Download web/src/views/document.php from datamatters24/research-document-archive: direct link, hf CLI and curl.
- Browser
- Download file 37.1 kB
-
https://huggingface.co/datamatters24/research-document-archive/resolve/main/web/src/views/document.php
- Command line
-
hf download hf://datamatters24/research-document-archive/web/src/views/document.php
-
curl -L -o document.php https://huggingface.co/datamatters24/research-document-archive/resolve/main/web/src/views/document.php
37.1 kB
| /** | |
| * Document Viewer - Split Pane | |
| * | |
| * Variables: | |
| * $document - Document record with: id, file_path, pages, file_size, source_section, processed_at | |
| * $pages - Array of page metadata: page_number, has_ocr, has_entities | |
| * $entities - Array of entities: entity_text, entity_type, page_number, confidence | |
| */ | |
| $docId = $document['id'] ?? 0; | |
| $filename = basename($document['file_path'] ?? 'Unknown Document'); | |
| $totalPages = $document['pages'] ?? 0; | |
| $sourceSection = $document['source_section'] ?? ''; | |
| $collectionNames = [ | |
| 'doj_disclosures' => 'DOJ Disclosures (Epstein Files)', | |
| 'lincoln_archives' => 'Lincoln Archives', | |
| 'house_resolutions' => 'House Resolutions', | |
| 'area51_cia' => 'Area 51 / CIA Declassified', | |
| 'jfk_assassination' => 'JFK Assassination Records', | |
| 'court_records' => 'Court Records', | |
| 'foia' => 'FOIA Releases', | |
| 'house_oversight' => 'House Oversight', | |
| ]; | |
| $collectionName = $collectionNames[$sourceSection] ?? ucwords(str_replace('_', ' ', $sourceSection)); | |
| // Group entities by type | |
| $entityGroups = []; | |
| if (!empty($entities)) { | |
| foreach ($entities as $ent) { | |
| $type = $ent['entity_type'] ?? 'OTHER'; | |
| if (!isset($entityGroups[$type])) { | |
| $entityGroups[$type] = []; | |
| } | |
| $entityGroups[$type][] = $ent; | |
| } | |
| ksort($entityGroups); | |
| } | |
| $entityTypeColors = [ | |
| 'PERSON' => 'bg-blue-100 text-blue-800', | |
| 'ORGANIZATION' => 'bg-purple-100 text-purple-800', | |
| 'ORG' => 'bg-purple-100 text-purple-800', | |
| 'LOCATION' => 'bg-green-100 text-green-800', | |
| 'LOC' => 'bg-green-100 text-green-800', | |
| 'GPE' => 'bg-green-100 text-green-800', | |
| 'DATE' => 'bg-amber-100 text-amber-800', | |
| 'MONEY' => 'bg-emerald-100 text-emerald-800', | |
| 'LAW' => 'bg-red-100 text-red-800', | |
| 'EVENT' => 'bg-pink-100 text-pink-800', | |
| 'NORP' => 'bg-indigo-100 text-indigo-800', | |
| 'FAC' => 'bg-teal-100 text-teal-800', | |
| 'PRODUCT' => 'bg-orange-100 text-orange-800', | |
| 'WORK_OF_ART' => 'bg-rose-100 text-rose-800', | |
| ]; | |
| function getEntityColor(string $type, array $colors): string { | |
| return $colors[$type] ?? 'bg-gray-100 text-gray-800'; | |
| } | |
| $title = htmlspecialchars($filename) . ' - Research Document Archive'; | |
| $content = ''; | |
| ob_start(); | |
| <!-- Breadcrumb --> | |
| <nav class="flex items-center text-sm text-gray-500 mb-4" aria-label="Breadcrumb"> | |
| <a href="/" class="hover:text-gray-700 transition-colors">Home</a> | |
| <svg class="h-4 w-4 mx-2 text-gray-400" fill="none" viewBox="0 0 24 24" stroke="currentColor" stroke-width="2"> | |
| <path stroke-linecap="round" stroke-linejoin="round" d="M8.25 4.5l7.5 7.5-7.5 7.5" /> | |
| </svg> | |
| <a href="/browse/<?= htmlspecialchars(urlencode($sourceSection)) ?>" class="hover:text-gray-700 transition-colors"> | |
| <?= htmlspecialchars($collectionName) ?> | |
| </a> | |
| <svg class="h-4 w-4 mx-2 text-gray-400" fill="none" viewBox="0 0 24 24" stroke="currentColor" stroke-width="2"> | |
| <path stroke-linecap="round" stroke-linejoin="round" d="M8.25 4.5l7.5 7.5-7.5 7.5" /> | |
| </svg> | |
