toolkit.py 12.9 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
"""toolkit.py - utility functions for DHParser

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.


Module ``toolkit`` contains utility functions and cross-sectional
functionality like logging support that is needed across several 
of the the other DHParser-Modules.

For logging functionality, the global variable LOGGING is defined which
contains the name of a directory where log files shall be placed. By
setting its value to the empty string "" logging can be turned off.

To read the directory name function ``LOGS_DIR()`` should be called
rather than reading the variable LOGGING. ``LOGS_DIR()`` makes sure
the directory exists and raises an error if a file with the same name
already exists.
"""

33
import codecs
Eckhart Arnold's avatar
Eckhart Arnold committed
34
import contextlib
35
36
import hashlib
import os
37

di68kap's avatar
di68kap committed
38
39
40
41
try:
    import regex as re
except ImportError:
    import re
42
import sys
43

44
try:
45
    import typing
46
except ImportError:
47
    import DHParser.foreign_typing as typing
di68kap's avatar
di68kap committed
48
    sys.modules['typing'] = typing  # make it possible to import from typing
49

50
from typing import Any, Iterable, Sequence, Set, Union
51

52
__all__ = ('logging',
Eckhart Arnold's avatar
Eckhart Arnold committed
53
           'is_logging',
54
           'clear_logs',
Eckhart Arnold's avatar
Eckhart Arnold committed
55
           'log_dir',
56
           'logfile_basename',
57
           'escape_re',
Eckhart Arnold's avatar
Eckhart Arnold committed
58
           'is_filename',
59
60
           'load_if_file',
           'is_python_code',
61
62
           'md5',
           'expand_table',
63
           'smart_list',
64
           'sane_parser_name',
65
           'identity')
66
67
68
69
70
71
72


#######################################################################
#
# logging
#
#######################################################################
73
74


Eckhart Arnold's avatar
Eckhart Arnold committed
75
76
77
def log_dir() -> str:
    """Creates a directory for log files (if it does not exist) and
    returns its path.
78

Eckhart Arnold's avatar
Eckhart Arnold committed
79
80
81
    WARNING: Any files in the log dir will eventually be overwritten.
    Don't use a directory name that could be the name of a directory
    for other purposes than logging.
82
83
84
85

    Returns:
        name of the logging directory
    """
Eckhart Arnold's avatar
Eckhart Arnold committed
86
    # the try-except clauses in the following are precautions for multiprocessing
87
    global LOGGING
Eckhart Arnold's avatar
Eckhart Arnold committed
88
    try:
89
        dirname = LOGGING  # raises a name error if LOGGING is not defined
Eckhart Arnold's avatar
Eckhart Arnold committed
90
91
92
        if not dirname:
            raise NameError  # raise a name error if LOGGING evaluates to False
    except NameError:
93
94
        raise NameError("No access to log directory before logging has been "
                        "turned on within the same thread/process.")
Eckhart Arnold's avatar
Eckhart Arnold committed
95
96
97
    if os.path.exists(dirname) and not os.path.isdir(dirname):
        raise IOError('"' + dirname + '" cannot be used as log directory, '
                                      'because it is not a directory!')
98
    else:
Eckhart Arnold's avatar
Eckhart Arnold committed
99
100
101
102
103
        try:
            os.mkdir(dirname)
        except FileExistsError:
            pass
    info_file_name = os.path.join(dirname, 'info.txt')
104
105
106
107
    if not os.path.exists(info_file_name):
        with open(info_file_name, 'w') as f:
            f.write("This directory has been created by DHParser to store log files from\n"
                    "parsing. ANY FILE IN THIS DIRECTORY CAN BE OVERWRITTEN! Therefore,\n"
108
109
                    "do not place any files here and do not bother editing files in this\n"
                    "directory as any changes will get lost.\n")
110
111
112
    return dirname


Eckhart Arnold's avatar
Eckhart Arnold committed
113
@contextlib.contextmanager
114
def logging(dirname="LOGS"):
Eckhart Arnold's avatar
Eckhart Arnold committed
115
116
117
118
119
120
121
122
    """Context manager. Log files within this context will be stored in
    directory ``dirname``. Logging is turned off if name is empty.
    
    Args:
        dirname: the name for the log directory or the empty string to
            turn logging of
    """
    global LOGGING
123
    if dirname and not isinstance(dirname, str):  dirname = "LOGS"  # be fail tolerant here...
Eckhart Arnold's avatar
Eckhart Arnold committed
124
125
126
127
128
129
130
131
132
    try:
        save = LOGGING
    except NameError:
        save = ""
    LOGGING = dirname
    yield
    LOGGING = save


133
def is_logging() -> bool:
Eckhart Arnold's avatar
Eckhart Arnold committed
134
135
136
137
138
139
140
141
    """-> True, if logging is turned on."""
    global LOGGING
    try:
        return bool(LOGGING)
    except NameError:
        return False


