From 89dc1efda75ada7633aa0c440c8bfe827461d6b8 Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 00:08:10 +0900
Subject: [PATCH 01/25] update template asynch

---
 .gitea/scripts/code_review.py    | 77 ++++++++++++++++++++------------
 .gitea/scripts/model.py          | 16 +++----
 .gitea/workflows/code-review.yml |  2 +-
 3 files changed, 57 insertions(+), 38 deletions(-)

diff --git a/.gitea/scripts/code_review.py b/.gitea/scripts/code_review.py
index e8d71fe..e87e86a 100644
--- a/.gitea/scripts/code_review.py
+++ b/.gitea/scripts/code_review.py
@@ -1,5 +1,6 @@
 """Code Reviewer for Gitea."""
 
+import asyncio
 import fnmatch
 import json
 import os
@@ -7,6 +8,7 @@ import re
 from typing import Any
 
 import requests
+import aiohttp
 from model import Model
 
 ACCESS_TOKEN = os.getenv("ACCESS_TOKEN", "")
@@ -133,7 +135,7 @@ def create_comment(
     return comments
 
 
-def analyze_single_chunks(
+async def analyze_single_chunks(
     single_chunk_model: Model, parsed_diff: list[dict[str, Any]]
 ) -> list[dict[str, Any]]:
     """Analyze single chunks and create comments.
@@ -145,29 +147,33 @@ def analyze_single_chunks(
     Returns:
         list[dict[str, Any]]: comments for single chunk review
     """
-    comments = []
-    title = EVENT_DATA["pull_request"]["title"]
-    description = EVENT_DATA["pull_request"]["body"]
-    for diff in parsed_diff:
+
+    async def process_single_chunk(diff: dict[str, Any]):
         file = diff["file"]
         chunk = diff["chunk"]
-        response = single_chunk_model.get_response_single_chunk(
+        response = await single_chunk_model.get_response_single_chunk(
             file, title, description, chunk
         )
         response = response.strip("`").lstrip("json").strip() or "[]"
 
         try:
             response_json = json.loads(response)
-            new_comments = create_comment(file, response_json)
-            comments.extend(new_comments)
+            return create_comment(file, response_json)
         except json.JSONDecodeError:
             print(f"Failed to parse response: {response}")
-            continue
+            return []
 
+    title = EVENT_DATA["pull_request"]["title"]
+    description = EVENT_DATA["pull_request"]["body"]
+    tasks = [process_single_chunk(diff) for diff in parsed_diff]
+    results = await asyncio.gather(*tasks)
+
+    # Flatten the list of comments
+    comments = [comment for result in results for comment in result]
     return comments
 
 
-def get_file_content(file: str) -> str | None:
+async def get_file_content(file: str) -> str | None:
     """Get file content from Gitea.
 
     Args:
@@ -183,15 +189,16 @@ def get_file_content(file: str) -> str | None:
     url = f"{repo_url}/raw/{branch}%2F{replaced_file}?ref={branch}"
 
     try:
-        response = requests.get(url, headers=HEADERS)
-        response.raise_for_status()
-        return response.text
-    except requests.RequestException as e:
+        async with aiohttp.ClientSession(headers=HEADERS) as session:
+            async with session.get(url) as response:
+                response.raise_for_status()
+                return await response.text()
+    except Exception as e:
         print(f"Failed to get file content: {e}")
         return None
 
 
-def analyze_full_context(
+async def analyze_full_context(
     full_context_model: Model, parsed_diff: list[dict[str, Any]]
 ) -> str:
     """Analyze full context and create review.
@@ -203,22 +210,26 @@ def analyze_full_context(
     Returns:
         str: review for full context
     """
-    file_contents = []
-    for diff in parsed_diff:
+
+    async def get_file_data(diff: dict[str, Any]):
         file = diff["file"]
         chunk = diff["chunk"]
-        content = get_file_content(file)
+        content = await get_file_content(file)
         if content is None:
-            continue
-        file_contents.append(f"File: {file}")
-        file_contents.append(content)
-        file_contents.append(f"Diff: {chunk}")
+            return None
+        return f"File: {file}\n{content}\nDiff: {chunk}"
+
+    tasks = [get_file_data(diff) for diff in parsed_diff]
+    file_contents_list = await asyncio.gather(*tasks)
+
+    file_contents = [item for item in file_contents_list if item is not None]
+
     if not file_contents:
         return ""
 
     title = EVENT_DATA["pull_request"]["title"]
     description = EVENT_DATA["pull_request"]["body"]
-    response = full_context_model.get_response_full_context(
+    response = await full_context_model.get_response_full_context(
         title, description, file_contents
     )
     response = response.strip("`").lstrip("markdown").strip()
@@ -248,10 +259,10 @@ def post_review(
     response.raise_for_status()
 
 
-def main() -> None:
-    """Code Reviewer for Gitea."""
+async def main() -> None:
+    """Code Reviewer for Gitea: Asynchronous version."""
     if EVENT_DATA["action"] not in ["opened", "synchronized"]:
-        print("Unsupproted event.")
+        print("Unsupported event.")
         return
 
     diff = get_diff()
@@ -273,10 +284,18 @@ def main() -> None:
     )
 
     parsed_diff = parse_diff(diff)
-    comments = analyze_single_chunks(single_chunk_model, parsed_diff)
-    full_context_response = analyze_full_context(full_context_model, parsed_diff)
+    comments_task = asyncio.create_task(
+        analyze_single_chunks(single_chunk_model, parsed_diff)
+    )
+    full_context_response_task = asyncio.create_task(
+        analyze_full_context(full_context_model, parsed_diff)
+    )
+
+    comments = await comments_task
+    full_context_response = await full_context_response_task
+
     post_review(full_context_response, comments)
 
 
 if __name__ == "__main__":
-    main()
+    asyncio.run(main())
diff --git a/.gitea/scripts/model.py b/.gitea/scripts/model.py
index 9004ab1..5773673 100644
--- a/.gitea/scripts/model.py
+++ b/.gitea/scripts/model.py
@@ -88,7 +88,7 @@ class Model:
             case ModelProvider.DEEPSEEK:
                 return OpenAI(api_key=api_key, base_url="https://api.deepseek.com")
 
-    def request(self, prompt: str) -> str:
+    async def request(self, prompt: str) -> str:
         """Request the model to generate a response.
 
