
||
import 'package:xianyan/core/utils/logger.dart';
|
||
|
||
/// RSS订阅源分类
|
||
enum RssCategory {
|
||
tech('科技'),
|
||
efficiency('效率'),
|
||
general('综合'),
|
||
techCommunity('技术社区'),
|
||
design('设计'),
|
||
finance('财经'),
|
||
news('新闻');
|
||
|
||
const RssCategory(this.label);
|
||
|
||
final String label;
|
||
}
|
||
|
||
/// RSS订阅源数据模型
|
||
class RssSubscription {
|
||
const RssSubscription({
|
||
required this.id,
|
||
required this.title,
|
||
required this.url,
|
||
this.description,
|
||
this.iconUrl,
|
||
this.category = RssCategory.general,
|
||
this.lastUpdated,
|
||
this.unreadCount = 0,
|
||
this.addedAt,
|
||
});
|
||
|
||
final String id;
|
||
final String title;
|
||
final String url;
|
||
final String? description;
|
||
final String? iconUrl;
|
||
final RssCategory category;
|
||
final DateTime? lastUpdated;
|
||
final int unreadCount;
|
||
final DateTime? addedAt;
|
||
|
||
RssSubscription copyWith({
|
||
String? title,
|
||
String? description,
|
||
String? iconUrl,
|
||
RssCategory? category,
|
||
DateTime? lastUpdated,
|
||
int? unreadCount,
|
||
}) {
|
||
return RssSubscription(
|
||
id: id,
|
||
title: title ?? this.title,
|
||
url: url,
|
||
description: description ?? this.description,
|
||
iconUrl: iconUrl ?? this.iconUrl,
|
||
category: category ?? this.category,
|
||
lastUpdated: lastUpdated ?? this.lastUpdated,
|
||
unreadCount: unreadCount ?? this.unreadCount,
|
||
addedAt: addedAt,
|
||
);
|
||
}
|
||
|
||
Map<String, dynamic> toMap() => {
|
||
'id': id,
|
||
'title': title,
|
||
'url': url,
|
||
'description': description,
|
||
'iconUrl': iconUrl,
|
||
'category': category.name,
|
||
'lastUpdated': lastUpdated?.toIso8601String(),
|
||
'unreadCount': unreadCount,
|
||
'addedAt': addedAt?.toIso8601String(),
|
||
};
|
||
|
||
factory RssSubscription.fromMap(Map<String, dynamic> map) {
|
||
return RssSubscription(
|
||
id: map['id'] as String? ?? '',
|
||
title: map['title'] as String? ?? '',
|
||
url: map['url'] as String? ?? '',
|
||
description: map['description'] as String?,
|
||
iconUrl: map['iconUrl'] as String?,
|
||
category: RssCategory.values.firstWhere(
|
||
(e) => e.name == map['category'],
|
||
orElse: () => RssCategory.general,
|
||
),
|
||
lastUpdated: RssService.parseDateTime(map['lastUpdated'] as String?),
|
||
unreadCount: map['unreadCount'] as int? ?? 0,
|
||
addedAt: RssService.parseDateTime(map['addedAt'] as String?),
|
||
);
|
||
}
|
||
}
|
||
|
||
/// RSS条目数据模型
|
||
class RssFeedItem {
|
||
const RssFeedItem({
|
||
required this.title,
|
||
this.description,
|
||
this.content,
|
||
this.link,
|
||
this.author,
|
||
this.pubDate,
|
||
this.imageUrl,
|
||
this.sourceId,
|
||
this.sourceTitle,
|
||
this.isRead = false,
|
||
});
|
||
|
||
final String title;
|
||
final String? description;
|
||
final String? content;
|
||
final String? link;
|
||
final String? author;
|
||
final DateTime? pubDate;
|
||
final String? imageUrl;
|
||
final String? sourceId;
|
||
final String? sourceTitle;
|
||
final bool isRead;
|
||
|
||
RssFeedItem copyWith({bool? isRead}) => RssFeedItem(
|
||
title: title,
|
||
description: description,
|
||
content: content,
|
||
link: link,
|
||
author: author,
|
||
pubDate: pubDate,
|
||
imageUrl: imageUrl,
|
||
sourceId: sourceId,
|
||
sourceTitle: sourceTitle,
|
||
isRead: isRead ?? this.isRead,
|
||
);
|
||
|
||
