spider_control.html 15 KB

Raw Blame History Permalink

<!DOCTYPE html>
<html lang="zh-CN">
<head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>爬虫控制面板</title>
    <link href="https://cdn.bootcdn.net/ajax/libs/twitter-bootstrap/5.0.2/css/bootstrap.min.css" rel="stylesheet">
    <link href="https://cdn.bootcdn.net/ajax/libs/font-awesome/5.15.4/css/all.min.css" rel="stylesheet">
    <style>
        .topic-item {
            margin: 5px;
            padding: 8px 15px;
            border-radius: 20px;
            background-color: #f8f9fa;
            display: inline-block;
            cursor: pointer;
        }
        .topic-item.selected {
            background-color: #0d6efd;
            color: white;
        }
        .custom-topic-input {
            margin: 10px 0;
        }
        .parameter-section {
            margin: 20px 0;
            padding: 20px;
            border-radius: 10px;
            background-color: #f8f9fa;
        }
    </style>
</head>
<body>
    <div class="container mt-5">
        <h2 class="mb-4">爬虫控制面板</h2>

        <!-- 话题选择区域 -->
        <div class="card mb-4">
            <div class="card-header">
                <h5 class="mb-0">选择话题类型</h5>
            </div>
            <div class="card-body">
                <div id="predefinedTopics" class="mb-3">
                    <!-- 预定义话题将通过JavaScript动态加载 -->
                </div>

                <div class="custom-topic-input">
                    <h6>添加自定义话题</h6>
                    <div class="input-group">
                        <input type="text" class="form-control" id="customTopic" placeholder="输入自定义话题">
                        <button class="btn btn-primary" onclick="addCustomTopic()">
                            <i class="fas fa-plus"></i> 添加
                        </button>
                    </div>
                </div>

                <div id="selectedTopics" class="mt-3">
                    <h6>已选择的话题：</h6>
                    <div id="selectedTopicsList" class="mt-2">
                        <!-- 已选择的话题将在这里显示 -->
                    </div>
                </div>
            </div>
        </div>

        <!-- 爬虫参数配置 -->
        <div class="card mb-4">
            <div class="card-header">
                <h5 class="mb-0">爬虫参数配置</h5>
            </div>
            <div class="card-body">
                <div class="row">
                    <div class="col-md-6">
                        <div class="mb-3">
                            <label for="crawlDepth" class="form-label">爬取深度</label>
                            <input type="number" class="form-control" id="crawlDepth" value="3" min="1" max="10">
                            <small class="text-muted">每个话题爬取的页数（1-10）</small>
                        </div>
                    </div>
                    <div class="col-md-6">
                        <div class="mb-3">
                            <label for="interval" class="form-label">爬取间隔(秒)</label>
                            <input type="number" class="form-control" id="interval" value="5" min="1">
                            <small class="text-muted">每次请求之间的间隔时间</small>
                        </div>
                    </div>
                </div>

                <div class="row">
                    <div class="col-md-6">
                        <div class="mb-3">
                            <label for="maxRetries" class="form-label">最大重试次数</label>
                            <input type="number" class="form-control" id="maxRetries" value="3" min="1">
                        </div>
                    </div>
                    <div class="col-md-6">
                        <div class="mb-3">
                            <label for="timeout" class="form-label">请求超时时间(秒)</label>
                            <input type="number" class="form-control" id="timeout" value="30" min="1">
                        </div>
                    </div>
                </div>
            </div>
        </div>

