catlog22/Claude-Code-Workflow
1
0
Fork 0
mirror of https://github.com/catlog22/Claude-Code-Workflow.git synced 2026-02-12 02:37:45 +08:00
Code Issues Packages Projects Releases Wiki Activity

Enhance FastEmbed Integration and GPU Management

Browse Source 
- Updated Windows platform guidelines for path formats and Bash rules.
- Refactored CodexLens routes to improve GPU detection and indexing cancellation logic.
- Added FastEmbed installation status handling in the dashboard, including UI updates for installation and reinstallation options.
- Implemented local model management with improved API responses for downloaded models.
- Enhanced GPU selection logic in the model mode configuration.
- Improved error handling and user feedback for FastEmbed installation processes.
- Adjusted Python environment checks to avoid shell escaping issues on Windows.
This commit is contained in:
catlog22
2026-01-06 14:42:00 +08:00
parent 08099cdcb9
commit 6cb819cb3a
6 changed files with 659 additions and 100 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
.claude/CLAUDE.md
View File

@@ -1,5 +1,6 @@
# Claude Instructions # Claude Instructions
- **CLI Tools Usage**: @~/.claude/workflows/cli-tools-usage.md - **CLI Tools Usage**: @~/.claude/workflows/cli-tools-usage.md
- **Coding Philosophy**: @~/.claude/workflows/coding-philosophy.md - **Coding Philosophy**: @~/.claude/workflows/coding-philosophy.md
- **Context Requirements**: @~/.claude/workflows/context-tools.md - **Context Requirements**: @~/.claude/workflows/context-tools.md

23
.claude/workflows/windows-platform.md
View File

@@ -1,16 +1,19 @@
# Windows Platform Guidelines # Windows Platform Guidelines
## Path Format Guidelines ## Path Format
### MCP Tools - **MCP Tools**: `D:\\path\\file.txt`
- Use double backslash: `D:\\path\\file.txt` - **Bash**: `D:/path/file.txt` or `/d/path/file.txt`
- Example: `read_file(paths="D:\\Claude_dms3\\src\\index.ts")` - **Relative**: `./src/index.ts`
### Bash Commands ## Bash Rules (Prevent Garbage Files)
- Use forward slash: `D:/path/file.txt` or `/d/path/file.txt`
- Example: `cd D:/Claude_dms3/src`
### Relative Paths 1. **Null redirect**: `command > NUL 2>&1`
- Universal format works in both MCP and Bash contexts 2. **Quote all**: `echo "$VAR"`, `cat "file name.txt"`
- Example: `./src/index.ts` or `../shared/utils.ts` 3. **Variable assignment**: `export VAR=value && command`
4. **Regex escape**: `grep -F "State<T>"` or `grep "State\<T\>"`
5. **Pipe output**: `command 2>&1 | ...` (avoid bare command output)
## Tool Priority
MCP Tools > PowerShell > Git Bash > cmd

95
ccw/src/core/routes/codexlens-routes.ts
View File

@@ -423,20 +423,14 @@ export async function handleCodexLensRoutes(ctx: RouteContext): Promise<boolean>
watcherProcess = null; watcherProcess = null;
} }
// Cancel any running indexing process // Cancel any running indexing process using exported function
if (currentIndexingProcess) { if (isIndexingInProgress()) {
console.log('[CodexLens] Cancelling indexing before uninstall...'); console.log('[CodexLens] Cancelling indexing before uninstall...');
try { try {
if (process.platform === 'win32') { cancelIndexing();
const { execSync } = require('child_process');
execSync(`taskkill /pid ${currentIndexingProcess.pid} /T /F`, { stdio: 'ignore' });
} else {
currentIndexingProcess.kill('SIGKILL');
}
} catch { } catch {
// Ignore errors // Ignore errors
} }
currentIndexingProcess = null;
} }
// Wait a moment for processes to fully exit and release handles // Wait a moment for processes to fully exit and release handles
@@ -977,27 +971,80 @@ export async function handleCodexLensRoutes(ctx: RouteContext): Promise<boolean>
// API: List available GPU devices for selection // API: List available GPU devices for selection
if (pathname === '/api/codexlens/gpu/list' && req.method === 'GET') { if (pathname === '/api/codexlens/gpu/list' && req.method === 'GET') {
try { try {
// Check if CodexLens is installed first (without auto-installing) // Try CodexLens gpu-list first if available
const venvStatus = await checkVenvStatus(); const venvStatus = await checkVenvStatus();
if (!venvStatus.ready) { if (venvStatus.ready) {
res.writeHead(200, { 'Content-Type': 'application/json' }); const result = await executeCodexLens(['gpu-list', '--json']);
res.end(JSON.stringify({ success: true, devices: [], selected_device_id: null })); if (result.success) {
return true; try {
const parsed = extractJSON(result.output);
if (parsed.devices && parsed.devices.length > 0) {
res.writeHead(200, { 'Content-Type': 'application/json' });
res.end(JSON.stringify(parsed));
return true;
}
} catch {
// Fall through to system detection
}
}
} }
const result = await executeCodexLens(['gpu-list', '--json']);
if (result.success) { // Fallback: Use system commands to detect GPUs
const devices: Array<{ name: string; type: string; index: number }> = [];
if (process.platform === 'win32') {
// Windows: Use WMIC to get GPU info
try { try {
const parsed = extractJSON(result.output); const { execSync } = await import('child_process');
res.writeHead(200, { 'Content-Type': 'application/json' }); const wmicOutput = execSync('wmic path win32_VideoController get name', {
res.end(JSON.stringify(parsed)); encoding: 'utf-8',
} catch { timeout: 10000,
res.writeHead(200, { 'Content-Type': 'application/json' }); stdio: ['pipe', 'pipe', 'pipe']
res.end(JSON.stringify({ success: true, devices: [], output: result.output })); });