/// 生成唯一ID（基于sourceId + link）
|
||
String get uid => '${sourceId ?? ""}_${link?.hashCode ?? title.hashCode}';
|
||
}
|
||
|
||
/// RSS订阅服务 — 解析RSS/Atom源 + 持久化
|
||
class RssService {
|
||
RssService._();
|
||
|
||
static final Dio _dio = Dio(
|
||
BaseOptions(
|
||
connectTimeout: const Duration(seconds: 10),
|
||
receiveTimeout: const Duration(seconds: 10),
|
||
responseType: ResponseType.plain,
|
||
),
|
||
);
|
||
|
||
static const String _boxName = 'rss_subscriptions';
|
||
static const String _readArticlesBox = 'rss_read_articles';
|
||
static Box<dynamic>? _box;
|
||
static Box<dynamic>? _readBox;
|
||
|
||
/// 初始化Hive存储
|
||
static Future<void> init() async {
|
||
try {
|
||
_box = await Hive.openBox(_boxName);
|
||
_readBox = await Hive.openBox(_readArticlesBox);
|
||
Log.i('RssService', 'Hive存储初始化完成');
|
||
} catch (e) {
|
||
Log.e('RssService', 'Hive存储初始化失败: $e');
|
||
}
|
||
}
|
||
|
||
/// 获取所有已保存的订阅源
|
||
static List<RssSubscription> getSavedSubscriptions() {
|
||
if (_box == null) return defaultSubscriptions.toList();
|
||
final saved = _box!.get('subscriptions');
|
||
if (saved == null) return defaultSubscriptions.toList();
|
||
try {
|
||
final list = jsonDecode(saved as String) as List<dynamic>;
|
||
return list
|
||
.map((e) => RssSubscription.fromMap(e as Map<String, dynamic>))
|
||
.toList();
|
||
} catch (e) {
|
||
Log.e('RssService', '读取订阅源失败: $e');
|
||
return defaultSubscriptions.toList();
|
||
}
|
||
}
|
||
|
||
/// 保存订阅源列表
|
||
static Future<void> saveSubscriptions(List<RssSubscription> subs) async {
|
||
if (_box == null) return;
|
||
try {
|
||
final encoded = jsonEncode(subs.map((e) => e.toMap()).toList());
|
||
await _box!.put('subscriptions', encoded);
|
||
} catch (e) {
|
||
Log.e('RssService', '保存订阅源失败: $e');
|
||
}
|
||
}
|
||
|
||
/// 添加订阅源
|
||
static Future<void> addSubscription(RssSubscription sub) async {
|
||
final subs = getSavedSubscriptions();
|
||
if (subs.any((s) => s.url == sub.url)) {
|
||
Log.w('RssService', '订阅源已存在: ${sub.url}');
|
||
return;
|
||
}
|
||
subs.add(sub);
|
||
await saveSubscriptions(subs);
|
||
}
|
||
|
||
/// 删除订阅源
|
||
static Future<void> removeSubscription(String id) async {
|
||
final subs = getSavedSubscriptions();
|
||
subs.removeWhere((s) => s.id == id);
|
||
await saveSubscriptions(subs);
|
||
}
|
||
|
||
/// 标记文章已读
|
||
static Future<void> markArticleRead(String articleUid) async {
|
||
if (_readBox == null) return;
|
||
await _readBox!.put(articleUid, true);
|
||
}
|
||
|
||
/// 检查文章是否已读
|
||
static bool isArticleRead(String articleUid) {
|
||
if (_readBox == null) return false;
|
||
return _readBox!.get(articleUid) == true;
|
||
}
|
||
|
||
/// 拉取并解析RSS/Atom订阅源
|
||
static Future<List<RssFeedItem>> fetchFeed(RssSubscription sub) async {
|
||
try {
|
||
final response = await _dio.get<String>(sub.url);
|
||
final xml = response.data ?? '';
|
||
|
||
List<RssFeedItem> items = [];
|
||
|
||
if (xml.contains('<rss') || xml.contains('<channel')) {
|
||
final feed = RssFeed.parse(xml);
|
||
items = feed.items
|
||
.map(
|
||
(item) => RssFeedItem(
|
||
title: item.title ?? '无标题',
|
||
description: item.description ?? item.content?.value,
|
||
content: item.content?.value ?? item.description,
|
||
link: item.link,
|
||
author: item.author ?? item.dc?.creator,
|
||
pubDate: parseDateTime(item.pubDate),
|
||
imageUrl:
|
||
item.enclosure?.url ??
