testing.py 12.3 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
"""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.
"""
18
import collections
19
import configparser
20
import copy
Eckhart Arnold's avatar
Eckhart Arnold committed
21
import fnmatch
di68kap's avatar
di68kap committed
22
import inspect
23
24
import json
import os
25
26
27
28
try:
    import regex as re
except ImportError:
    import re
29

30
from DHParser.toolkit import is_logging, clear_logs
31
from DHParser.syntaxtree import mock_syntax_tree, flatten_sxpr
32
from DHParser.error import is_error, error_messages
33

34
__all__ = ('unit_from_configfile',
35
36
37
38
39
40
41
           'unit_from_json',
           'unit_from_file',
           'get_report',
           'grammar_unit',
           'grammar_suite',
           'runner')

42
UNIT_STAGES = {'match', 'fail', 'ast', 'cst', '__ast__', '__cst__'}
43
44


45
def unit_from_configfile(config_filename):
46
47
    """
    Reads a grammar unit test from a config file.
48
    """
49
    cfg = configparser.ConfigParser(interpolation=None)
di68kap's avatar
di68kap committed
50
    cfg.read(config_filename, encoding="utf8")
51
52
    OD = collections.OrderedDict
    unit = OD()
53
54
    for section in cfg.sections():
        symbol, stage = section.split(':')
55
56
        if stage not in UNIT_STAGES:
            if symbol in UNIT_STAGES:
57
58
                symbol, stage = stage, symbol
            else:
59
                raise ValueError('Test stage %s not in: ' % (stage, str(UNIT_STAGES)))
60
        for testkey, testcode in cfg[section].items():
61
62
            if testcode[:3] + testcode[-3:] in {"''''''", '""""""'}:
                testcode = testcode[3:-3]
di68kap's avatar
di68kap committed
63
64
                # testcode = testcode.replace('\\#', '#')
                testcode = re.sub(r'(?<!\\)\\#', '#', testcode).replace('\\\\', '\\')
65
66
            elif testcode[:1] + testcode[-1:] in {"''", '""'}:
                testcode = testcode[1:-1]
67
            unit.setdefault(symbol, OD()).setdefault(stage, OD())[testkey] = testcode
68
69
    # print(json.dumps(unit, sort_keys=True, indent=4))
    return unit
70

71

72
def unit_from_json(json_filename):
73
74
    """
    Reads a grammar unit test from a json file.
75
    """
di68kap's avatar
di68kap committed
76
    with open(json_filename, 'r', encoding='utf8') as f:
77
78
79
80
81
82
83
        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

84
# TODO: add support for yaml, cson, toml
85
86


87
def unit_from_file(filename):
88
89
    """
    Reads a grammar unit test from a file. The format of the file is
90
91
    determined by the ending of its name.
    """
92
93
94
95
    if filename.endswith(".json"):
        return unit_from_json(filename)
    elif filename.endswith(".ini"):
        return unit_from_configfile(filename)
96
    else:
97
        raise ValueError("Unknown unit test file type: " + filename[filename.rfind('.'):])
98
99


100
def get_report(test_unit):
101
102
    """
    Returns a text-report of the results of a grammar unit test.
103
    """
104
105
106
107
    def indent(txt):
        lines = txt.split('\n')
        lines[0] = '    ' + lines[0]
        return "\n    ".join(lines)
108
109
110
111
112
113
114
115
    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:')
116
            report.append(indent(test_code))
117
118
119
120
121
            error = tests.get('__err__', {}).get(test_name, "")
            if error:
                report.append('\n### Error:')
                report.append(error)
            ast = tests.get('__ast__', {}).get(test_name, None)
122
123
124
            cst = tests.get('__cst__', {}).get(test_name, None)
            if cst and (not ast or cst == ast):
                report.append('\n### CST')
125
                report.append(indent(cst.as_sxpr()))
126
            elif ast:
127
                report.append('\n### AST')
128
                report.append(indent(ast.as_sxpr()))
129
130
131
    return '\n'.join(report)


