feat(gpt5): first draft (#242)

fix #241

Author: devthejo

.env.example

@@ -33,3 +33,9 @@ PROXY_REVIEWER_TOKEN=ghp_your_token_here
 # Server Configuration (Probot v14)
 HOST=0.0.0.0
 PORT=3000
+
+# OpenAI Configuration
+# Required when llmProvider is set to "openai"
+OPENAI_API_KEY=xxx
+# Optional: override model (defaults to "gpt-5")
+OPENAI_MODEL=gpt-5

CHANGELOG.md

@@ -68,6 +68,39 @@ WEBHOOK_PROXY_URL=https://smee.io/your-url
 
 See [.env.example](.env.example) for an example.
 
+## Choosing a provider (Anthropic or OpenAI)
+
+Revu supports both Anthropic and OpenAI. Select the provider in `config.json`:
+
+```json
+{
+  "promptStrategy": "line-comments",
+  "thinkingEnabled": true,
+  "llmProvider": "openai"
+}
+```
+
+- llmProvider: "anthropic" (default) or "openai"
+
+Environment variables per provider:
+
+- Anthropic (default):
+  - Required: ANTHROPIC_API_KEY
+  - Optional: ANTHROPIC_MODEL (default: claude-sonnet-4-5-20250929)
+  - Optional: ANTHROPIC_EXTENDED_CONTEXT=true to enable 1M context (beta API)
+
+- OpenAI (official endpoint):
+  - Required: OPENAI_API_KEY
+  - Optional: OPENAI_MODEL (default: gpt-5)
+
+Example OpenAI env:
+
+```env
+OPENAI_API_KEY=your_openai_key
+# Optional model override
+OPENAI_MODEL=gpt-5
+```
+
 ## Running Revu
 
 ### Local Development
@@ -106,8 +139,8 @@ thinkingEnabled: true
 
 # Custom coding guidelines
 codingGuidelines:
-  - "Use descriptive variable names"
-  - "Add comments for complex logic"
+  - 'Use descriptive variable names'
+  - 'Add comments for complex logic'
 