|
||
item.media?.thumbnails.firstOrNull?.url,
|
||
sourceId: sub.id,
|
||
sourceTitle: sub.title,
|
||
isRead: isArticleRead(
|
||
'${sub.id}_${item.link?.hashCode ?? item.title.hashCode}',
|
||
),
|
||
),
|
||
)
|
||
.toList();
|
||
} else if (xml.contains('<feed')) {
|
||
final feed = AtomFeed.parse(xml);
|
||
items = feed.items
|
||
.map(
|
||
(entry) => RssFeedItem(
|
||
title: entry.title ?? '无标题',
|
||
description: entry.summary ?? entry.content,
|
||
content: entry.content ?? entry.summary,
|
||
link: entry.links.firstOrNull?.href,
|
||
author: entry.authors.firstOrNull?.name,
|
||
pubDate: parseDateTime(entry.published ?? entry.updated),
|
||
imageUrl: entry.media?.thumbnails.firstOrNull?.url,
|
||
sourceId: sub.id,
|
||
sourceTitle: sub.title,
|
||
isRead: isArticleRead(
|
||
'${sub.id}_${entry.links.firstOrNull?.href.hashCode ?? entry.title.hashCode}',
|
||
),
|
||
),
|
||
)
|
||
.toList();
|
||
}
|
||
|
||
return items;
|
||
} catch (e) {
|
||
Log.e('RssService', 'RSS解析失败 [${sub.url}]: $e');
|
||
return [];
|
||
}
|
||
}
|
||
|
||
/// 发现RSS源（输入URL自动解析）
|
||
static Future<RssSubscription?> discoverFeed(String url) async {
|
||
try {
|
||
final normalizedUrl = _normalizeUrl(url);
|
||
final response = await _dio.get<String>(normalizedUrl);
|
||
final xml = response.data ?? '';
|
||
|
||
if (xml.contains('<rss') || xml.contains('<channel')) {
|
||
final feed = RssFeed.parse(xml);
|
||
return RssSubscription(
|
||
id: 'rss_${DateTime.now().millisecondsSinceEpoch}',
|
||
title: feed.title ?? normalizedUrl,
|
||
url: normalizedUrl,
|
||
description: feed.description,
|
||
iconUrl: feed.image?.url,
|
||
category: _guessCategory(feed.title ?? ''),
|
||
addedAt: DateTime.now(),
|
||
);
|
||
} else if (xml.contains('<feed')) {
|
||
final feed = AtomFeed.parse(xml);
|
||
return RssSubscription(
|
||
id: 'rss_${DateTime.now().millisecondsSinceEpoch}',
|
||
title: feed.title ?? normalizedUrl,
|
||
url: normalizedUrl,
|
||
description: feed.subtitle,
|
||
iconUrl: feed.icon,
|
||
category: _guessCategory(feed.title ?? ''),
|
||
addedAt: DateTime.now(),
|
||
);
|
||
}
|
||
return null;
|
||
} catch (e) {
|
||
Log.e('RssService', '发现RSS源失败: $e');
|
||
return null;
|
||
}
|
||
}
|
||
|
||
/// 批量拉取所有订阅源
|
||
static Future<Map<String, List<RssFeedItem>>> fetchAllFeeds(
|
||
List<RssSubscription> subs,
|
||
) async {
|
||
final results = <String, List<RssFeedItem>>{};
|
||
await Future.wait(
|
||
subs.map((sub) async {
|
||
final items = await fetchFeed(sub);
|
||
results[sub.id] = items;
|
||
}),
|
||
);
|
||
return results;
|
||
}
|
||
|
||
/// 默认推荐订阅源
|
||
static const List<RssSubscription> defaultSubscriptions = [
|
||
RssSubscription(
|
||
id: '36kr',
|
||
title: '36氪',
|
||
url: 'https://36kr.com/feed',
|
||
description: '科技创业资讯',
|
||
category: RssCategory.tech,
|
||
),
|
||
RssSubscription(
|
||
id: 'ifanr',
|
||
title: '爱范儿',