132
def grammar_unit(test_unit, parser_factory, transformer_factory, report=True, verbose=False):
133
134
    """
    Unit tests for a grammar-parser and ast transformations.
135
    """
136
    if isinstance(test_unit, str):
137
        unit_dir, unit_name = os.path.split(os.path.splitext(test_unit)[0])
138
        test_unit = unit_from_file(test_unit)
139
    else:
140
        unit_dir = ""
141
        unit_name = str(id(test_unit))
142
143
    if verbose:
        print("\nUnit: " + unit_name)
144
145
146
    errata = []
    parser = parser_factory()
    transform = transformer_factory()
147
148
    for parser_name, tests in test_unit.items():
        assert set(tests.keys()).issubset(UNIT_STAGES)
149
150
        if verbose:
            print('  Match-Tests for parser "' + parser_name + '"')
151
        for test_name, test_code in tests.get('match', dict()).items():
152
153
154
            if verbose:
                infostr = '    match-test "' + test_name + '" ... '
                errflag = len(errata)
155
            cst = parser(test_code, parser_name)
Eckhart Arnold's avatar
Eckhart Arnold committed
156
            cst.log("match_%s_%s.cst" % (parser_name, test_name))
157
            tests.setdefault('__cst__', {})[test_name] = cst
158
            if "ast" in tests or report:
159
160
161
                ast = copy.deepcopy(cst)
                transform(ast)
                tests.setdefault('__ast__', {})[test_name] = ast
Eckhart Arnold's avatar
Eckhart Arnold committed
162
                ast.log("match_%s_%s.ast" % (parser_name, test_name))
163
            if is_error(cst.error_flag):
Eckhart Arnold's avatar
Eckhart Arnold committed
164
                errata.append('Match test "%s" for parser "%s" failed:\n\tExpr.:  %s\n\n\t%s\n\n' %
165
                              (test_name, parser_name, '\n\t'.join(test_code.split('\n')),
166
167
                               '\n\t'.join(m.replace('\n', '\n\t\t') for m in
                                           error_messages(test_code, cst.collect_errors()))))
168
                tests.setdefault('__err__', {})[test_name] = errata[-1]
169
170
                # write parsing-history log only in case of failure!
                parser.log_parsing_history__("match_%s_%s.log" % (parser_name, test_name))
171
172
            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' %
173
                                  (test_name, parser_name, cst.as_sxpr()))
174
175
176
177
178
179
            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')),
Eckhart Arnold's avatar
Eckhart Arnold committed
180
181
                                     flatten_sxpr(compare.as_sxpr()),
                                     flatten_sxpr(ast.as_sxpr())))
182
                    tests.setdefault('__err__', {})[test_name] = errata[-1]
183
            if verbose:
184
                print(infostr + ("OK" if len(errata) == errflag else "FAIL"))
185

186
        if verbose and 'fail' in tests:
187
            print('  Fail-Tests for parser "' + parser_name + '"')
188
        for test_name, test_code in tests.get('fail', dict()).items():
189
190
191
            if verbose:
                infostr = '    fail-test  "' + test_name + '" ... '
                errflag = len(errata)
192
            cst = parser(test_code, parser_name)
193
            if not is_error(cst.error_flag):
194
195
                errata.append('Fail test "%s" for parser "%s" yields match instead of '
                              'expected failure!' % (test_name, parser_name))
196
                tests.setdefault('__err__', {})[test_name] = errata[-1]
197
198
                # write parsing-history log only in case of test-failure
                parser.log_parsing_history__("fail_%s_%s.log" % (parser_name, test_name))
199
            if verbose:
Eckhart Arnold's avatar
Eckhart Arnold committed
200
                print(infostr + ("OK" if len(errata) == errflag else "FAIL"))
201

202
203
    # write test-report
    if report:
204
        report_dir = "REPORT"
205
206
        if not os.path.exists(report_dir):
            os.mkdir(report_dir)
di68kap's avatar
di68kap committed
207
        with open(os.path.join(report_dir, unit_name + '.md'), 'w', encoding='utf8') as f:
208
            f.write(get_report(test_unit))
209

210
211
212
    return errata


Eckhart Arnold's avatar
Eckhart Arnold committed
213
214
215
216
def grammar_suite(directory, parser_factory, transformer_factory,
                  fn_patterns=['*test*'],
                  ignore_unknown_filetypes=False,
                  report=True, verbose=True):
217
218
    """
    Runs all grammar unit tests in a directory. A file is considered a test
219
220
    unit, if it has the word "test" in its name.
    """
221
    if not isinstance(fn_patterns, collections.abc.Iterable):
Eckhart Arnold's avatar
Eckhart Arnold committed
222
        fn_patterns = [fn_patterns]
223
    all_errors = collections.OrderedDict()
224
225
    if verbose:
        print("\nScanning test-directory: " + directory)
226
227
    save_cwd = os.getcwd()
    os.chdir(directory)
Eckhart Arnold's avatar
Eckhart Arnold committed
228
    if is_logging():    clear_logs()
229
    for filename in sorted(os.listdir()):
Eckhart Arnold's avatar
Eckhart Arnold committed
230
        if any(fnmatch.fnmatch(filename, pattern) for pattern in fn_patterns):
231
            try:
232
233
                if verbose:
                    print("\nRunning grammar tests from: " + filename)
234
235
                errata = grammar_unit(filename, parser_factory,
                                      transformer_factory, report, verbose)
236
237
238
                if errata:
                    all_errors[filename] = errata
            except ValueError as e:
239
                if not ignore_unknown_filetypes or str(e).find("Unknown") < 0:
240
                    raise e
241
    os.chdir(save_cwd)
242
243
244
245
246
247
248
    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:
249
        return ('Test suite "%s" revealed some errors:\n\n' % directory) + '\n'.join(error_report)
250
251
252
    return ''


253
def runner(tests, namespace):
254
255
    """
    Runs all or some selected Python unit tests found in the
256
257
    namespace. To run all tests in a module, call 
    ``runner("", globals())`` from within that module.
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276

    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__":
di68kap's avatar
di68kap committed
277
            from DHParser.testing import runner
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
            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()