@layer-ai/core 2.0.19 → 2.0.21

package/dist/lib/spending-tracker.js

@@ -0,0 +1,89 @@
+import { db } from './db/postgres.js';
+import { cache } from './db/redis.js';
+export const spendingTracker = {
+    async trackSpending(userId, cost) {
+        try {
+            const newSpending = await cache.incrementUserSpending(userId, cost);
+            const spendingInfo = await db.getUserSpending(userId);
+            if (!spendingInfo) {
+                return { userId, cost, newSpending };
+            }
+            const { limit, status } = spendingInfo;
+            const exceeded = limit !== null && newSpending > limit;
+            if (exceeded && status === 'active') {
+                await db.setUserStatus(userId, 'over_limit');
+                console.log(`[Spending] User ${userId} exceeded limit: ${newSpending} > ${limit}`);
+            }
+            await this.checkAlertThresholds(userId, newSpending, limit);
+            return { userId, cost, exceeded, newSpending };
+        }
+        catch (error) {
+            console.error('[Spending] Redis error, falling back to DB:', error);
+            return await this.trackSpendingDB(userId, cost);
+        }
+    },
+    async trackSpendingDB(userId, cost) {
+        const result = await db.incrementUserSpending(userId, cost);
+        if (result.exceeded) {
+            await db.setUserStatus(userId, 'over_limit');
+            console.log(`[Spending] User ${userId} exceeded limit: ${result.newSpending} > ${result.limit}`);
+        }
+        await this.checkAlertThresholds(userId, result.newSpending, result.limit);
+        return {
+            userId,
+            cost,
+            exceeded: result.exceeded,
+            newSpending: result.newSpending,
+        };
+    },
+    async checkAlertThresholds(userId, currentSpending, limit) {
+        if (!limit || limit === 0)
+            return;
+        const percentage = (currentSpending / limit) * 100;
+        const thresholds = [50, 80, 95, 100];
+        for (const threshold of thresholds) {
+            if (percentage >= threshold) {
+                await this.sendAlertIfNeeded(userId, threshold, currentSpending, limit);
+                break;
+            }
+        }
+    },
+    async sendAlertIfNeeded(userId, threshold, currentSpending, limit) {
+        console.log(`[Spending] Alert: User ${userId} at ${threshold}% of limit ($${currentSpending}/$${limit})`);
+        await db.recordSpendingAlert(userId);
+    },
+    async syncSpendingToDB(userId) {
+        try {
+            const cachedSpending = await cache.getUserSpending(userId);
+            if (cachedSpending !== null) {
+                await db.updateUserSpending(userId, cachedSpending);
+                console.log(`[Spending] Synced user ${userId}: $${cachedSpending}`);
+            }
+        }
+        catch (error) {
+            console.error(`[Spending] Sync error for user ${userId}:`, error);
+        }
+    },
+    async syncAllSpending() {
+        try {
+            const userIds = await cache.getAllCachedSpendingUsers();
+            console.log(`[Spending] Syncing ${userIds.length} users to DB`);
+            await Promise.all(userIds.map(userId => this.syncSpendingToDB(userId)));
+            console.log('[Spending] Sync complete');
+        }
+        catch (error) {
+            console.error('[Spending] Bulk sync error:', error);
+        }
+    },
+    async warmCache(userId) {
+        try {
+            const spendingInfo = await db.getUserSpending(userId);
+            if (spendingInfo) {
+                await cache.setUserSpending(userId, spendingInfo.currentSpending);
+            }
+        }
+        catch (error) {
+            console.error(`[Spending] Cache warm error for user ${userId}:`, error);
+        }
+    },
+};