const lines = wmicOutput.split('\n')
.map(line => line.trim())
.filter(line => line && line !== 'Name');
lines.forEach((name, index) => {
if (name) {
const isIntegrated = name.toLowerCase().includes('intel') ||
name.toLowerCase().includes('integrated');
devices.push({
name: name,
type: isIntegrated ? 'integrated' : 'discrete',
index: index
});
}
});
} catch (e) {
console.warn('[CodexLens] WMIC GPU detection failed:', (e as Error).message);
} }
} else { } else {
res.writeHead(500, { 'Content-Type': 'application/json' }); // Linux/Mac: Try nvidia-smi for NVIDIA GPUs
res.end(JSON.stringify({ success: false, error: result.error })); try {
const { execSync } = await import('child_process');
const nvidiaOutput = execSync('nvidia-smi --query-gpu=name --format=csv,noheader', {
encoding: 'utf-8',
timeout: 10000,
stdio: ['pipe', 'pipe', 'pipe']
});
const lines = nvidiaOutput.split('\n').filter(line => line.trim());
lines.forEach((name, index) => {
devices.push({
name: name.trim(),
type: 'discrete',
index: index
});
});
} catch {
// NVIDIA not available, that's fine
}
} }
res.writeHead(200, { 'Content-Type': 'application/json' });
res.end(JSON.stringify({ success: true, devices: devices, selected_device_id: null }));
} catch (err) { } catch (err) {
res.writeHead(500, { 'Content-Type': 'application/json' }); res.writeHead(500, { 'Content-Type': 'application/json' });
res.end(JSON.stringify({ success: false, error: err.message })); res.end(JSON.stringify({ success: false, error: err.message }));

52
ccw/src/templates/dashboard-js/i18n.js
View File

@@ -436,6 +436,32 @@ const i18n = {
'codexlens.modelListError': 'Failed to load models', 'codexlens.modelListError': 'Failed to load models',
'codexlens.noModelsAvailable': 'No models available', 'codexlens.noModelsAvailable': 'No models available',
// FastEmbed Installation
'codexlens.fastembedNotInstalled': 'FastEmbed not installed',
'codexlens.fastembedDesc': 'FastEmbed provides local embedding models for semantic search',
'codexlens.selectMode': 'Select Mode',
'codexlens.cpuModeShort': 'Standard',
'codexlens.directmlModeShort': 'Windows GPU',
'codexlens.cudaModeShort': 'NVIDIA GPU',
'codexlens.installFastembed': 'Install FastEmbed',
'codexlens.installingFastembed': 'Installing FastEmbed...',
'codexlens.installMayTakeTime': 'This may take several minutes...',
'codexlens.fastembedInstalled': 'FastEmbed Installed',
'codexlens.fastembedInstallFailed': 'FastEmbed installation failed',
'codexlens.installFastembedFirst': 'Install FastEmbed above to manage local embedding models',
'codexlens.detectedGpus': 'Detected GPUs',
'codexlens.reinstallOptions': 'Reinstall Options',
'codexlens.reinstallDesc': 'Reinstall with a different GPU mode:',
'codexlens.confirmReinstall': 'This will reinstall FastEmbed. Continue?',
'codexlens.fastembedReinstalled': 'FastEmbed reinstalled',
'codexlens.reinstallingFastembed': 'Reinstalling FastEmbed...',
'codexlens.activeAccelerator': 'Active Accelerator',
'codexlens.active': 'Active',
'codexlens.downloadedModels': 'Downloaded Models',
'codexlens.noLocalModels': 'No models downloaded',
'codexlens.configuredModels': 'Configured in API Settings',
'codexlens.commonModels': 'Common Models',
// Model Download Progress // Model Download Progress
'codexlens.downloadingModel': 'Downloading', 'codexlens.downloadingModel': 'Downloading',
'codexlens.connectingToHuggingFace': 'Connecting to Hugging Face...', 'codexlens.connectingToHuggingFace': 'Connecting to Hugging Face...',
@@ -2418,6 +2444,32 @@ const i18n = {
'codexlens.modelListError': '加载模型列表失败', 'codexlens.modelListError': '加载模型列表失败',
'codexlens.noModelsAvailable': '没有可用模型', 'codexlens.noModelsAvailable': '没有可用模型',
// FastEmbed 安装
'codexlens.fastembedNotInstalled': 'FastEmbed 未安装',
'codexlens.fastembedDesc': 'FastEmbed 提供本地嵌入模型用于语义搜索',
'codexlens.selectMode': '选择模式',
'codexlens.cpuModeShort': '标准',
'codexlens.directmlModeShort': 'Windows GPU',
'codexlens.cudaModeShort': 'NVIDIA GPU',
'codexlens.installFastembed': '安装 FastEmbed',
'codexlens.installingFastembed': '正在安装 FastEmbed...',
'codexlens.installMayTakeTime': '这可能需要几分钟...',
'codexlens.fastembedInstalled': 'FastEmbed 已安装',
'codexlens.fastembedInstallFailed': 'FastEmbed 安装失败',
'codexlens.installFastembedFirst': '请先在上方安装 FastEmbed 以管理本地嵌入模型',
'codexlens.detectedGpus': '检测到的 GPU',
'codexlens.reinstallOptions': '重新安装选项',
'codexlens.reinstallDesc': '使用不同的 GPU 模式重新安装：',
'codexlens.confirmReinstall': '这将重新安装 FastEmbed。是否继续？',
'codexlens.fastembedReinstalled': 'FastEmbed 已重新安装',
'codexlens.reinstallingFastembed': '正在重新安装 FastEmbed...',
'codexlens.activeAccelerator': '当前加速器',
'codexlens.active': '使用中',
'codexlens.downloadedModels': '已下载模型',
'codexlens.noLocalModels': '暂无已下载模型',
'codexlens.configuredModels': '已配置的 API 模型',
'codexlens.commonModels': '常用模型',
// 模型下载进度 // 模型下载进度
'codexlens.downloadingModel': '正在下载', 'codexlens.downloadingModel': '正在下载',
'codexlens.connectingToHuggingFace': '正在连接 Hugging Face...', 'codexlens.connectingToHuggingFace': '正在连接 Hugging Face...',

583
ccw/src/templates/dashboard-js/views/codexlens-manager.js
View File

@@ -511,6 +511,9 @@ function initCodexLensConfigEvents(currentConfig) {
}; };
} }
// Load FastEmbed installation status (show/hide install card)
loadFastEmbedInstallStatus();
// Load semantic dependencies status // Load semantic dependencies status
loadSemanticDepsStatus(); loadSemanticDepsStatus();
@@ -749,11 +752,13 @@ async function loadEnvVariables() {
