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

Eckhart Arnold's avatar
Eckhart Arnold committed
30
from DHParser.toolkit import is_logging, clear_logs, error_messages
Eckhart Arnold's avatar
Eckhart Arnold committed
31
from DHParser.syntaxtree import mock_syntax_tree, flatten_sxpr
32

33
__all__ = ('unit_from_configfile',
34
35
36
37
38
39
40
           'unit_from_json',
           'unit_from_file',
           'get_report',
           'grammar_unit',
           'grammar_suite',
           'runner')

41
UNIT_STAGES = {'match', 'fail', 'ast', 'cst', '__ast__', '__cst__'}
42
43


44
def unit_from_configfile(config_filename):
45
46
    """
    Reads a grammar unit test from a config file.
47
    """
48
    cfg = configparser.ConfigParser(interpolation=None)
di68kap's avatar
di68kap committed
49
    cfg.read(config_filename, encoding="utf8")
50
51
    OD = collections.OrderedDict
    unit = OD()
52
53
    for section in cfg.sections():
        symbol, stage = section.split(':')
54
55
        if stage not in UNIT_STAGES:
            if symbol in UNIT_STAGES:
56
57
                symbol, stage = stage, symbol
            else:
58
                raise ValueError('Test stage %s not in: ' % (stage, str(UNIT_STAGES)))
59
        for testkey, testcode in cfg[section].items():
60
61
            if testcode[:3] + testcode[-3:] in {"''''''", '""""""'}:
                testcode = testcode[3:-3]
di68kap's avatar
di68kap committed
62
63
                # testcode = testcode.replace('\\#', '#')
                testcode = re.sub(r'(?<!\\)\\#', '#', testcode).replace('\\\\', '\\')
64
65
            elif testcode[:1] + testcode[-1:] in {"''", '""'}:
                testcode = testcode[1:-1]
66
            unit.setdefault(symbol, OD()).setdefault(stage, OD())[testkey] = testcode
67
68
    # print(json.dumps(unit, sort_keys=True, indent=4))
    return unit
69

70

71
def unit_from_json(json_filename):
72
73
    """
    Reads a grammar unit test from a json file.
74
    """
di68kap's avatar
di68kap committed
75
    with open(json_filename, 'r', encoding='utf8') as f:
76
77
78
79
80
81
82
        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

83
# TODO: add support for yaml, cson, toml
84
85


86
def unit_from_file(filename):
87
88
    """
    Reads a grammar unit test from a file. The format of the file is
89
90
    determined by the ending of its name.
    """
91
92
93
94
    if filename.endswith(".json"):
        return unit_from_json(filename)
    elif filename.endswith(".ini"):
        return unit_from_configfile(filename)
95
    else:
96
        raise ValueError("Unknown unit test file type: " + filename[filename.rfind('.'):])
97
98


99
def get_report(test_unit):
100
101
    """
    Returns a text-report of the results of a grammar unit test.
102
    """
103
104
105
106
    def indent(txt):
        lines = txt.split('\n')
        lines[0] = '    ' + lines[0]
        return "\n    ".join(lines)
107
108
109
110
111
112
113
114
    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:')
115
            report.append(indent(test_code))
116
117
118
119
120
            error = tests.get('__err__', {}).get(test_name, "")
            if error:
                report.append('\n### Error:')
                report.append(error)
            ast = tests.get('__ast__', {}).get(test_name, None)
121
122
123
            cst = tests.get('__cst__', {}).get(test_name, None)
            if cst and (not ast or cst == ast):
                report.append('\n### CST')
124
                report.append(indent(cst.as_sxpr()))
125
            elif ast:
126
                report.append('\n### AST')
127
                report.append(indent(ast.as_sxpr()))
128
129
130
    return '\n'.join(report)