|
||
url: 'https://www.ifanr.com/feed',
|
||
description: '科技媒体',
|
||
category: RssCategory.tech,
|
||
),
|
||
RssSubscription(
|
||
id: 'sspai',
|
||
title: '少数派',
|
||
url: 'https://sspai.com/feed',
|
||
description: '高效工作和品质生活',
|
||
category: RssCategory.efficiency,
|
||
),
|
||
RssSubscription(
|
||
id: 'zhihu_daily',
|
||
title: '知乎日报',
|
||
url: 'https://daily.zhihu.com/rss',
|
||
description: '知乎精选',
|
||
),
|
||
RssSubscription(
|
||
id: 'ruanyifeng',
|
||
title: '阮一峰',
|
||
url: 'http://www.ruanyifeng.com/blog/atom.xml',
|
||
description: '科技博客',
|
||
category: RssCategory.tech,
|
||
),
|
||
RssSubscription(
|
||
id: 'v2ex',
|
||
title: 'V2EX',
|
||
url: 'https://www.v2ex.com/index.xml',
|
||
description: '创意工作者社区',
|
||
category: RssCategory.techCommunity,
|
||
),
|
||
];
|
||
|
||
/// 解析日期字符串（公开方法，供fromMap使用）
|
||
static DateTime? parseDateTime(String? dateStr) {
|
||
if (dateStr == null || dateStr.isEmpty) return null;
|
||
return DateTime.tryParse(dateStr);
|
||
}
|
||
|
||
/// URL标准化
|
||
static String _normalizeUrl(String url) {
|
||
var normalized = url.trim();
|
||
if (!normalized.startsWith('http://') &&
|
||
!normalized.startsWith('https://')) {
|
||
normalized = 'https://$normalized';
|
||
}
|
||
return normalized;
|
||
}
|
||
|
||
/// 根据标题猜测分类
|
||
static RssCategory _guessCategory(String title) {
|
||
final lower = title.toLowerCase();
|
||
if (lower.contains('设计') || lower.contains('design')) {
|
||
return RssCategory.design;
|
||
}
|
||
if (lower.contains('财经') ||
|
||
lower.contains('金融') ||
|
||
lower.contains('finance')) {
|
||
return RssCategory.finance;
|
||
}
|
||
if (lower.contains('新闻') || lower.contains('news')) {
|
||
return RssCategory.news;
|
||
}
|
||
if (lower.contains('效率') ||
|
||
lower.contains('少数派') ||
|
||
lower.contains('sspai')) {
|
||
return RssCategory.efficiency;
|
||
}
|
||
if (lower.contains('v2ex') ||
|
||
lower.contains('社区') ||
|
||
lower.contains('forum')) {
|
||
return RssCategory.techCommunity;
|
||
}
|
||
if (lower.contains('科技') ||
|
||
lower.contains('tech') ||
|
||
lower.contains('氪') ||
|
||
lower.contains('范儿')) {
|
||
return RssCategory.tech;
|
||
}
|
||
return RssCategory.general;
|
||
}
|
||
|
||
// ============================================================
|
||
// OPML 导入导出
|
||
// ============================================================
|
||
|
||
/// 导出订阅源为OPML格式XML
|
||
static String exportToOpml() {
|
||
final subs = getSavedSubscriptions();
|
||
final buffer = StringBuffer();
|
||
|
||
buffer.writeln('<?xml version="1.0" encoding="UTF-8"?>');
|
||
buffer.writeln('<opml version="2.0">');
|
||
buffer.writeln(' <head>');
|
||
buffer.writeln(' <title>闲言APP RSS订阅</title>');
|
||
buffer.writeln(
|
||
' <dateCreated>${DateTime.now().toIso8601String()}</dateCreated>',
|
||
);
|
||
buffer.writeln(' </head>');
|
||
buffer.writeln(' <body>');
|
||
|
||
