Upload 5 files

README.md

@@ -1,12 +1,10 @@
 ---
-title: Hf Search
-emoji: 🏆
-colorFrom: blue
-colorTo: yellow
+title: HF Search
+emoji: 🔍
+colorFrom: gray
+colorTo: gray
 sdk: gradio
 sdk_version: 5.13.2
 app_file: app.py
 pinned: false
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+---

app.py

@@ -0,0 +1,129 @@
+from html import escape
+import os
+import shutil
+import time
+from datetime import datetime, timedelta
+import bleach
+import requests
+from flask import Flask, request, jsonify, render_template
+from whoosh.index import create_in, open_dir
+from whoosh.fields import Schema, TEXT, ID, BOOLEAN
+from whoosh.qparser import QueryParser
+
+app = Flask(__name__)
+
+# Configure index directory and schema
+BASE_INDEX_DIR = "discussion_indices"
+CACHE_DURATION = timedelta(hours=24)
+
+schema = Schema(
+    discussion_id=ID(stored=True),
+    title=TEXT(stored=True),
+    content=TEXT(stored=True),
+    author=TEXT(stored=True),
+    is_pr=BOOLEAN(stored=True),
+    is_open=BOOLEAN(stored=True)
+)
+
+def get_repo_index_dir(repo_name):
+    # Convert repo name to safe directory name
+    safe_name = repo_name.replace('/', '_')
+    return os.path.join(BASE_INDEX_DIR, safe_name)
+
+def get_repo_last_indexed_file(repo_name):
+    return os.path.join(get_repo_index_dir(repo_name), 'last_indexed.txt')
+
+def needs_reindex(repo_name):
+    last_indexed_file = get_repo_last_indexed_file(repo_name)
+    if not os.path.exists(last_indexed_file):
+        return True
+    
+    with open(last_indexed_file, 'r') as f:
+        last_indexed = datetime.fromtimestamp(float(f.read().strip()))
+    
+    return datetime.now() - last_indexed > CACHE_DURATION
+
+def index_discussions(repo_name):
+    index_dir = get_repo_index_dir(repo_name)
+    
+    # Clear and recreate index directory
+    if os.path.exists(index_dir):
+        shutil.rmtree(index_dir)
+    os.makedirs(index_dir, exist_ok=True)
+    
+    # Create index
+    ix = create_in(index_dir, schema)
+    writer = ix.writer()
+    
+    # Fetch and index discussions
+    discussions = requests.get(f'https://huggingface.co/api/{repo_name}/discussions').json()
+    
+    for discussion in discussions['discussions']:
+        comments = requests.get(
+            f'https://huggingface.co/api/{repo_name}/discussions/{discussion["num"]}'
+        ).json()
+        
+        # Combine all comments into one content string
+        content = []
+        for comment in comments['events']:
+            if comment['type'] == 'comment':
+                content.append(f'{comment["author"]["name"]}: {comment["data"]["latest"]["raw"]}')
+        writer.add_document(
+            discussion_id=str(discussion["num"]),
+            title=discussion["title"],
+            content=f"Title: {discussion['title']}\n\n" + '\n'.join(content),
+            author=discussion["author"]["name"],
+            is_pr=discussion["isPullRequest"],
+            is_open=discussion["status"] == "open"
+        )
+    
+    writer.commit()
+    
+    # Update last indexed timestamp
+    with open(get_repo_last_indexed_file(repo_name), 'w') as f:
+        f.write(str(time.time()))
+
+@app.route('/')
+def index():
+    repo_name = request.args.get('repo')
+    query = request.args.get('query')
+    if not repo_name:
+        return render_template('no_repo.html')
+    return render_template('index.html', repo_name=repo_name, query=query)
+
+@app.route('/search', methods=['POST'])
+def search():
+    data = request.json
+    query = data.get('query')
+    repo_name = data.get('repo')
+    if not repo_name:
+        return jsonify({'error': 'No repository provided'}), 400
+    
+    if not query:
+        return jsonify({'error': 'No query provided'}), 400
+        
+    # Check if we need to reindex
+    if needs_reindex(repo_name):
+        index_discussions(repo_name)
+    
+    # Search the index
+    ix = open_dir(get_repo_index_dir(repo_name))
+    with ix.searcher() as searcher:
+        query_parser = QueryParser("content", ix.schema)
+        q = query_parser.parse(query)
+        results = searcher.search(q)
+        
+        # Format results
+        formatted_results = [{
+            'discussion_id': escape(result['discussion_id']),
+            'title': escape(result['title']), 
+            'author': escape(result['author']),
+            'excerpt': bleach.clean(result.highlights("content"), tags=['b'], strip=True),
+            'is_pr': result['is_pr'],
+            'is_open': result['is_open']
+        } for result in results]
+        
+        return jsonify({'results': formatted_results})
+
+if __name__ == '__main__':
+    app.run(host='0.0.0.0', port=7860)