| <span class="text-gray-900 font-medium truncate max-w-xs"><?= htmlspecialchars($filename) ?></span> | |
| </nav> | |
| <!-- Document Header --> | |
| <div class="flex flex-col sm:flex-row sm:items-center sm:justify-between gap-3 mb-6"> | |
| <div> | |
| <h1 class="text-2xl font-bold text-gray-900 break-all"><?= htmlspecialchars($filename) ?></h1> | |
| <div class="mt-1 flex items-center space-x-3 text-sm text-gray-500"> | |
| <span class="inline-flex items-center px-2.5 py-0.5 rounded-full text-xs font-medium bg-gray-100 text-gray-700"> | |
| <?= htmlspecialchars($collectionName) ?> | |
| </span> | |
| <span><?= $totalPages ?> page<?= $totalPages !== 1 ? 's' : '' ?></span> | |
| </div> | |
| </div> | |
| <a href="/pdf/<?= (int)$docId ?>" target="_blank" | |
| class="inline-flex items-center px-4 py-2 text-sm font-medium text-gray-700 bg-white border border-gray-300 rounded-lg hover:bg-gray-50 transition-colors"> | |
| <svg class="h-4 w-4 mr-2" fill="none" viewBox="0 0 24 24" stroke="currentColor" stroke-width="2"> | |
| <path stroke-linecap="round" stroke-linejoin="round" d="M3 16.5v2.25A2.25 2.25 0 005.25 21h13.5A2.25 2.25 0 0021 18.75V16.5M16.5 12L12 16.5m0 0L7.5 12m4.5 4.5V3" /> | |
| </svg> | |
| Download PDF | |
| </a> | |
| </div> | |
| <!-- Page Navigation Bar --> | |
| <div class="bg-white border border-gray-200 rounded-lg shadow-sm p-3 mb-4"> | |
| <div class="flex items-center justify-between"> | |
| <button id="prev-page-btn" disabled | |
| class="inline-flex items-center px-3 py-1.5 text-sm font-medium text-gray-700 bg-gray-100 border border-gray-300 rounded-md hover:bg-gray-200 disabled:opacity-50 disabled:cursor-not-allowed transition-colors"> | |
| <svg class="h-4 w-4 mr-1" fill="none" viewBox="0 0 24 24" stroke="currentColor" stroke-width="2"> | |
| <path stroke-linecap="round" stroke-linejoin="round" d="M15 19l-7-7 7-7" /> | |
| </svg> | |
| Previous | |
| </button> | |
| <div class="flex items-center space-x-2"> | |
| <span class="text-sm text-gray-500">Page</span> | |
| <select id="page-selector" class="rounded-md border-gray-300 text-sm py-1.5 pl-3 pr-8 focus:border-blue-500 focus:ring-blue-500"> | |
| <?php for ($i = 1; $i <= $totalPages; $i++): ?> | |
| <option value="<?= $i ?>"><?= $i ?></option> | |
| <?php endfor; ?> | |
| </select> | |
| <span class="text-sm text-gray-500">of <?= $totalPages ?></span> | |
| </div> | |
| <div class="flex items-center space-x-2"> | |
| <a id="find-similar-btn" href="/similar/0" target="_blank" | |
| class="inline-flex items-center px-3 py-1.5 text-sm font-medium text-purple-700 bg-purple-50 border border-purple-200 rounded-md hover:bg-purple-100 transition-colors"> | |
| <svg class="h-4 w-4 mr-1" fill="none" viewBox="0 0 24 24" stroke="currentColor" stroke-width="2"> | |
| <path stroke-linecap="round" stroke-linejoin="round" d="M7.5 21L3 16.5m0 0L7.5 12M3 16.5h13.5m0-13.5L21 7.5m0 0L16.5 12M21 7.5H7.5" /> | |
| </svg> | |
| Find Similar | |
| </a> | |
| <button id="next-page-btn" | |
| class="inline-flex items-center px-3 py-1.5 text-sm font-medium text-gray-700 bg-gray-100 border border-gray-300 rounded-md hover:bg-gray-200 disabled:opacity-50 disabled:cursor-not-allowed transition-colors"> | |
| Next | |
| <svg class="h-4 w-4 ml-1" fill="none" viewBox="0 0 24 24" stroke="currentColor" stroke-width="2"> | |
| <path stroke-linecap="round" stroke-linejoin="round" d="M9 5l7 7-7 7" /> | |
| </svg> | |
| </button> | |
| </div> | |
| </div> | |
| </div> | |
| <!-- Split Pane Layout --> | |
| <div class="flex flex-col lg:flex-row gap-4" style="min-height: 70vh;"> | |
| <!-- Left: PDF Viewer --> | |
| <div class="lg:w-3/5 bg-white border border-gray-200 rounded-lg shadow-sm overflow-hidden flex flex-col"> | |
| <iframe id="pdf-viewer" | |
| src="/pdf/<?= (int)$docId ?>" | |