         Args:
@@ -99,7 +99,7 @@ class Model:
         """
         match self.provider:
             case ModelProvider.OPENAI | ModelProvider.DEEPSEEK:
-                response = self.session.chat.completions.create(
+                response = await self.session.chat.completions.create(
                     model=self.model,
                     messages=[
                         {"role": "system", "content": self.system_prompt},
@@ -113,7 +113,7 @@ class Model:
                 )
                 return response.choices[0].message.content.strip()
             case ModelProvider.ANTHROPIC:
-                response = self.session.messages.create(
+                response = await self.session.messages.create(
                     model=self.model,
                     messages=[{"role": "user", "content": prompt}],
                     system=[
@@ -128,10 +128,10 @@ class Model:
                 )
                 return response.content[0].text.strip()
             case ModelProvider.GOOGLE:
-                response = self.session.generate_content(prompt)
+                response = await self.session.generate_content(prompt)
                 return response.text.strip()
 
-    def get_response_single_chunk(
+    async def get_response_single_chunk(
         self, file: str, title: str, description: str, chunk: str
     ) -> str:
         """Get the response for a single chunk.
@@ -146,9 +146,9 @@ class Model:
             str: The response.
         """
         prompt = SINGLE_CHUNK_USER_PROMPT.format(file, title, description, chunk)
-        return self.request(prompt)
+        return await self.request(prompt)
 
-    def get_response_full_context(
+    async def get_response_full_context(
         self, title: str, description: str, file_contents: list[str]
     ) -> str:
         """Get the response for full context.
@@ -165,7 +165,7 @@ class Model:
             prompt = FULL_CONTEXT_USER_PROMPT.format(
                 title, description, "\n".join(file_contents)
             )
-            return self.request(prompt)
+            return await self.request(prompt)
         except Exception as e:
             print(f"Error during full context response: {e}")
             print(prompt)
diff --git a/.gitea/workflows/code-review.yml b/.gitea/workflows/code-review.yml
index be7d67d..2c3928e 100644
--- a/.gitea/workflows/code-review.yml
+++ b/.gitea/workflows/code-review.yml
@@ -21,7 +21,7 @@ jobs:
       - name: Install dependencies
         run: |
           python -m pip install --upgrade pip
-          pip install requests py-gitea openai anthropic google-generativeai
+          pip install aiohttp requests py-gitea openai anthropic google-generativeai
 
       - name: Run Code Review
         env:
-- 
2.49.1


From 37b93207989ccfad96ff6a6164ca262f4f20e0b9 Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 00:14:49 +0900
Subject: [PATCH 02/25] update aynch client

---
 .gitea/scripts/model.py | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/.gitea/scripts/model.py b/.gitea/scripts/model.py
index 5773673..572dd72 100644
--- a/.gitea/scripts/model.py
+++ b/.gitea/scripts/model.py
@@ -4,8 +4,8 @@ from enum import Enum
 from typing import Any
 
 import google.generativeai as genai
-from anthropic import Anthropic
-from openai import OpenAI
+from anthropic import AsyncAnthropic
+from openai import AsyncOpenAI
 
 
 class ModelProvider(Enum):
@@ -79,14 +79,14 @@ class Model:
         """
         match self.provider:
             case ModelProvider.OPENAI:
-                return OpenAI(api_key=api_key)
+                return AsyncOpenAI(api_key=api_key)
             case ModelProvider.ANTHROPIC:
-                return Anthropic(api_key=api_key)
+                return AsyncAnthropic(api_key=api_key)
             case ModelProvider.GOOGLE:
                 genai.configure(api_key=api_key)
                 return genai.GenerativeModel(model=self.model, api_key=api_key)
             case ModelProvider.DEEPSEEK:
-                return OpenAI(api_key=api_key, base_url="https://api.deepseek.com")
+                return AsyncOpenAI(api_key=api_key, base_url="https://api.deepseek.com")
 
     async def request(self, prompt: str) -> str:
         """Request the model to generate a response.
@@ -128,7 +128,7 @@ class Model:
                 )
                 return response.content[0].text.strip()
             case ModelProvider.GOOGLE:
-                response = await self.session.generate_content(prompt)
+                response = await self.session.generate_content_async(prompt)
                 return response.text.strip()
 
     async def get_response_single_chunk(
-- 
2.49.1


From d82ff134391184ee6d61e74542b6c4e2e4729035 Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 00:15:15 +0900
Subject: [PATCH 03/25] change test model