final grouped = <RssCategory, List<RssSubscription>>{};
|
||
for (final sub in subs) {
|
||
grouped.putIfAbsent(sub.category, () => []).add(sub);
|
||
}
|
||
|
||
for (final entry in grouped.entries) {
|
||
buffer.writeln(' <outline text="${_xmlEscape(entry.key.label)}">');
|
||
for (final sub in entry.value) {
|
||
buffer.writeln(
|
||
' <outline type="rss" '
|
||
'text="${_xmlEscape(sub.title)}" '
|
||
'title="${_xmlEscape(sub.title)}" '
|
||
'xmlUrl="${_xmlEscape(sub.url)}"'
|
||
'${sub.description != null ? ' description="${_xmlEscape(sub.description!)}"' : ''}'
|
||
' />',
|
||
);
|
||
}
|
||
buffer.writeln(' </outline>');
|
||
}
|
||
|
||
buffer.writeln(' </body>');
|
||
buffer.writeln('</opml>');
|
||
|
||
return buffer.toString();
|
||
}
|
||
|
||
/// 从OPML格式XML导入订阅源
|
||
static Future<int> importFromOpml(String opmlXml) async {
|
||
try {
|
||
final subs = <RssSubscription>[];
|
||
final outlineRegex = RegExp(
|
||
r'<outline[^>]*type="rss"[^>]*>',
|
||
caseSensitive: false,
|
||
);
|
||
final textRegex = RegExp(r'text="([^"]*)"');
|
||
final titleRegex = RegExp(r'title="([^"]*)"');
|
||
final xmlUrlRegex = RegExp(r'xmlUrl="([^"]*)"');
|
||
final descRegex = RegExp(r'description="([^"]*)"');
|
||
|
||
for (final match in outlineRegex.allMatches(opmlXml)) {
|
||
final outline = match.group(0)!;
|
||
|
||
final xmlUrl = xmlUrlRegex.firstMatch(outline)?.group(1);
|
||
if (xmlUrl == null || xmlUrl.isEmpty) continue;
|
||
|
||
final title =
|
||
titleRegex.firstMatch(outline)?.group(1) ??
|
||
textRegex.firstMatch(outline)?.group(1) ??
|
||
xmlUrl;
|
||
final description = descRegex.firstMatch(outline)?.group(1);
|
||
|
||
subs.add(
|
||
RssSubscription(
|
||
id: 'import_${DateTime.now().millisecondsSinceEpoch}_${subs.length}',
|
||
title: _xmlUnescape(title),
|
||
url: _xmlUnescape(xmlUrl),
|
||
description: _xmlUnescape(description),
|
||
category: _guessCategory(title),
|
||
addedAt: DateTime.now(),
|
||
),
|
||
);
|
||
}
|
||
|
||
final existing = getSavedSubscriptions();
|
||
final existingUrls = existing.map((s) => s.url).toSet();
|
||
var added = 0;
|
||
|
||
for (final sub in subs) {
|
||
if (!existingUrls.contains(sub.url)) {
|
||
existing.add(sub);
|
||
existingUrls.add(sub.url);
|
||
added++;
|
||
}
|
||
}
|
||
|
||
if (added > 0) {
|
||
await saveSubscriptions(existing);
|
||
}
|
||
|
||
Log.i('RssService', 'OPML导入完成: ${subs.length}个源，新增$added个');
|
||
return added;
|
||
} catch (e) {
|
||
Log.e('RssService', 'OPML导入失败: $e');
|
||
return 0;
|
||
}
|
||
}
|
||
|
||
/// XML特殊字符转义
|
||
static String _xmlEscape(String input) {
|
||
return input
|
||
.replaceAll('&', '&')
|
||
.replaceAll('"', '"')
|
||
.replaceAll("'", ''')
|
||
.replaceAll('<', '<')
|
||
.replaceAll('>', '>');
|
||
}
|
||
|
||
/// XML特殊字符反转义
|
||
static String _xmlUnescape(String? input) {
|
||
if (input == null) return '';
|
||
return input
|
||
