This repository contains the testrunner for the Kantan Programming Language
python3 src/main.py <path-to-compiler> <path-to-test-suite> [--valgrind=True]
- kantan-test scans the directory for .kan and .py files.
- files listed in kantan-test-ignore are filtered out
- a python object is created for each test case (.py file) which should extend
testcase.Code - the .kan files are compiled and the output is checked by the objects from the last step
- if there were errors those are printed to the console
- (optional) if the --valgrind option is set, memory leaks inside the compiler are listed separately
from typing import Optional
from testcase import Code
from errors import TestError, Severity
from output import KantanOutput
class Test(Code):
# overwrite this method of the Code class. The argument is the compiler output, which
# has already been parsed
def check(self, output: KantanOutput) -> Optional[TestError]:
"""Checks if the correct error is printed for struct-decl-access-error.kan"""
if len(output.errors) != 1:
msg = 'expected 1 error, but got {}'.format(len(output.errors))
return self.create_error(msg)
error = output.errors[0]
expected_rsn = "'struct declaration' cannot be accessed with '.' operator"
actual_rsn = error.reason
if expected_rsn not in actual_rsn:
msg = 'wrong reason, expected <{}>, but got {}'.format(expected_rsn, actual_rsn)
return self.create_error(msg)
if error.line != 5:
return self.create_error('wrong line')
if error.col != 1:
print(error.col)
return self.create_error('wrong column')
return None