| class="flex-1 w-full" | |
| style="min-height: 600px; border: none;" | |
| title="PDF Document Viewer"> | |
| </iframe> | |
| </div> | |
| <!-- Right: OCR Text / Entities Tabs --> | |
| <div class="lg:w-2/5 bg-white border border-gray-200 rounded-lg shadow-sm flex flex-col overflow-hidden"> | |
| <!-- Tab Headers --> | |
| <div class="flex border-b border-gray-200 bg-gray-50"> | |
| <button id="tab-ocr" | |
| class="flex-1 px-4 py-3 text-sm font-medium text-blue-600 border-b-2 border-blue-600 bg-white transition-colors" | |
| data-tab="ocr"> | |
| <svg class="inline h-4 w-4 mr-1.5 -mt-0.5" fill="none" viewBox="0 0 24 24" stroke="currentColor" stroke-width="2"> | |
| <path stroke-linecap="round" stroke-linejoin="round" d="M3.75 6.75h16.5M3.75 12h16.5m-16.5 5.25H12" /> | |
| </svg> | |
| OCR Text | |
| </button> | |
| <button id="tab-entities" | |
| class="flex-1 px-4 py-3 text-sm font-medium text-gray-500 border-b-2 border-transparent hover:text-gray-700 transition-colors" | |
| data-tab="entities"> | |
| <svg class="inline h-4 w-4 mr-1.5 -mt-0.5" fill="none" viewBox="0 0 24 24" stroke="currentColor" stroke-width="2"> | |
| <path stroke-linecap="round" stroke-linejoin="round" d="M9.568 3H5.25A2.25 2.25 0 003 5.25v4.318c0 .597.237 1.17.659 1.591l9.581 9.581c.699.699 1.78.872 2.607.33a18.095 18.095 0 005.223-5.223c.542-.827.369-1.908-.33-2.607L11.16 3.66A2.25 2.25 0 009.568 3z" /> | |
| </svg> | |
| Entities | |
| <?php if (!empty($entities)): ?> | |
| <span class="ml-1 inline-flex items-center px-1.5 py-0.5 rounded-full text-xs font-medium bg-gray-200 text-gray-700"> | |
| <?= count($entities) ?> | |
| </span> | |
| <?php endif; ?> | |
| </button> | |
| <button id="tab-topics" | |
| class="flex-1 px-4 py-3 text-sm font-medium text-gray-500 border-b-2 border-transparent hover:text-gray-700 transition-colors" | |
| data-tab="topics"> | |
| <svg class="inline h-4 w-4 mr-1.5 -mt-0.5" fill="none" viewBox="0 0 24 24" stroke="currentColor" stroke-width="2"> | |
| <path stroke-linecap="round" stroke-linejoin="round" d="M9.568 3H5.25A2.25 2.25 0 003 5.25v4.318c0 .597.237 1.17.659 1.591l9.581 9.581c.699.699 1.78.872 2.607.33a18.095 18.095 0 005.223-5.223c.542-.827.369-1.908-.33-2.607L11.16 3.66A2.25 2.25 0 009.568 3z" /> | |
| <path stroke-linecap="round" stroke-linejoin="round" d="M6 6h.008v.008H6V6z" /> | |
| </svg> | |
| Topics | |
| <?php if (!empty($topics)): ?> | |
| <span class="ml-1 inline-flex items-center px-1.5 py-0.5 rounded-full text-xs font-medium bg-indigo-100 text-indigo-700"> | |
| <?= count(array_filter($topics, fn($s) => $s > 0.3)) ?> | |
| </span> | |
| <?php endif; ?> | |
| </button> | |
| <button id="tab-crisis" | |
| class="flex-1 px-4 py-3 text-sm font-medium text-gray-500 border-b-2 border-transparent hover:text-gray-700 transition-colors" | |
| data-tab="crisis"> | |
| <svg class="inline h-4 w-4 mr-1.5 -mt-0.5" fill="none" viewBox="0 0 24 24" stroke="currentColor" stroke-width="2"> | |
| <path stroke-linecap="round" stroke-linejoin="round" d="M12 6v6h4.5m4.5 0a9 9 0 11-18 0 9 9 0 0118 0z" /> | |
| </svg> | |
| Crisis | |
| <?php if (!empty($crisisEvents)): ?> | |
| <span class="ml-1 inline-flex items-center px-1.5 py-0.5 rounded-full text-xs font-medium bg-red-100 text-red-700"> | |
| <?= count($crisisEvents) ?> | |
| </span> | |
| <?php endif; ?> | |
| </button> | |
| </div> | |
| <!-- OCR Text Panel --> | |
| <div id="panel-ocr" class="flex-1 overflow-auto p-4"> | |
| <div id="ocr-loading" class="flex items-center justify-center py-12"> | |
| <svg class="animate-spin h-6 w-6 text-blue-500 mr-3" fill="none" viewBox="0 0 24 24"> | |
| <circle class="opacity-25" cx="12" cy="12" r="10" stroke="currentColor" stroke-width="4"></circle> | |