container.innerHTML = '<div class="text-xs text-muted-foreground animate-pulse">Loading...</div>'; container.innerHTML = '<div class="text-xs text-muted-foreground animate-pulse">Loading...</div>';
try { try {
// Fetch env vars and configured models in parallel // Fetch env vars, configured models, and local models in parallel
var [envResponse, embeddingPoolResponse, rerankerPoolResponse] = await Promise.all([ var [envResponse, embeddingPoolResponse, rerankerPoolResponse, localModelsResponse, localRerankerModelsResponse] = await Promise.all([
fetch('/api/codexlens/env'), fetch('/api/codexlens/env'),
fetch('/api/litellm-api/embedding-pool').catch(function() { return null; }), fetch('/api/litellm-api/embedding-pool').catch(function() { return null; }),
fetch('/api/litellm-api/reranker-pool').catch(function() { return null; }) fetch('/api/litellm-api/reranker-pool').catch(function() { return null; }),
fetch('/api/codexlens/models').catch(function() { return null; }),
fetch('/api/codexlens/reranker/models').catch(function() { return null; })
]); ]);
var result = await envResponse.json(); var result = await envResponse.json();
@@ -777,6 +782,26 @@ async function loadEnvVariables() {
configuredRerankerModels = rerankerData.availableModels || []; configuredRerankerModels = rerankerData.availableModels || [];
} }
// Get local downloaded embedding models
var localEmbeddingModels = [];
if (localModelsResponse && localModelsResponse.ok) {
var localData = await localModelsResponse.json();
if (localData.success && localData.models) {
// Filter to only downloaded models
localEmbeddingModels = localData.models.filter(function(m) { return m.downloaded; });
}
}
// Get local downloaded reranker models
var localRerankerModels = [];
if (localRerankerModelsResponse && localRerankerModelsResponse.ok) {
var localRerankerData = await localRerankerModelsResponse.json();
if (localRerankerData.success && localRerankerData.models) {
// Filter to only downloaded models
localRerankerModels = localRerankerData.models.filter(function(m) { return m.downloaded; });
}
}
var env = result.env || {}; var env = result.env || {};
var settings = result.settings || {}; // Current settings from settings.json var settings = result.settings || {}; // Current settings from settings.json
var html = '<div class="space-y-4">'; var html = '<div class="space-y-4">';
@@ -842,9 +867,12 @@ async function loadEnvVariables() {
var backendKey = isEmbedding ? 'CODEXLENS_EMBEDDING_BACKEND' : 'CODEXLENS_RERANKER_BACKEND'; var backendKey = isEmbedding ? 'CODEXLENS_EMBEDDING_BACKEND' : 'CODEXLENS_RERANKER_BACKEND';
var isApiBackend = env[backendKey] === 'litellm' || env[backendKey] === 'api'; var isApiBackend = env[backendKey] === 'litellm' || env[backendKey] === 'api';
// Choose model list based on backend type // Get actual downloaded local models
var modelList = isApiBackend ? (config.apiModels || config.models || []) : (config.localModels || config.models || []); var actualLocalModels = isEmbedding ? localEmbeddingModels : localRerankerModels;
// Get configured API models
var configuredModels = isEmbedding ? configuredEmbeddingModels : configuredRerankerModels; var configuredModels = isEmbedding ? configuredEmbeddingModels : configuredRerankerModels;
// Fallback preset list for API models
var apiModelList = config.apiModels || [];
html += '<div class="flex items-center gap-2">' + html += '<div class="flex items-center gap-2">' +
'<label class="text-xs text-muted-foreground w-28 flex-shrink-0" title="' + escapeHtml(key) + '">' + escapeHtml(config.label) + '</label>' + '<label class="text-xs text-muted-foreground w-28 flex-shrink-0" title="' + escapeHtml(key) + '">' + escapeHtml(config.label) + '</label>' +
@@ -856,32 +884,45 @@ async function loadEnvVariables() {
'</div>' + '</div>' +
'<datalist id="' + datalistId + '">'; '<datalist id="' + datalistId + '">';
// For API backend: show configured models from API settings first if (isApiBackend) {
if (isApiBackend && configuredModels.length > 0) { // For API backend: show configured models from API settings first
html += '<option value="" disabled>-- Configured in API Settings --</option>'; if (configuredModels.length > 0) {
configuredModels.forEach(function(model) { html += '<option value="" disabled>-- ' + (t('codexlens.configuredModels') || 'Configured in API Settings') + ' --</option>';
var providers = model.providers ? model.providers.join(', ') : ''; configuredModels.forEach(function(model) {
html += '<option value="' + escapeHtml(model.modelId) + '">' + var providers = model.providers ? model.providers.join(', ') : '';
escapeHtml(model.modelName || model.modelId) + html += '<option value="' + escapeHtml(model.modelId) + '">' +
(providers ? ' (' + escapeHtml(providers) + ')' : '') + escapeHtml(model.modelName || model.modelId) +
'</option>'; (providers ? ' (' + escapeHtml(providers) + ')' : '') +
}); '</option>';
if (modelList.length > 0) { });
html += '<option value="" disabled>-- Common Models --</option>'; }
// Then show common API models as suggestions
if (apiModelList.length > 0) {
html += '<option value="" disabled>-- ' + (t('codexlens.commonModels') || 'Common Models') + ' --</option>';
apiModelList.forEach(function(group) {
group.items.forEach(function(model) {
// Skip if already in configured list
var exists = configuredModels.some(function(m) { return m.modelId === model; });
if (!exists) {