        <!-- AI配置助手 -->
        <div class="card mb-4">
            <div class="card-header">
                <h5 class="mb-0">
                    <i class="fas fa-robot"></i> AI配置助手
                </h5>
            </div>
            <div class="card-body">
                <div class="mb-3">
                    <label for="aiPrompt" class="form-label">用自然语言描述您的爬虫需求</label>
                    <textarea class="form-control" id="aiPrompt" rows="3"
                        placeholder="例如：我想爬取最近一周关于人工智能的热门微博，重点关注转发量超过1000的内容，每个话题爬取前5页内容。"></textarea>
                </div>
                <div class="d-flex justify-content-between align-items-center">
                    <button class="btn btn-primary" onclick="generateConfig()">
                        <i class="fas fa-magic"></i> 生成配置
                    </button>
                    <div class="form-check">
                        <input class="form-check-input" type="checkbox" id="autoApply" checked>
                        <label class="form-check-label" for="autoApply">
                            自动应用生成的配置
                        </label>
                    </div>
                </div>
                <div id="aiResponse" class="mt-3" style="display: none;">
                    <div class="alert alert-info">
                        <h6 class="alert-heading">AI助手建议：</h6>
                        <p id="aiSuggestion" class="mb-0"></p>
                    </div>
                </div>
            </div>
        </div>

        <!-- 操作按钮 -->
        <div class="d-flex justify-content-between mb-5">
            <button class="btn btn-primary" onclick="startCrawling()">
                <i class="fas fa-play"></i> 开始爬取
            </button>
            <button class="btn btn-secondary" onclick="saveConfig()">
                <i class="fas fa-save"></i> 保存配置
            </button>
        </div>

        <!-- 爬虫状态和日志 -->
        <div class="card">
            <div class="card-header">
                <h5 class="mb-0">爬虫状态</h5>
            </div>
            <div class="card-body">
                <div class="progress mb-3">
                    <div id="crawlProgress" class="progress-bar" role="progressbar" style="width: 0%"></div>
                </div>
                <div class="border p-3 bg-light" style="height: 200px; overflow-y: auto;">
                    <pre id="crawlLog" class="mb-0"></pre>
                </div>
            </div>
        </div>
    </div>

    <script src="https://cdn.bootcdn.net/ajax/libs/jquery/3.6.0/jquery.min.js"></script>
    <script src="https://cdn.bootcdn.net/ajax/libs/twitter-bootstrap/5.0.2/js/bootstrap.bundle.min.js"></script>
    <script>
        // 预定义话题列表
        const predefinedTopics = [
            '热门', '社会', '科技', '娱乐', '体育', '财经',
            '教育', '健康', '军事', '文化', '汽车', '美食'
        ];

        // 已选择的话题
        let selectedTopics = new Set();

        // 初始化页面
        window.onload = function() {
            loadPredefinedTopics();
        };

        // 加载预定义话题
        function loadPredefinedTopics() {
            const topicsDiv = document.getElementById('predefinedTopics');
            predefinedTopics.forEach(topic => {
                const topicElement = document.createElement('span');
                topicElement.className = 'topic-item';
                topicElement.textContent = topic;
                topicElement.onclick = () => toggleTopic(topic, topicElement);
                topicsDiv.appendChild(topicElement);
            });
        }

        // 切换话题选择状态
        function toggleTopic(topic, element) {
            if (selectedTopics.has(topic)) {
                selectedTopics.delete(topic);
                element.classList.remove('selected');
            } else {
                selectedTopics.add(topic);
                element.classList.add('selected');
            }
            updateSelectedTopicsList();
        }

        // 添加自定义话题
        function addCustomTopic() {
            const input = document.getElementById('customTopic');
            const topic = input.value.trim();
            if (topic) {
                selectedTopics.add(topic);
                input.value = '';
                updateSelectedTopicsList();
            }
        }

        // 更新已选择的话题列表
        function updateSelectedTopicsList() {
            const listDiv = document.getElementById('selectedTopicsList');
            listDiv.innerHTML = '';
            selectedTopics.forEach(topic => {
                const topicElement = document.createElement('span');
                topicElement.className = 'topic-item selected';
                topicElement.textContent = topic;
                topicElement.onclick = () => {
                    selectedTopics.delete(topic);
                    updateSelectedTopicsList();
                };
                listDiv.appendChild(topicElement);
            });
        }

