testing.py

"""testing.py - test support for DHParser based grammars and compilers

Copyright 2016  by Eckhart Arnold (arnold@badw.de)
                Bavarian Academy of Sciences an Humanities (badw.de)

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied.  See the License for the specific language governing
permissions and limitations under the License.
"""
import collections
import configparser
import copy
import fnmatch
import inspect
import json
import os
try:
    import regex as re
except ImportError:
    import re

from DHParser.toolkit import is_logging, clear_logs, error_messages
from DHParser.syntaxtree import mock_syntax_tree, flatten_sxpr

__all__ = ('unit_from_configfile',
           'unit_from_json',
           'unit_from_file',
           'get_report',
           'grammar_unit',
           'grammar_suite',
           'runner')

UNIT_STAGES = {'match', 'fail', 'ast', 'cst', '__ast__', '__cst__'}


def unit_from_configfile(config_filename):
    """
    Reads a grammar unit test from a config file.
    """
    cfg = configparser.ConfigParser(interpolation=None)
    cfg.read(config_filename, encoding="utf8")
    OD = collections.OrderedDict
    unit = OD()
    for section in cfg.sections():
        symbol, stage = section.split(':')
        if stage not in UNIT_STAGES:
            if symbol in UNIT_STAGES:
                symbol, stage = stage, symbol
            else:
                raise ValueError('Test stage %s not in: ' % (stage, str(UNIT_STAGES)))
        for testkey, testcode in cfg[section].items():
            if testcode[:3] + testcode[-3:] in {"''''''", '""""""'}:
                testcode = testcode[3:-3]
                # testcode = testcode.replace('\\#', '#')
                testcode = re.sub(r'(?<!\\)\\#', '#', testcode).replace('\\\\', '\\')
            elif testcode[:1] + testcode[-1:] in {"''", '""'}:
                testcode = testcode[1:-1]
            unit.setdefault(symbol, OD()).setdefault(stage, OD())[testkey] = testcode
    # print(json.dumps(unit, sort_keys=True, indent=4))
    return unit


def unit_from_json(json_filename):
    """
    Reads a grammar unit test from a json file.
    """
    with open(json_filename, 'r', encoding='utf8') as f:
        unit = json.load(f)
    for symbol in unit:
        for stage in unit[symbol]:
            if stage not in UNIT_STAGES:
                raise ValueError('Test stage %s not in: ' % (stage, str(UNIT_STAGES)))
    return unit

# TODO: add support for yaml, cson, toml


def unit_from_file(filename):
    """
    Reads a grammar unit test from a file. The format of the file is
    determined by the ending of its name.
    """
    if filename.endswith(".json"):
        return unit_from_json(filename)
    elif filename.endswith(".ini"):
        return unit_from_configfile(filename)
    else:
        raise ValueError("Unknown unit test file type: " + filename[filename.rfind('.'):])


def get_report(test_unit):
    """
    Returns a text-report of the results of a grammar unit test.
    """
    def indent(txt):
        lines = txt.split('\n')
        lines[0] = '    ' + lines[0]
        return "\n    ".join(lines)
    report = []
    for parser_name, tests in test_unit.items():
        heading = 'Test of parser: "%s"' % parser_name
        report.append('\n\n%s\n%s\n' % (heading, '=' * len(heading)))
        for test_name, test_code in tests.get('match', dict()).items():
            heading = 'Match-test "%s"' % test_name
            report.append('\n%s\n%s\n' % (heading, '-' * len(heading)))
            report.append('### Test-code:')
            report.append(indent(test_code))
            error = tests.get('__err__', {}).get(test_name, "")
            if error:
                report.append('\n### Error:')
                report.append(error)
            ast = tests.get('__ast__', {}).get(test_name, None)
            cst = tests.get('__cst__', {}).get(test_name, None)
            if cst and (not ast or cst == ast):
                report.append('\n### CST')
                report.append(indent(cst.as_sxpr()))
            elif ast:
                report.append('\n### AST')
                report.append(indent(ast.as_sxpr()))
    return '\n'.join(report)


def grammar_unit(test_unit, parser_factory, transformer_factory, report=True, verbose=False):
    """
    Unit tests for a grammar-parser and ast transformations.
    """
    if isinstance(test_unit, str):
        unit_dir, unit_name = os.path.split(os.path.splitext(test_unit)[0])
        test_unit = unit_from_file(test_unit)
    else:
        unit_dir = ""
        unit_name = str(id(test_unit))
    if verbose:
        print("\nUnit: " + unit_name)
    errata = []
    parser = parser_factory()
    transform = transformer_factory()
    for parser_name, tests in test_unit.items():
        assert set(tests.keys()).issubset(UNIT_STAGES)
        if verbose:
            print('  Match-Tests for parser "' + parser_name + '"')
        for test_name, test_code in tests.get('match', dict()).items():
            if verbose:
                infostr = '    match-test "' + test_name + '" ... '
                errflag = len(errata)
            cst = parser(test_code, parser_name)
            cst.log("match_%s_%s.cst" % (parser_name, test_name))
            tests.setdefault('__cst__', {})[test_name] = cst
            if "ast" in tests or report:
                ast = copy.deepcopy(cst)
                transform(ast)
                tests.setdefault('__ast__', {})[test_name] = ast
                ast.log("match_%s_%s.ast" % (parser_name, test_name))
            if cst.error_flag:
                errata.append('Match test "%s" for parser "%s" failed:\n\tExpr.:  %s\n\n\t%s\n\n' %
                              (test_name, parser_name, '\n\t'.join(test_code.split('\n')),
                               '\n\t'.join(m.replace('\n', '\n\t\t') for m in
                                           error_messages(test_code, cst.collect_errors()))))
                tests.setdefault('__err__', {})[test_name] = errata[-1]
                # write parsing-history log only in case of failure!
                parser.log_parsing_history__("match_%s_%s.log" % (parser_name, test_name))
            elif "cst" in tests and mock_syntax_tree(tests["cst"][test_name]) != cst:
                    errata.append('Concrete syntax tree test "%s" for parser "%s" failed:\n%s' %
                                  (test_name, parser_name, cst.as_sxpr()))
            elif "ast" in tests:
                compare = mock_syntax_tree(tests["ast"][test_name])
                if compare != ast:
                    errata.append('Abstract syntax tree test "%s" for parser "%s" failed:'
                                  '\n\tExpr.:     %s\n\tExpected:  %s\n\tReceived:  %s'
                                  % (test_name, parser_name, '\n\t'.join(test_code.split('\n')),
                                     flatten_sxpr(compare.as_sxpr()),
                                     flatten_sxpr(ast.as_sxpr())))
                    tests.setdefault('__err__', {})[test_name] = errata[-1]
            if verbose:
                print(infostr + ("OK" if len(errata) == errflag else "FAIL"))