package/dist/middleware/auth.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"auth.d.ts","sourceRoot":"","sources":["../../src/middleware/auth.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,OAAO,EAAE,QAAQ,EAAE,YAAY,EAAE,MAAM,SAAS,CAAC;AAK1D,OAAO,CAAC,MAAM,CAAC;IACb,UAAU,OAAO,CAAC;QAChB,UAAU,OAAO;YACf,MAAM,CAAC,EAAE,MAAM,CAAC;YAChB,QAAQ,CAAC,EAAE,MAAM,CAAC;YAClB,UAAU,CAAC,EAAE,MAAM,CAAC;SACrB;KACF;CACF;AAED;;;;;GAKG;AACH,wBAAsB,YAAY,CAChC,GAAG,EAAE,OAAO,EACZ,GAAG,EAAE,QAAQ,EACb,IAAI,EAAE,YAAY,GACjB,OAAO,CAAC,IAAI,CAAC,CAyHf;AAED;;;GAGG;AACH,wBAAgB,YAAY,CAC1B,GAAG,EAAE,OAAO,EACZ,GAAG,EAAE,QAAQ,EACb,IAAI,EAAE,YAAY,GACjB,IAAI,CAWN"}
+{"version":3,"file":"auth.d.ts","sourceRoot":"","sources":["../../src/middleware/auth.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,OAAO,EAAE,QAAQ,EAAE,YAAY,EAAE,MAAM,SAAS,CAAC;AAK1D,OAAO,CAAC,MAAM,CAAC;IACb,UAAU,OAAO,CAAC;QAChB,UAAU,OAAO;YACf,MAAM,CAAC,EAAE,MAAM,CAAC;YAChB,QAAQ,CAAC,EAAE,MAAM,CAAC;YAClB,UAAU,CAAC,EAAE,MAAM,CAAC;SACrB;KACF;CACF;AAED;;;;;GAKG;AACH,wBAAsB,YAAY,CAChC,GAAG,EAAE,OAAO,EACZ,GAAG,EAAE,QAAQ,EACb,IAAI,EAAE,YAAY,GACjB,OAAO,CAAC,IAAI,CAAC,CAiJf;AAED;;;GAGG;AACH,wBAAgB,YAAY,CAC1B,GAAG,EAAE,OAAO,EACZ,GAAG,EAAE,QAAQ,EACb,IAAI,EAAE,YAAY,GACjB,IAAI,CAWN"}

package/dist/middleware/auth.js

@@ -64,6 +64,17 @@ export async function authenticate(req, res, next) {
                 });
                 return;
             }
+            // Check spending limits if user has exceeded and enforcement is set to block
+            if (userStatus === 'over_limit') {
+                const spendingInfo = await db.getUserSpending(apiKeyRecord.userId);
+                if (spendingInfo?.limitEnforcementType === 'block') {
+                    res.status(403).json({
+                        error: 'spending_limit_exceeded',
+                        message: 'You have exceeded your spending limit. Requests are blocked until your next billing period or until you increase your limit.',
+                    });
+                    return;
+                }
+            }
             // Attach userId to request for downstream handlers
             req.userId = apiKeyRecord.userId;
             req.apiKeyId = apiKeyRecord.id;
@@ -94,6 +105,17 @@ export async function authenticate(req, res, next) {
                 });
                 return;
             }
+            // Check spending limits if user has exceeded and enforcement is set to block
+            if (userStatus === 'over_limit') {
+                const spendingInfo = await db.getUserSpending(sessionKey.userId);
+                if (spendingInfo?.limitEnforcementType === 'block') {
+                    res.status(403).json({
+                        error: 'spending_limit_exceeded',
+                        message: 'You have exceeded your spending limit. Requests are blocked until your next billing period or until you increase your limit.',
+                    });
+                    return;
+                }
+            }
             req.userId = sessionKey.userId;
             next();
             return;

package/dist/routes/tests/test-openai-endpoint.d.ts

@@ -0,0 +1,3 @@
+#!/usr/bin/env npx tsx
+export {};
+//# sourceMappingURL=test-openai-endpoint.d.ts.map

package/dist/routes/tests/test-openai-endpoint.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"test-openai-endpoint.d.ts","sourceRoot":"","sources":["../../../src/routes/tests/test-openai-endpoint.ts"],"names":[],"mappings":""}

