python3中调用的进程错误后的子进程输出

python3中调用的进程错误后的子进程输出,python,python-3.x,subprocess,Python,Python 3.x,Subprocess,我正在尝试使用子流程从Python中运行bandit。我有一些这样的代码: import subprocess command = ['bandit', '-r', 'goodcode'] output = subprocess.check_output(command) print(output.decode()) Run started:2016-09-27 10:37:17.567678 Test results: No issues identified. Cod

我正在尝试使用
子流程从Python中运行
bandit
。我有一些这样的代码:

import subprocess

command = ['bandit', '-r', 'goodcode']
output = subprocess.check_output(command)
print(output.decode())
Run started:2016-09-27 10:37:17.567678

Test results:
        No issues identified.

Code scanned:
        Total lines of code: 940
        Total lines skipped (#nosec): 0

Run metrics:
        Total issues (by severity):
                Undefined: 0
                Low: 0
                Medium: 0
                High: 0
        Total issues (by confidence):
                Undefined: 0
                Low: 0
                Medium: 0
                High: 0
Files skipped (0):
这很好,并提供如下字符串输出:

import subprocess

command = ['bandit', '-r', 'goodcode']
output = subprocess.check_output(command)
print(output.decode())
Run started:2016-09-27 10:37:17.567678

Test results:
        No issues identified.

Code scanned:
        Total lines of code: 940
        Total lines skipped (#nosec): 0

Run metrics:
        Total issues (by severity):
                Undefined: 0
                Low: 0
                Medium: 0
                High: 0
        Total issues (by confidence):
                Undefined: 0
                Low: 0
                Medium: 0
                High: 0
Files skipped (0):
…但是,当我在
bandit
返回一些错误的目录上运行它时,
bandit
进程本身返回1。因此,我必须捕获被调用的
进程错误
,如下所示:

import subprocess

command = ['bandit', '-r', 'badcode']

try:
    output = subprocess.check_output(command)
except subprocess.CalledProcessError as e:
    output = e.output

print(output.decode())
…这给了我以下结果:

b"Run started:2016-09-27 10:42:26.616123\n\nTest results:\n>> Issue: [B110:try_except_pass] Try, Except, Pass detected.\n   Severity: Low   Confidence: High\n   Location: badcode/conf/settings_development.py:93\n92\t    from .settings_local import *\n93\texcept:\n94\t    pass\n\n--------------------------------------------------\n>> Issue: [B105:hardcoded_password_string] Possible hardcoded password: 'password'\n   Severity: Low   Confidence: Medium\n   Location: badcode/frontend/tests/test_views.py:21\n20\t        form['username'] = self.user.username\n21\t        form['password'] = 'password'\n22\t\n\n--------------------------------------------------\n>> Issue: [B105:hardcoded_password_string] Possible hardcoded password: 'password'\n   Severity: Low   Confidence: Medium\n   Location: badcode/frontend/tests/test_views.py:35\n34\t        form['username'] = self.user.username\n35\t        form['password'] = 'password'\n36\t\n\n--------------------------------------------------\n>> Issue: [B110:try_except_pass] Try, Except, Pass detected.\n   Severity: Low   Confidence: High\n   Location: badcode/reasons/views.py:234\n233\t                nr = subject.number\n234\t            except:\n235\t                pass\n\n--------------------------------------------------\n>> Issue: [B110:try_except_pass] Try, Except, Pass detected.\n   Severity: Low   Confidence: High\n   Location: badcode/reasons/views.py:277\n276\t                nr = event.number\n277\t            except:\n278\t                pass\n\n--------------------------------------------------\n>> Issue: [B110:try_except_pass] Try, Except, Pass detected.\n   Severity: Low   Confidence: High\n   Location: badcode/retention/migrations/0010_auto_20160527_1603.py:13\n12\t            retention.save()\n13\t        except:\n14\t            pass\n\n--------------------------------------------------\n>> Issue: [B110:try_except_pass] Try, Except, Pass detected.\n   Severity: Low   Confidence: High\n   Location: badcode/retention/migrations/0015_auto_20160623_1051.py:13\n12\t            retention.save()\n13\t        except:\n14\t            pass\n\n--------------------------------------------------\n>> Issue: [B108:hardcoded_tmp_directory] Probable insecure usage of temp file/directory.\n   Severity: Medium   Confidence: Medium\n   Location: badcode/utils/views.py:322\n321\t        css = '{}/static/badcode/css/screen.css'.format(settings.ROOT_DIR)\n322\t        location = '/tmp/{}.pdf'.format(filename)\n323\t\n\n--------------------------------------------------\n\nCode scanned:\n\tTotal lines of code: 15287\n\tTotal lines skipped (#nosec): 0\n\nRun metrics:\n\tTotal issues (by severity):\n\t\tUndefined: 0.0\n\t\tLow: 7.0\n\t\tMedium: 1.0\n\t\tHigh: 0.0\n\tTotal issues (by confidence):\n\t\tUndefined: 0.0\n\t\tLow: 0.0\n\t\tMedium: 3.0\n\t\tHigh: 5.0\nFiles skipped (0):\n"

请注意,
b”“
位于字符串内部,因此
output[0]='b'
output[1]='''
。这是为什么?如果进程返回0,那么
e.output
不应该将输出编码为与
output
相同吗?

似乎您在
decode>之前已经查看了
output
变量
方法在第二种情况下被调用。
输出
在这两种情况下引用相同的bytestring

>>> import subprocess, sys
>>> output1 = subprocess.check_output([sys.executable, '-c', 'print("abc")'])
>>> try:
...   subprocess.check_output([sys.executable, '-c', 
...                            'print("abc"); import sys;sys.exit(1)'])
... except subprocess.CalledProcessError as e:
...   output2 = e.output
... 
>>> output1 == output2
True
如果非零退出状态在您的情况下不是错误,那么您可以使用在这种情况下不会引发CalledProcessError的函数:

#!/usr/bin/env python3
from subprocess import run, PIPE

finished_process = run(command, stdout=PIPE, universal_newlines=True)
print("{p.returncode}, {p.stdout}".format(p=finished_process))
# e.g.: 1, abc
universal\u newlines=True
是此处“启用文本模式”的一种模糊拼写