Eckhart Arnold's avatar
Eckhart Arnold committed
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
def clear_logs(logfile_types={'.cst', '.ast', '.log'}):
    """Removes all logs from the log-directory and removes the
    log-directory if it is empty.
    """
    log_dirname = log_dir()
    files = os.listdir(log_dirname)
    only_log_files = True
    for file in files:
        path = os.path.join(log_dirname, file)
        if os.path.splitext(file)[1] in logfile_types or file == 'info.txt':
            os.remove(path)
        else:
            only_log_files = False
    if only_log_files:
        os.rmdir(log_dirname)


159
def escape_re(s) -> str:
160
161
    """Returns `s` with all regular expression special characters escaped.
    """
Eckhart Arnold's avatar
Eckhart Arnold committed
162
    # assert isinstance(s, str)
163
164
165
166
167
168
    re_chars = r"\.^$*+?{}[]()#<>=|!"
    for esc_ch in re_chars:
        s = s.replace(esc_ch, '\\' + esc_ch)
    return s


169
def is_filename(s) -> bool:
Eckhart Arnold's avatar
Eckhart Arnold committed
170
171
    """Tries to guess whether string ``s`` is a file name."""
    return s.find('\n') < 0 and s[:1] != " " and s[-1:] != " " \
172
173
           and all(s.find(ch) < 0 for ch in '*?"<>|')
           # and s.find('*') < 0 and s.find('?') < 0
Eckhart Arnold's avatar
Eckhart Arnold committed
174
175


176
def logfile_basename(filename_or_text, function_or_class_or_instance) -> str:
Eckhart Arnold's avatar
Eckhart Arnold committed
177
178
179
180
181
182
183
184
185
186
187
188
189
190
    """Generates a reasonable logfile-name (without extension) based on
    the given information.
    """
    if is_filename(filename_or_text):
        return os.path.basename(os.path.splitext(filename_or_text)[0])
    else:
        try:
            s = function_or_class_or_instance.__qualname.__
        except AttributeError:
            s = function_or_class_or_instance.__class__.__name__
        i = s.find('.')
        return s[:i] + '_out' if i >= 0 else s


191
192
193
194
195
196
197
#######################################################################
#
# loading and compiling
#
#######################################################################


198
199
200
def load_if_file(text_or_file) -> str:
    """Reads and returns content of a text-file if parameter
    `text_or_file` is a file name (i.e. a single line string),
201
    otherwise (i.e. if `text_or_file` is a multi-line string)
202
    `text_or_file` is returned.
203
    """
Eckhart Arnold's avatar
Eckhart Arnold committed
204
    if is_filename(text_or_file):
205
206
207
208
        try:
            with open(text_or_file, encoding="utf-8") as f:
                content = f.read()
            return content
209
        except FileNotFoundError:
210
            if re.fullmatch(r'[\w/:. \\]+', text_or_file):
211
212
                raise FileNotFoundError('Not a valid file: ' + text_or_file + '!\n(Add "\\n" '
                                        'to distinguish source data from a file name.)')
213
214
            else:
                return text_or_file
215
216
217
218
    else:
        return text_or_file


219
def is_python_code(text_or_file: str) -> bool:
220
221
222
    """Checks whether 'text_or_file' is python code or the name of a file that
    contains python code.
    """
Eckhart Arnold's avatar
Eckhart Arnold committed
223
    if is_filename(text_or_file):
224
225
226
227
228
229
230
231
232
        return text_or_file[-3:].lower() == '.py'
    try:
        compile(text_or_file, '<string>', 'exec')
        return True
    except (SyntaxError, ValueError, OverflowError):
        pass
    return False


233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
def has_fenced_code(text_or_file: str, info_strings = ('ebnf', 'test')) -> bool:
    """Checks whether `text_or_file` contains fenced code blocks, which are
    marked by one of the given info strings.
    See http://spec.commonmark.org/0.28/#fenced-code-blocks for more
    information on fenced code blocks in common mark documents.
    """
    if is_filename(text_or_file):
        with open(text_or_file, 'r', encoding='utf-8') as f:
            markdown = f.read()
    else:
        markdown = text_or_file

    if markdown.find('\n~~~') < 0 and markdown.find('\n```') < 0:
        return False

    if isinstance(info_strings, str):  info_strings = (info_strings,)
    FENCE_TMPL = '\n(?:(?:``[`]*[ ]*(?:%s)(?=[ .\-:\n])[^`\n]*\n)|(?:~~[~]*[ ]*(?:%s)(?=[ .\-:\n])[\n]*\n))'
    LABEL_RE = '|'.join('(?:%s)' % s for s in info_strings)
    RX_FENCE = re.compile(FENCE_TMPL % (LABEL_RE, LABEL_RE), flags=re.IGNORECASE)

    for m in RX_FENCE.finditer(markdown):
        s = re.match('(?:\n`+)|(?:\n~+)', m.group(0)).group(0)
        if markdown.find(s, m.end()) >= 0:
            return True
        else:
            return False