        // 开始爬取
        function startCrawling() {
            if (selectedTopics.size === 0) {
                alert('请至少选择一个话题！');
                return;
            }

            const config = {
                topics: Array.from(selectedTopics),
                parameters: {
                    crawlDepth: parseInt(document.getElementById('crawlDepth').value),
                    interval: parseInt(document.getElementById('interval').value),
                    maxRetries: parseInt(document.getElementById('maxRetries').value),
                    timeout: parseInt(document.getElementById('timeout').value)
                }
            };

            // 发送爬虫配置到后端
            fetch('/api/spider/start', {
                method: 'POST',
                headers: {
                    'Content-Type': 'application/json'
                },
                body: JSON.stringify(config)
            })
            .then(response => response.json())
            .then(data => {
                if (data.success) {
                    updateCrawlLog('爬虫任务已启动...');
                } else {
                    updateCrawlLog('启动失败：' + data.message);
                }
            })
            .catch(error => {
                updateCrawlLog('错误：' + error.message);
            });
        }

        // 保存配置
        function saveConfig() {
            const config = {
                topics: Array.from(selectedTopics),
                parameters: {
                    crawlDepth: parseInt(document.getElementById('crawlDepth').value),
                    interval: parseInt(document.getElementById('interval').value),
                    maxRetries: parseInt(document.getElementById('maxRetries').value),
                    timeout: parseInt(document.getElementById('timeout').value)
                }
            };

            fetch('/api/spider/save-config', {
                method: 'POST',
                headers: {
                    'Content-Type': 'application/json'
                },
                body: JSON.stringify(config)
            })
            .then(response => response.json())
            .then(data => {
                if (data.success) {
                    alert('配置已保存！');
                } else {
                    alert('保存失败：' + data.message);
                }
            })
            .catch(error => {
                alert('保存出错：' + error.message);
            });
        }

        // 更新爬虫日志
        function updateCrawlLog(message) {
            const log = document.getElementById('crawlLog');
            const timestamp = new Date().toLocaleTimeString();
            log.innerHTML += `[${timestamp}] ${message}\n`;
            log.scrollTop = log.scrollHeight;
        }

        // WebSocket连接用于实时更新爬虫状态
        const ws = new WebSocket(`ws://${window.location.host}/ws/spider-status`);

        ws.onmessage = function(event) {
            const data = JSON.parse(event.data);
            if (data.type === 'progress') {
                document.getElementById('crawlProgress').style.width = data.value + '%';
            } else if (data.type === 'log') {
                updateCrawlLog(data.message);
            }
        };

        // AI配置生成
        async function generateConfig() {
            const prompt = document.getElementById('aiPrompt').value.trim();
            if (!prompt) {
                alert('请输入您的爬虫需求描述！');
                return;
            }

            const aiResponse = document.getElementById('aiResponse');
            const aiSuggestion = document.getElementById('aiSuggestion');

            try {
                const response = await fetch('/api/spider/ai-config', {
                    method: 'POST',
                    headers: {
                        'Content-Type': 'application/json'
                    },
                    body: JSON.stringify({ prompt })
                });

                const data = await response.json();
                if (data.success) {
                    // 显示AI建议
                    aiSuggestion.textContent = data.suggestion;
                    aiResponse.style.display = 'block';

                    // 如果选择自动应用配置
                    if (document.getElementById('autoApply').checked) {
                        // 清除现有选择
                        selectedTopics.clear();

                        // 应用新的话题
                        data.config.topics.forEach(topic => {
                            selectedTopics.add(topic);
                        });

                        // 更新参数
                        document.getElementById('crawlDepth').value = data.config.parameters.crawlDepth;
                        document.getElementById('interval').value = data.config.parameters.interval;
                        document.getElementById('maxRetries').value = data.config.parameters.maxRetries;
                        document.getElementById('timeout').value = data.config.parameters.timeout;

                        // 更新UI
                        updateSelectedTopicsList();

                        // 添加提示
                        updateCrawlLog('AI配置已自动应用');
                    }
                } else {
                    throw new Error(data.message);
                }
            } catch (error) {
                aiSuggestion.textContent = '生成配置时出错：' + error.message;
                aiResponse.style.display = 'block';
            }
        }
    </script>
</body>
</html>