requirements.txt

@@ -0,0 +1,4 @@
+bleach
+whoosh
+flask
+requests

templates/index.html

@@ -0,0 +1,385 @@
+<!DOCTYPE html>
+<html>
+
+<head>
+    <meta name="viewport" content="width=device-width, initial-scale=1">
+    <title>HF Discussion Search</title>
+    <style>
+        :root {
+            --bg-primary: #ffffff;
+            --bg-secondary: #f3f4f6;
+            --text-primary: #111827;
+            --text-secondary: #4b5563;
+            --border-color: #e5e7eb;
+            --accent-color: #2563eb;
+            --accent-hover: #1d4ed8;
+            --focus-ring: #3b82f680;
+        }
+
+        @media (prefers-color-scheme: dark) {
+            :root {
+                --bg-primary: #1f2937;
+                --bg-secondary: #111827;
+                --text-primary: #f9fafb;
+                --text-secondary: #9ca3af;
+                --border-color: #374151;
+                --accent-color: #3b82f6;
+                --accent-hover: #2563eb;
+            }
+        }
+
+        * {
+            box-sizing: border-box;
+            font-family: system-ui, -apple-system, sans-serif;
+        }
+
+        /* Custom scrollbar styles */
+        ::-webkit-scrollbar {
+            width: 8px;
+        }
+
+        ::-webkit-scrollbar-track {
+            background: var(--bg-secondary);
+        }
+
+        ::-webkit-scrollbar-thumb {
+            background: var(--text-secondary);
+            border-radius: 4px;
+        }
+
+        ::-webkit-scrollbar-thumb:hover {
+            background: var(--text-primary);
+        }
+
+        /* Firefox scrollbar */
+        * {
+            scrollbar-width: thin;
+            scrollbar-color: var(--text-secondary) var(--bg-secondary);
+        }
+
+        body {
+            margin: 0;
+            padding: 20px;
+            background: var(--bg-primary);
+            color: var(--text-primary);
+            min-height: 100vh;
+        }
+
+        .container {
+            max-width: 800px;
+            margin: 0 auto;
+        }
+
+        h1 {
+            font-size: 24px;
+            margin: 0 0 20px 0;
+            color: var(--text-primary);
+        }
+
+        .search-container {
+            position: relative;
+            margin-bottom: 30px;
+        }
+
+        .search-input {
+            width: 100%;
+            padding: 12px 16px 12px 40px;
+            border: 1px solid var(--border-color);
+            border-radius: 8px;
+            background: var(--bg-secondary);
+            color: var(--text-primary);
+            font-size: 16px;
+            transition: border-color 0.2s, box-shadow 0.2s;
+        }
+
+        .search-icon {
+            position: absolute;
+            left: 12px;
+            top: 50%;
+            transform: translateY(-50%);
+            color: var(--text-secondary);
+            width: 20px;
+            height: 20px;
+        }
+
+        .search-input:focus {
+            outline: none;
+            border-color: var(--accent-color);
+            box-shadow: 0 0 0 3px var(--focus-ring);
+        }
+
+        .results {
+            display: grid;
+            gap: 16px;
+        }
+
+        .result-card {
+            padding: 16px;
+            background: var(--bg-secondary);
+            border: 1px solid var(--border-color);
+            border-radius: 8px;
+            transition: transform 0.2s;
+            position: relative;
+            display: flex;
+            gap: 16px;
+        }
+
+        .type-icon-container {
+            flex-shrink: 0;
+            width: 24px;
+            height: 24px;
+            color: var(--text-secondary);
+        }
+
+        .result-content {
+            flex-grow: 1;
+            min-width: 0;
+        }
+
+        .result-card.closed {
+            opacity: 0.75;
+        }
+
+        .result-card:hover {
+            transform: translateY(-2px);
+        }
+
+        .result-title {
+            font-size: 18px;
+            font-weight: 600;
+            margin-bottom: 8px;
+            color: var(--accent-color);
+        }
+
+        .result-author {
+            font-size: 14px;
+            color: var(--text-secondary);
+            margin-bottom: 12px;
+        }
+
+        .result-excerpt {
+            font-size: 14px;
+            line-height: 1.5;
+            color: var(--text-primary);
+        }
+
+        .result-excerpt mark {