131
def grammar_unit(test_unit, parser_factory, transformer_factory, report=True, verbose=False):
132
133
    """
    Unit tests for a grammar-parser and ast transformations.
134
    """
135
    if isinstance(test_unit, str):
136
        unit_dir, unit_name = os.path.split(os.path.splitext(test_unit)[0])
137
        test_unit = unit_from_file(test_unit)
138
    else:
139
        unit_dir = ""
140
        unit_name = str(id(test_unit))
141
142
    if verbose:
        print("\nUnit: " + unit_name)
143
144
145
    errata = []
    parser = parser_factory()
    transform = transformer_factory()
146
147
    for parser_name, tests in test_unit.items():
        assert set(tests.keys()).issubset(UNIT_STAGES)
148
149
        if verbose:
            print('  Match-Tests for parser "' + parser_name + '"')
150
        for test_name, test_code in tests.get('match', dict()).items():
151
152
153
            if verbose:
                infostr = '    match-test "' + test_name + '" ... '
                errflag = len(errata)
154
            cst = parser(test_code, parser_name)
Eckhart Arnold's avatar
Eckhart Arnold committed
155
            cst.log("match_%s_%s.cst" % (parser_name, test_name))
156
            tests.setdefault('__cst__', {})[test_name] = cst
157
            if "ast" in tests or report:
158
159
160
                ast = copy.deepcopy(cst)
                transform(ast)
                tests.setdefault('__ast__', {})[test_name] = ast
Eckhart Arnold's avatar
Eckhart Arnold committed
161
                ast.log("match_%s_%s.ast" % (parser_name, test_name))
162
            if cst.error_flag:
Eckhart Arnold's avatar
Eckhart Arnold committed
163
                errata.append('Match test "%s" for parser "%s" failed:\n\tExpr.:  %s\n\n\t%s\n\n' %
164
                              (test_name, parser_name, '\n\t'.join(test_code.split('\n')),
165
166
                               '\n\t'.join(m.replace('\n', '\n\t\t') for m in
                                           error_messages(test_code, cst.collect_errors()))))
167
                tests.setdefault('__err__', {})[test_name] = errata[-1]
168
169
                # write parsing-history log only in case of failure!
                parser.log_parsing_history__("match_%s_%s.log" % (parser_name, test_name))
170
171
            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' %
172
                                  (test_name, parser_name, cst.as_sxpr()))
173
174
175
176
177
178
            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
179
180
                                     flatten_sxpr(compare.as_sxpr()),
                                     flatten_sxpr(ast.as_sxpr())))
181
                    tests.setdefault('__err__', {})[test_name] = errata[-1]
182
            if verbose:
183
                print(infostr + ("OK" if len(errata) == errflag else "FAIL"))
184

185
        if verbose and 'fail' in tests:
186
            print('  Fail-Tests for parser "' + parser_name + '"')
187
        for test_name, test_code in tests.get('fail', dict()).items():
188
189
190
            if verbose:
                infostr = '    fail-test  "' + test_name + '" ... '
                errflag = len(errata)
191
192
193
194
            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))
195
                tests.setdefault('__err__', {})[test_name] = errata[-1]
196
197
                # write parsing-history log only in case of test-failure
                parser.log_parsing_history__("fail_%s_%s.log" % (parser_name, test_name))
198
            if verbose:
Eckhart Arnold's avatar
Eckhart Arnold committed
199
                print(infostr + ("OK" if len(errata) == errflag else "FAIL"))
200

201
202
    # write test-report
    if report:
203
        report_dir = "REPORT"
204
205
        if not os.path.exists(report_dir):
            os.mkdir(report_dir)
di68kap's avatar
di68kap committed
206
        with open(os.path.join(report_dir, unit_name + '.md'), 'w', encoding='utf8') as f:
207
            f.write(get_report(test_unit))
208

209
210
211
    return errata


Eckhart Arnold's avatar
Eckhart Arnold committed
212
213
214
215
def grammar_suite(directory, parser_factory, transformer_factory,
                  fn_patterns=['*test*'],
                  ignore_unknown_filetypes=False,
                  report=True, verbose=True):
216
217
    """
    Runs all grammar unit tests in a directory. A file is considered a test
218
219
    unit, if it has the word "test" in its name.
    """
Eckhart Arnold's avatar
Eckhart Arnold committed
220
221
    if not isinstance(fn_patterns, collections.abc.Collection):
        fn_patterns = [fn_patterns]
222
    all_errors = collections.OrderedDict()
223
224
    if verbose:
        print("\nScanning test-directory: " + directory)
225
226
    save_cwd = os.getcwd()
    os.chdir(directory)
Eckhart Arnold's avatar
Eckhart Arnold committed
227
    if is_logging():    clear_logs()
228
    for filename in sorted(os.listdir()):
Eckhart Arnold's avatar
Eckhart Arnold committed
229
        if any(fnmatch.fnmatch(filename, pattern) for pattern in fn_patterns):
230
            try:
231
232
                if verbose:
                    print("\nRunning grammar tests from: " + filename)
233
234
                errata = grammar_unit(filename, parser_factory,
                                      transformer_factory, report, verbose)
235
236
237
                if errata:
                    all_errors[filename] = errata
            except ValueError as e:
238
                if not ignore_unknown_filetypes or str(e).find("Unknown") < 0:
239
                    raise e
240
    os.chdir(save_cwd)
241
242
243
244
245
246
247
    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:
248
        return ('Test suite "%s" revealed some errors:\n\n' % directory) + '\n'.join(error_report)
249
250
251
    return ''


252
def runner(tests, namespace):
253
254
    """
    Runs all or some selected Python unit tests found in the
255
256
    namespace. To run all tests in a module, call 
    ``runner("", globals())`` from within that module.
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275

    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
276
            from DHParser.testing import runner
277
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
            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()