html += '<option value="' + escapeHtml(model) + '">' + escapeHtml(group.group) + ': ' + escapeHtml(model) + '</option>';
}
});
});
}
} else {
// For local backend (fastembed): show actually downloaded models
if (actualLocalModels.length > 0) {
html += '<option value="" disabled>-- ' + (t('codexlens.downloadedModels') || 'Downloaded Models') + ' --</option>';
actualLocalModels.forEach(function(model) {
var modelId = model.model_id || model.id || model.name;
var displayName = model.display_name || model.name || modelId;
html += '<option value="' + escapeHtml(modelId) + '">' + escapeHtml(displayName) + '</option>';
});
} else {
html += '<option value="" disabled>-- ' + (t('codexlens.noLocalModels') || 'No models downloaded') + ' --</option>';
} }
} }
// Add model list (local or API based on backend)
modelList.forEach(function(group) {
group.items.forEach(function(model) {
// Skip if already in configured list
var exists = configuredModels.some(function(m) { return m.modelId === model; });
if (!exists) {
html += '<option value="' + escapeHtml(model) + '">' + escapeHtml(group.group) + ': ' + escapeHtml(model) + '</option>';
}
});
});
html += '</datalist></div>'; html += '</datalist></div>';
} else { } else {
var inputType = config.type || 'text'; var inputType = config.type || 'text';
@@ -1558,6 +1599,432 @@ async function installSplade(gpu) {
// MODEL MANAGEMENT // MODEL MANAGEMENT
// ============================================================ // ============================================================
/**
* Build FastEmbed installation card UI with GPU mode options
* @param {Array} gpuDevices - List of detected GPU devices
*/
function buildFastEmbedInstallCardUI(gpuDevices) {
gpuDevices = gpuDevices || [];
// Build GPU devices info section
var gpuInfoHtml = '';
if (gpuDevices.length > 0) {
gpuInfoHtml =
'<div class="mb-4 p-3 bg-muted/30 rounded-lg">' +
'<div class="text-xs font-medium text-muted-foreground mb-2">' +
'<i data-lucide="monitor" class="w-3.5 h-3.5 inline mr-1"></i>' +
(t('codexlens.detectedGpus') || 'Detected GPUs') + ':' +
'</div>' +
'<div class="space-y-1">';
gpuDevices.forEach(function(device) {
var typeIcon = device.type === 'integrated' ? 'cpu' : 'zap';
var typeClass = device.type === 'integrated' ? 'text-muted-foreground' : 'text-green-500';
gpuInfoHtml +=
'<div class="flex items-center gap-2 text-sm">' +
'<i data-lucide="' + typeIcon + '" class="w-3.5 h-3.5 ' + typeClass + '"></i>' +
'<span>' + escapeHtml(device.name) + '</span>' +
'<span class="text-xs text-muted-foreground">(' + (device.type === 'integrated' ? 'Integrated' : 'Discrete') + ')</span>' +
'</div>';
});
gpuInfoHtml += '</div></div>';
}
return '<div class="bg-card border border-warning/30 rounded-lg overflow-hidden">' +
// Header
'<div class="bg-warning/10 border-b border-warning/20 px-4 py-3">' +
'<div class="flex items-center gap-2">' +
'<i data-lucide="alert-circle" class="w-5 h-5 text-warning"></i>' +
'<h4 class="font-semibold">' + (t('codexlens.fastembedNotInstalled') || 'FastEmbed Not Installed') + '</h4>' +
'</div>' +
'</div>' +
// Content
'<div class="p-4 space-y-4">' +
'<p class="text-sm text-muted-foreground">' +
(t('codexlens.fastembedDesc') || 'FastEmbed provides local embedding models for semantic search. Select your preferred acceleration mode below.') +
'</p>' +
// Show detected GPUs
gpuInfoHtml +
// GPU Mode Cards
'<div class="space-y-2">' +
'<div class="text-xs font-medium text-muted-foreground mb-2">' +
(t('codexlens.selectMode') || 'Select Acceleration Mode') + ':' +
'</div>' +
'<div class="grid grid-cols-1 gap-2">' +
// CPU Option Card
'<label class="group flex items-center gap-3 p-3 border-2 border-border rounded-lg cursor-pointer transition-all hover:border-primary/50 has-[:checked]:border-primary has-[:checked]:bg-primary/5">' +
'<input type="radio" name="fastembedMode" value="cpu" class="sr-only" checked />' +
'<div class="flex items-center justify-center w-10 h-10 rounded-lg bg-muted group-has-[:checked]:bg-primary/20">' +
'<i data-lucide="cpu" class="w-5 h-5 text-muted-foreground group-has-[:checked]:text-primary"></i>' +
'</div>' +
'<div class="flex-1">' +
'<div class="font-medium">CPU</div>' +
'<div class="text-xs text-muted-foreground">' + (t('codexlens.cpuModeDesc') || 'Standard CPU processing, works on all systems') + '</div>' +
'</div>' +
'<div class="w-5 h-5 rounded-full border-2 border-muted group-has-[:checked]:border-primary group-has-[:checked]:bg-primary flex items-center justify-center">' +
'<div class="w-2 h-2 rounded-full bg-white opacity-0 group-has-[:checked]:opacity-100"></div>' +
'</div>' +
'</label>' +
// DirectML Option Card
'<label class="group flex items-center gap-3 p-3 border-2 border-border rounded-lg cursor-pointer transition-all hover:border-primary/50 has-[:checked]:border-primary has-[:checked]:bg-primary/5">' +
'<input type="radio" name="fastembedMode" value="directml" class="sr-only" />' +
'<div class="flex items-center justify-center w-10 h-10 rounded-lg bg-muted group-has-[:checked]:bg-primary/20">' +
'<i data-lucide="monitor" class="w-5 h-5 text-muted-foreground group-has-[:checked]:text-primary"></i>' +
'</div>' +
'<div class="flex-1">' +
'<div class="font-medium">DirectML</div>' +
'<div class="text-xs text-muted-foreground">' + (t('codexlens.directmlModeDesc') || 'Windows GPU acceleration (NVIDIA/AMD/Intel)') + '</div>' +
'</div>' +