| <path class="opacity-75" fill="currentColor" d="M4 12a8 8 0 018-8V0C5.373 0 0 5.373 0 12h4zm2 5.291A7.962 7.962 0 014 12H0c0 3.042 1.135 5.824 3 7.938l3-2.647z"></path> | |
| </svg> | |
| <span class="text-sm text-gray-500">Loading OCR text...</span> | |
| </div> | |
| <div id="ocr-content" class="hidden"> | |
| <pre id="ocr-text" class="text-sm text-gray-800 whitespace-pre-wrap font-sans leading-relaxed"></pre> | |
| </div> | |
| <div id="ocr-empty" class="hidden text-center py-12"> | |
| <svg class="mx-auto h-10 w-10 text-gray-400" fill="none" viewBox="0 0 24 24" stroke="currentColor" stroke-width="1.5"> | |
| <path stroke-linecap="round" stroke-linejoin="round" d="M3.75 6.75h16.5M3.75 12h16.5m-16.5 5.25H12" /> | |
| </svg> | |
| <p class="mt-3 text-sm text-gray-500">No OCR text available for this page.</p> | |
| </div> | |
| <div id="ocr-error" class="hidden text-center py-12"> | |
| <svg class="mx-auto h-10 w-10 text-red-400" fill="none" viewBox="0 0 24 24" stroke="currentColor" stroke-width="1.5"> | |
| <path stroke-linecap="round" stroke-linejoin="round" d="M12 9v3.75m9-.75a9 9 0 11-18 0 9 9 0 0118 0zm-9 3.75h.008v.008H12v-.008z" /> | |
| </svg> | |
| <p class="mt-3 text-sm text-red-600">Failed to load OCR text.</p> | |
| </div> | |
| </div> | |
| <!-- Entities Panel --> | |
| <div id="panel-entities" class="flex-1 overflow-auto p-4 hidden"> | |
| <?php if (!empty($entityGroups)): ?> | |
| <div class="space-y-5"> | |
| <?php foreach ($entityGroups as $type => $ents): ?> | |
| <div> | |
| <h4 class="text-xs font-semibold text-gray-500 uppercase tracking-wider mb-2"> | |
| <?= htmlspecialchars($type) ?> | |
| <span class="text-gray-400 font-normal">(<?= count($ents) ?>)</span> | |
| </h4> | |
| <div class="flex flex-wrap gap-2"> | |
| <?php | |
| // Deduplicate entities by text | |
| $seen = []; | |
| foreach ($ents as $ent): | |
| $text = $ent['entity_text'] ?? ''; | |
| $key = strtolower(trim($text)); | |
| if (isset($seen[$key])) continue; | |
| $seen[$key] = true; | |
| $colorClass = getEntityColor($type, $entityTypeColors); | |
| ?> | |
| <span class="inline-flex items-center px-2.5 py-1 rounded-full text-xs font-medium <?= $colorClass ?>"> | |
| <?= htmlspecialchars($text) ?> | |
| <?php if (!empty($ent['page_number'])): ?> | |
| <span class="ml-1.5 opacity-60">p.<?= (int)$ent['page_number'] ?></span> | |
| <?php endif; ?> | |
| </span> | |
| <?php endforeach; ?> | |
| </div> | |
| </div> | |
| <?php endforeach; ?> | |
| </div> | |
| <?php else: ?> | |
| <div class="text-center py-12"> | |
| <svg class="mx-auto h-10 w-10 text-gray-400" fill="none" viewBox="0 0 24 24" stroke="currentColor" stroke-width="1.5"> | |
| <path stroke-linecap="round" stroke-linejoin="round" d="M9.568 3H5.25A2.25 2.25 0 003 5.25v4.318c0 .597.237 1.17.659 1.591l9.581 9.581c.699.699 1.78.872 2.607.33a18.095 18.095 0 005.223-5.223c.542-.827.369-1.908-.33-2.607L11.16 3.66A2.25 2.25 0 009.568 3z" /> | |
| </svg> | |
| <p class="mt-3 text-sm text-gray-500">No entities extracted for this document.</p> | |
| </div> | |
| <?php endif; ?> | |
| </div> | |
| <!-- Topics & Keywords Panel --> | |
| <div id="panel-topics" class="flex-1 overflow-auto p-4 hidden"> | |
| <?php if (!empty($topics)): ?> | |
| <div class="space-y-6"> | |
| <!-- Topic Classifications --> | |
| <div> | |
| <h4 class="text-xs font-semibold text-gray-500 uppercase tracking-wider mb-3">Topic Classifications</h4> | |
| <div class="space-y-2"> | |
| <?php | |
| $topicColors = [ | |
| 'intelligence operations' => 'bg-red-500', | |
| 'national security' => 'bg-orange-500', | |
| 'military operations' => 'bg-amber-600', | |
| 'surveillance' => 'bg-yellow-500', | |
| 'assassination' => 'bg-red-700', | |
| 'congressional legislation' => 'bg-blue-500', | |
| 'government oversight' => 'bg-indigo-500', | |
