mirror of
https://github.com/codeflash-ai/codeflash-internal.git
synced 2026-05-04 18:25:18 +00:00
359 lines
9.9 KiB
Text
359 lines
9.9 KiB
Text
{
|
|
"cells": [
|
|
{
|
|
"cell_type": "code",
|
|
"id": "e4d33681-e7f7-4508-b453-68c8c57c1f45",
|
|
"metadata": {
|
|
"ExecuteTime": {
|
|
"end_time": "2025-09-11T02:32:16.058137Z",
|
|
"start_time": "2025-09-11T02:32:16.052210Z"
|
|
}
|
|
},
|
|
"source": [
|
|
"import json\n",
|
|
"import os\n",
|
|
"from typing import Any, Dict, Optional\n",
|
|
"import re\n",
|
|
"import pandas as pd\n",
|
|
"from pandas import DataFrame\n",
|
|
"from scipy.stats import hmean\n",
|
|
"from sqlalchemy import create_engine\n",
|
|
"import difflib\n",
|
|
"from dotenv import load_dotenv\n",
|
|
"# Load environment variables from .env file\n",
|
|
"load_dotenv(dotenv_path='../django/aiservice/.env')\n",
|
|
"\n",
|
|
"# Access the environment variables\n",
|
|
"DATABASE_URI = os.getenv(\"DATABASE_URL\")\n",
|
|
"trace_id = 'your trace id'\n"
|
|
],
|
|
"outputs": [],
|
|
"execution_count": 11
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"id": "3c900e6c5dd188d9",
|
|
"metadata": {
|
|
"ExecuteTime": {
|
|
"end_time": "2025-09-11T02:32:28.565731Z",
|
|
"start_time": "2025-09-11T02:32:28.561708Z"
|
|
}
|
|
},
|
|
"source": [
|
|
"def load_data(\n",
|
|
" trace_id: str, database_uri: str\n",
|
|
") -> pd.DataFrame:\n",
|
|
" engine = create_engine(database_uri)\n",
|
|
" with engine.connect() as connection:\n",
|
|
" query = \"\"\"\n",
|
|
" SELECT * FROM optimization_features\n",
|
|
" WHERE created_at > 'time you want to look at'\n",
|
|
" AND user_id LIKE 'git hub user id'\n",
|
|
" \"\"\"\n",
|
|
" return pd.read_sql_query(\n",
|
|
" query, connection, params=(trace_id,)\n",
|
|
" )\n"
|
|
],
|
|
"outputs": [],
|
|
"execution_count": 14
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"id": "3642113faff9515b",
|
|
"metadata": {
|
|
"ExecuteTime": {
|
|
"end_time": "2025-09-11T02:32:33.436079Z",
|
|
"start_time": "2025-09-11T02:32:29.373321Z"
|
|
}
|
|
},
|
|
"source": "data = load_data(trace_id, DATABASE_URI)\n",
|
|
"outputs": [],
|
|
"execution_count": 15
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"id": "bf7a941e66d5ed97",
|
|
"metadata": {},
|
|
"source": [
|
|
"data"
|
|
],
|
|
"outputs": [],
|
|
"execution_count": null
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"id": "95820a54-c15c-48a9-9f3a-5858855681fa",
|
|
"metadata": {},
|
|
"source": "excel_out = {'diffs':[],'optimization_ids':[],'speedups':[], 'trace_id':[]}",
|
|
"outputs": [],
|
|
"execution_count": null
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"id": "48a7deca-e4be-4f57-bc60-af73fa73a65e",
|
|
"metadata": {
|
|
"ExecuteTime": {
|
|
"end_time": "2025-09-11T02:32:44.274590Z",
|
|
"start_time": "2025-09-11T02:32:44.270598Z"
|
|
}
|
|
},
|
|
"source": [
|
|
"def extract(text: str) -> str:\n",
|
|
" # Regex to capture content inside triple backticks\n",
|
|
" match = re.search(r\"```(?:[\\w+:.\\-]*)\\n(.*?)```\", text, re.DOTALL)\n",
|
|
" if match:\n",
|
|
" return match.group(1).strip()\n",
|
|
" return \"\"\n",
|
|
"def unified_diff_strings(code1: str, code2: str, fromfile=\"original\", tofile=\"modified\") -> str:\n",
|
|
" \"\"\"\n",
|
|
" Return the unified diff between two code strings as a single string.\n",
|
|
"\n",
|
|
" :param code1: First code string (original).\n",
|
|
" :param code2: Second code string (modified).\n",
