methods/ALGO/count.py

import sys
sys.path.append("/home/i-luoxianzhen/data/TestCase-Gen/methods/utils")
from dataset_all import get_datasets_by_name
from config import cfg
import os

dataset = get_datasets_by_name("ours")
total_wrong_code = 0
empty = 0

for item in dataset:
    tcb_id = item['tcb_id']
    total_wrong_code += len(item['wrong_code'])
    if not os.path.exists(cfg.file_algo.format(tcb_id)):
        empty += len(item['wrong_code'])

print(f"empty {(empty / total_wrong_code) * 100:.2f}%")

from config import cfg
import sys
sys.path.append("/home/i-luoxianzhen/data/TestCase-Gen/methods/utils")
from dataset_all import get_datasets_by_name

def see_all():
    all_count = 0
    for batch in range(8):
        al_dataset = get_datasets_by_name("ours")
        start_pos = (batch) * 100
        end_pos = (batch + 1) * 100 if (batch + 1) * 100 <= len(al_dataset) else len(al_dataset)
        al_dataset = al_dataset[start_pos: end_pos]
        import os
        total_count = 0
        for item in al_dataset:
            if os.path.exists(cfg.file_algo.format(item['tcb_id'])):
                total_count += 1
                all_count += 1

        print(total_count)

    print(f"共计：{all_count}")

def find_index(batch, tcb_id):
    al_dataset = get_datasets_by_name(cfg.dataset_name)
    # start_pos = (batch) * 100
    # end_pos = (batch + 1) * 100 if (batch + 1) * 100 <= len(al_dataset) else len(al_dataset)
    # al_dataset = al_dataset[start_pos: end_pos]
    import os
    total_count = 0
    for item in al_dataset:
        if tcb_id == item['tcb_id']:
            break
        total_count += 1
    print(total_count)

print("执行进度")
see_all()


def find_unexist():
    all_count = 0
    for batch in range(8):
        al_dataset = get_datasets_by_name(cfg.dataset_name)
        start_pos = (batch) * 100
        end_pos = (batch + 1) * 100 if (batch + 1) * 100 <= len(al_dataset) else len(al_dataset)
        al_dataset = al_dataset[start_pos: end_pos]
        import os
        total_count = 0
        for item in al_dataset:
            if not os.path.exists(cfg.file_algo.format(item['tcb_id'])):
                print(item['tcb_id'])
                all_count += 1

    print(f"共计：{all_count}")     

# tcb = "毒瘤题加强版"
# print(f"寻找idx {tcb}")
# find_index(7, tcb)
# find_unexist()

import json

# 读取多个jsonl文件并合并为一个数组
def read_jsonl_files():
    all_data = []
    for i in range(8):
        with open(f"/home/i-luoxianzhen/data/TestCase-Gen/saved_tests/predo/test_pass_rate-{i}.json", 'r', encoding='utf-8') as f:
            for line in f:
                all_data.append(json.loads(line))  # 将每一行解析为一个字典并添加到all_data列表中
    return all_data

# 计算 right_num != 0 的占比
def calculate_right_num_percentage(data):
    count_non_zero = sum(1 for item in data if item.get('right_num', 0) != 0)
    total_count = len(data)
    return count_non_zero / total_count if total_count > 0 else 0


# # 读取文件并合并
# all_data = read_jsonl_files()

# # 计算 right_num != 0 的占比
# percentage = calculate_right_num_percentage(all_data)

# print(f"The percentage of items with right_num != 0 is: {percentage:.2%}")