---
 .gitea/workflows/code-review.yml | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/.gitea/workflows/code-review.yml b/.gitea/workflows/code-review.yml
index 2c3928e..4c5b042 100644
--- a/.gitea/workflows/code-review.yml
+++ b/.gitea/workflows/code-review.yml
@@ -28,8 +28,8 @@ jobs:
           ACCESS_TOKEN: ${{ secrets.ACCESS_TOKEN }}
           FULL_CONTEXT_MODEL: gpt-4o
           FULL_CONTEXT_API_KEY: ${{ secrets.OPENAI_API_KEY }}
-          SINGLE_CHUNK_MODEL: gpt-4o
-          SINGLE_CHUNK_API_KEY: ${{ secrets.OPENAI_API_KEY }}
+          SINGLE_CHUNK_MODEL: claude-3-5-sonnet-20240620
+          SINGLE_CHUNK_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
           EXCLUDE: "*.yml,*.yaml"
         run: python .gitea/scripts/code_review.py
 
-- 
2.49.1


From 0089b581bbb842becbc996209d633c10a723f08a Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 00:20:40 +0900
Subject: [PATCH 04/25] change model

---
 .gitea/workflows/code-review.yml | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/.gitea/workflows/code-review.yml b/.gitea/workflows/code-review.yml
index 4c5b042..68f1147 100644
--- a/.gitea/workflows/code-review.yml
+++ b/.gitea/workflows/code-review.yml
@@ -26,9 +26,9 @@ jobs:
       - name: Run Code Review
         env:
           ACCESS_TOKEN: ${{ secrets.ACCESS_TOKEN }}
-          FULL_CONTEXT_MODEL: gpt-4o
-          FULL_CONTEXT_API_KEY: ${{ secrets.OPENAI_API_KEY }}
-          SINGLE_CHUNK_MODEL: claude-3-5-sonnet-20240620
+          FULL_CONTEXT_MODEL: deepseek-reasoner
+          FULL_CONTEXT_API_KEY: ${{ secrets.DEEPSEEK_API_KEY }}
+          SINGLE_CHUNK_MODEL: gemini-2.0-flash-exp
           SINGLE_CHUNK_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
           EXCLUDE: "*.yml,*.yaml"
         run: python .gitea/scripts/code_review.py
-- 
2.49.1


From f1b4e2ce617fc5714dbde27405d6b5cea18da4d4 Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 00:24:11 +0900
Subject: [PATCH 05/25] remove model parameter

---
 .gitea/scripts/model.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.gitea/scripts/model.py b/.gitea/scripts/model.py
index 572dd72..8065795 100644
--- a/.gitea/scripts/model.py
+++ b/.gitea/scripts/model.py
@@ -84,7 +84,7 @@ class Model:
                 return AsyncAnthropic(api_key=api_key)
             case ModelProvider.GOOGLE:
                 genai.configure(api_key=api_key)
-                return genai.GenerativeModel(model=self.model, api_key=api_key)
+                return genai.GenerativeModel(self.model, api_key=api_key)
             case ModelProvider.DEEPSEEK:
                 return AsyncOpenAI(api_key=api_key, base_url="https://api.deepseek.com")
 
-- 
2.49.1


From 3ff0e65564a8c4e089992eb904448f56551467d7 Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 00:25:36 +0900
Subject: [PATCH 06/25] remove api key

---
 .gitea/scripts/model.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.gitea/scripts/model.py b/.gitea/scripts/model.py
index 8065795..be8a903 100644
--- a/.gitea/scripts/model.py
+++ b/.gitea/scripts/model.py
@@ -84,7 +84,7 @@ class Model:
                 return AsyncAnthropic(api_key=api_key)
             case ModelProvider.GOOGLE:
                 genai.configure(api_key=api_key)
-                return genai.GenerativeModel(self.model, api_key=api_key)
+                return genai.GenerativeModel(self.model)
             case ModelProvider.DEEPSEEK:
                 return AsyncOpenAI(api_key=api_key, base_url="https://api.deepseek.com")
 
-- 
2.49.1


From b5d2df6b2f3a7fcc3a5c2fade8efe0c45a76b338 Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 00:26:58 +0900
Subject: [PATCH 07/25] change api key

---
 .gitea/workflows/code-review.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.gitea/workflows/code-review.yml b/.gitea/workflows/code-review.yml
index 68f1147..21649e5 100644
--- a/.gitea/workflows/code-review.yml
+++ b/.gitea/workflows/code-review.yml
@@ -29,7 +29,7 @@ jobs:
           FULL_CONTEXT_MODEL: deepseek-reasoner
           FULL_CONTEXT_API_KEY: ${{ secrets.DEEPSEEK_API_KEY }}
           SINGLE_CHUNK_MODEL: gemini-2.0-flash-exp
-          SINGLE_CHUNK_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
+          SINGLE_CHUNK_API_KEY: ${{ secrets.GOOGLE_API_KEY }}
           EXCLUDE: "*.yml,*.yaml"
         run: python .gitea/scripts/code_review.py
 
-- 
2.49.1


From 94c892c4a771a6a21b6176f6cc97aa73b882e470 Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 00:34:28 +0900
Subject: [PATCH 08/25] add exeception

---
 .gitea/scripts/code_review.py | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/.gitea/scripts/code_review.py b/.gitea/scripts/code_review.py
index e87e86a..667908b 100644
--- a/.gitea/scripts/code_review.py
+++ b/.gitea/scripts/code_review.py
@@ -193,9 +193,11 @@ async def get_file_content(file: str) -> str | None:
             async with session.get(url) as response:
                 response.raise_for_status()
                 return await response.text()
-    except Exception as e:
-        print(f"Failed to get file content: {e}")
-        return None
+    except aiohttp.ClientError as e:  # More specific exception handling
+        print(f"Network error fetching {file}: {e}")
+    except asyncio.TimeoutError:
+        print(f"Timeout fetching {file}")
+    return None
 
 
 async def analyze_full_context(
-- 
2.49.1


From 1970f27dceb65cfbf9e6a387efcc25d5e7ce3097 Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 00:39:52 +0900
Subject: [PATCH 09/25] update yml

