testing.py 12.9 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 Node, error_messages
30
from DHParser.toolkit import compact_sexpr, is_logging, log_dir
31
32
33
from DHParser.syntaxtree import MockParser
from DHParser.ebnf import grammar_changed
from DHParser.dsl import compile_on_disk
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51


def mock_syntax_tree(sexpr):
    """Generates a tree of nodes from an S-expression.

    Example: 
    >>> mock_syntax_tree("(a (b c))").as_sexpr()
    (a 
        (b 
            "c" 
        )
    )
    """
    def next_block(s):
        s = s.strip()
        while s[0] != ')':
            if s[0] != '(': raise ValueError('"(" expected, not ' + s[:10])
            # assert s[0] == '(', s
52
            level = 1
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
            i = 1
            while level > 0:
                if s[i] == '(':
                    level += 1
                elif s[i] == ')':
                    level -= 1
                i += 1
            yield s[:i]
            s = s[i:].strip()

    sexpr = sexpr.strip()
    if sexpr[0] != '(': raise ValueError('"(" expected, not ' + sexpr[:10])
    # assert sexpr[0] == '(', sexpr
    sexpr = sexpr[1:].strip()
    m = re.match('[\w:]+', sexpr)
    name, class_name = (sexpr[:m.end()].split(':') + [''])[:2]
    sexpr = sexpr[m.end():].strip()
    if sexpr[0] == '(':
        result = tuple(mock_syntax_tree(block) for block in next_block(sexpr))
    else:
        lines = []
        while sexpr and sexpr[0] != ')':
            for qm in ['"""', "'''", '"', "'"]:
                m = re.match(qm + r'.*?' + qm, sexpr)
                if m:
                    i = len(qm)
                    lines.append(sexpr[i:m.end() - i])
                    sexpr = sexpr[m.end():].strip()
                    break
            else:
                m = re.match(r'(?:(?!\)).)*', sexpr)
                lines.append(sexpr[:m.end()])
                sexpr = sexpr[m.end():]
        result = "\n".join(lines)
    return Node(MockParser(name, ':' + class_name), result)


90
91
def recompile_grammar(ebnf_filename, query_remove_error_files=True):
    """Recompiles an ebnf-grammar if necessary, that is if either no
92
    corresponding 'XXXXCompiler.py'-file exists or if that file is
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
    outdated.
    
    Parameters:
        ebnf_filename(str):  The filename of the ebnf-source of the 
            grammar. In case this is a directory and not a file all
            files within this directory ending with .ebnf will be 
            compiled.
    """
    if os.path.isdir(ebnf_filename):
        for entry in os.listdir(ebnf_filename):
            if entry.lower().endswith('.ebnf') and os.path.isfile(entry):
                recompile_grammar(entry)
        return

    base, ext = os.path.splitext(ebnf_filename)
108
    compiler_name = base + 'Compiler.py'
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
    errors = []
    if (not os.path.exists(compiler_name) or
        grammar_changed(compiler_name, ebnf_filename)):
        # print("recompiling parser for: " + ebnf_filename)
        errors = compile_on_disk(ebnf_filename)
        if errors:
            # print("Errors while compiling: " + ebnf_filename + '!')
            with open(base + '_errors.txt', 'w') as f:
                for e in errors:
                    f.write(e)
                    f.write('\n')

    if not errors:
        if os.path.exists(base + '_errors.txt'):
            if query_remove_error_files:
                answer = input('Remove obsolete file ' + base + '_errors.txt (y/n)? ').lower()
                if answer not in {'y', 'yes'}:
                    return
            os.remove(base + '_errors.txt')