+            background: var(--accent-color);
+            color: white;
+            padding: 0 2px;
+            border-radius: 2px;
+        }
+
+        .loader {
+            display: none;
+            justify-content: center;
+            margin: 40px 0;
+        }
+
+        .loader::after {
+            content: "";
+            width: 30px;
+            height: 30px;
+            border: 3px solid var(--border-color);
+            border-radius: 50%;
+            border-top-color: var(--accent-color);
+            animation: spin 1s linear infinite;
+        }
+
+        @keyframes spin {
+            to {
+                transform: rotate(360deg);
+            }
+        }
+
+        .no-results {
+            text-align: center;
+            color: var(--text-secondary);
+            padding: 40px 0;
+        }
+
+        .status-indicator {
+            position: absolute;
+            top: 16px;
+            right: 16px;
+            display: flex;
+            align-items: center;
+            gap: 4px;
+            font-size: 12px;
+            padding: 4px 8px;
+            border-radius: 12px;
+        }
+
+        .status-open {
+            background: #22c55e20;
+            color: #22c55e;
+        }
+
+        .status-closed {
+            background: #ef444420;
+            color: #ef4444;
+        }
+
+        .filter-controls {
+            display: flex;
+            gap: 8px;
+            margin-bottom: 16px;
+        }
+
+        .filter-button {
+            padding: 6px 12px;
+            border: 1px solid var(--border-color);
+            border-radius: 6px;
+            background: var(--bg-secondary);
+            color: var(--text-secondary);
+            cursor: pointer;
+        }
+
+        .filter-button.active {
+            background: var(--accent-color);
+            color: white;
+            border-color: var(--accent-color);
+        }
+
+        .type-icon {
+            width: 24px;
+            height: 24px;
+        }
+    </style>
+</head>
+
+<body>
+    <div class="container">
+        <div class="search-container">
+            <svg class="search-icon" viewBox="0 0 20 20" fill="currentColor">
+                <path fill-rule="evenodd" d="M8 4a4 4 0 100 8 4 4 0 000-8zM2 8a6 6 0 1110.89 3.476l4.817 4.817a1 1 0 01-1.414 1.414l-4.816-4.816A6 6 0 012 8z" clip-rule="evenodd" />
+            </svg>
+            <input type="text" id="searchInput" class="search-input" placeholder="Search discussions..." autofocus>
+        </div>
+        <div class="filter-controls">
+            <button class="filter-button active" data-filter="all">All</button>
+            <button class="filter-button" data-filter="open">Open</button>
+            <button class="filter-button" data-filter="closed">Closed</button>
+        </div>
+        <div id="loader" class="loader"></div>
+        <div id="results" class="results"></div>
+    </div>
+
+    <script>
+        const searchInput = document.getElementById('searchInput');
+        const loader = document.getElementById('loader');
+        const results = document.getElementById('results');
+        const filterButtons = document.querySelectorAll('.filter-button');
+        let currentFilter = 'all';
+        let searchTimeout;
+        let currentResults = [];
+
+        function getTypeIcon(isPR) {
+            return isPR ? 
+                `<svg class="type-icon" viewBox="0 0 16 16" fill="currentColor">
+                    <path fill-rule="evenodd" d="M7.177 3.073L9.573.677A.25.25 0 0110 .854v4.792a.25.25 0 01-.427.177L7.177 3.427a.25.25 0 010-.354zM3.75 2.5a.75.75 0 100 1.5.75.75 0 000-1.5zm-2.25.75a2.25 2.25 0 113 2.122v5.256a2.251 2.251 0 11-1.5 0V5.372A2.25 2.25 0 011.5 3.25zM11 2.5h-1V4h1a1 1 0 011 1v5.628a2.251 2.251 0 101.5 0V5A2.5 2.5 0 0011 2.5zm1 10.25a.75.75 0 111.5 0 .75.75 0 01-1.5 0zM3.75 12a.75.75 0 100 1.5.75.75 0 000-1.5z"/>
+                </svg>` :
+                `<svg class="type-icon" viewBox="0 0 16 16" fill="currentColor">
+                    <path fill-rule="evenodd" d="M1.5 2.75a.25.25 0 01.25-.25h8.5a.25.25 0 01.25.25v5.5a.25.25 0 01-.25.25h-3.5a.75.75 0 00-.53.22L3.5 11.44V9.25a.75.75 0 00-.75-.75h-1a.25.25 0 01-.25-.25v-5.5z"/>
+                </svg>`;
+        }
+
+        function renderResults(results) {
+            const filteredResults = results.filter(result => {
