toolkit.py 11.7 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
    from typing import Any, List, Tuple, Iterable, Sequence, Set, Union, Optional, TypeVar
46
except ImportError:
47
    from .typing34 import Any, List, Tuple, Iterable, Sequence, Set, Union, Optional, TypeVar
48

49
__all__ = ('logging',
Eckhart Arnold's avatar
Eckhart Arnold committed
50
           'is_logging',
51
           'clear_logs',
Eckhart Arnold's avatar
Eckhart Arnold committed
52
           'log_dir',
53
           'logfile_basename',
54
           'escape_re',
Eckhart Arnold's avatar
Eckhart Arnold committed
55
           'is_filename',
56
57
           'load_if_file',
           'is_python_code',
58
59
           'md5',
           'expand_table',
60
           'smart_list',
61
           'sane_parser_name',
62
           'identity')
63
64
65
66
67
68
69


#######################################################################
#
# logging
#
#######################################################################
70
71


Eckhart Arnold's avatar
Eckhart Arnold committed
72
73
74
def log_dir() -> str:
    """Creates a directory for log files (if it does not exist) and
    returns its path.
75

Eckhart Arnold's avatar
Eckhart Arnold committed
76
77
78
    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.
79
80
81
82

    Returns:
        name of the logging directory
    """
Eckhart Arnold's avatar
Eckhart Arnold committed
83
    # the try-except clauses in the following are precautions for multiprocessing
84
    global LOGGING
Eckhart Arnold's avatar
Eckhart Arnold committed
85
    try:
86
        dirname = LOGGING  # raises a name error if LOGGING is not defined
Eckhart Arnold's avatar
Eckhart Arnold committed
87
88
89
90
91
92
93
94
        if not dirname:
            raise NameError  # raise a name error if LOGGING evaluates to False
    except NameError:
        raise NameError("No access to log directory before logging has been turned "
                        "on within the same thread/process.")
    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!')
95
    else:
Eckhart Arnold's avatar
Eckhart Arnold committed
96
97
98
99
100
        try:
            os.mkdir(dirname)
        except FileExistsError:
            pass
    info_file_name = os.path.join(dirname, 'info.txt')
101
102
103
104
    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"
105
106
                    "do not place any files here and do not bother editing files in this\n"
                    "directory as any changes will get lost.\n")
107
108
109
    return dirname


Eckhart Arnold's avatar
Eckhart Arnold committed
110
@contextlib.contextmanager
111
def logging(dirname="LOGS"):
Eckhart Arnold's avatar
Eckhart Arnold committed
112
113
114
115
116
117
118
119
    """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
120
    if dirname and not isinstance(dirname, str):  dirname = "LOGS"  # be fail tolerant here...
Eckhart Arnold's avatar
Eckhart Arnold committed
121
122
123
124
125
126
127
128
129
    try:
        save = LOGGING
    except NameError:
        save = ""
    LOGGING = dirname
    yield
    LOGGING = save


130
def is_logging() -> bool:
Eckhart Arnold's avatar
Eckhart Arnold committed
131
132
133
134
135
136
137
138
    """-> True, if logging is turned on."""
    global LOGGING
    try:
        return bool(LOGGING)
    except NameError:
        return False


Eckhart Arnold's avatar
Eckhart Arnold committed
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
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)


156
def escape_re(s) -> str:
157
158
    """Returns `s` with all regular expression special characters escaped.
    """
Eckhart Arnold's avatar
Eckhart Arnold committed
159
    # assert isinstance(s, str)
160
161
162
163
164
165
    re_chars = r"\.^$*+?{}[]()#<>=|!"
    for esc_ch in re_chars:
        s = s.replace(esc_ch, '\\' + esc_ch)
    return s


166
def is_filename(s) -> bool:
Eckhart Arnold's avatar
Eckhart Arnold committed
167
168
169
170
171
    """Tries to guess whether string ``s`` is a file name."""
    return s.find('\n') < 0 and s[:1] != " " and s[-1:] != " " \
           and s.find('*') < 0 and s.find('?') < 0


172
def logfile_basename(filename_or_text, function_or_class_or_instance) -> str:
Eckhart Arnold's avatar
Eckhart Arnold committed
173
174
175
176
177
178
179
180
181
182
183
184
185
186
    """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