 # PR validation rules
 validation:
@@ -117,9 +150,9 @@ validation:
 # Branch filtering
 branches:
   patterns:
-    - "!**"
-    - "main"
-    - "release/*"
+    - '!**'
+    - 'main'
+    - 'release/*'
 ```
 
 ## Contributing

README.md

@@ -1,5 +1,5 @@
 import { beforeEach, describe, expect, it, vi } from 'vitest'
-import { lineCommentsSender } from '../../src/anthropic-senders/line-comments-sender.ts'
+import { anthropicLineCommentsSender } from '../../src/senders/providers/anthropic/line-comments-sender.ts'
 
 // Mock the Anthropic SDK
 vi.mock('@anthropic-ai/sdk', () => {
@@ -113,7 +113,7 @@ const mockMixedResponse = (
   target.mockResolvedValue({ content })
 }
 
-describe('lineCommentsSender', () => {
+describe('anthropicLineCommentsSender', () => {
   let mockAnthropic: {
     messages: {
       create: ReturnType<typeof vi.fn>
@@ -156,7 +156,7 @@ describe('lineCommentsSender', () => {
 
     mockToolUseResponse(mockAnthropic, expectedResponse)
 
-    const result = await lineCommentsSender('test prompt')
+    const result = await anthropicLineCommentsSender('test prompt')
     expect(result).toBe(JSON.stringify(expectedResponse))
   })
 
@@ -165,7 +165,7 @@ describe('lineCommentsSender', () => {
 
     mockToolUseResponse(mockAnthropic, expectedResponse)
 
-    const result = await lineCommentsSender('test prompt')
+    const result = await anthropicLineCommentsSender('test prompt')
     expect(result).toBe(JSON.stringify(expectedResponse))
   })
 
@@ -184,7 +184,7 @@ This is the code review result.`
 
     mockTextResponse(mockAnthropic, textResponse)
 
-    const result = await lineCommentsSender('test prompt')
+    const result = await anthropicLineCommentsSender('test prompt')
     expect(result).toBe(JSON.stringify(expectedJson, null, 2))
   })
 
@@ -196,7 +196,7 @@ This is the code review result.`
     const jsonResponse = JSON.stringify(expectedJson)
     mockTextResponse(mockAnthropic, jsonResponse)
 
-    const result = await lineCommentsSender('test prompt')
+    const result = await anthropicLineCommentsSender('test prompt')
     expect(result).toBe(jsonResponse)
   })
 
@@ -215,7 +215,7 @@ Some text after that should be ignored`
 
     mockTextResponse(mockAnthropic, textWithJson)
 
-    const result = await lineCommentsSender('test prompt')
+    const result = await anthropicLineCommentsSender('test prompt')
     expect(result).toBe(JSON.stringify(expectedJson))
   })
 
@@ -224,7 +224,7 @@ Some text after that should be ignored`
 
     mockTextResponse(mockAnthropic, plainTextResponse)
 
-    const result = await lineCommentsSender('test prompt')
+    const result = await anthropicLineCommentsSender('test prompt')
     expect(result).toBe(plainTextResponse)
   })
 
@@ -237,15 +237,15 @@ Some text after that should be ignored`
       }
     ])
 
-    await expect(lineCommentsSender('test prompt')).rejects.toThrow(
+    await expect(anthropicLineCommentsSender('test prompt')).rejects.toThrow(
       'Unexpected tool name: unexpected_tool'
     )
   })
 
   it('should throw error when no content is found', async () => {
     mockMixedResponse(mockAnthropic, [])
 
-    await expect(lineCommentsSender('test prompt')).rejects.toThrow(
+    await expect(anthropicLineCommentsSender('test prompt')).rejects.toThrow(
       'Unexpected response format from Anthropic inline comment - no content found'
     )
   })
@@ -270,7 +270,7 @@ This should fallback to the entire text since JSON is invalid.`
 
     mockTextResponse(mockAnthropic, invalidJsonInCodeBlock)
 
-    const result = await lineCommentsSender('test prompt')
+    const result = await anthropicLineCommentsSender('test prompt')
 
     // Should return the invalid JSON string (not the whole text)
     // The validation happens later in lineCommentsHandler
@@ -283,7 +283,7 @@ This should fallback to the entire text since JSON is invalid.`
 
     mockTextResponse(mockAnthropic, invalidJsonText)
 
-    const result = await lineCommentsSender('test prompt')
+    const result = await anthropicLineCommentsSender('test prompt')
 
     // Should return the invalid JSON string as-is
     // The validation/parsing happens in lineCommentsHandler
@@ -295,7 +295,7 @@ This should fallback to the entire text since JSON is invalid.`
 
     mockTextResponse(mockAnthropic, malformedJson)
 
-    const result = await lineCommentsSender('test prompt')
+    const result = await anthropicLineCommentsSender('test prompt')
 
     // Should return the malformed JSON as-is
     expect(result).toBe(malformedJson)
@@ -318,7 +318,7 @@ More text after.`
 
     mockTextResponse(mockAnthropic, textWithValidJsonInCodeBlock)
 
-    const result = await lineCommentsSender('test prompt')
+    const result = await anthropicLineCommentsSender('test prompt')
 
     // Should return the valid JSON from the code block, not the malformed JSON-like text
     expect(result).toBe(JSON.stringify(validJson, null, 2))
@@ -336,7 +336,7 @@ More text after.`
         createToolUseContent(expectedResponse)
       ])
 
-      const result = await lineCommentsSender('test prompt', true)
+      const result = await anthropicLineCommentsSender('test prompt', true)
 
       // Should extract only the tool_use result, ignoring thinking blocks
       expect(result).toBe(JSON.stringify(expectedResponse))
@@ -360,7 +360,7 @@ More text after.`
 
       mockToolUseResponse(mockAnthropic, expectedResponse)
 
-      const result = await lineCommentsSender('test prompt', false)
+      const result = await anthropicLineCommentsSender('test prompt', false)
 
       expect(result).toBe(JSON.stringify(expectedResponse))
 
@@ -381,7 +381,7 @@ More text after.`
 
       mockToolUseResponse(mockAnthropic, expectedResponse)
 
-      const result = await lineCommentsSender('test prompt')
+      const result = await anthropicLineCommentsSender('test prompt')
 
       expect(result).toBe(JSON.stringify(expectedResponse))
 
@@ -401,7 +401,7 @@ More text after.`
 
       mockToolUseResponse(mockAnthropic, expectedResponse)
 
-      const result = await lineCommentsSender('test prompt')
+      const result = await anthropicLineCommentsSender('test prompt')
 
       expect(result).toBe(JSON.stringify(expectedResponse))
 
@@ -423,7 +423,7 @@ More text after.`
 
       mockToolUseResponse(mockAnthropic, expectedResponse)
 
-      const result = await lineCommentsSender('test prompt')
+      const result = await anthropicLineCommentsSender('test prompt')
 
       expect(result).toBe(JSON.stringify(expectedResponse))
 
@@ -448,7 +448,7 @@ More text after.`
 
       mockToolUseResponse(mockAnthropic, expectedResponse, false)
 
-      const result = await lineCommentsSender('test prompt')
+      const result = await anthropicLineCommentsSender('test prompt')
 
       expect(result).toBe(JSON.stringify(expectedResponse))
 
@@ -474,7 +474,7 @@ More text after.`
         createToolUseContent(expectedResponse)
       ])
 
-      const result = await lineCommentsSender('test prompt', true)
+      const result = await anthropicLineCommentsSender('test prompt', true)
 
       expect(result).toBe(JSON.stringify(expectedResponse))
 
@@ -507,7 +507,7 @@ More text after.`
         false
       )
 
-      const result = await lineCommentsSender('test prompt', true)
+      const result = await anthropicLineCommentsSender('test prompt', true)
 
       expect(result).toBe(JSON.stringify(expectedResponse))