+                if (currentFilter === 'all') return true;
+                if (currentFilter === 'open') return result.is_open;
+                if (currentFilter === 'closed') return !result.is_open;
+            });
+
+            if (filteredResults.length === 0) {
+                if (currentFilter === 'all') {
+                    return '<div class="no-results">No results found</div>';
+                } else {
+                    return `<div class="no-results">No ${currentFilter} discussions found</div>`;
+                }
+            }
+
+            return filteredResults
+                .map(result => `
+                    <a href="https://huggingface.co/{{ repo_name }}/discussions/${result.discussion_id}" 
+                       class="result-card${!result.is_open ? ' closed' : ''}" 
+                       target="_blank" style="text-decoration: none;">
+                        <div class="type-icon-container">
+                            ${getTypeIcon(result.is_pr)}
+                        </div>
+                        <div class="result-content">
+                            <div class="result-title">${result.title}</div>
+                            <div class="result-author">by ${result.author}</div>
+                            <div class="result-excerpt">${result.excerpt}</div>
+                            <div class="status-indicator ${result.is_open ? 'status-open' : 'status-closed'}">
+                                ${result.is_open ? 'Open' : 'Closed'}
+                            </div>
+                        </div>
+                    </a>
+                `).join('');
+        }
+
+        filterButtons.forEach(button => {
+            button.addEventListener('click', () => {
+                filterButtons.forEach(btn => btn.classList.remove('active'));
+                button.classList.add('active');
+                currentFilter = button.dataset.filter;
+                results.innerHTML = renderResults(currentResults);
+            });
+        });
+
+        async function performSearch(query) {
+            loader.style.display = 'flex';
+            results.innerHTML = '';
+
+            try {
+                const response = await fetch('/search', {
+                    method: 'POST',
+                    headers: {
+                        'Content-Type': 'application/json',
+                    },
+                    body: JSON.stringify({
+                        query: query,
+                        repo: '{{ repo_name }}'
+                    })
+                });
+
+                const data = await response.json();
+                currentResults = data.results;
+                results.innerHTML = renderResults(currentResults);
+            } catch (error) {
+                results.innerHTML = '<div class="no-results">An error occurred while searching</div>';
+            } finally {
+                loader.style.display = 'none';
+            }
+        }
+
+        searchInput.addEventListener('input', (e) => {
+            clearTimeout(searchTimeout);
+            const query = e.target.value.trim();
+
+            if (query.length === 0) {
+                results.innerHTML = '';
+                return;
+            }
+
+            if (query.length < 2) return;
+
+            searchTimeout = setTimeout(() => {
+                performSearch(query);
+            }, 300);
+        });
+
+        // Pre-fill search input and trigger search if query parameter exists
+        window.addEventListener('DOMContentLoaded', () => {
+            const urlParams = new URLSearchParams(window.location.search);
+            const query = urlParams.get('query');
+            if (query) {
+                searchInput.value = query;
+                performSearch(query);
+            }
+        });
+    </script>
+</body>
+
+</html>

templates/no_repo.html

@@ -0,0 +1,13 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>No repository provided</title>
+</head>
+<body>
+    <h1>No repository provided</h1>
+    <p>If you are trying to access the Search API, please provide a repository name in the URL.</p>
+    <p>This is not intended to be used as a public/standalone service. Please see <a href="https://github.com/fakerybakery/hf-tools">the GitHub repository</a> for more information on how to use this tool.</p>
+</body>
+</html>