toolkit.py 12.1 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
34
import collections
Eckhart Arnold's avatar
Eckhart Arnold committed
35
import contextlib
36
37
import hashlib
import os
38

di68kap's avatar
di68kap committed
39
40
41
42
try:
    import regex as re
except ImportError:
    import re
43
import sys
44

45
try:
46
    from typing import Any, List, Tuple
47
except ImportError:
48
    from .typing34 import Any, List, Tuple
49

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


Eckhart Arnold's avatar
Eckhart Arnold committed
69
70
71
def log_dir() -> str:
    """Creates a directory for log files (if it does not exist) and
    returns its path.
72

Eckhart Arnold's avatar
Eckhart Arnold committed
73
74
75
    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.
76
77
78
79

    Returns:
        name of the logging directory
    """
Eckhart Arnold's avatar
Eckhart Arnold committed
80
    # the try-except clauses in the following are precautions for multiprocessing
81
    global LOGGING
Eckhart Arnold's avatar
Eckhart Arnold committed
82
    try:
83
        dirname = LOGGING  # raises a name error if LOGGING is not defined
Eckhart Arnold's avatar
Eckhart Arnold committed
84
85
86
87
88
89
90
91
        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!')
92
    else:
Eckhart Arnold's avatar
Eckhart Arnold committed
93
94
95
96
97
        try:
            os.mkdir(dirname)
        except FileExistsError:
            pass
    info_file_name = os.path.join(dirname, 'info.txt')
98
99
100
101
    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"
102
103
                    "do not place any files here and do not bother editing files in this\n"
                    "directory as any changes will get lost.\n")
104
105
106
    return dirname


Eckhart Arnold's avatar
Eckhart Arnold committed
107
@contextlib.contextmanager
108
def logging(dirname="LOGS"):
Eckhart Arnold's avatar
Eckhart Arnold committed
109
110
111
112
113
114
115
116
    """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
117
    if dirname and not isinstance(dirname, str):  dirname = "LOGS"  # be fail tolerant here...
Eckhart Arnold's avatar
Eckhart Arnold committed
118
119
120
121
122
123
124
125
126
    try:
        save = LOGGING
    except NameError:
        save = ""
    LOGGING = dirname
    yield
    LOGGING = save


127
def is_logging() -> bool:
Eckhart Arnold's avatar
Eckhart Arnold committed
128
129
130
131
132
133
134
135
    """-> True, if logging is turned on."""
    global LOGGING
    try:
        return bool(LOGGING)
    except NameError:
        return False


136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
# def repr_call(f, parameter_list) -> str:
#     """Turns a list of items into a string resembling the parameter
#     list of a function call by omitting default values at the end:
#     >>> def f(a, b=1):    print(a, b)
#     >>> repr_call(f, (5,1))
#     'f(5)'
#     >>> repr_call(f, (5,2))
#     'f(5, 2)'
#     """
#     i = 0
#     defaults = f.__defaults__ if f.__defaults__ is not None else []
#     for parameter, default in zip(reversed(parameter_list), reversed(defaults)):
#         if parameter != default:
#             break
#         i -= 1
#     if i < 0:
#         parameter_list = parameter_list[:i]
#     name = f.__self__.__class__.__name__ if f.__name__ == '__init__' else f.__name__
154
#     return "%s(%s)" % (name, ", ".merge_children(repr(item) for item in parameter_list))
155
156


157
def line_col(text: str, pos: int) -> Tuple[int, int]:
158
159
    """Returns the position within a text as (line, column)-tuple.
    """
160
    assert pos <= len(text), str(pos) + " > " + str(len(text))  # can point one character after EOF
161
162
163
164
165
    line = text.count("\n", 0, pos) + 1
    column = pos - text.rfind("\n", 0, pos)
    return line, column


166
def error_messages(source_text, errors) -> List[str]:
167
168
169
170
171
172
173
174
175
176
177
    """Returns the sequence or iterator of error objects as an intertor
    of error messages with line and column numbers at the beginning.
    
