testing.py 11.8 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
di68kap's avatar
di68kap committed
21
import inspect
22
23
import json
import os
24
25
26
27
try:
    import regex as re
except ImportError:
    import re
28

29
from DHParser import error_messages
Eckhart Arnold's avatar
Eckhart Arnold committed
30
from DHParser.syntaxtree import mock_syntax_tree, flatten_sxpr
31

32
__all__ = ('unit_from_configfile',
33
34
35
36
37
38
39
           'unit_from_json',
           'unit_from_file',
           'get_report',
           'grammar_unit',
           'grammar_suite',
           'runner')

40
UNIT_STAGES = {'match', 'fail', 'ast', 'cst', '__ast__', '__cst__'}
41
42


43
def unit_from_configfile(config_filename):
44
45
    """
    Reads a grammar unit test from a config file.
46
    """
47
    cfg = configparser.ConfigParser(interpolation=None)
48
    cfg.read(config_filename)
49
50
    OD = collections.OrderedDict
    unit = OD()
51
52
    for section in cfg.sections():
        symbol, stage = section.split(':')
53
54
        if stage not in UNIT_STAGES:
            if symbol in UNIT_STAGES:
55
56
                symbol, stage = stage, symbol
            else:
57
                raise ValueError('Test stage %s not in: ' % (stage, str(UNIT_STAGES)))
58
        for testkey, testcode in cfg[section].items():
59
60
61
62
            if testcode[:3] + testcode[-3:] in {"''''''", '""""""'}:
                testcode = testcode[3:-3]
            elif testcode[:1] + testcode[-1:] in {"''", '""'}:
                testcode = testcode[1:-1]
63
            unit.setdefault(symbol, OD()).setdefault(stage, OD())[testkey] = testcode
64
65
    # print(json.dumps(unit, sort_keys=True, indent=4))
    return unit
66

67

68
def unit_from_json(json_filename):
69
70
    """
    Reads a grammar unit test from a json file.
71
    """
72
    with open(json_filename, 'r') as f:
73
74
75
76
77
78
79
        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

80
# TODO: add support for yaml, cson, toml
81
82


83
def unit_from_file(filename):
84
85
    """
    Reads a grammar unit test from a file. The format of the file is
86
87
    determined by the ending of its name.
    """
88
89
90
91
    if filename.endswith(".json"):
        return unit_from_json(filename)
    elif filename.endswith(".ini"):
        return unit_from_configfile(filename)
92
    else:
93
        raise ValueError("Unknown unit test file type: " + filename[filename.rfind('.'):])
94
95


96
def get_report(test_unit):
97
98
    """
    Returns a text-report of the results of a grammar unit test.
99
    """
100
101
102
103
    def indent(txt):
        lines = txt.split('\n')
        lines[0] = '    ' + lines[0]
        return "\n    ".join(lines)
104
105
106
107
108
109
110
111
    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:')
112
            report.append(indent(test_code))
113
114
115
116
117
            error = tests.get('__err__', {}).get(test_name, "")
            if error:
                report.append('\n### Error:')
                report.append(error)
            ast = tests.get('__ast__', {}).get(test_name, None)
118
119
120
            cst = tests.get('__cst__', {}).get(test_name, None)
            if cst and (not ast or cst == ast):
                report.append('\n### CST')
121
                report.append(indent(cst.as_sxpr()))
122
            elif ast:
123
                report.append('\n### AST')
124
                report.append(indent(ast.as_sxpr()))
125
126
127
    return '\n'.join(report)