'<div class="w-5 h-5 rounded-full border-2 border-muted group-has-[:checked]:border-primary group-has-[:checked]:bg-primary flex items-center justify-center">' +
'<div class="w-2 h-2 rounded-full bg-white opacity-0 group-has-[:checked]:opacity-100"></div>' +
'</div>' +
'</label>' +
// CUDA Option Card
'<label class="group flex items-center gap-3 p-3 border-2 border-border rounded-lg cursor-pointer transition-all hover:border-primary/50 has-[:checked]:border-primary has-[:checked]:bg-primary/5">' +
'<input type="radio" name="fastembedMode" value="cuda" class="sr-only" />' +
'<div class="flex items-center justify-center w-10 h-10 rounded-lg bg-muted group-has-[:checked]:bg-primary/20">' +
'<i data-lucide="zap" class="w-5 h-5 text-muted-foreground group-has-[:checked]:text-primary"></i>' +
'</div>' +
'<div class="flex-1">' +
'<div class="font-medium">CUDA</div>' +
'<div class="text-xs text-muted-foreground">' + (t('codexlens.cudaModeDesc') || 'NVIDIA GPU acceleration (requires CUDA Toolkit)') + '</div>' +
'</div>' +
'<div class="w-5 h-5 rounded-full border-2 border-muted group-has-[:checked]:border-primary group-has-[:checked]:bg-primary flex items-center justify-center">' +
'<div class="w-2 h-2 rounded-full bg-white opacity-0 group-has-[:checked]:opacity-100"></div>' +
'</div>' +
'</label>' +
'</div>' +
'</div>' +
// Install Button
'<button class="btn btn-primary w-full" onclick="installFastEmbed()">' +
'<i data-lucide="download" class="w-4 h-4 mr-2"></i> ' +
(t('codexlens.installFastembed') || 'Install FastEmbed') +
'</button>' +
'</div>' +
'</div>';
}
/**
* Build FastEmbed status card UI (when installed)
* @param {Object} status - Semantic status object
* @param {Array} gpuDevices - List of detected GPU devices
* @param {Object} litellmStatus - LiteLLM installation status from API settings endpoint
*/
function buildFastEmbedStatusCardUI(status, gpuDevices, litellmStatus) {
gpuDevices = gpuDevices || [];
litellmStatus = litellmStatus || {};
// Determine accelerator info
var accelerator = status.accelerator || 'CPU';
var acceleratorIcon = accelerator === 'CPU' ? 'cpu' :
accelerator.includes('CUDA') ? 'zap' : 'monitor';
var acceleratorClass = accelerator === 'CPU' ? 'text-muted-foreground' : 'text-green-500';
var acceleratorBgClass = accelerator === 'CPU' ? 'bg-muted/50' :
accelerator.includes('CUDA') ? 'bg-green-500/20' : 'bg-blue-500/20';
// Check if LiteLLM (ccw-litellm) is installed - use the same check as API Settings
var isLitellmInstalled = litellmStatus.installed === true;
// Build GPU devices section with active indicator
var gpuInfoHtml = '';
if (gpuDevices.length > 0) {
gpuInfoHtml =
'<div class="mb-3">' +
'<div class="text-xs font-medium text-muted-foreground mb-2">' +
'<i data-lucide="monitor" class="w-3 h-3 inline mr-1"></i>' +
(t('codexlens.detectedGpus') || 'Detected GPUs') +
'</div>' +
'<div class="space-y-1.5">';
gpuDevices.forEach(function(device, index) {
var isActive = false;
// Determine if this GPU matches the active accelerator
if (accelerator === 'CUDA' && device.type === 'discrete' && device.name.toLowerCase().includes('nvidia')) {
isActive = true;
} else if (accelerator === 'DirectML' && device.type === 'discrete') {
isActive = true;
} else if (accelerator === 'CPU' && device.type === 'integrated') {
isActive = index === 0; // First integrated GPU is likely active
}
var typeIcon = device.type === 'integrated' ? 'cpu' : 'zap';
var activeClass = isActive ? 'border-green-500 bg-green-500/10' : 'border-border bg-muted/30';
var activeBadge = isActive ?
'<span class="text-[10px] px-1.5 py-0.5 rounded bg-green-500/20 text-green-600 font-medium">' +
(t('codexlens.active') || 'Active') +
'</span>' : '';
gpuInfoHtml +=
'<div class="flex items-center justify-between p-2 rounded border ' + activeClass + '">' +
'<div class="flex items-center gap-2">' +
'<i data-lucide="' + typeIcon + '" class="w-3.5 h-3.5 ' + (isActive ? 'text-green-500' : 'text-muted-foreground') + '"></i>' +
'<div>' +
'<div class="text-xs font-medium">' + escapeHtml(device.name) + '</div>' +
'<div class="text-[10px] text-muted-foreground">' + (device.type === 'integrated' ? 'Integrated' : 'Discrete') + '</div>' +
'</div>' +
'</div>' +
activeBadge +
'</div>';
});
gpuInfoHtml += '</div></div>';
}
// Active accelerator section
var activeAcceleratorHtml =
'<div class="p-3 rounded-lg ' + acceleratorBgClass + ' mb-3">' +
'<div class="flex items-center justify-between">' +
'<div class="flex items-center gap-2">' +
'<i data-lucide="' + acceleratorIcon + '" class="w-5 h-5 ' + acceleratorClass + '"></i>' +
'<div>' +
'<div class="text-xs text-muted-foreground">' + (t('codexlens.activeAccelerator') || 'Active Accelerator') + '</div>' +
'<div class="font-semibold">' + accelerator + '</div>' +
'</div>' +
'</div>' +
'<div class="text-right text-xs">' +
'<div class="text-muted-foreground">Backend</div>' +
'<div class="font-medium">' + (status.backend || 'fastembed') + '</div>' +
'</div>' +
'</div>' +
'</div>';
return '<div class="bg-card border border-green-500/30 rounded-lg overflow-hidden">' +
// Header
'<div class="bg-green-500/10 border-b border-green-500/20 px-4 py-2">' +
'<div class="flex items-center justify-between">' +
'<div class="flex items-center gap-2">' +
'<i data-lucide="check-circle" class="w-4 h-4 text-green-500"></i>' +
'<h4 class="font-medium text-sm">' + (t('codexlens.fastembedInstalled') || 'FastEmbed Installed') + '</h4>' +
'</div>' +
'<div class="flex items-center gap-2">' +
// LiteLLM status badge
(isLitellmInstalled ?