187
188
189
190
191
192
193
#######################################################################
#
# loading and compiling
#
#######################################################################


194
195
196
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),
197
    otherwise (i.e. if `text_or_file` is a multi-line string)
198
    `text_or_file` is returned.
199
    """
Eckhart Arnold's avatar
Eckhart Arnold committed
200
    if is_filename(text_or_file):
201
202
203
204
        try:
            with open(text_or_file, encoding="utf-8") as f:
                content = f.read()
            return content
205
        except FileNotFoundError:
206
            if re.fullmatch(r'[\w/:. \\]+', text_or_file):
207
208
                raise FileNotFoundError('Not a valid file: ' + text_or_file + '!\n(Add "\\n" '
                                        'to distinguish source data from a file name.)')
209
210
            else:
                return text_or_file
211
212
213
214
    else:
        return text_or_file


215
def is_python_code(text_or_file) -> bool:
216
217
218
    """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
219
    if is_filename(text_or_file):
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
        return text_or_file[-3:].lower() == '.py'
    try:
        compile(text_or_file, '<string>', 'exec')
        return True
    except (SyntaxError, ValueError, OverflowError):
        pass
    return False


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()


239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
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
269
270
# def smart_list(arg: Union[str, Iterable[T]]) -> Union[Sequence[str], Sequence[T]]:
def smart_list(arg: Union[str, Iterable, Any]) -> Sequence:
271
    """Returns the argument as list, depending on its type and content.
272
273
274
275
    
    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.
276
277
278
279
280
281
282
    >>> smart_list('1; 2, 3; 4')
    ['1', '2, 3', '4']
    >>> smart_list('2, 3')
    ['2', '3']
    >>> smart_list('a b cd')
    ['a', 'b', 'cd']

283
284
    If the argument is a collection other than a string, it will be
    returned as is, e.g.
285
286
287
288
289
    >>> smart_list((1, 2, 3))
    (1, 2, 3)
    >>> smart_list({1, 2, 3})
    {1, 2, 3}

290
291
    If the argument is another iterable than a collection, it will
    be converted into a list, e.g.
292
293
294
    >>> smart_list(i for i in {1,2,3})
    [1, 2, 3]

295
296
    Finally, if none of the above is true, the argument will be 
    wrapped in a list and returned, e.g.
297
298
    >>> smart_list(125)
    [125]
299
300
301
302
303
    """
    if isinstance(arg, str):
        for delimiter in (';', ','):
            lst = arg.split(delimiter)
            if len(lst) > 1:
304
305
                return [s.strip() for s in lst]
        return [s.strip() for s in arg.strip().split(' ')]
306
    elif isinstance(arg, Sequence) or isinstance(arg, Set):
307
        return arg
Eckhart Arnold's avatar
Eckhart Arnold committed
308
    elif isinstance(arg, Iterable):
309
        return list(arg)
310
311
312
313
    else:
        return [arg]


314
315
316
317
318
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:
319
320
    >>> expand_table({"a, b": 1, ('d','e','f'):5, "c":3})
    {'a': 1, 'b': 1, 'd': 5, 'e': 5, 'f': 5, 'c': 3}
321
322
323
324
325
    """
    expanded_table = {}
    keys = list(compact_table.keys())
    for key in keys:
        value = compact_table[key]
326
        for k in smart_list(key):
327
            if k in expanded_table:
328
                raise KeyError('Key "%s" used more than once in compact table!' % key)
329
            expanded_table[k] = value
330
    return expanded_table
331
332


333
334
335
336
337
338
339
#######################################################################
#
# miscellaneous
#
#######################################################################


340
def sane_parser_name(name) -> bool:
341
    """Checks whether given name is an acceptable parser name. Parser names
342
    must not be preceded or succeeded by a double underscore '__'!
343
    """
di68kap's avatar
di68kap committed
344
345
346
    return name and name[:2] != '__' and name[-2:] != '__'


347
348
349
350
def identity(anything: Any) -> Any:
    return anything


351
352
353
354
355
356
357
#######################################################################
#
# initialization
#
#######################################################################


358
359
360
361
362
363
364
365
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