如何在 Python 中隔离 ANTLR 解析器的测试用例？

如何解决如何在 Python 中隔离 ANTLR 解析器的测试用例？

我想使用 Python 和一些测试用例来测试我的 ANTLR 解析器 antlr v4 运行时。我遇到的问题是关于隔离测试用例。在每个测试用例中，应该解析另一个文件。但解析器（或流阅读器——我不知道）似乎连接到目前为止读取的文件。所以在测试用例n中，解析结果是案例 n-1 和案例 n 的结果的串联（并且递归地回到第一个测试用例）。

这是一个最小的工作示例：

文件Minimal.g4：

grammar Minimal ;

WS : [ \t\n\r\f] ;

WORD : (~ ([ \n\r\t\f]) )+ ;

text : ( token | WS+ )* ;

token : WORD ;

文件minimal/minimal.py：

from antlr4 import *
from minimal.MinimalListener import MinimalListener

class MinimalListListener(MinimalListener):

    tokens = []

    def exitToken(self,ctx):
        self.tokens.append(ctx.WORD().getText())

文件test_minimal.py：

import unittest
import os
from antlr4 import *
from minimal.MinimalListener import MinimalListener
from minimal.MinimalLexer import MinimalLexer
from minimal.MinimalParser import MinimalParser

from minimal.minimal import MinimalListListener


class MinimalTstBase(unittest.TestCase):

    fname = None
    
    def setUp(self):
        self.parsed = []
        input_stream = FileStream(os.path.join("samples",self.fname),encoding="utf-8")
        lexer = MinimalLexer(input_stream)
        stream = CommonTokenStream(lexer)
        parser = MinimalParser(stream)
        tree = parser.text()    # start rule
        printer = MinimalListListener()
        walker = ParseTreeWalker()
        walker.walk(printer,tree)
        self.parsed = printer.tokens # get result stored in the listener

    def tearDown(self):
        self.parsed = []
        
class TestCite1(MinimalTstBase):

    fname = "cite1.txt"

    def test_parsed(self):
        self.assertEqual(
            self.parsed,["A","B"])

class TestCite2(MinimalTstBase):

    fname = "cite2.txt"

    def test_parsed(self):
        self.assertEqual(
            self.parsed,['c','d'])

文件samples/cite1.txt：

A B

文件samples/cite2.txt：

c d

File minimal/__init__.py 只是一个空文件。文件 setup.py 只是样板文件：

from setuptools import find_packages,setup

setup(
    name = "minimal-antlr-testing-example",version = "0.0.1",description = "Minimal testing example",classifiers = [
        "Programming Language :: Python :: 3","License :: OSI Approved :: MIT License","Operating System :: OS Independent",],packages = find_packages(where="."),python_requires = ">=3.6",install_requires = [
        "wheel","setuptools","antlr4-python3-runtime==4.9.1"
        ],test_suite = 'minimal',)

文件requirements.txt：

wheel==0.36.2
antlr4-python3-runtime==4.9.1

运行后

java -cp antlr-4.9.1-complete.jar org.antlr.v4.Tool -o minimal/ \
-Xexact-output-dir -Dlanguage="python3" Minimal.g4

有以下文件：

./minimal/Minimal.tokens
./minimal/MinimalListener.py
./minimal/MinimalParser.py
./minimal/Minimal.interp
./minimal/MinimalLexer.tokens
./minimal/MinimalLexer.py
./minimal/MinimalLexer.interp
./minimal/test_minimal.py
./minimal/minimal.py
./minimal/__init__.py
./samples/cite2.txt
./samples/cite1.txt
./Minimal.g4
./requirements.txt
./setup.py

安装和运行测试：

python3 -m venv env
source env/bin/activate
python -m pip install -r requirements.txt
python setup.py bdist_wheel
python setup.py test

...

test_parsed (minimal.test_minimal.TestCite1) ... ok
test_parsed (minimal.test_minimal.TestCite2) ... FAIL

======================================================================
FAIL: test_parsed (minimal.test_minimal.TestCite2)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/clueck/src/minimal/antlr/minimal/test_minimal.py",line 46,in test_parsed
    ['c','d'])
AssertionError: Lists differ: ['A','B','c','d'] != ['c','d']

First differing element 0:
'A'
'c'

First list contains 2 additional elements.
First extra element 2:
'c'

- ['A','d']
+ ['c','d']

在运行单元测试时如何隔离解析器运行？

亲切的问候

解决方法

class MinimalListListener(MinimalListener):

    tokens = []

这将创建一个属于 MinimalListListener 类的列表，该列表在该类的所有实例之间共享。因此，如果您创建该类的两个实例，这两个实例都将元素添加到列表中，则列表最终将包含来自这两个实例的元素。

要在 Python 中创建特定于每个实例的变量，请将变量设置在 __init__ 中，而不是类主体：

class MinimalListListener(MinimalListener):
    def __init__(self):
        self.tokens = []