'<span class="text-[10px] px-1.5 py-0.5 rounded-full bg-blue-500/20 text-blue-600" title="LiteLLM API Available">' +
'<i data-lucide="cloud" class="w-2.5 h-2.5 inline"></i> LiteLLM' +
'</span>' : '') +
'<span class="text-xs px-2 py-0.5 rounded-full bg-green-500/20 text-green-600">' +
'<i data-lucide="' + acceleratorIcon + '" class="w-3 h-3 inline mr-1"></i>' +
accelerator +
'</span>' +
'</div>' +
'</div>' +
'</div>' +
// Content
'<div class="p-3 space-y-3">' +
// Active accelerator section
activeAcceleratorHtml +
// GPU devices
gpuInfoHtml +
// Reinstall option (collapsed by default)
'<details class="text-xs">' +
'<summary class="cursor-pointer text-muted-foreground hover:text-foreground">' +
'<i data-lucide="settings" class="w-3 h-3 inline mr-1"></i>' +
(t('codexlens.reinstallOptions') || 'Reinstall Options') +
'</summary>' +
'<div class="mt-2 p-2 bg-muted/30 rounded space-y-2">' +
'<p class="text-muted-foreground">' + (t('codexlens.reinstallDesc') || 'Reinstall with a different GPU mode:') + '</p>' +
'<div class="flex gap-2">' +
'<button class="btn-xs ' + (accelerator === 'CPU' ? 'btn-primary' : 'btn-outline') + '" onclick="reinstallFastEmbed(\'cpu\')" ' + (accelerator === 'CPU' ? 'disabled' : '') + '>' +
'<i data-lucide="cpu" class="w-3 h-3 mr-1"></i>CPU' +
'</button>' +
'<button class="btn-xs ' + (accelerator === 'DirectML' ? 'btn-primary' : 'btn-outline') + '" onclick="reinstallFastEmbed(\'directml\')" ' + (accelerator === 'DirectML' ? 'disabled' : '') + '>' +
'<i data-lucide="monitor" class="w-3 h-3 mr-1"></i>DirectML' +
'</button>' +
'<button class="btn-xs ' + (accelerator === 'CUDA' ? 'btn-primary' : 'btn-outline') + '" onclick="reinstallFastEmbed(\'cuda\')" ' + (accelerator === 'CUDA' ? 'disabled' : '') + '>' +
'<i data-lucide="zap" class="w-3 h-3 mr-1"></i>CUDA' +
'</button>' +
'</div>' +
'</div>' +
'</details>' +
'</div>' +
'</div>';
}
/**
* Reinstall FastEmbed with specified GPU mode
* @param {string} mode - GPU mode: cpu, directml, cuda
*/
async function reinstallFastEmbed(mode) {
if (!confirm((t('codexlens.confirmReinstall') || 'This will reinstall FastEmbed with ' + mode + ' mode. Continue?'))) {
return;
}
var card = document.getElementById('fastembedInstallCard');
if (!card) return;
var modeLabels = {
cpu: 'CPU',
cuda: 'NVIDIA CUDA',
directml: 'DirectML'
};
// Show reinstalling state
card.innerHTML =
'<div class="bg-card border border-primary/30 rounded-lg overflow-hidden">' +
'<div class="bg-primary/10 border-b border-primary/20 px-4 py-2">' +
'<div class="flex items-center gap-2">' +
'<div class="animate-spin w-4 h-4 border-2 border-primary border-t-transparent rounded-full"></div>' +
'<h4 class="font-medium text-sm">' + (t('codexlens.reinstallingFastembed') || 'Reinstalling FastEmbed...') + '</h4>' +
'</div>' +
'</div>' +
'<div class="p-3 text-xs text-muted-foreground">' +
(t('codexlens.installingMode') || 'Installing with') + ': ' + modeLabels[mode] +
'</div>' +
'</div>';
try {
var response = await fetch('/api/codexlens/semantic/install', {
method: 'POST',
headers: { 'Content-Type': 'application/json' },
body: JSON.stringify({ gpuMode: mode })
});
var result = await response.json();
if (result.success) {
showRefreshToast((t('codexlens.fastembedReinstalled') || 'FastEmbed reinstalled') + ' (' + modeLabels[mode] + ')', 'success');
// Reload status
loadFastEmbedInstallStatus();
} else {
showRefreshToast((t('codexlens.fastembedInstallFailed') || 'FastEmbed reinstall failed') + ': ' + result.error, 'error');
loadFastEmbedInstallStatus();
}
} catch (err) {
showRefreshToast((t('common.error') || 'Error') + ': ' + err.message, 'error');
loadFastEmbedInstallStatus();
}
}
/**
* Load FastEmbed installation status and show card
* Card is always visible - shows install UI or status UI based on state
*/
async function loadFastEmbedInstallStatus() {
console.log('[CodexLens] loadFastEmbedInstallStatus called');
var card = document.getElementById('fastembedInstallCard');
console.log('[CodexLens] fastembedInstallCard element:', card);
if (!card) {
console.warn('[CodexLens] fastembedInstallCard element not found!');
return;
}
try {
// Load semantic status, GPU list, and LiteLLM status in parallel
console.log('[CodexLens] Fetching semantic status, GPU list, and LiteLLM status...');
var [semanticResponse, gpuResponse, litellmResponse] = await Promise.all([
fetch('/api/codexlens/semantic/status'),
fetch('/api/codexlens/gpu/list'),
fetch('/api/litellm-api/ccw-litellm/status').catch(function() { return { ok: false }; })
]);
var result = await semanticResponse.json();
var gpuResult = await gpuResponse.json();
var gpuDevices = gpuResult.devices || [];
// Get LiteLLM status (same endpoint as API Settings page)
var litellmStatus = {};
if (litellmResponse.ok) {
try {
litellmStatus = await litellmResponse.json();
} catch (e) {
console.warn('[CodexLens] Failed to parse LiteLLM status:', e);
}
}
console.log('[CodexLens] Semantic status:', result);
console.log('[CodexLens] GPU devices:', gpuDevices);
console.log('[CodexLens] LiteLLM status:', litellmStatus);
if (result.available) {
// FastEmbed is installed - show status card
console.log('[CodexLens] FastEmbed available, showing status card');
card.innerHTML = buildFastEmbedStatusCardUI(result, gpuDevices, litellmStatus);
card.classList.remove('hidden');
if (window.lucide) lucide.createIcons();
} else {
// FastEmbed not installed - show install card with GPU devices
console.log('[CodexLens] FastEmbed NOT available, showing install card');
card.innerHTML = buildFastEmbedInstallCardUI(gpuDevices);