package/dist/routes/tests/test-openai-endpoint.js

@@ -0,0 +1,292 @@
+#!/usr/bin/env npx tsx
+console.log('='.repeat(80));
+console.log('OPENAI-COMPATIBLE ENDPOINT TESTS');
+console.log('='.repeat(80));
+console.log('');
+const BASE_URL = process.env.API_URL || 'http://localhost:3004';
+const API_KEY = process.env.LAYER_API_KEY;
+const GATE_ID = process.env.TEST_GATE_ID;
+if (!API_KEY) {
+    console.error('❌ Error: LAYER_API_KEY environment variable not set');
+    process.exit(1);
+}
+if (!GATE_ID) {
+    console.error('❌ Error: TEST_GATE_ID environment variable not set');
+    process.exit(1);
+}
+async function testNonStreamingBasic() {
+    console.log('Test 1: Non-streaming basic chat completion');
+    console.log('-'.repeat(80));
+    const request = {
+        model: 'gpt-4o',
+        messages: [
+            { role: 'user', content: 'Say "test passed" and nothing else.' }
+        ],
+        max_tokens: 10,
+        gateId: GATE_ID,
+    };
+    const response = await fetch(`${BASE_URL}/v1/chat/completions`, {
+        method: 'POST',
+        headers: {
+            'Content-Type': 'application/json',
+            'Authorization': `Bearer ${API_KEY}`,
+        },
+        body: JSON.stringify(request),
+    });
+    if (!response.ok) {
+        const error = await response.json();
+        throw new Error(`Request failed: ${JSON.stringify(error)}`);
+    }
+    const data = await response.json();
+    console.log('  Response ID:', data.id);
+    console.log('  Model:', data.model);
+    console.log('  Content:', data.choices[0].message.content);
+    console.log('  Finish reason:', data.choices[0].finish_reason);
+    console.log('  Usage:', data.usage);
+    console.log('  ✅ Non-streaming basic test passed\n');
+}
+async function testNonStreamingWithGateIdInHeader() {
+    console.log('Test 2: Non-streaming with gateId in header');
+    console.log('-'.repeat(80));
+    const request = {
+        model: 'gpt-4o',
+        messages: [
+            { role: 'user', content: 'Say "header test passed" and nothing else.' }
+        ],
+        max_tokens: 10,
+    };
+    const response = await fetch(`${BASE_URL}/v1/chat/completions`, {
+        method: 'POST',
+        headers: {
+            'Content-Type': 'application/json',
+            'Authorization': `Bearer ${API_KEY}`,
+            'X-Layer-Gate-Id': GATE_ID,
+        },
+        body: JSON.stringify(request),
+    });
+    if (!response.ok) {
+        const error = await response.json();
+        throw new Error(`Request failed: ${JSON.stringify(error)}`);
+    }
+    const data = await response.json();
+    console.log('  Content:', data.choices[0].message.content);
+    console.log('  ✅ Header gateId test passed\n');
+}
+async function testStreamingBasic() {
+    console.log('Test 3: Streaming basic chat completion');
+    console.log('-'.repeat(80));
+    const request = {
+        model: 'gpt-4o',
+        messages: [
+            { role: 'user', content: 'Count from 1 to 3, one number per line.' }
+        ],
+        max_tokens: 50,
+        stream: true,
+        gateId: GATE_ID,
+    };
+    const response = await fetch(`${BASE_URL}/v1/chat/completions`, {
+        method: 'POST',
+        headers: {
+            'Content-Type': 'application/json',
+            'Authorization': `Bearer ${API_KEY}`,
+        },
+        body: JSON.stringify(request),
+    });
+    if (!response.ok) {
+        const error = await response.json();
+        throw new Error(`Request failed: ${JSON.stringify(error)}`);
+    }
+    let chunkCount = 0;
+    let fullContent = '';
+    let finalUsage = null;
+    const reader = response.body?.getReader();
+    const decoder = new TextDecoder();
+    if (!reader) {
+        throw new Error('No response body reader');
+    }
+    while (true) {
+        const { done, value } = await reader.read();
+        if (done)
+            break;
+        const text = decoder.decode(value);
+        const lines = text.split('\n').filter(line => line.trim().startsWith('data: '));
+        for (const line of lines) {
+            const data = line.replace('data: ', '').trim();
+            if (data === '[DONE]') {
+                continue;
+            }
+            try {
+                const chunk = JSON.parse(data);
+                chunkCount++;
+                if (chunk.choices[0].delta.content) {
+                    fullContent += chunk.choices[0].delta.content;
+                }
+                if (chunk.usage) {
+                    finalUsage = chunk.usage;
+                }
+            }
+            catch (e) {
+                // Skip invalid JSON
+            }
+        }
+    }
+    console.log('  Chunks received:', chunkCount);
+    console.log('  Full content:', fullContent.trim());
+    console.log('  Final usage:', finalUsage);
+    console.log('  ✅ Streaming basic test passed\n');
+}
+async function testWithToolCalls() {
+    console.log('Test 4: Non-streaming with tool calls');
+    console.log('-'.repeat(80));
+    const request = {
+        model: 'gpt-4o',
+        messages: [
+            { role: 'user', content: 'What is the weather in Paris?' }
+        ],
+        tools: [
+            {
+                type: 'function',
+                function: {
+                    name: 'get_weather',
+                    description: 'Get the current weather for a location',
+                    parameters: {
+                        type: 'object',
+                        properties: {