| 'civil rights' => 'bg-purple-500', | |
| 'foreign policy' => 'bg-teal-500', | |
| 'law enforcement' => 'bg-cyan-600', | |
| 'financial regulation' => 'bg-emerald-500', | |
| 'public health' => 'bg-green-500', | |
| 'human experimentation' => 'bg-rose-600', | |
| 'scientific research' => 'bg-violet-500', | |
| 'judicial proceedings' => 'bg-slate-500', | |
| ]; | |
| foreach ($topics as $topic => $score): | |
| if ($score < 0.1) continue; | |
| $pct = min(100, $score * 100); | |
| $barColor = $topicColors[$topic] ?? 'bg-gray-500'; | |
| ?> | |
| <div> | |
| <div class="flex justify-between items-center mb-1"> | |
| <span class="text-sm text-gray-700 capitalize"><?= htmlspecialchars($topic) ?></span> | |
| <span class="text-xs font-mono text-gray-500"><?= number_format($pct, 1) ?>%</span> | |
| </div> | |
| <div class="w-full bg-gray-200 rounded-full h-2"> | |
| <div class="<?= $barColor ?> h-2 rounded-full transition-all" style="width: <?= $pct ?>%"></div> | |
| </div> | |
| </div> | |
| <?php endforeach; ?> | |
| </div> | |
| </div> | |
| <?php if (!empty($keywords)): ?> | |
| <!-- Keywords --> | |
| <div> | |
| <h4 class="text-xs font-semibold text-gray-500 uppercase tracking-wider mb-3">Extracted Keywords</h4> | |
| <div class="flex flex-wrap gap-2"> | |
| <?php foreach ($keywords as $kw): | |
| $opacity = max(0.4, min(1.0, (float)$kw['score'])); | |
| ?> | |
| <span class="inline-flex items-center px-2.5 py-1 rounded-full text-xs font-medium bg-blue-50 text-blue-700 border border-blue-200" | |
| style="opacity: <?= $opacity ?>"> | |
| <?= htmlspecialchars($kw['keyword']) ?> | |
| </span> | |
| <?php endforeach; ?> | |
| </div> | |
| </div> | |
| <?php endif; ?> | |
| <?php | |
| $stamps = $forensic['stamps'] ?? []; | |
| $pdfMeta = $forensic['pdf_metadata'] ?? []; | |
| if (!empty($stamps) || !empty($pdfMeta) || !empty($redactionSummary)): | |
| ?> | |
| <!-- Forensic Analysis --> | |
| <div> | |
| <h4 class="text-xs font-semibold text-gray-500 uppercase tracking-wider mb-3">Forensic Analysis</h4> | |
| <?php if (!empty($stamps)): ?> | |
| <div class="mb-3"> | |
| <p class="text-[10px] font-medium text-gray-400 uppercase mb-1">Classification Stamps</p> | |
| <div class="flex flex-wrap gap-1.5"> | |
| <?php | |
| $stampColors = [ | |
| 'TOP SECRET' => 'bg-red-100 text-red-800 border-red-200', | |
| 'SECRET' => 'bg-orange-100 text-orange-800 border-orange-200', | |
| 'CONFIDENTIAL' => 'bg-yellow-100 text-yellow-800 border-yellow-200', | |
| 'CLASSIFIED' => 'bg-amber-100 text-amber-800 border-amber-200', | |
| 'UNCLASSIFIED' => 'bg-green-100 text-green-800 border-green-200', | |
| 'DECLASSIFIED' => 'bg-emerald-100 text-emerald-800 border-emerald-200', | |
| 'EYES ONLY' => 'bg-red-100 text-red-800 border-red-200', | |
| 'NOFORN' => 'bg-rose-100 text-rose-800 border-rose-200', | |
| 'REDACTED' => 'bg-gray-800 text-white border-gray-900', | |
| ]; | |
| foreach ($stamps as $s): | |
| $color = $stampColors[$s['stamp']] ?? 'bg-gray-100 text-gray-700 border-gray-200'; | |
| ?> | |
| <span class="inline-flex items-center px-2 py-0.5 rounded text-[10px] font-bold border <?= $color ?>"> | |
| <?= htmlspecialchars($s['stamp']) ?> | |
| <?php if ($s['count'] > 1): ?> | |
| <span class="ml-1 opacity-60">x<?= $s['count'] ?></span> | |
| <?php endif; ?> | |
| </span> | |
| <?php endforeach; ?> | |
| </div> | |
| </div> | |
| <?php endif; ?> | |
| <?php if (!empty($redactionSummary) && ($redactionSummary['total_redactions'] ?? 0) > 0): ?> | |
| <div class="mb-3"> | |
| <p class="text-[10px] font-medium text-gray-400 uppercase mb-1">Redaction Detection</p> | |
| <div class="bg-gray-50 rounded-lg p-3 border border-gray-200"> | |
| <div class="flex items-center justify-between"> | |