card.classList.remove('hidden');
if (window.lucide) lucide.createIcons();
}
} catch (err) {
// On error, show install card without GPU info
console.error('[CodexLens] Error loading FastEmbed status:', err);
card.innerHTML = buildFastEmbedInstallCardUI([]);
card.classList.remove('hidden');
if (window.lucide) lucide.createIcons();
}
}
/**
* Install FastEmbed with selected GPU mode
*/
async function installFastEmbed() {
var card = document.getElementById('fastembedInstallCard');
if (!card) return;
// Get selected GPU mode
var selectedMode = 'cpu';
var radios = document.querySelectorAll('input[name="fastembedMode"]');
radios.forEach(function(radio) {
if (radio.checked) {
selectedMode = radio.value;
}
});
var modeLabels = {
cpu: 'CPU',
cuda: 'NVIDIA CUDA',
directml: 'DirectML'
};
// Show installing state in card
card.innerHTML =
'<div class="bg-card border border-primary/30 rounded-lg overflow-hidden">' +
'<div class="bg-primary/10 border-b border-primary/20 px-4 py-3">' +
'<div class="flex items-center gap-2">' +
'<div class="animate-spin w-5 h-5 border-2 border-primary border-t-transparent rounded-full"></div>' +
'<h4 class="font-semibold">' + (t('codexlens.installingFastembed') || 'Installing FastEmbed...') + '</h4>' +
'</div>' +
'</div>' +
'<div class="p-4 space-y-2">' +
'<div class="text-sm">' +
(t('codexlens.installingMode') || 'Installing with') + ': <span class="font-medium">' + modeLabels[selectedMode] + '</span>' +
'</div>' +
'<div class="text-xs text-muted-foreground">' +
(t('codexlens.installMayTakeTime') || 'This may take several minutes. Please do not close this page.') +
'</div>' +
'<div class="w-full bg-muted rounded-full h-1.5 mt-3">' +
'<div class="bg-primary h-1.5 rounded-full animate-pulse" style="width: 30%"></div>' +
'</div>' +
'</div>' +
'</div>';
try {
var response = await fetch('/api/codexlens/semantic/install', {
method: 'POST',
headers: { 'Content-Type': 'application/json' },
body: JSON.stringify({ gpuMode: selectedMode })
});
var result = await response.json();
if (result.success) {
showRefreshToast((t('codexlens.fastembedInstalled') || 'FastEmbed installed') + ' (' + modeLabels[selectedMode] + ')', 'success');
// Hide card and reload status
await loadFastEmbedInstallStatus();
await loadSemanticDepsStatus();
await loadModelList();
} else {
showRefreshToast((t('codexlens.fastembedInstallFailed') || 'FastEmbed installation failed') + ': ' + result.error, 'error');
await loadFastEmbedInstallStatus();
}
} catch (err) {
showRefreshToast(t('common.error') + ': ' + err.message, 'error');
await loadFastEmbedInstallStatus();
}
}
/** /**
* Copy text to clipboard * Copy text to clipboard
*/ */
@@ -1602,7 +2069,11 @@ async function loadModelList() {
if (!result.success) { if (!result.success) {
var errorMsg = result.error || ''; var errorMsg = result.error || '';
if (errorMsg.includes('fastembed not installed') || errorMsg.includes('Semantic')) { if (errorMsg.includes('fastembed not installed') || errorMsg.includes('Semantic')) {
html += '<div class="text-sm text-muted-foreground">' + t('codexlens.semanticNotInstalled') + '</div>'; // Just show a simple message - installation UI is in the separate card above
html += '<div class="flex items-center gap-2 text-sm text-muted-foreground p-3 bg-muted/30 rounded">' +
'<i data-lucide="info" class="w-4 h-4"></i>' +
'<span>' + (t('codexlens.installFastembedFirst') || 'Install FastEmbed above to manage local embedding models') + '</span>' +
'</div>';
} else { } else {
html += '<div class="text-sm text-error">' + escapeHtml(errorMsg || t('common.unknownError')) + '</div>'; html += '<div class="text-sm text-error">' + escapeHtml(errorMsg || t('common.unknownError')) + '</div>';
} }
@@ -2141,19 +2612,8 @@ function switchCodexLensModelTab(tabName) {
* Update model mode (Local vs API) * Update model mode (Local vs API)
*/ */
function updateModelMode(mode) { function updateModelMode(mode) {
var gpuContainer = document.getElementById('gpuSelectContainer');
var modeSelect = document.getElementById('modelModeSelect'); var modeSelect = document.getElementById('modelModeSelect');
// Show/hide GPU selector based on mode
if (gpuContainer) {
if (mode === 'local') {
gpuContainer.classList.remove('hidden');
loadGpuDevicesForModeSelector();
} else {
gpuContainer.classList.add('hidden');
}
}
// Store mode preference (will be saved when locked) // Store mode preference (will be saved when locked)
if (modeSelect) { if (modeSelect) {
modeSelect.setAttribute('data-current-mode', mode); modeSelect.setAttribute('data-current-mode', mode);
@@ -2165,6 +2625,7 @@ function updateModelMode(mode) {
*/ */
async function loadGpuDevicesForModeSelector() { async function loadGpuDevicesForModeSelector() {
var gpuSelect = document.getElementById('gpuDeviceSelect'); var gpuSelect = document.getElementById('gpuDeviceSelect');
var gpuSection = document.getElementById('gpuConfigSection');
if (!gpuSelect) return; if (!gpuSelect) return;
try { try {
@@ -2172,19 +2633,28 @@ async function loadGpuDevicesForModeSelector() {
if (!response.ok) { if (!response.ok) {
console.warn('[CodexLens] GPU list endpoint returned:', response.status); console.warn('[CodexLens] GPU list endpoint returned:', response.status);
gpuSelect.innerHTML = '<option value="auto">Auto</option>'; gpuSelect.innerHTML = '<option value="auto">Auto</option>';
// Hide section if no GPU devices available
if (gpuSection) gpuSection.classList.add('hidden');
return; return;
} }
var result = await response.json(); var result = await response.json();