.replaceAll('>', '>')
|
||
.replaceAll('<', '<')
|
||
.replaceAll(''', "'")
|
||
.replaceAll('"', '"')
|
||
.replaceAll('&', '&');
|
||
}
|
||
|
||
// ============================================================
|
||
// 全文提取（阅读模式）
|
||
// ============================================================
|
||
|
||
/// 从文章URL提取全文内容（阅读模式）
|
||
///
|
||
/// 使用简易 Readability 算法：
|
||
/// 1. 获取网页HTML
|
||
/// 2. 移除导航/侧边栏/页脚等非正文区域
|
||
/// 3. 提取最可能是正文的区域
|
||
/// 4. 清理HTML标签，返回纯文本
|
||
static Future<RssFullTextResult> fetchFullText(String url) async {
|
||
try {
|
||
final response = await _dio.get<String>(url);
|
||
final html = response.data ?? '';
|
||
if (html.isEmpty) {
|
||
return const RssFullTextResult(error: '页面内容为空');
|
||
}
|
||
|
||
final title = _extractTitle(html);
|
||
final content = _extractContent(html);
|
||
final images = _extractContentImages(html);
|
||
|
||
if (content.isEmpty) {
|
||
return RssFullTextResult(
|
||
error: '无法提取正文内容',
|
||
title: title,
|
||
);
|
||
}
|
||
|
||
return RssFullTextResult(
|
||
success: true,
|
||
title: title,
|
||
content: content,
|
||
images: images,
|
||
sourceUrl: url,
|
||
);
|
||
} catch (e) {
|
||
Log.e('RssService', '全文提取失败 [$url]: $e');
|
||
return RssFullTextResult(error: '加载失败: $e');
|
||
}
|
||
}
|
||
|
||
/// 提取页面标题
|
||
static String _extractTitle(String html) {
|
||
final ogTitle = RegExp(r'<meta[^>]*property=["\x27]og:title["\x27][^>]*content=["\x27]([^"\x27]*)["\x27]', caseSensitive: false)
|
||
.firstMatch(html);
|
||
if (ogTitle != null && ogTitle.group(1)!.isNotEmpty) {
|
||
return _decodeHtmlEntities(ogTitle.group(1)!);
|
||
}
|
||
final titleMatch = RegExp(r'<title[^>]*>(.*?)</title>', caseSensitive: false, dotAll: true)
|
||
.firstMatch(html);
|
||
if (titleMatch != null && titleMatch.group(1)!.isNotEmpty) {
|
||
return _decodeHtmlEntities(titleMatch.group(1)!.trim());
|
||
}
|
||
return '';
|
||
}
|
||
|
||
/// 提取正文内容（简易 Readability）
|
||
static String _extractContent(String html) {
|
||
var cleaned = html;
|
||
|
||
// 移除脚本和样式
|
||
cleaned = cleaned.replaceAll(
|
||
RegExp(r'<script[^>]*>.*?</script>', caseSensitive: false, dotAll: true),
|
||
'',
|
||
);
|
||
cleaned = cleaned.replaceAll(
|
||
RegExp(r'<style[^>]*>.*?</style>', caseSensitive: false, dotAll: true),
|
||
'',
|
||
);
|
||
cleaned = cleaned.replaceAll(
|
||
RegExp(r'<nav[^>]*>.*?</nav>', caseSensitive: false, dotAll: true),
|
||
'',
|
||
);
|
||
cleaned = cleaned.replaceAll(
|
||
RegExp(r'<footer[^>]*>.*?</footer>', caseSensitive: false, dotAll: true),
|
||
'',
|
||
);
|
||
cleaned = cleaned.replaceAll(
|
||
RegExp(r'<header[^>]*>.*?</header>', caseSensitive: false, dotAll: true),
|
||
'',
|
||
);
|
||
cleaned = cleaned.replaceAll(
|
||
RegExp(r'<aside[^>]*>.*?</aside>', caseSensitive: false, dotAll: true),
|
||
'',
|
||
);
|
||
cleaned = cleaned.replaceAll(
|
||
RegExp(r'<noscript[^>]*>.*?</noscript>', caseSensitive: false, dotAll: true),