|
|
" :param fromfile: Label for the first code string.\n",
|
|
" :param tofile: Label for the second code string.\n",
|
|
" :return: Unified diff as a string.\n",
|
|
" \"\"\"\n",
|
|
" code1_lines = code1.splitlines(keepends=True)\n",
|
|
" code2_lines = code2.splitlines(keepends=True)\n",
|
|
"\n",
|
|
" diff = difflib.unified_diff(\n",
|
|
" code1_lines,\n",
|
|
" code2_lines,\n",
|
|
" fromfile=fromfile,\n",
|
|
" tofile=tofile,\n",
|
|
" lineterm=\"\"\n",
|
|
" )\n",
|
|
"\n",
|
|
" return \"\".join(diff)\n",
|
|
"def diff_length(a: str, b: str) -> int:\n",
|
|
" \"\"\"Compute the length (in characters) of the unified diff between two strings.\n",
|
|
"\n",
|
|
" Args:\n",
|
|
" a (str): Original string.\n",
|
|
" b (str): Modified string.\n",
|
|
"\n",
|
|
" Returns:\n",
|
|
" int: Total number of characters in the diff.\n",
|
|
"\n",
|
|
" \"\"\"\n",
|
|
" # Split input strings into lines for line-by-line diff\n",
|
|
" a_lines = a.splitlines(keepends=True)\n",
|
|
" b_lines = b.splitlines(keepends=True)\n",
|
|
"\n",
|
|
" # Compute unified diff\n",
|
|
" diff_lines = list(difflib.unified_diff(a_lines, b_lines, lineterm=\"\"))\n",
|
|
"\n",
|
|
" # Join all lines with newline to calculate total diff length\n",
|
|
" diff_text = \"\\n\".join(diff_lines)\n",
|
|
"\n",
|
|
" return len(diff_text)\n",
|
|
"\n",
|
|
"\n",
|
|
"def count_changed_lines(code1: str, code2: str) -> int:\n",
|
|
" \"\"\"\n",
|
|
" Count the number of changed lines between two code strings.\n",
|
|
"\n",
|
|
" Args:\n",
|
|
" code1 (str): The first code string.\n",
|
|
" code2 (str): The second code string.\n",
|
|
"\n",
|
|
" Returns:\n",
|
|
" int: The number of changed lines (additions + deletions).\n",
|
|
" \"\"\"\n",
|
|
" lines1 = code1.splitlines()\n",
|
|
" lines2 = code2.splitlines()\n",
|
|
"\n",
|
|
" diff = difflib.ndiff(lines1, lines2)\n",
|
|
"\n",
|
|
" # Count lines starting with '+' (added) or '-' (removed)\n",
|
|
" changed_lines = sum(1 for line in diff if line.startswith(('+', '-')))\n",
|
|
"\n",
|
|
" return changed_lines"
|
|
],
|
|
"outputs": [],
|
|
"execution_count": 16
|
|
},
|
|
{
|
|
"metadata": {},
|
|
"cell_type": "code",
|
|
"source": "data['trace_id']",
|
|
"id": "cce7246ce5aa2b0f",
|
|
"outputs": [],
|
|
"execution_count": null
|
|
},
|
|
{
|
|
"metadata": {
|
|
"ExecuteTime": {
|
|
"end_time": "2025-09-11T02:36:38.194171Z",
|
|
"start_time": "2025-09-11T02:36:38.190926Z"
|
|
}
|
|
},
|
|
"cell_type": "code",
|
|
"source": [
|
|
"print(data['original_code'][0])\n",
|
|
"#extract(data['original_code'][0])"
|
|
],
|
|
"id": "aac5f37de751c812",
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"```python:langgraph/pregel/debug.py\n",
|
|
"from __future__ import annotations\n",
|
|
"\n",
|
|
"def get_colored_text(text: str, color: str) -> str:\n",
|
|
" \"\"\"Get colored text.\"\"\"\n",
|
|
" return f\"\\033[1;3{COLOR_MAPPING[color]}m{text}\\033[0m\"\n",
|
|
"```\n"
|
|
]
|
|
}
|
|
],
|
|
"execution_count": 20
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"id": "5dc620ce-2be6-4aca-829c-ea026792f5f0",
|
|
"metadata": {
|
|
"ExecuteTime": {
|
|
"end_time": "2025-09-11T02:39:02.539570Z",
|
|
"start_time": "2025-09-11T02:39:02.495813Z"
|
|
}
|
|
},
|
|
"source": [
|
|
"excel_out = {'diffs':[],'diff_lens':[],'speedups':[], 'explanations':[]}\n",
|
|
"i=0\n",
|
|
"difflens=[]\n",
|
|
"difflens_lines = []\n",