var html = '<option value="auto">Auto</option>'; var html = '<option value="auto">Auto</option>';
if (result.devices && result.devices.length > 0) { if (result.devices && result.devices.length > 1) {
// Only show section if multiple GPUs available
result.devices.forEach(function(device, index) { result.devices.forEach(function(device, index) {
html += '<option value="' + index + '">' + escapeHtml(device.name) + '</option>'; html += '<option value="' + index + '">' + escapeHtml(device.name) + '</option>';
}); });
gpuSelect.innerHTML = html;
if (gpuSection) gpuSection.classList.remove('hidden');
} else {
// Single or no GPU - hide section
gpuSelect.innerHTML = html;
if (gpuSection) gpuSection.classList.add('hidden');
} }
gpuSelect.innerHTML = html;
} catch (err) { } catch (err) {
console.error('Failed to load GPU devices:', err); console.error('Failed to load GPU devices:', err);
if (gpuSection) gpuSection.classList.add('hidden');
} }
} }
@@ -2259,7 +2729,6 @@ async function toggleModelModeLock() {
*/ */
async function initModelModeFromConfig() { async function initModelModeFromConfig() {
var modeSelect = document.getElementById('modelModeSelect'); var modeSelect = document.getElementById('modelModeSelect');
var gpuContainer = document.getElementById('gpuSelectContainer');
if (!modeSelect) return; if (!modeSelect) return;
@@ -2272,16 +2741,6 @@ async function initModelModeFromConfig() {
modeSelect.value = mode; modeSelect.value = mode;
modeSelect.setAttribute('data-current-mode', mode); modeSelect.setAttribute('data-current-mode', mode);
// Show GPU selector for local mode
if (gpuContainer) {
if (mode === 'local') {
gpuContainer.classList.remove('hidden');
loadGpuDevicesForModeSelector();
} else {
gpuContainer.classList.add('hidden');
}
}
} catch (err) { } catch (err) {
console.error('Failed to load model mode config:', err); console.error('Failed to load model mode config:', err);
} }
@@ -3112,6 +3571,9 @@ async function renderCodexLensManager() {
// Wait for LiteLLM config before loading semantic deps (it may need provider info) // Wait for LiteLLM config before loading semantic deps (it may need provider info)
await litellmPromise; await litellmPromise;
// Load FastEmbed installation status (show/hide install card)
loadFastEmbedInstallStatus();
// Always load semantic deps status - it needs GPU detection and device list // Always load semantic deps status - it needs GPU detection and device list
// which are not included in the aggregated endpoint // which are not included in the aggregated endpoint
loadSemanticDepsStatus(); loadSemanticDepsStatus();
@@ -3121,7 +3583,6 @@ async function renderCodexLensManager() {
// Initialize model mode and semantic status badge // Initialize model mode and semantic status badge
updateSemanticStatusBadge(); updateSemanticStatusBadge();
loadGpuDevicesForModeSelector();
// Initialize file watcher status // Initialize file watcher status
initWatcherStatus(); initWatcherStatus();
@@ -3238,6 +3699,10 @@ function buildCodexLensManagerPage(config) {
'</div>' + '</div>' +
// Right Column // Right Column
'<div class="space-y-6">' + '<div class="space-y-6">' +
// FastEmbed Installation Card (shown when not installed)
'<div id="fastembedInstallCard" class="hidden">' +
// Content will be populated by loadFastEmbedInstallStatus()
'</div>' +
// Combined: Semantic Status + Model Management with Tabs // Combined: Semantic Status + Model Management with Tabs
'<div class="bg-card border border-border rounded-lg overflow-hidden">' + '<div class="bg-card border border-border rounded-lg overflow-hidden">' +
// Compact Header with Semantic Status // Compact Header with Semantic Status
@@ -3249,16 +3714,6 @@ function buildCodexLensManagerPage(config) {
'</div>' + '</div>' +
'</div>' + '</div>' +
'</div>' + '</div>' +
// GPU Config Section (for local mode)
'<div id="gpuConfigSection" class="px-4 py-3 border-b border-border bg-muted/10">' +
'<div class="flex items-center gap-3">' +
'<span class="text-xs text-muted-foreground">GPU:</span>' +
'<select id="gpuDeviceSelect" class="text-sm border rounded px-2 py-1 bg-background flex-1">' +
'<option value="auto">Auto</option>' +
'</select>' +
'</div>' +
'<p class="text-xs text-muted-foreground mt-1.5">Backend configured in Environment Variables below</p>' +
'</div>' +
// Tabs for Embedding / Reranker // Tabs for Embedding / Reranker
'<div class="border-b border-border">' + '<div class="border-b border-border">' +
'<div class="flex">' + '<div class="flex">' +

3
ccw/src/tools/codex-lens.ts
View File

@@ -389,7 +389,8 @@ async function checkPythonEnvForDirectML(): Promise<PythonEnvInfo> {
try { try {
// Get Python version and architecture in one call // Get Python version and architecture in one call
const checkScript = `import sys, struct; print(f"{sys.version_info.major}.{sys.version_info.minor}.{sys.version_info.micro}|{struct.calcsize('P') * 8}")`; // Use % formatting instead of f-string to avoid Windows shell escaping issues with curly braces
const checkScript = `import sys, struct; print('%d.%d.%d|%d' % (sys.version_info.major, sys.version_info.minor, sys.version_info.micro, struct.calcsize('P') * 8))`;
const result = execSync(`"${pythonPath}" -c "${checkScript}"`, { encoding: 'utf-8', timeout: 10000 }).trim(); const result = execSync(`"${pythonPath}" -c "${checkScript}"`, { encoding: 'utf-8', timeout: 10000 }).trim();
const [version, archStr] = result.split('|'); const [version, archStr] = result.split('|');
const architecture = parseInt(archStr, 10); const architecture = parseInt(archStr, 10);