---
 .gitea/workflows/python-lint.yml | 30 +++++++++++++++++++++
 .gitea/workflows/rust-lint.yml   | 45 ++++++++++++++++++++++++++++++++
 2 files changed, 75 insertions(+)

diff --git a/.gitea/workflows/python-lint.yml b/.gitea/workflows/python-lint.yml
index e69de29..7806f1d 100644
--- a/.gitea/workflows/python-lint.yml
+++ b/.gitea/workflows/python-lint.yml
@@ -0,0 +1,30 @@
+name: mint_ci
+
+on:
+  push:
+    branches: [ "main"]
+  pull_request:
+    types: [unlabeled, opened, synchronize, reopened]
+
+env:
+  PYTHON_VERSION: "3.12.3"
+
+jobs:
+  lint:
+    name: Check Python code using ruff
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v4
+      with:
+        token: ${{ secrets.ACCESS_TOKEN }}
+    - uses: astral-sh/ruff-action@v1
+      with:
+        version: 0.7.4
+        args: check . --select=E5,F4,F8,D400,D403,D417,D100,D102,D103,D101,ANN001,ANN201 --output-format=full --exclude='**/test_*.py','**/__init__.py' --force-exclude
+        changed-files: 'true'
+    - uses: astral-sh/ruff-action@v1
+      with:
+        version: 0.7.4
+        args: check . --select=I --output-format=full --force-exclude
+        changed-files: 'true'
+      continue-on-error: true
\ No newline at end of file
diff --git a/.gitea/workflows/rust-lint.yml b/.gitea/workflows/rust-lint.yml
index e69de29..c1b627d 100644
--- a/.gitea/workflows/rust-lint.yml
+++ b/.gitea/workflows/rust-lint.yml
@@ -0,0 +1,45 @@
+on:
+  push:
+    branches: [main]
+  pull_request:
+    types: [unlabeled, opened, synchronize, reopened]
+  merge_group:
+
+name: Rust-lint
+
+
+jobs:
+  rust_tests:
+    env:
+      RUST_BACKTRACE: full
+    name: Run rust tests
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: dtolnay/rust-toolchain@stable
+        with:
+          components: clippy
+      - uses: Swatinem/rust-cache@v2
+
+      - name: run clippy
+        run: cargo clippy --workspace  --exclude rustpython_wasm -- -Dwarnings
+
+      - name: run rust tests
+        run: cargo test --workspace --verbose
+
+      - name: check compilation without threading
+        run: cargo check
+
+
+  lint:
+    name: Check Rust code with rustfmt and clippy
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: dtolnay/rust-toolchain@stable
+        with:
+            components: rustfmt, clippy
+      - name: run rustfmt
+        run: cargo fmt --check
+
+  
\ No newline at end of file
-- 
2.49.1


From 457f5ba18ea715479dac94a246ef125ab146f078 Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 00:54:45 +0900
Subject: [PATCH 10/25] update google config

---
 .gitea/scripts/model.py | 18 ++++++++++++++++--
 1 file changed, 16 insertions(+), 2 deletions(-)

diff --git a/.gitea/scripts/model.py b/.gitea/scripts/model.py
index be8a903..c1ed4d7 100644
--- a/.gitea/scripts/model.py
+++ b/.gitea/scripts/model.py
@@ -6,6 +6,14 @@ from typing import Any
 import google.generativeai as genai
 from anthropic import AsyncAnthropic
 from openai import AsyncOpenAI
+import typing_extensions as typing
+
+
+class GoogleReponse(typing.TypedDict):
+    """The response from Google model."""
+
+    lineNumber: int
+    reviewComment: str
 
 
 class ModelProvider(Enum):
@@ -128,7 +136,13 @@ class Model:
                 )
                 return response.content[0].text.strip()
             case ModelProvider.GOOGLE:
-                response = await self.session.generate_content_async(prompt)
+                response = await self.session.generate_content_async(
+                    prompt,
+                    generation_config=genai.GenerationConfig(
+                        response_mime_type="application/json",
+                        response_schema=list[GoogleReponse],
+                    ),
+                )
                 return response.text.strip()
 
     async def get_response_single_chunk(
@@ -175,7 +189,7 @@ class Model:
 SINGLE_CHUNK_SYSTEM_PROMPT = (
     "Your task is to review pull requests. Instructions:\n"
     "- Provide the response in the following JSON format:  "
-    """[{{"lineNumber":  <line_number>, "reviewComment": "<review comment>"}}] \n"""
+    """[{{"lineNumber": int, "reviewComment": str}}] \n"""
     "- lineNumber is about the line number of the code that in new file. \n"
     "- Do not give positive comments or compliments. \n"
     "- Provide comments and suggestions ONLY if there is something to improve"
-- 
2.49.1


From 3febb99f5b369020356df45325bd2d3d8ea3a909 Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 00:59:12 +0900
Subject: [PATCH 11/25] update system prompt