| <span class="text-sm text-gray-700"> | |
| <span class="font-semibold text-red-700"><?= $redactionSummary['total_redactions'] ?></span> | |
| redacted region<?= $redactionSummary['total_redactions'] != 1 ? 's' : '' ?> detected | |
| </span> | |
| <span class="text-xs text-gray-500"> | |
| max <?= number_format($redactionSummary['max_page_area_pct'] ?? 0, 1) ?>% area | |
| </span> | |
| </div> | |
| <div class="mt-1 w-full bg-gray-200 rounded-full h-1.5"> | |
| <div class="bg-red-600 h-1.5 rounded-full" style="width: <?= min(100, ($redactionSummary['max_page_area_pct'] ?? 0)) ?>%"></div> | |
| </div> | |
| </div> | |
| </div> | |
| <?php endif; ?> | |
| <?php if (!empty($pdfMeta)): ?> | |
| <div> | |
| <p class="text-[10px] font-medium text-gray-400 uppercase mb-1">PDF Metadata</p> | |
| <dl class="text-xs space-y-1"> | |
| <?php foreach ($pdfMeta as $key => $val): ?> | |
| <div class="flex"> | |
| <dt class="w-24 text-gray-500 flex-shrink-0"><?= htmlspecialchars(ucfirst($key)) ?></dt> | |
| <dd class="text-gray-700 truncate"><?= htmlspecialchars($val) ?></dd> | |
| </div> | |
| <?php endforeach; ?> | |
| </dl> | |
| </div> | |
| <?php endif; ?> | |
| </div> | |
| <?php endif; ?> | |
| <?php if (!empty($sentiment) && empty($sentiment['note'])): ?> | |
| <!-- Sentiment --> | |
| <div> | |
| <h4 class="text-xs font-semibold text-gray-500 uppercase tracking-wider mb-3">Sentiment</h4> | |
| <?php | |
| $pol = $sentiment['polarity'] ?? 0; | |
| $sub = $sentiment['subjectivity'] ?? 0; | |
| if ($pol > 0.1) { $polLabel = 'Positive'; $polColor = 'text-green-700'; } | |
| elseif ($pol < -0.1) { $polLabel = 'Negative'; $polColor = 'text-red-700'; } | |
| else { $polLabel = 'Neutral'; $polColor = 'text-gray-600'; } | |
| <div class="flex gap-4"> | |
| <div class="flex-1 bg-gray-50 rounded-lg p-2.5 border border-gray-200 text-center"> | |
| <p class="text-[10px] text-gray-400 uppercase">Polarity</p> | |
| <p class="text-sm font-bold <?= $polColor ?>"><?= $polLabel ?></p> | |
| <p class="text-[10px] text-gray-500 font-mono"><?= number_format($pol, 3) ?></p> | |
| </div> | |
| <div class="flex-1 bg-gray-50 rounded-lg p-2.5 border border-gray-200 text-center"> | |
| <p class="text-[10px] text-gray-400 uppercase">Subjectivity</p> | |
| <p class="text-sm font-bold text-gray-700"><?= number_format($sub * 100, 0) ?>%</p> | |
| <p class="text-[10px] text-gray-500"><?= $sub > 0.5 ? 'Subjective' : 'Objective' ?></p> | |
| </div> | |
| </div> | |
| </div> | |
| <?php endif; ?> | |
| </div> | |
| <?php else: ?> | |
| <div class="text-center py-12"> | |
| <svg class="mx-auto h-10 w-10 text-gray-400" fill="none" viewBox="0 0 24 24" stroke="currentColor" stroke-width="1.5"> | |
| <path stroke-linecap="round" stroke-linejoin="round" d="M9.568 3H5.25A2.25 2.25 0 003 5.25v4.318c0 .597.237 1.17.659 1.591l9.581 9.581c.699.699 1.78.872 2.607.33a18.095 18.095 0 005.223-5.223c.542-.827.369-1.908-.33-2.607L11.16 3.66A2.25 2.25 0 009.568 3z" /> | |
| </svg> | |
| <p class="mt-3 text-sm text-gray-500">No topic classifications available yet.</p> | |
| <p class="mt-1 text-xs text-gray-400">The ML pipeline may still be processing.</p> | |
| </div> | |
| <?php endif; ?> | |
| </div> | |
| <!-- Crisis Context Panel --> | |
| <div id="panel-crisis" class="flex-1 overflow-auto p-4 hidden"> | |
| <?php if (!empty($crisisEvents)): ?> | |
| <div class="space-y-4"> | |
| <?php | |
| $crisisMethodColors = [ | |
| 'date' => 'bg-blue-50 text-blue-700', | |
| 'keyword' => 'bg-green-50 text-green-700', | |
| 'entity' => 'bg-purple-50 text-purple-700', | |
| 'collection' => 'bg-amber-50 text-amber-700', | |
| ]; | |
| foreach ($crisisEvents as $ce): | |
| $methods = $ce['match_methods'] ?? '[]'; | |
| if (is_string($methods)) $methods = json_decode($methods, true) ?: []; | |