        if verbose and 'fail' in tests:
            print('  Fail-Tests for parser "' + parser_name + '"')
        for test_name, test_code in tests.get('fail', dict()).items():
            if verbose:
                infostr = '    fail-test  "' + test_name + '" ... '
                errflag = len(errata)
            cst = parser(test_code, parser_name)
            if not cst.error_flag:
                errata.append('Fail test "%s" for parser "%s" yields match instead of '
                              'expected failure!' % (test_name, parser_name))
                tests.setdefault('__err__', {})[test_name] = errata[-1]
                # write parsing-history log only in case of test-failure
                parser.log_parsing_history__("fail_%s_%s.log" % (parser_name, test_name))
            if verbose:
                print(infostr + ("OK" if len(errata) == errflag else "FAIL"))

    # write test-report
    if report:
        report_dir = "REPORT"
        if not os.path.exists(report_dir):
            os.mkdir(report_dir)
        with open(os.path.join(report_dir, unit_name + '.md'), 'w', encoding='utf8') as f:
            f.write(get_report(test_unit))

    return errata


def grammar_suite(directory, parser_factory, transformer_factory,
                  fn_patterns=['*test*'],
                  ignore_unknown_filetypes=False,
                  report=True, verbose=True):
    """
    Runs all grammar unit tests in a directory. A file is considered a test
    unit, if it has the word "test" in its name.
    """
    if not isinstance(fn_patterns, collections.abc.Collection):
        fn_patterns = [fn_patterns]
    all_errors = collections.OrderedDict()
    if verbose:
        print("\nScanning test-directory: " + directory)
    save_cwd = os.getcwd()
    os.chdir(directory)
    if is_logging():    clear_logs()
    for filename in sorted(os.listdir()):
        if any(fnmatch.fnmatch(filename, pattern) for pattern in fn_patterns):
            try:
                if verbose:
                    print("\nRunning grammar tests from: " + filename)
                errata = grammar_unit(filename, parser_factory,
                                      transformer_factory, report, verbose)
                if errata:
                    all_errors[filename] = errata
            except ValueError as e:
                if not ignore_unknown_filetypes or str(e).find("Unknown") < 0:
                    raise e
    os.chdir(save_cwd)
    error_report = []
    if all_errors:
        for filename in all_errors:
            error_report.append('Errors found by unit test "%s":' % filename)
            for error in all_errors[filename]:
                error_report.append('\t' + '\n\t'.join(error.split('\n')))
    if error_report:
        return ('Test suite "%s" revealed some errors:\n\n' % directory) + '\n'.join(error_report)
    return ''


def runner(tests, namespace):
    """
    Runs all or some selected Python unit tests found in the
    namespace. To run all tests in a module, call 
    ``runner("", globals())`` from within that module.

    Args:
        tests: Either a string or a list of strings that contains the
            names of test or test classes. Each test and, in the case
            of a test class, all tests within the test class will be
            run.
        namespace: The namespace for running the test, usually 
            ``globals()`` should be used.
            
    Example:
        class TestSomething()
            def setup(self):
                pass
            def teardown(self):
                pass
            def test_something(self):
                pass
                
        if __name__ == "__main__":
            from DHParser.testing import runner
            runner("", globals())        
    """
    def instantiate(cls_name):
        exec("obj = " + cls_name + "()", namespace)
        obj = namespace["obj"]
        if "setup" in dir(obj):
            obj.setup()
        return obj

    if tests:
        if isinstance(tests, str):
            tests = tests.split(" ")
    else:
        # collect all test classes, in case no methods or classes have been passed explicitly
        tests = []
        for name in namespace.keys():
            if name.lower().startswith('test') and inspect.isclass(namespace[name]):
                tests.append(name)

    obj = None
    for test in tests:
        try:
            if test.find('.') >= 0:
                cls_name, method_name = test.split('.')
                obj = instantiate(cls_name)
                print("Running " + cls_name + "." + method_name)
                exec('obj.' + method_name + '()')
            else:
                obj = instantiate(test)
                for name in dir(obj):
                    if name.lower().startswith("test"):
                        print("Running " + test + "." + name)
                        exec('obj.' + name + '()')
        finally:
            if "teardown" in dir(obj):
                obj.teardown()