---
 .gitea/scripts/model.py | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/.gitea/scripts/model.py b/.gitea/scripts/model.py
index c1ed4d7..3cc6873 100644
--- a/.gitea/scripts/model.py
+++ b/.gitea/scripts/model.py
@@ -4,9 +4,9 @@ from enum import Enum
 from typing import Any
 
 import google.generativeai as genai
+import typing_extensions as typing
 from anthropic import AsyncAnthropic
 from openai import AsyncOpenAI
-import typing_extensions as typing
 
 
 class GoogleReponse(typing.TypedDict):
@@ -92,7 +92,9 @@ class Model:
                 return AsyncAnthropic(api_key=api_key)
             case ModelProvider.GOOGLE:
                 genai.configure(api_key=api_key)
-                return genai.GenerativeModel(self.model)
+                return genai.GenerativeModel(
+                    model_name=self.model, system_instruction=self.system_prompt
+                )
             case ModelProvider.DEEPSEEK:
                 return AsyncOpenAI(api_key=api_key, base_url="https://api.deepseek.com")
 
-- 
2.49.1


From 9454ba7df36874c907f1056c5988ee73742a53cb Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 01:02:00 +0900
Subject: [PATCH 12/25] add line number information

---
 .gitea/scripts/code_review.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/.gitea/scripts/code_review.py b/.gitea/scripts/code_review.py
index 667908b..08fd9c1 100644
--- a/.gitea/scripts/code_review.py
+++ b/.gitea/scripts/code_review.py
@@ -89,10 +89,10 @@ def parse_diff(diff: str) -> list[dict[str, Any]]:
         diff_text = []
         for line in remain_text.splitlines():
             if line.startswith("-"):
-                diff_text.append(f"{old_idx} {line}")
+                diff_text.append(f"old_line_number:{old_idx} {line}")
                 old_idx += 1
             elif line.startswith("+"):
-                diff_text.append(f"{new_idx} {line}")
+                diff_text.append(f"new_line_number:{new_idx} {line}")
                 new_idx += 1
             else:
                 diff_text.append(line)
-- 
2.49.1


From a3904537f7c0a539c67f09c9a0593a3680158c58 Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 01:04:55 +0900
Subject: [PATCH 13/25] add position

---
 .gitea/scripts/code_review.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/.gitea/scripts/code_review.py b/.gitea/scripts/code_review.py
index 08fd9c1..4dba884 100644
--- a/.gitea/scripts/code_review.py
+++ b/.gitea/scripts/code_review.py
@@ -127,7 +127,8 @@ def create_comment(
     for ai_response in ai_response:
         comments.append(
             {
-                "body": f"[REVIEW] {ai_response['reviewComment']}",
+                "body": f"[REVIEW] {ai_response['reviewComment']},"
+                f"at line {ai_response['lineNumber']}",
                 "path": file,
                 "new_position": int(ai_response["lineNumber"]),
             }
-- 
2.49.1


From 939af86aa6da7b53b1370e800382a92cfef84101 Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 01:08:21 +0900
Subject: [PATCH 14/25] update system prompt to indicate linenumber precisely

---
 .gitea/scripts/code_review.py | 3 +--
 .gitea/scripts/model.py       | 1 +
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/.gitea/scripts/code_review.py b/.gitea/scripts/code_review.py
index 4dba884..08fd9c1 100644
--- a/.gitea/scripts/code_review.py
+++ b/.gitea/scripts/code_review.py
@@ -127,8 +127,7 @@ def create_comment(
     for ai_response in ai_response:
         comments.append(
             {
-                "body": f"[REVIEW] {ai_response['reviewComment']},"
-                f"at line {ai_response['lineNumber']}",
+                "body": f"[REVIEW] {ai_response['reviewComment']}",
                 "path": file,
                 "new_position": int(ai_response["lineNumber"]),
             }
diff --git a/.gitea/scripts/model.py b/.gitea/scripts/model.py
index 3cc6873..78d0aa5 100644
--- a/.gitea/scripts/model.py
+++ b/.gitea/scripts/model.py
@@ -193,6 +193,7 @@ SINGLE_CHUNK_SYSTEM_PROMPT = (
     "- Provide the response in the following JSON format:  "
     """[{{"lineNumber": int, "reviewComment": str}}] \n"""
     "- lineNumber is about the line number of the code that in new file. \n"
+    "- lineNumber can be cound at the front of each line. \n"
     "- Do not give positive comments or compliments. \n"
     "- Provide comments and suggestions ONLY if there is something to improve"
     "otherwise return an empty array. \n"
-- 
2.49.1


From 2a5f2491c2ac8c78b195781f90213c80fba2b949 Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 11:44:27 +0900
Subject: [PATCH 15/25] add prompt log

---
 .gitea/scripts/model.py | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/.gitea/scripts/model.py b/.gitea/scripts/model.py
index 78d0aa5..aa57827 100644
--- a/.gitea/scripts/model.py
+++ b/.gitea/scripts/model.py
@@ -162,7 +162,10 @@ class Model:
             str: The response.
         """
         prompt = SINGLE_CHUNK_USER_PROMPT.format(file, title, description, chunk)
-        return await self.request(prompt)
+        response = await self.request(prompt)
+        print(f"prompt: {prompt}")
+        print(f"response: {response}")
+        return response
 
     async def get_response_full_context(
         self, title: str, description: str, file_contents: list[str]
@@ -181,7 +184,10 @@ class Model:
             prompt = FULL_CONTEXT_USER_PROMPT.format(
                 title, description, "\n".join(file_contents)
             )
-            return await self.request(prompt)
+            response = await self.request(prompt)
+            print(f"prompt: {prompt}")
+            print(f"response: {response}")
+            return response
         except Exception as e:
             print(f"Error during full context response: {e}")
             print(prompt)
-- 
2.49.1


From ccf60f3c7e622243a46e3b0bcdaeb4d6387fd74c Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 11:49:43 +0900
Subject: [PATCH 16/25] remove response