128
def grammar_unit(test_unit, parser_factory, transformer_factory, report=True, verbose=False):
129
130
    """
    Unit tests for a grammar-parser and ast transformations.
131
    """
132
    if isinstance(test_unit, str):
133
        unit_dir, unit_name = os.path.split(os.path.splitext(test_unit)[0])
134
        test_unit = unit_from_file(test_unit)
135
    else:
136
        unit_dir = ""
137
        unit_name = str(id(test_unit))
138
139
    if verbose:
        print("\nUnit: " + unit_name)
140
141
142
    errata = []
    parser = parser_factory()
    transform = transformer_factory()
143
144
    for parser_name, tests in test_unit.items():
        assert set(tests.keys()).issubset(UNIT_STAGES)
145
146
        if verbose:
            print('  Match-Tests for parser "' + parser_name + '"')
147
        for test_name, test_code in tests.get('match', dict()).items():
148
149
150
            if verbose:
                infostr = '    match-test "' + test_name + '" ... '
                errflag = len(errata)
151
            cst = parser(test_code, parser_name)
Eckhart Arnold's avatar
Eckhart Arnold committed
152
            cst.log("%s_match_%s_%s.cst" % (unit_name, parser_name, test_name))
153
            tests.setdefault('__cst__', {})[test_name] = cst
154
            if "ast" in tests or report:
155
156
157
                ast = copy.deepcopy(cst)
                transform(ast)
                tests.setdefault('__ast__', {})[test_name] = ast
Eckhart Arnold's avatar
Eckhart Arnold committed
158
                ast.log("%s_match_%s_%s.ast" % (unit_name, parser_name, test_name))
159
            if cst.error_flag:
Eckhart Arnold's avatar
Eckhart Arnold committed
160
                errata.append('Match test "%s" for parser "%s" failed:\n\tExpr.:  %s\n\n\t%s\n\n' %
161
                              (test_name, parser_name, '\n\t'.join(test_code.split('\n')),
162
163
                               '\n\t'.join(m.replace('\n', '\n\t\t') for m in
                                           error_messages(test_code, cst.collect_errors()))))
164
                tests.setdefault('__err__', {})[test_name] = errata[-1]
165
166
                # write parsing-history log only in case of failure!
                parser.log_parsing_history__("match_%s_%s.log" % (parser_name, test_name))
167
168
            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' %
169
                                  (test_name, parser_name, cst.as_sxpr()))
170
171
172
173
174
175
            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
176
177
                                     flatten_sxpr(compare.as_sxpr()),
                                     flatten_sxpr(ast.as_sxpr())))
178
                    tests.setdefault('__err__', {})[test_name] = errata[-1]
179
            if verbose:
180
                print(infostr + ("OK" if len(errata) == errflag else "FAIL"))
181

182
        if verbose and 'fail' in tests:
183
            print('  Fail-Tests for parser "' + parser_name + '"')
184
        for test_name, test_code in tests.get('fail', dict()).items():
185
186
187
            if verbose:
                infostr = '    fail-test  "' + test_name + '" ... '
                errflag = len(errata)
188
189
190
191
            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))
192
                tests.setdefault('__err__', {})[test_name] = errata[-1]
193
194
                # write parsing-history log only in case of test-failure
                parser.log_parsing_history__("fail_%s_%s.log" % (parser_name, test_name))
195
196
            if verbose:
                print(infostr + "OK" if len(errata) == errflag else "FAIL")
197

198
199
    # write test-report
    if report:
200
        report_dir = "REPORT"
201
202
        if not os.path.exists(report_dir):
            os.mkdir(report_dir)
203
        with open(os.path.join(report_dir, unit_name + '.md'), 'w') as f:
204
            f.write(get_report(test_unit))
205

206
207
208
    return errata


209
def grammar_suite(directory, parser_factory, transformer_factory, ignore_unknown_filetypes=False,
210
                  report=True, verbose=False):
211
212
    """
    Runs all grammar unit tests in a directory. A file is considered a test
213
214
215
    unit, if it has the word "test" in its name.
    """
    all_errors = collections.OrderedDict()
216
217
    if verbose:
        print("\nScanning test-directory: " + directory)
218
219
220
    save_cwd = os.getcwd()
    os.chdir(directory)
    for filename in sorted(os.listdir()):
221
222
        if filename.lower().find("test") >= 0:
            try:
223
224
                if verbose:
                    print("\nRunning grammar tests from: " + filename)
225
226
                errata = grammar_unit(filename, parser_factory,
                                      transformer_factory, report, verbose)
227
228
229
                if errata:
                    all_errors[filename] = errata
            except ValueError as e:
230
                if not ignore_unknown_filetypes or str(e).find("Unknown") < 0:
231
                    raise e
232
    os.chdir(save_cwd)
233
234
235
236
237
238
239
    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:
240
        return ('Test suite "%s" revealed some errors:\n\n' % directory) + '\n'.join(error_report)
241
242
243
    return ''


244
def runner(tests, namespace):
245
246
    """
    Runs all or some selected Python unit tests found in the
247
248
    namespace. To run all tests in a module, call 
    ``runner("", globals())`` from within that module.
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267

    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
268
            from DHParser.testing import runner
269
270
271
272
273
274
275
276
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
            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()