Python 实战：自动化测试日志转 XML 系统实践

1
2
3
4
5
6

┌─────────────┐    ┌──────────────┐    ┌──────────────┐
│  原始 Log   │ -> │  解析转换    │ -> │  XML 输出     │
│  (测试设备)  │    │  (Python)    │    │  (EDA 系统)   │
└─────────────┘    └──────────────┘    └──────────────┘
       ↓                   ↓                   ↓
  产线测试           正则提取 + 逻辑处理      数据入库分析

1
2
3
4

log_to_xml/
├── py_test_modify_V03.py    # 核心转换脚本
├── 实现方法.md               # 实施文档
└── README.md                 # 项目说明

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11

# 提取测试状态
status_pattern = r"runner\s+-\s+INFO\s+-\s+Status:\s+(PASS|FAIL)"

# 提取序列号
serial_number_pattern = r"runner\s+-\s+INFO\s+-\s+Serial\s+Number:\s+(\S+)"

# 提取测试数据
test_data_pattern = r"runner\s+-\s+INFO\s+-\s+Test\s+(\d+)\s+-\s+([\w\-_.]+)\s+\(.*?\)\s+\.\.\.\s+(pass|fail)\s+\(([\dhms]+)\)"

# 提取测量数据
measurement_pattern = r"runner\s+-\s+INFO\s+-\s+teststep:(\S+)\s+testname:(.*?)\s+value:(\S+)\s+unit:(\S+)\s+judgetype:(\S+)\s+lowlimit:(\S+)\s+uplimit:(\S+)\s+datatype:(\S+)"

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12

def determine_model_name(region_str):
    """根据区域字符串确定型号名称"""
    if 'nperear' in region_str:
        return 'NPE_REAR'
    elif 'npefront' in region_str:
        return 'NPE_FRONT'
    elif 'niorear' in region_str or 'fct_rearfct' in region_str:
        return 'NPD_REAR'
    elif 'niofront' in region_str or 'fct_frontfct' in region_str:
        return 'NPD_FRONT'
    else:
        return 'NA'

 1
 2
 3
 4
 5
 6
 7
 8
 9
10

def escape_xml_special_chars(s):
    """转义 XML 特殊字符"""
    special_chars = {
        '&': '&',
        '<': '&lt;',
        '>': '&gt;',
        '"': '&quot;',
        "'": '&apos;'
    }
    return ''.join(special_chars.get(char, char) for char in s)

1
2
3
4
5
6

def convert_timestamp(timestamp_str):
    """转换时间戳为带时区的 ISO 格式"""
    dt = datetime.strptime(timestamp_str, "%Y-%m-%d %H:%M:%S,%f")
    tz = timezone(timedelta(hours=8))  # 北京时间
    dt_with_tz = dt.replace(tzinfo=tz)
    return dt_with_tz.isoformat(timespec='milliseconds')

1
2
3
4
5

# ❌ 错误做法：直接替换冒号
measurement_name.replace(':', '_')

# ✅ 正确做法：转义 XML 特殊字符
escape_xml_special_chars(measurement_name)

1
2
3
4
5
6
7
8
9

# 示例 "20241105082202245147" 转换为 "2024-11-05_08-22-02"
def convert_time_format(time_str):
    year = time_str[:4]
    month = time_str[4:6]
    day = time_str[6:8]
    hour = time_str[8:10]
    minute = time_str[10:12]
    second = time_str[12:14]
    return f"{year}-{month}-{day}_{hour}-{minute}-{second}"

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20

# 完整判断逻辑
if units in ['boolean', 'Boolean', 'boolen', 'string', '']:
    comp_type = 'LOG'
elif units in ['unknown', 'g', 'A', 'ma', 'v', 'Hz', '%', 's', 'digital']:
    try:
        result_val = float(result)
        lowlimit_val = float(lowlimit) if lowlimit else None
        uplimit_val = float(uplimit) if uplimit else None
        
        if lowlimit_val is not None and uplimit_val is not None:
            if lowlimit_val == uplimit_val:
                comp_type = 'EQ'  # 上下限相等
            else:
                comp_type = 'GELE'  # 上下限不相等
        else:
            comp_type = 'NA'
    except ValueError:
        comp_type = 'NA'
else:
    comp_type = 'NA'

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11

# ❌ 错误：只匹配小写
test_time_pattern = r"\.\.\.\s+(pass|fail)\s+"

# ✅ 正确：匹配大小写
test_time_pattern = r"\.\.\.\s+(pass|FAIL|skip)\s+"

# 提取测试状态
if status_keyword == "pass":
    test_status = "Pass"
elif status_keyword == "FAIL":  # 注意大写
    test_status = "Fail"