---
 .gitea/scripts/model.py | 10 ++--------
 1 file changed, 2 insertions(+), 8 deletions(-)

diff --git a/.gitea/scripts/model.py b/.gitea/scripts/model.py
index aa57827..78d0aa5 100644
--- a/.gitea/scripts/model.py
+++ b/.gitea/scripts/model.py
@@ -162,10 +162,7 @@ class Model:
             str: The response.
         """
         prompt = SINGLE_CHUNK_USER_PROMPT.format(file, title, description, chunk)
-        response = await self.request(prompt)
-        print(f"prompt: {prompt}")
-        print(f"response: {response}")
-        return response
+        return await self.request(prompt)
 
     async def get_response_full_context(
         self, title: str, description: str, file_contents: list[str]
@@ -184,10 +181,7 @@ class Model:
             prompt = FULL_CONTEXT_USER_PROMPT.format(
                 title, description, "\n".join(file_contents)
             )
-            response = await self.request(prompt)
-            print(f"prompt: {prompt}")
-            print(f"response: {response}")
-            return response
+            return await self.request(prompt)
         except Exception as e:
             print(f"Error during full context response: {e}")
             print(prompt)
-- 
2.49.1


From 76d543c766afb86b7f90f99bc3316a7b9999aab3 Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 13:42:46 +0900
Subject: [PATCH 17/25] print for debug

---
 .gitea/scripts/code_review.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/.gitea/scripts/code_review.py b/.gitea/scripts/code_review.py
index 08fd9c1..9bf6a93 100644
--- a/.gitea/scripts/code_review.py
+++ b/.gitea/scripts/code_review.py
@@ -285,7 +285,9 @@ async def main() -> None:
         is_full_context=False,
     )
 
+    print("diff: ", diff)
     parsed_diff = parse_diff(diff)
+    print("parsed_diff: ", parsed_diff)
     comments_task = asyncio.create_task(
         analyze_single_chunks(single_chunk_model, parsed_diff)
     )
-- 
2.49.1


From a7ae4de9bf25c7c1905f40500a50a51bc86ea74a Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 14:55:32 +0900
Subject: [PATCH 18/25] update line number tag

---
 .gitea/scripts/code_review.py | 41 +++++++++++++++++------------------
 .gitea/scripts/model.py       |  3 ++-
 2 files changed, 22 insertions(+), 22 deletions(-)

diff --git a/.gitea/scripts/code_review.py b/.gitea/scripts/code_review.py
index 9bf6a93..53c7a5e 100644
--- a/.gitea/scripts/code_review.py
+++ b/.gitea/scripts/code_review.py
@@ -60,7 +60,7 @@ def parse_diff(diff: str) -> list[dict[str, Any]]:
     )
     old_new_pattern = re.compile(r"(?m)^(---|\+\+\+)\s+(.*)$")
     hunk_pattern = re.compile(
-        r"^@@ -(\d+)(?:,(\d+))? \+(\d+)(?:,(\d+))? @@(.*?)(?=^@@ |$)",
+        r"@@ -(\d+)(?:,(\d+))? \+(\d+)(?:,(\d+))? @@(.*?)?(?=@@ -\d+(?:,\d+)? \+\d+(?:,\d+)? @@|\Z)",
         re.MULTILINE | re.DOTALL,
     )
     list_diff = []
@@ -79,33 +79,32 @@ def parse_diff(diff: str) -> list[dict[str, Any]]:
             print("Neglict deleted file")
             continue
         new_file = new_file.lstrip("b/")
-
-        hunk_match = hunk_pattern.search(diff_text)
-        if hunk_match is None:
-            continue
-        old_idx = int(hunk_match.group(1))
-        new_idx = int(hunk_match.group(3))
-        remain_text = diff_text[hunk_match.end() + 1 :]
-        diff_text = []
-        for line in remain_text.splitlines():
-            if line.startswith("-"):
-                diff_text.append(f"old_line_number:{old_idx} {line}")
-                old_idx += 1
-            elif line.startswith("+"):
-                diff_text.append(f"new_line_number:{new_idx} {line}")
-                new_idx += 1
-            else:
-                diff_text.append(line)
-        diff_text = "\n".join(diff_text)
-
         if any(fnmatch.fnmatch(new_file, pattern) for pattern in EXCLUDE_PATTERNS):
             print(f"Exclude file {new_file}")
             continue
 
+        output_diff_text = []
+        for hunk_match in hunk_pattern.finditer(diff_text):
+            old_idx = int(hunk_match.group(1))
+            new_idx = int(hunk_match.group(3))
+            remain_text = hunk_match.group(5).splitlines()
+            for line in remain_text:
+                if line.startswith("-"):
+                    output_diff_text.append(f"{old_idx} \t {line}")
+                    old_idx += 1
+                elif line.startswith("+"):
+                    output_diff_text.append(f"\t {new_idx} {line}")
+                    new_idx += 1
+                else:
+                    output_diff_text.append(f"{old_idx} {new_idx} {line}")
+                    old_idx += 1
+                    new_idx += 1
+
+        output_diff_text = "\n".join(output_diff_text)
         list_diff.append(
             {
                 "file": new_file,
-                "chunk": diff_text,
+                "chunk": output_diff_text,
             }
         )
     return list_diff
diff --git a/.gitea/scripts/model.py b/.gitea/scripts/model.py
index 78d0aa5..b484170 100644
--- a/.gitea/scripts/model.py
+++ b/.gitea/scripts/model.py
@@ -193,7 +193,8 @@ SINGLE_CHUNK_SYSTEM_PROMPT = (
     "- Provide the response in the following JSON format:  "
     """[{{"lineNumber": int, "reviewComment": str}}] \n"""
     "- lineNumber is about the line number of the code that in new file. \n"
-    "- lineNumber can be cound at the front of each line. \n"
+    "- lineNumber can be found at the front of each line. \n"
+    "- At the first number is old line number, the second number is new line number. \n"
     "- Do not give positive comments or compliments. \n"
     "- Provide comments and suggestions ONLY if there is something to improve"
     "otherwise return an empty array. \n"
-- 
2.49.1