261
262
263
264
265
266
267
268
269
270
def md5(*txt):
    """Returns the md5-checksum for `txt`. This can be used to test if
    some piece of text, for example a grammar source file, has changed.
    """
    md5_hash = hashlib.md5()
    for t in txt:
        md5_hash.update(t.encode('utf8'))
    return md5_hash.hexdigest()


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
def compile_python_object(python_src, catch_obj_regex=""):
    """Compiles the python source code and returns the (first) object
    the name of which is matched by ``catch_obj_regex``. If catch_obj
    is the empty string, the namespace dictionary will be returned.
    """
    if isinstance(catch_obj_regex, str):
        catch_obj_regex = re.compile(catch_obj_regex)
    code = compile(python_src, '<string>', 'exec')
    namespace = {}
    exec(code, namespace)  # safety risk?
    if catch_obj_regex:
        matches = [key for key in namespace.keys() if catch_obj_regex.match(key)]
        if len(matches) == 0:
            raise ValueError("No object matching /%s/ defined in source code." %
                             catch_obj_regex.pattern)
        elif len(matches) > 1:
            raise ValueError("Ambiguous matches for %s : %s" %
                             (str(catch_obj_regex), str(matches)))
        return namespace[matches[0]] if matches else None
    else:
        return namespace


#######################################################################
#
# smart lists and multi-keyword tables
#
#######################################################################


Eckhart Arnold's avatar
Eckhart Arnold committed
301
# def smart_list(arg: Union[str, Iterable[T]]) -> Union[Sequence[str], Sequence[T]]:
302
def smart_list(arg: Union[str, Iterable, Any]) -> Union[Sequence, Set]:
303
    """Returns the argument as list, depending on its type and content.
304
305
306
307
    
    If the argument is a string, it will be interpreted as a list of
    comma separated values, trying ';', ',', ' ' as possible delimiters
    in this order, e.g.
308
309
310
311
312
313
314
    >>> smart_list('1; 2, 3; 4')
    ['1', '2, 3', '4']
    >>> smart_list('2, 3')
    ['2', '3']
    >>> smart_list('a b cd')
    ['a', 'b', 'cd']

315
316
    If the argument is a collection other than a string, it will be
    returned as is, e.g.
317
318
319
320
321
    >>> smart_list((1, 2, 3))
    (1, 2, 3)
    >>> smart_list({1, 2, 3})
    {1, 2, 3}

322
323
    If the argument is another iterable than a collection, it will
    be converted into a list, e.g.
324
325
326
    >>> smart_list(i for i in {1,2,3})
    [1, 2, 3]

327
328
    Finally, if none of the above is true, the argument will be 
    wrapped in a list and returned, e.g.
329
330
    >>> smart_list(125)
    [125]
331
332
333
334
335
    """
    if isinstance(arg, str):
        for delimiter in (';', ','):
            lst = arg.split(delimiter)
            if len(lst) > 1:
336
337
                return [s.strip() for s in lst]
        return [s.strip() for s in arg.strip().split(' ')]
338
    elif isinstance(arg, Sequence) or isinstance(arg, Set):
339
        return arg
Eckhart Arnold's avatar
Eckhart Arnold committed
340
    elif isinstance(arg, Iterable):
341
        return list(arg)
342
343
344
345
    else:
        return [arg]


346
347
348
349
350
def expand_table(compact_table):
    """Expands a table by separating keywords that are tuples or strings
    containing comma separated words into single keyword entries with
    the same values. Returns the expanded table.
    Example:
351
352
    >>> expand_table({"a, b": 1, ('d','e','f'):5, "c":3})
    {'a': 1, 'b': 1, 'd': 5, 'e': 5, 'f': 5, 'c': 3}
353
354
355
356
357
    """
    expanded_table = {}
    keys = list(compact_table.keys())
    for key in keys:
        value = compact_table[key]
358
        for k in smart_list(key):
359
            if k in expanded_table:
360
                raise KeyError('Key "%s" used more than once in compact table!' % key)
361
            expanded_table[k] = value
362
    return expanded_table
363
364


365
366
367
368
369
370
371
#######################################################################
#
# miscellaneous
#
#######################################################################


372
def sane_parser_name(name) -> bool:
373
    """Checks whether given name is an acceptable parser name. Parser names
374
    must not be preceded or succeeded by a double underscore '__'!
375
    """
di68kap's avatar
di68kap committed
376
377
378
    return name and name[:2] != '__' and name[-2:] != '__'


379
380
381
382
def identity(anything: Any) -> Any:
    return anything


383
384
385
386
387
388
389
#######################################################################
#
# initialization
#
#######################################################################


390
391
392
393
394
395
396
397
try:
    if sys.stdout.encoding.upper() != "UTF-8":
        # make sure that `print()` does not raise an error on 
        # non-ASCII characters:
        sys.stdout = codecs.getwriter("utf-8")(sys.stdout.detach())
except AttributeError:
    # somebody has already taken care of this !?
    pass