1
2
3
4
5
6
7
8
9

# 正则表达式
errorcode_pattern = r'find errorcode\..*?errorcode:(\w+)'
errormessage_pattern = r'ERROR - Failing test due to: (.*?):'
errordetails_pattern = r'ERROR - Failing test due to: (.*)'
errortestname_pattern = r'INFO - Testname: (.*)'
errorsuptestname_pattern = r'Failing test due to:(.*?)\n.*? - +INFO - (.*?)\(.*?\)'

# 组合 ErrorFullTestName
errorfulltestname = convert_suptestname_string(errorsuptestname).strip() + "_" + errortestname

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

# Factory 版本
version_pattern = r"Build at\s+:\s+(\w+\s+\d+\s+\d{4},\s+\d{2}:\d{2}:\d{2})"

# Shipping 版本（需要从特定日志段提取）
def get_ship_version(log_content):
    lines = log_content.splitlines()
    for i, line in enumerate(lines):
        if "GUI Progress:  check Ship switch version" in line:
            for j in range(i + 1, i + 21):
                if "Alive, info" in lines[j]:
                    match = re.search(r'Alive, info\s*:(.*)', lines[j])
                    if match:
                        return match.group(1).strip()
    return None

# 合并逻辑
if factory_version is None and shipping_version is None:
    version_name = "NA"
elif factory_version and shipping_version is None:
    version_name = factory_version
elif factory_version is None and shipping_version:
    version_name = shipping_version
else:
    version_name = factory_version + ";" + shipping_version

1
2
3
4
5
6

def transform_string(input_string):
    """将 StationId 统一转换为 00001~00009 格式"""
    for i in range(1, 10):
        if f"0{i}" in input_string:
            return f"{i:05d}"  # 格式化为 5 位数字
    return "NA"

1
2
3
4
5

# 原始名称
'H Bridge J2600-1-25-27 forward & reverse'

# 转义后
'H Bridge J2600-1-25-27 forward & reverse'

1
2
3
4
5

def convert_timestamp(timestamp_str):
    dt = datetime.strptime(timestamp_str, "%Y-%m-%d %H:%M:%S,%f")
    tz = timezone(timedelta(hours=8))  # 北京时间
    dt_with_tz = dt.replace(tzinfo=tz)
    return dt_with_tz.isoformat(timespec='milliseconds')

1
2
3
4
5

# 上下限为空时的处理
if not lowlimit and not uplimit:
    # TDD 确认：没有影响，Status 设为 Pass
    status = 'Pass'
    comp_type = 'NA'

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38

<?xml version="1.0" encoding="UTF-8"?>
<TestResults>
  <OverallResult>
    <Result>Pass</Result>
    <ErrorCode></ErrorCode>
    <LogErrorMessage></LogErrorMessage>
    <ErrorTestName></ErrorTestName>
    <ErrorFullTestName></ErrorFullTestName>
    <ErrorDetails></ErrorDetails>
  </OverallResult>
  
  <UnitSerialNumber>NRNWXPH077C482</UnitSerialNumber>
  <StartDate>2024-12-09T13:34:30.276+08:00</StartDate>
  <StopDate>2024-12-09T13:36:45.123+08:00</StopDate>
  <TestCycleTime>2h15m30s</TestCycleTime>
  <ModelName>NPD_REAR</ModelName>
  <StationType>TT</StationType>
  <StationId>00003</StationId>
  <StationLine>FA11</StationLine>
  <DutSwVersion>Build at : Nov 15 2024, 10:30:45</DutSwVersion>
  
  <TestDatas>
    <TestData Name="Test ID 001 - H Bridge Test" Status="Pass" TestTime="45">
      <Measurements>
        <MeasurementData 
          Name="H Bridge J2600-1-25-27 forward" 
          Status="Pass" 
          Result="0.0012" 
          LowerLimit="-13.5" 
          UpperLimit="-12" 
          Units="v" 
          CompType="GELE"
          LogCount="0" 
          LogTime="2024-11-05T03:57:12.217+08:00"/>
      </Measurements>
    </TestData>
  </TestDatas>
</TestResults>

1
2
3
4

# 关键步骤都添加日志
print(f"Found {len(test_matches)} Test IDs.")
print(f"Processing: {measurement_name}, Result: {result}")
print(f"Execution time: {execution_time:.2f} seconds")

1
2
3
4
5
6

# 主脚本使用多进程并行处理
from multiprocessing import Pool

def process_batch(log_files):
    with Pool(processes=4) as pool:
        pool.map(process_log_file, log_files)

1
2
3
4

start_time = time.time()
# ... 处理逻辑 ...
end_time = time.time()
print(f"Execution time: {end_time - start_time:.2f} seconds")