From 3e97b34e2eeab027d9a712d7b5034e0ae764e8fc Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 14:56:36 +0900
Subject: [PATCH 19/25] remove print for log

---
 .gitea/scripts/code_review.py | 2 --
 1 file changed, 2 deletions(-)

diff --git a/.gitea/scripts/code_review.py b/.gitea/scripts/code_review.py
index 53c7a5e..0e3edf5 100644
--- a/.gitea/scripts/code_review.py
+++ b/.gitea/scripts/code_review.py
@@ -284,9 +284,7 @@ async def main() -> None:
         is_full_context=False,
     )
 
-    print("diff: ", diff)
     parsed_diff = parse_diff(diff)
-    print("parsed_diff: ", parsed_diff)
     comments_task = asyncio.create_task(
         analyze_single_chunks(single_chunk_model, parsed_diff)
     )
-- 
2.49.1


From 133437192d1b697889b6722d0d6355c355e95dd6 Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 15:04:42 +0900
Subject: [PATCH 20/25] simplify regex pattern

---
 .gitea/scripts/code_review.py | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/.gitea/scripts/code_review.py b/.gitea/scripts/code_review.py
index 0e3edf5..3a3246c 100644
--- a/.gitea/scripts/code_review.py
+++ b/.gitea/scripts/code_review.py
@@ -60,7 +60,7 @@ def parse_diff(diff: str) -> list[dict[str, Any]]:
     )
     old_new_pattern = re.compile(r"(?m)^(---|\+\+\+)\s+(.*)$")
     hunk_pattern = re.compile(
-        r"@@ -(\d+)(?:,(\d+))? \+(\d+)(?:,(\d+))? @@(.*?)?(?=@@ -\d+(?:,\d+)? \+\d+(?:,\d+)? @@|\Z)",
+        r"@@ -(\d+)(?:,(\d+))? \+(\d+)(?:,(\d+))? @@(.*?)?(?=@@|\Z)",
         re.MULTILINE | re.DOTALL,
     )
     list_diff = []
@@ -90,10 +90,10 @@ def parse_diff(diff: str) -> list[dict[str, Any]]:
             remain_text = hunk_match.group(5).splitlines()
             for line in remain_text:
                 if line.startswith("-"):
-                    output_diff_text.append(f"{old_idx} \t {line}")
+                    output_diff_text.append(f"{old_idx} None {line}")
                     old_idx += 1
                 elif line.startswith("+"):
-                    output_diff_text.append(f"\t {new_idx} {line}")
+                    output_diff_text.append(f"None {new_idx} {line}")
                     new_idx += 1
                 else:
                     output_diff_text.append(f"{old_idx} {new_idx} {line}")
-- 
2.49.1


From 443c8271c06b7ed463047733e7692c07f50e804d Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 22:36:45 +0900
Subject: [PATCH 21/25] full context only for opened

---
 .gitea/scripts/code_review.py | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/.gitea/scripts/code_review.py b/.gitea/scripts/code_review.py
index 3a3246c..127cebd 100644
--- a/.gitea/scripts/code_review.py
+++ b/.gitea/scripts/code_review.py
@@ -288,13 +288,16 @@ async def main() -> None:
     comments_task = asyncio.create_task(
         analyze_single_chunks(single_chunk_model, parsed_diff)
     )
-    full_context_response_task = asyncio.create_task(
-        analyze_full_context(full_context_model, parsed_diff)
-    )
+
+    if EVENT_DATA["action"] == "opened":
+        full_context_response_task = asyncio.create_task(
+            analyze_full_context(full_context_model, parsed_diff)
+        )
+        full_context_response = await full_context_response_task
+    else:
+        full_context_response = ""
 
     comments = await comments_task
-    full_context_response = await full_context_response_task
-
     post_review(full_context_response, comments)
 
 
-- 
2.49.1


From b8fd0bd36128d64646af3be8e7f2ecea526ae154 Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 22:37:05 +0900
Subject: [PATCH 22/25] change hunk for more descriptive name

---
 .gitea/scripts/code_review.py | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/.gitea/scripts/code_review.py b/.gitea/scripts/code_review.py
index 127cebd..769ee74 100644
--- a/.gitea/scripts/code_review.py
+++ b/.gitea/scripts/code_review.py
@@ -59,7 +59,7 @@ def parse_diff(diff: str) -> list[dict[str, Any]]:
         r"(?s)diff --git a/(.+?) b/(.*?)\r?\n(.*?)(?=diff --git a/|$)", re.S
     )
     old_new_pattern = re.compile(r"(?m)^(---|\+\+\+)\s+(.*)$")
-    hunk_pattern = re.compile(
+    chunk_range_pattern = re.compile(
         r"@@ -(\d+)(?:,(\d+))? \+(\d+)(?:,(\d+))? @@(.*?)?(?=@@|\Z)",
         re.MULTILINE | re.DOTALL,
     )
@@ -84,10 +84,10 @@ def parse_diff(diff: str) -> list[dict[str, Any]]:
             continue
 
         output_diff_text = []
-        for hunk_match in hunk_pattern.finditer(diff_text):
-            old_idx = int(hunk_match.group(1))
-            new_idx = int(hunk_match.group(3))
-            remain_text = hunk_match.group(5).splitlines()
+        for chunk_range_match in chunk_range_pattern.finditer(diff_text):
+            old_idx = int(chunk_range_match.group(1))
+            new_idx = int(chunk_range_match.group(3))
+            remain_text = chunk_range_match.group(5).splitlines()
             for line in remain_text:
                 if line.startswith("-"):
                     output_diff_text.append(f"{old_idx} None {line}")
-- 
2.49.1