+                            location: {
+                                type: 'string',
+                                description: 'The city and state, e.g. Paris, France',
+                            },
+                        },
+                        required: ['location'],
+                    },
+                },
+            },
+        ],
+        tool_choice: 'auto',
+        gateId: GATE_ID,
+    };
+    const response = await fetch(`${BASE_URL}/v1/chat/completions`, {
+        method: 'POST',
+        headers: {
+            'Content-Type': 'application/json',
+            'Authorization': `Bearer ${API_KEY}`,
+        },
+        body: JSON.stringify(request),
+    });
+    if (!response.ok) {
+        const error = await response.json();
+        throw new Error(`Request failed: ${JSON.stringify(error)}`);
+    }
+    const data = await response.json();
+    console.log('  Finish reason:', data.choices[0].finish_reason);
+    if (data.choices[0].message.tool_calls && data.choices[0].message.tool_calls.length > 0) {
+        console.log('  Tool calls:', JSON.stringify(data.choices[0].message.tool_calls, null, 2));
+        console.log('  ✅ Tool calls test passed\n');
+    }
+    else {
+        console.log('  ⚠️  No tool calls received (model may have chosen not to use tools)\n');
+    }
+}
+async function testClaudeModel() {
+    console.log('Test 5: OpenAI format with Claude model');
+    console.log('-'.repeat(80));
+    const request = {
+        model: 'claude-3-7-sonnet-20250219',
+        messages: [
+            { role: 'user', content: 'Say "claude via openai format works" and nothing else.' }
+        ],
+        max_tokens: 20,
+        gateId: GATE_ID,
+    };
+    const response = await fetch(`${BASE_URL}/v1/chat/completions`, {
+        method: 'POST',
+        headers: {
+            'Content-Type': 'application/json',
+            'Authorization': `Bearer ${API_KEY}`,
+        },
+        body: JSON.stringify(request),
+    });
+    if (!response.ok) {
+        const error = await response.json();
+        throw new Error(`Request failed: ${JSON.stringify(error)}`);
+    }
+    const data = await response.json();
+    console.log('  Model:', data.model);
+    console.log('  Content:', data.choices[0].message.content);
+    console.log('  ✅ Claude model test passed\n');
+}
+async function testGeminiModel() {
+    console.log('Test 6: OpenAI format with Gemini model');
+    console.log('-'.repeat(80));
+    const request = {
+        model: 'gemini-2.0-flash',
+        messages: [
+            { role: 'user', content: 'Say "gemini via openai format works" and nothing else.' }
+        ],
+        max_tokens: 20,
+        gateId: GATE_ID,
+    };
+    const response = await fetch(`${BASE_URL}/v1/chat/completions`, {
+        method: 'POST',
+        headers: {
+            'Content-Type': 'application/json',
+            'Authorization': `Bearer ${API_KEY}`,
+        },
+        body: JSON.stringify(request),
+    });
+    if (!response.ok) {
+        const error = await response.json();
+        throw new Error(`Request failed: ${JSON.stringify(error)}`);
+    }
+    const data = await response.json();
+    console.log('  Model:', data.model);
+    console.log('  Content:', data.choices[0].message.content);
+    console.log('  ✅ Gemini model test passed\n');
+}
+async function testMistralModel() {
+    console.log('Test 7: OpenAI format with Mistral model');
+    console.log('-'.repeat(80));
+    const request = {
+        model: 'mistral-small-2501',
+        messages: [
+            { role: 'user', content: 'Say "mistral via openai format works" and nothing else.' }
+        ],
+        max_tokens: 20,
+        gateId: GATE_ID,
+    };
+    const response = await fetch(`${BASE_URL}/v1/chat/completions`, {
+        method: 'POST',
+        headers: {
+            'Content-Type': 'application/json',
+            'Authorization': `Bearer ${API_KEY}`,
+        },
+        body: JSON.stringify(request),
+    });
+    if (!response.ok) {
+        const error = await response.json();
+        throw new Error(`Request failed: ${JSON.stringify(error)}`);
+    }
+    const data = await response.json();
+    console.log('  Model:', data.model);
+    console.log('  Content:', data.choices[0].message.content);
+    console.log('  ✅ Mistral model test passed\n');
+}
+(async () => {
+    try {
+        await testNonStreamingBasic();
+        await testNonStreamingWithGateIdInHeader();
+        await testStreamingBasic();
+        await testWithToolCalls();
+        await testClaudeModel();
+        await testGeminiModel();
+        await testMistralModel();
+        console.log('='.repeat(80));
+        console.log('✅ ALL OPENAI-COMPATIBLE ENDPOINT TESTS PASSED');
+        console.log('='.repeat(80));
+    }
+    catch (error) {
+        console.error('❌ Test failed:', error);
+        process.exit(1);
+    }
+})();
+export {};

package/dist/routes/v1/chat-completions.d.ts

@@ -0,0 +1,4 @@
+import type { Router as RouterType } from 'express';
+declare const router: RouterType;
+export default router;
+//# sourceMappingURL=chat-completions.d.ts.map

package/dist/routes/v1/chat-completions.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"chat-completions.d.ts","sourceRoot":"","sources":["../../../src/routes/v1/chat-completions.ts"],"names":[],"mappings":"AACA,OAAO,KAAK,EAAE,MAAM,IAAI,UAAU,EAAE,MAAM,SAAS,CAAC;AAcpD,QAAA,MAAM,MAAM,EAAE,UAAqB,CAAC;AAqSpC,eAAe,MAAM,CAAC"}