    Args:
        source_text (str):  The source text on which the errors occurred.
            (Needed in order to determine the line and column numbers.)
        errors (list):  The list of errors as returned by the method 
            ``collect_errors()`` of a Node object     
    Returns:
        a list that contains all error messages in string form. Each
        string starts with "line: [Line-No], column: [Column-No]
178
    """
Eckhart Arnold's avatar
Eckhart Arnold committed
179
    return ["line: %i, column: %i" % line_col(source_text, err.pos) + ", error: %s" % err.msg
180
            for err in sorted(list(errors))]
181
182


183
def escape_re(s) -> str:
184
185
186
187
188
189
190
191
192
    """Returns `s` with all regular expression special characters escaped.
    """
    assert isinstance(s, str)
    re_chars = r"\.^$*+?{}[]()#<>=|!"
    for esc_ch in re_chars:
        s = s.replace(esc_ch, '\\' + esc_ch)
    return s


193
def is_filename(s) -> bool:
Eckhart Arnold's avatar
Eckhart Arnold committed
194
195
196
197
198
    """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


199
def logfile_basename(filename_or_text, function_or_class_or_instance) -> str:
Eckhart Arnold's avatar
Eckhart Arnold committed
200
201
202
203
204
205
206
207
208
209
210
211
212
213
    """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


214
215
216
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),
217
    otherwise (i.e. if `text_or_file` is a multi-line string)
218
    `text_or_file` is returned.
219
    """
Eckhart Arnold's avatar
Eckhart Arnold committed
220
    if is_filename(text_or_file):
221
222
223
224
        try:
            with open(text_or_file, encoding="utf-8") as f:
                content = f.read()
            return content
225
        except FileNotFoundError:
226
            if re.fullmatch(r'[\w/:. \\]+', text_or_file):
227
228
                raise FileNotFoundError('Not a valid file: ' + text_or_file + '!\n(Add "\\n" '
                                        'to distinguish source data from a file name.)')
229
230
            else:
                return text_or_file
231
232
233
234
    else:
        return text_or_file


235
def is_python_code(text_or_file) -> bool:
236
237
238
    """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
239
    if is_filename(text_or_file):
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
        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()


259
260
def smart_list(arg) -> list:
    """Returns the argument as list, depending on its type and content.
261
262
263
264
    
    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.
265
266
267
268
269
270
271
    >>> smart_list('1; 2, 3; 4')
    ['1', '2, 3', '4']
    >>> smart_list('2, 3')
    ['2', '3']
    >>> smart_list('a b cd')
    ['a', 'b', 'cd']

272
273
    If the argument is a collection other than a string, it will be
    returned as is, e.g.
274
275
276
277
278
    >>> smart_list((1, 2, 3))
    (1, 2, 3)
    >>> smart_list({1, 2, 3})
    {1, 2, 3}

279
280
    If the argument is another iterable than a collection, it will
    be converted into a list, e.g.
281
282
283
    >>> smart_list(i for i in {1,2,3})
    [1, 2, 3]

284
285
    Finally, if none of the above is true, the argument will be 
    wrapped in a list and returned, e.g.
286
287
    >>> smart_list(125)
    [125]
288
289
290
291
292
    """
    if isinstance(arg, str):
        for delimiter in (';', ','):
            lst = arg.split(delimiter)
            if len(lst) > 1:
293
294
                return [s.strip() for s in lst]
        return [s.strip() for s in arg.strip().split(' ')]
295
296
    elif isinstance(arg, collections.abc.Container):
        return arg
297
    elif isinstance(arg, collections.abc.Iterable):
298
        return list(arg)
299
300
301
302
    else:
        return [arg]


303
304
305
306
307
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:
308
309
    >>> expand_table({"a, b": 1, ('d','e','f'):5, "c":3})
    {'a': 1, 'b': 1, 'd': 5, 'e': 5, 'f': 5, 'c': 3}
310
311
312
313
314
    """
    expanded_table = {}
    keys = list(compact_table.keys())
    for key in keys:
        value = compact_table[key]
315
        for k in smart_list(key):
316
            if k in expanded_table:
317
                raise KeyError('Key "%s" used more than once in compact table!' % key)
318
            expanded_table[k] = value
319
    return expanded_table
320
321


322
def sane_parser_name(name) -> bool:
323
    """Checks whether given name is an acceptable parser name. Parser names
324
    must not be preceded or succeeded by a double underscore '__'!
325
    """
di68kap's avatar
di68kap committed
326
327
328
    return name and name[:2] != '__' and name[-2:] != '__'


329
330
331
332
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.
di68kap's avatar
di68kap committed
333
334
335
336
337
338
    """
    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?
339
340
341
342
343
344
    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:
345
            raise ValueError("Ambiguous matches for %s : %s" %
346
347
348
349
                             (str(catch_obj_regex), str(matches)))
        return namespace[matches[0]] if matches else None
    else:
        return namespace
350
351


352
353
354
355
def identity(anything: Any) -> Any:
    return anything


356
357
358
359
360
361
362
363
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