129

130
UNIT_STAGES = {'match', 'fail', 'ast', 'cst', '__ast__', '__cst__'}
131
132


133
134
def unit_from_configfile(config_filename):
    """Reads a grammar unit test from a config file.
135
    """
136
    cfg = configparser.ConfigParser()
137
    cfg.read(config_filename)
138
139
    OD = collections.OrderedDict
    unit = OD()
140
141
    for section in cfg.sections():
        symbol, stage = section.split(':')
142
143
        if stage not in UNIT_STAGES:
            if symbol in UNIT_STAGES:
144
145
                symbol, stage = stage, symbol
            else:
146
                raise ValueError('Test stage %s not in: ' % (stage, str(UNIT_STAGES)))
147
        for testkey, testcode in cfg[section].items():
148
149
150
151
            if testcode[:3] + testcode[-3:] in {"''''''", '""""""'}:
                testcode = testcode[3:-3]
            elif testcode[:1] + testcode[-1:] in {"''", '""'}:
                testcode = testcode[1:-1]
152
            unit.setdefault(symbol, OD()).setdefault(stage, OD())[testkey] = testcode
153
154
    # print(json.dumps(unit, sort_keys=True, indent=4))
    return unit
155

156
157
158
159
160
161
162
163
164
165
166
167

def unit_from_json(config_filename):
    """Reads a grammar unit test from a json file.
    """
    with open(config_filename, 'r') 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

168
# TODO: add support for yaml, cson, toml
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183


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


184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
def report(test_unit):
    """Returns a text-report of the results of a grammar unit test.
    """
    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(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)
            if ast:
                report.append('\n### AST')
                report.append(ast.as_sexpr())
    return '\n'.join(report)


207
def grammar_unit(test_unit, parser_factory, transformer_factory):
208
209
    """Unit tests for a grammar-parser and ast transformations.
    """
210
    if isinstance(test_unit, str):
211
        unit_name = os.path.basename(os.path.splitext(test_unit)[0])
212
        test_unit = unit_from_file(test_unit)
213
214
    else:
        unit_name = str(id(test_unit))
215
216
217
    errata = []
    parser = parser_factory()
    transform = transformer_factory()
218
219
    for parser_name, tests in test_unit.items():
        assert set(tests.keys()).issubset(UNIT_STAGES)
220
221
222
223

        for test_name, test_code in tests.get('match', dict()).items():
            cst = parser(test_code, parser_name)
            tests.setdefault('__cst__', {})[test_name] = cst
224
225
226
227
            if "ast" in tests or is_logging():
                ast = copy.deepcopy(cst)
                transform(ast)
                tests.setdefault('__ast__', {})[test_name] = ast
228
229
230
231
            if cst.error_flag:
                errata.append('Match test "%s" for parser "%s" failed:\n\tExpr.:  %s\n\t%s' %
                              (test_name, parser_name, '\n\t'.join(test_code.split('\n')),
                               '\n\t'.join(error_messages(test_code, cst.collect_errors()))))
232
                tests.setdefault('__err__', {})[test_name] = errata[-1]
233
234
235
236
237
238
239
240
241
242
243
            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_sexpr()))
            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')),
                                     compact_sexpr(compare.as_sexpr()),
                                     compact_sexpr(ast.as_sexpr())))
244
                    tests.setdefault('__err__', {})[test_name] = errata[-1]
245
246
247
248
249
250

        for test_name, test_code in tests.get('fail', dict()).items():
            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))
251
252
253
254
255
256
                tests.setdefault('__err__', {})[test_name] = errata[-1]

    if is_logging():
        with open(os.path.join(log_dir(), unit_name + '.report'), 'w') as f:
            f.write(report(test_unit))

257
258
259
    return errata


260
261
262
263
264
265
266
267
268
269
270
271
272
273
def grammar_suite(directory, parser_factory, transformer_factory, ignore_unknown_filetypes=False):
    """Runs all grammar unit tests in a directory. A file is considered a test
    unit, if it has the word "test" in its name.
    """
    all_errors = collections.OrderedDict()
    for filename in sorted(os.listdir(directory)):
        if filename.lower().find("test") >= 0:
            try:
                print("Running grammar tests in: " + filename)
                errata = grammar_unit(os.path.join(directory, filename),
                                      parser_factory, transformer_factory)
                if errata:
                    all_errors[filename] = errata
            except ValueError as e:
274
                if not ignore_unknown_filetypes or str(e).find("Unknown") < 0:
275
276
277
278
279
280
281
282
283
284
285
286
                    raise e
    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' %directory) + '\n'.join(error_report)
    return ''


287
def runner(tests, namespace):
288
289
290
    """ 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.
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309

    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
310
            from DHParser.testing import runner
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
            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()