|
|
"for _, row in data.iterrows():\n",
|
|
" #print(row)\n",
|
|
" if row['ranking']: #if it exists\n",
|
|
" for opt_id in row['ranking']['ranking']:\n",
|
|
" raw_original = row['original_code']\n",
|
|
" raw_modified = row['optimizations_post'][opt_id]\n",
|
|
" excel_out['diffs'].append(unified_diff_strings(raw_original,raw_modified))\n",
|
|
" excel_out['diff_lens'].append(diff_length(raw_original,raw_modified))\n",
|
|
" #print(row['trace_id'],row['speedup_ratio'])\n",
|
|
" excel_out['speedups'].append(row['speedup_ratio'][opt_id])\n",
|
|
" excel_out['explanations'].append(row['ranking']['explanation'])\n",
|
|
"df = pd.DataFrame(excel_out)\n",
|
|
"df.to_excel(\"output_newsep10.xlsx\", index=False)"
|
|
],
|
|
"outputs": [],
|
|
"execution_count": 21
|
|
},
|
|
{
|
|
"metadata": {},
|
|
"cell_type": "code",
|
|
"source": "!uv pip install openpyxl\n",
|
|
"id": "d2b1c5601a786c32",
|
|
"outputs": [],
|
|
"execution_count": null
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"id": "b612aa33-0f5d-4c8d-9cba-0426af654299",
|
|
"metadata": {},
|
|
"source": [
|
|
"import json\n",
|
|
"# Convert the dictionary to a JSON string\n",
|
|
"json_string = json.dumps(excel_out)\n",
|
|
"\n",
|
|
"# Print the resulting JSON string\n",
|
|
"print(json_string)"
|
|
],
|
|
"outputs": [],
|
|
"execution_count": null
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"id": "226cedec-2f7e-4a45-a789-e0d233170678",
|
|
"metadata": {},
|
|
"source": [
|
|
"with open('excel_out.json', 'r') as f:\n",
|
|
" excel_out_json = json.load(f)"
|
|
],
|
|
"outputs": [],
|
|
"execution_count": null
|
|
},
|
|
{
|
|
"metadata": {},
|
|
"cell_type": "code",
|
|
"source": [
|
|
"print(excel_out[''][11],excel_out['diffs'][12])\n",
|
|
"\n"
|
|
],
|
|
"id": "b26e7119057f215",
|
|
"outputs": [],
|
|
"execution_count": null
|
|
},
|
|
{
|
|
"metadata": {},
|
|
"cell_type": "code",
|
|
"source": "print(difflens[11],difflens[12])",
|
|
"id": "c4633c670ce48202",
|
|
"outputs": [],
|
|
"execution_count": null
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"id": "ae19abe6-37f7-4a13-9659-e60c06e217e0",
|
|
"metadata": {},
|
|
"source": [
|
|
"df = pd.DataFrame(excel_out)\n",
|
|
"df.to_excel(\"output.xlsx\", index=False)"
|
|
],
|
|
"outputs": [],
|
|
"execution_count": null
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"id": "5f9cf122-54db-4377-99fa-57a668729deb",
|
|
"metadata": {},
|
|
"source": "print(difflens_lines[11],difflens_lines[12])",
|
|
"outputs": [],
|
|
"execution_count": null
|
|
},
|
|
{
|
|
"metadata": {},
|
|
"cell_type": "code",
|
|
"source": [
|
|
"print(excel_out['speedups'][11])\n",
|
|
"print(excel_out['speedups'][12])"
|
|
],
|
|
"id": "cef80d598f330dca",
|
|
"outputs": [],
|
|
"execution_count": null
|
|
},
|
|
{
|
|
"metadata": {},
|
|
"cell_type": "code",
|
|
"source": "",
|
|
"id": "f2001959c761c8e0",
|
|
"outputs": [],
|
|
"execution_count": null
|
|
}
|
|
],
|
|
"metadata": {
|
|
"kernelspec": {
|
|
"display_name": "Python 3 (ipykernel)",
|
|
"language": "python",
|
|
"name": "python3"
|
|
},
|
|
"language_info": {
|
|
"codemirror_mode": {
|
|
"name": "ipython",
|
|
"version": 3
|
|
},
|
|
"file_extension": ".py",
|
|
"mimetype": "text/x-python",
|
|
"name": "python",
|
|
"nbconvert_exporter": "python",
|
|
"pygments_lexer": "ipython3",
|
|
"version": "3.12.10"
|
|
}
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 5
|
|
}
|