| $details = $ce['details'] ?? '{}'; | |
| if (is_string($details)) $details = json_decode($details, true) ?: []; | |
| $score = (float)($ce['relevance_score'] ?? 0); | |
| $scorePercent = min(100, $score * 100); | |
| $startDate = $ce['start_date'] ?? ''; | |
| $endDate = $ce['end_date'] ?? ''; | |
| $dateStr = $startDate ? date('M j, Y', strtotime($startDate)) : ''; | |
| if ($endDate && $endDate !== $startDate) { | |
| $dateStr .= ' — ' . date('M j, Y', strtotime($endDate)); | |
| } | |
| ?> | |
| <a href="/crisis/<?= (int)$ce['id'] ?>" | |
| class="block bg-gray-50 rounded-lg p-4 hover:bg-gray-100 transition-colors border border-gray-200"> | |
| <div class="flex items-start justify-between gap-2"> | |
| <div> | |
| <h4 class="text-sm font-semibold text-gray-900"> | |
| <?= htmlspecialchars($ce['event_name'] ?? '') ?> | |
| </h4> | |
| <p class="text-xs text-gray-500 mt-0.5"><?= $dateStr ?></p> | |
| </div> | |
| <span class="flex-shrink-0 inline-flex items-center px-2 py-1 rounded-full text-xs font-bold | |
| <?= $scorePercent >= 60 ? 'bg-red-100 text-red-800' : | |
| ($scorePercent >= 30 ? 'bg-yellow-100 text-yellow-800' : 'bg-gray-100 text-gray-700') ?>"> | |
| <?= number_format($scorePercent, 0) ?>% | |
| </span> | |
| </div> | |
| <!-- Relevance bar --> | |
| <div class="mt-2"> | |
| <div class="w-full bg-gray-200 rounded-full h-1.5"> | |
| <div class="bg-blue-500 h-1.5 rounded-full" style="width: <?= $scorePercent ?>%"></div> | |
| </div> | |
| </div> | |
| <!-- Match methods --> | |
| <div class="mt-2 flex flex-wrap gap-1"> | |
| <?php foreach ($methods as $m): | |
| $mColor = $crisisMethodColors[$m] ?? 'bg-gray-50 text-gray-700'; | |
| ?> | |
| <span class="inline-flex items-center px-1.5 py-0.5 rounded text-[10px] font-medium <?= $mColor ?>"> | |
| <?= htmlspecialchars($m) ?> | |
| </span> | |
| <?php endforeach; ?> | |
| </div> | |
| <!-- Matched keywords if present --> | |
| <?php if (!empty($details['matched_keywords'])): ?> | |
| <div class="mt-2"> | |
| <p class="text-[10px] font-medium text-gray-400 uppercase mb-1">Matched Keywords</p> | |
| <div class="flex flex-wrap gap-1"> | |
| <?php foreach ($details['matched_keywords'] as $kw): ?> | |
| <span class="inline-flex items-center px-1.5 py-0.5 rounded-full text-[10px] bg-green-50 text-green-700"> | |
| <?= htmlspecialchars($kw) ?> | |
| </span> | |
| <?php endforeach; ?> | |
| </div> | |
| </div> | |
| <?php endif; ?> | |
| </a> | |
| <?php endforeach; ?> | |
| </div> | |
| <?php else: ?> | |
| <div class="text-center py-12"> | |
| <svg class="mx-auto h-10 w-10 text-gray-400" fill="none" viewBox="0 0 24 24" stroke="currentColor" stroke-width="1.5"> | |
| <path stroke-linecap="round" stroke-linejoin="round" d="M12 6v6h4.5m4.5 0a9 9 0 11-18 0 9 9 0 0118 0z" /> | |
| </svg> | |
| <p class="mt-3 text-sm text-gray-500">No crisis correlations found for this document.</p> | |
| <p class="mt-1 text-xs text-gray-400">The ML pipeline may still be processing.</p> | |
| </div> | |
| <?php endif; ?> | |
| </div> | |
| </div> | |
| </div> | |
| <script> | |
| (function() { | |
| const docId = <?= (int)$docId ?>; | |
| const totalPages = <?= (int)$totalPages ?>; | |
| let currentPage = 1; | |
| // Page metadata from server (for "Find Similar" page IDs) | |
| const pagesMeta = <?= json_encode($pages ?? []) ?>; | |
| // DOM elements | |
| const pageSelector = document.getElementById('page-selector'); | |
| const prevBtn = document.getElementById('prev-page-btn'); | |
| const nextBtn = document.getElementById('next-page-btn'); | |
| const findSimilarBtn = document.getElementById('find-similar-btn'); | |
| const ocrLoading = document.getElementById('ocr-loading'); | |
| const ocrContent = document.getElementById('ocr-content'); | |