|
||
'',
|
||
);
|
||
|
||
// 查找 article 标签
|
||
final articleMatch = RegExp(r'<article[^>]*>(.*?)</article>', caseSensitive: false, dotAll: true)
|
||
.firstMatch(cleaned);
|
||
if (articleMatch != null) {
|
||
cleaned = articleMatch.group(1)!;
|
||
} else {
|
||
// 查找 class 含 article/content/post/entry 的 div
|
||
final contentDiv = RegExp(
|
||
r'<div[^>]*class=["\x27][^\x27]*(?:article|content|post-body|entry-content|post-content|story-body|article-body|rich-text|markdown-body)[^\x27]*["\x27][^>]*>(.*?)</div>',
|
||
caseSensitive: false,
|
||
dotAll: true,
|
||
).firstMatch(cleaned);
|
||
if (contentDiv != null) {
|
||
cleaned = contentDiv.group(1)!;
|
||
}
|
||
}
|
||
|
||
// 清理HTML标签，保留段落结构
|
||
var text = cleaned;
|
||
text = text.replaceAll(RegExp(r'<br\s*/?>', caseSensitive: false), '\n');
|
||
text = text.replaceAll(RegExp(r'<p[^>]*>', caseSensitive: false), '\n');
|
||
text = text.replaceAll(RegExp(r'</p>', caseSensitive: false), '\n');
|
||
text = text.replaceAll(RegExp(r'<h[1-6][^>]*>', caseSensitive: false), '\n\n');
|
||
text = text.replaceAll(RegExp(r'</h[1-6]>', caseSensitive: false), '\n');
|
||
text = text.replaceAll(RegExp(r'<li[^>]*>', caseSensitive: false), '• ');
|
||
text = text.replaceAll(RegExp(r'<blockquote[^>]*>', caseSensitive: false), '\n> ');
|
||
text = text.replaceAll(RegExp(r'<[^>]*>'), '');
|
||
text = text.replaceAll(RegExp(r' '), ' ');
|
||
text = text.replaceAll(RegExp(r'&'), '&');
|
||
text = text.replaceAll(RegExp(r'<'), '<');
|
||
text = text.replaceAll(RegExp(r'>'), '>');
|
||
text = text.replaceAll(RegExp(r'"'), '"');
|
||
text = text.replaceAll(RegExp(r'&#\d+;'), '');
|
||
text = text.replaceAll(RegExp(r'\n{3,}'), '\n\n');
|
||
|
||
return text.trim();
|
||
}
|
||
|
||
/// 提取正文中的图片URL
|
||
static List<String> _extractContentImages(String html) {
|
||
final imgRegex = RegExp(r'<img[^>]+src\s*=\s*["\x27]([^"\x27]+)["\x27]', dotAll: true);
|
||
return imgRegex
|
||
.allMatches(html)
|
||
.map((m) => m.group(1) ?? '')
|
||
.where((url) => url.isNotEmpty && !url.endsWith('.svg') && !url.contains('avatar') && !url.contains('icon') && !url.contains('logo'))
|
||
.take(10)
|
||
.toList();
|
||
}
|
||
|
||
/// HTML实体解码
|
||
static String _decodeHtmlEntities(String text) {
|
||
return text
|
||
.replaceAll('&', '&')
|
||
.replaceAll('<', '<')
|
||
.replaceAll('>', '>')
|
||
.replaceAll('"', '"')
|
||
.replaceAll(''', "'")
|
||
.replaceAll(' ', ' ');
|
||
}
|
||
}
|
||
|
||
/// 全文提取结果
|
||
class RssFullTextResult {
|
||
const RssFullTextResult({
|
||
this.success = false,
|
||
this.title,
|
||
this.content,
|
||
this.images = const [],
|
||
this.sourceUrl,
|
||
this.error,
|
||
});
|
||
|
||
final bool success;
|
||
final String? title;
|
||
final String? content;
|
||
final List<String> images;
|
||
final String? sourceUrl;
|
||
final String? error;
|
||
}
|