From 810b46b0b4015511b8814ea02aa719d427ae2fa0 Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 23:03:50 +0900
Subject: [PATCH 23/25] adapt comment

---
 .gitea/scripts/code_review.py | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/.gitea/scripts/code_review.py b/.gitea/scripts/code_review.py
index 769ee74..0a1683b 100644
--- a/.gitea/scripts/code_review.py
+++ b/.gitea/scripts/code_review.py
@@ -87,8 +87,7 @@ def parse_diff(diff: str) -> list[dict[str, Any]]:
         for chunk_range_match in chunk_range_pattern.finditer(diff_text):
             old_idx = int(chunk_range_match.group(1))
             new_idx = int(chunk_range_match.group(3))
-            remain_text = chunk_range_match.group(5).splitlines()
-            for line in remain_text:
+            for line in chunk_range_match.group(5).splitlines():
                 if line.startswith("-"):
                     output_diff_text.append(f"{old_idx} None {line}")
                     old_idx += 1
@@ -150,7 +149,7 @@ async def analyze_single_chunks(
     async def process_single_chunk(diff: dict[str, Any]):
         file = diff["file"]
         chunk = diff["chunk"]
-        response = await single_chunk_model.get_response_single_chunk(
+        response = single_chunk_model.get_response_single_chunk(
             file, title, description, chunk
         )
         response = response.strip("`").lstrip("json").strip() or "[]"
@@ -215,7 +214,7 @@ async def analyze_full_context(
     async def get_file_data(diff: dict[str, Any]):
         file = diff["file"]
         chunk = diff["chunk"]
-        content = await get_file_content(file)
+        content = get_file_content(file)
         if content is None:
             return None
         return f"File: {file}\n{content}\nDiff: {chunk}"
@@ -230,7 +229,7 @@ async def analyze_full_context(
 
     title = EVENT_DATA["pull_request"]["title"]
     description = EVENT_DATA["pull_request"]["body"]
-    response = await full_context_model.get_response_full_context(
+    response = full_context_model.get_response_full_context(
         title, description, file_contents
     )
     response = response.strip("`").lstrip("markdown").strip()
-- 
2.49.1


From 284e0735a71e6a28989d0fec6971f6a794af3bc4 Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 23:06:42 +0900
Subject: [PATCH 24/25] add await

---
 .gitea/scripts/code_review.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/.gitea/scripts/code_review.py b/.gitea/scripts/code_review.py
index 0a1683b..bb75b9b 100644
--- a/.gitea/scripts/code_review.py
+++ b/.gitea/scripts/code_review.py
@@ -149,7 +149,7 @@ async def analyze_single_chunks(
     async def process_single_chunk(diff: dict[str, Any]):
         file = diff["file"]
         chunk = diff["chunk"]
-        response = single_chunk_model.get_response_single_chunk(
+        response = await single_chunk_model.get_response_single_chunk(
             file, title, description, chunk
         )
         response = response.strip("`").lstrip("json").strip() or "[]"
@@ -229,7 +229,7 @@ async def analyze_full_context(
 
     title = EVENT_DATA["pull_request"]["title"]
     description = EVENT_DATA["pull_request"]["body"]
-    response = full_context_model.get_response_full_context(
+    response = await full_context_model.get_response_full_context(
         title, description, file_contents
     )
     response = response.strip("`").lstrip("markdown").strip()
-- 
2.49.1


From c51f3060cd39b9cc687e6cf0f0a9cef22c30f44f Mon Sep 17 00:00:00 2001
From: Myeongseon Choi <key262yek@gmail.com>
Date: Mon, 27 Jan 2025 23:24:28 +0900
Subject: [PATCH 25/25] change instruction

---
 .gitea/scripts/model.py | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/.gitea/scripts/model.py b/.gitea/scripts/model.py
index b484170..dbe344e 100644
--- a/.gitea/scripts/model.py
+++ b/.gitea/scripts/model.py
@@ -195,12 +195,17 @@ SINGLE_CHUNK_SYSTEM_PROMPT = (
     "- lineNumber is about the line number of the code that in new file. \n"
     "- lineNumber can be found at the front of each line. \n"
     "- At the first number is old line number, the second number is new line number. \n"
+    "- If the line starts with `+`, it means the line is added. \n"
+    "- If the line starts with `-`, it means the line is deleted. \n"
+    "- Evaluate whether the code changes and additions are appropriate "
+    "and if the new code structure is suitable. \n"
     "- Do not give positive comments or compliments. \n"
     "- Provide comments and suggestions ONLY if there is something to improve"
     "otherwise return an empty array. \n"
     "- Write the comment in GitHub Markdown format. \n"
     "- Use the given description only for the overall context "
     "and only comment the code. \n"
+    "- Do not suggest type hint or naming convention. \n"
     "- IMPORTANT: NEVER suggest adding comments to the code. \n"
 )
 SINGLE_CHUNK_USER_PROMPT = (
-- 
2.49.1