| const ocrText = document.getElementById('ocr-text'); | |
| const ocrEmpty = document.getElementById('ocr-empty'); | |
| const ocrError = document.getElementById('ocr-error'); | |
| // Tab elements | |
| const tabs = { | |
| ocr: { tab: document.getElementById('tab-ocr'), panel: document.getElementById('panel-ocr') }, | |
| entities: { tab: document.getElementById('tab-entities'), panel: document.getElementById('panel-entities') }, | |
| topics: { tab: document.getElementById('tab-topics'), panel: document.getElementById('panel-topics') }, | |
| crisis: { tab: document.getElementById('tab-crisis'), panel: document.getElementById('panel-crisis') }, | |
| }; | |
| // Tab switching | |
| function switchTab(active) { | |
| Object.keys(tabs).forEach(function(key) { | |
| var t = tabs[key]; | |
| if (key === active) { | |
| t.tab.classList.add('text-blue-600', 'border-blue-600', 'bg-white'); | |
| t.tab.classList.remove('text-gray-500', 'border-transparent'); | |
| t.panel.classList.remove('hidden'); | |
| } else { | |
| t.tab.classList.remove('text-blue-600', 'border-blue-600', 'bg-white'); | |
| t.tab.classList.add('text-gray-500', 'border-transparent'); | |
| t.panel.classList.add('hidden'); | |
| } | |
| }); | |
| } | |
| Object.keys(tabs).forEach(function(key) { | |
| tabs[key].tab.addEventListener('click', function() { switchTab(key); }); | |
| }); | |
| // Load OCR text for a page | |
| function loadOcrText(pageNum) { | |
| ocrLoading.classList.remove('hidden'); | |
| ocrContent.classList.add('hidden'); | |
| ocrEmpty.classList.add('hidden'); | |
| ocrError.classList.add('hidden'); | |
| fetch('/api/page/' + docId + '/' + pageNum) | |
| .then(function(response) { | |
| if (!response.ok) throw new Error('HTTP ' + response.status); | |
| return response.json(); | |
| }) | |
| .then(function(data) { | |
| ocrLoading.classList.add('hidden'); | |
| var text = data.ocr_text || data.text || ''; | |
| if (text.trim().length > 0) { | |
| ocrText.textContent = text; | |
| ocrContent.classList.remove('hidden'); | |
| } else { | |
| ocrEmpty.classList.remove('hidden'); | |
| } | |
| }) | |
| .catch(function() { | |
| ocrLoading.classList.add('hidden'); | |
| ocrError.classList.remove('hidden'); | |
| }); | |
| } | |
| // Update page navigation state | |
| function updateNavigation() { | |
| prevBtn.disabled = (currentPage <= 1); | |
| nextBtn.disabled = (currentPage >= totalPages); | |
| pageSelector.value = currentPage; | |
| // Update "Find Similar" link | |
| var pageId = 0; | |
| for (var i = 0; i < pagesMeta.length; i++) { | |
| if (pagesMeta[i].page_number == currentPage) { | |
| pageId = pagesMeta[i].id || 0; | |
| break; | |
| } | |
| } | |
| findSimilarBtn.href = '/similar/' + pageId; | |
| } | |
| // Navigate to a page | |
| function goToPage(pageNum) { | |
| if (pageNum < 1 || pageNum > totalPages) return; | |
| currentPage = pageNum; | |
| updateNavigation(); | |
| loadOcrText(currentPage); | |
| } | |
| // Event listeners | |
| prevBtn.addEventListener('click', function() { goToPage(currentPage - 1); }); | |
| nextBtn.addEventListener('click', function() { goToPage(currentPage + 1); }); | |
| pageSelector.addEventListener('change', function() { goToPage(parseInt(this.value, 10)); }); | |
| // Keyboard navigation | |
| document.addEventListener('keydown', function(e) { | |
| if (e.target.tagName === 'INPUT' || e.target.tagName === 'TEXTAREA' || e.target.tagName === 'SELECT') return; | |
| if (e.key === 'ArrowLeft') { goToPage(currentPage - 1); } | |
| if (e.key === 'ArrowRight') { goToPage(currentPage + 1); } | |
| }); | |
| // Initial load | |
| goToPage(1); | |
| })(); | |
| </script> | |
| <?php | |
| $content = ob_get_clean(); | |
| include __DIR__ . '/layout.php'; | |
| ?> | |