|
1
|
|
|
"""Converter classes for extensions to builtin types.""" |
|
2
|
|
|
|
|
3
|
1 |
|
import re |
|
4
|
|
|
|
|
5
|
1 |
|
from .standard import String, Integer, Float, Boolean |
|
6
|
1 |
|
from .containers import Dictionary, List |
|
7
|
1 |
|
from ._representers import LiteralString |
|
8
|
|
|
|
|
9
|
|
|
|
|
10
|
|
|
# NULLABLE BUILTINS ############################################################ |
|
11
|
|
|
|
|
12
|
|
|
|
|
13
|
1 |
|
class NullableString(String): |
|
14
|
|
|
"""Converter for the `str` type with `None` as default.""" |
|
15
|
|
|
|
|
16
|
1 |
|
DEFAULT = None |
|
17
|
|
|
|
|
18
|
|
|
|
|
19
|
1 |
|
class NullableInteger(Integer): |
|
20
|
|
|
"""Converter for the `int` type with `None` as default.""" |
|
21
|
|
|
|
|
22
|
1 |
|
DEFAULT = None |
|
23
|
|
|
|
|
24
|
|
|
|
|
25
|
1 |
|
class NullableFloat(Float): |
|
26
|
|
|
"""Converter for the `float` type with `None` as default.""" |
|
27
|
|
|
|
|
28
|
1 |
|
DEFAULT = None |
|
29
|
|
|
|
|
30
|
|
|
|
|
31
|
1 |
|
class NullableBoolean(Boolean): |
|
32
|
|
|
"""Converter for the `bool` type with `None` as default.""" |
|
33
|
|
|
|
|
34
|
1 |
|
DEFAULT = None |
|
35
|
|
|
|
|
36
|
|
|
|
|
37
|
|
|
# CUSTOM TYPES ################################################################# |
|
38
|
|
|
|
|
39
|
|
|
|
|
40
|
1 |
|
class Markdown(String): |
|
41
|
|
|
"""Converter for a `str` type that contains Markdown.""" |
|
42
|
|
|
|
|
43
|
1 |
|
REGEX_MARKDOWN_SPACES = re.compile(r""" |
|
44
|
|
|
|
|
45
|
|
|
([^\n ]) # any character but a newline or space |
|
46
|
|
|
|
|
47
|
|
|
(\ ?\n) # optional space + single newline |
|
48
|
|
|
|
|
49
|
|
|
(?! # none of the following: |
|
50
|
|
|
|
|
51
|
|
|
(?:\s) # whitespace |
|
52
|
|
|
| |
|
53
|
|
|
(?:[-+*]\s) # unordered list separator + whitespace |
|
54
|
|
|
| |
|
55
|
|
|
(?:\d+\.\s) # number + period + whitespace |
|
56
|
|
|
|
|
57
|
|
|
) |
|
58
|
|
|
|
|
59
|
|
|
([^\n]) # any character but a newline |
|
60
|
|
|
|
|
61
|
|
|
""", re.VERBOSE | re.IGNORECASE) |
|
62
|
|
|
|
|
63
|
|
|
# based on: http://en.wikipedia.org/wiki/Sentence_boundary_disambiguation |
|
64
|
1 |
|
REGEX_SENTENCE_BOUNDARIES = re.compile(r""" |
|
65
|
|
|
|
|
66
|
|
|
( # one of the following: |
|
67
|
|
|
|
|
68
|
|
|
(?<=[a-z)][.?!]) # lowercase letter + punctuation |
|
69
|
|
|
| |
|
70
|
|
|
(?<=[a-z0-9][.?!]\") # lowercase letter/number + punctuation + quote |
|
71
|
|
|
|
|
72
|
|
|
) |
|
73
|
|
|
|
|
74
|
|
|
(\s) # any whitespace |
|
75
|
|
|
|
|
76
|
|
|
(?=\"?[A-Z]) # optional quote + an upppercase letter |
|
77
|
|
|
|
|
78
|
|
|
""", re.VERBOSE) |
|
79
|
|
|
|
|
80
|
1 |
|
@classmethod |
|
81
|
|
|
def to_value(cls, obj): |
|
82
|
|
|
"""Join non-meaningful line breaks.""" |
|
83
|
1 |
|
value = String.to_value(obj) |
|
84
|
1 |
|
return cls._join(value) |
|
85
|
|
|
|
|
86
|
1 |
|
@classmethod |
|
87
|
|
|
def to_data(cls, obj): |
|
88
|
|
|
"""Break a string at sentences and dump as a literal string.""" |
|
89
|
1 |
|
value = String.to_value(obj) |
|
90
|
1 |
|
data = String.to_data(value) |
|
91
|
1 |
|
split = cls._split(data) |
|
92
|
1 |
|
return LiteralString(split) |
|
93
|
|
|
|
|
94
|
1 |
|
@classmethod |
|
95
|
|
|
def _join(cls, text): |
|
96
|
|
|
r"""Convert single newlines (ignored by Markdown) to spaces. |
|
97
|
|
|
|
|
98
|
|
|
>>> Markdown._join("abc\n123") |
|
99
|
|
|
'abc 123' |
|
100
|
|
|
|
|
101
|
|
|
>>> Markdown._join("abc\n\n123") |
|
102
|
|
|
'abc\n\n123' |
|
103
|
|
|
|
|
104
|
|
|
>>> Markdown._join("abc \n123") |
|
105
|
|
|
'abc 123' |
|
106
|
|
|
|
|
107
|
|
|
""" |
|
108
|
1 |
|
return cls.REGEX_MARKDOWN_SPACES.sub(r'\1 \3', text).strip() |
|
109
|
|
|
|
|
110
|
1 |
|
@classmethod |
|
111
|
1 |
|
def _split(cls, text, end='\n'): |
|
112
|
|
|
r"""Replace sentence boundaries with newlines and append a newline. |
|
113
|
|
|
|
|
114
|
|
|
:param text: string to line break at sentences |
|
115
|
|
|
:param end: appended to the end of the update text |
|
116
|
|
|
|
|
117
|
|
|
>>> Markdown._split("Hello, world!", end='') |
|
118
|
|
|
'Hello, world!' |
|
119
|
|
|
|
|
120
|
|
|
>>> Markdown._split("Hello, world! How are you? I'm fine. Good.") |
|
121
|
|
|
"Hello, world!\nHow are you?\nI'm fine.\nGood.\n" |
|
122
|
|
|
|
|
123
|
|
|
""" |
|
124
|
1 |
|
stripped = text.strip() |
|
125
|
1 |
|
if stripped: |
|
126
|
1 |
|
return cls.REGEX_SENTENCE_BOUNDARIES.sub('\n', stripped) + end |
|
127
|
|
|
else: |
|
128
|
1 |
|
return '' |
|
129
|
|
|
|
|
130
|
|
|
|
|
131
|
|
|
# CUSTOM CONTAINERS ############################################################ |
|
132
|
|
|
|
|
133
|
|
|
|
|
134
|
1 |
|
class AttributeDictionary(Dictionary): |
|
135
|
|
|
"""Dictionary converter with keys available as attributes.""" |
|
136
|
|
|
|
|
137
|
1 |
|
def __init__(self, *args, **kwargs): |
|
138
|
1 |
|
super().__init__(*args, **kwargs) |
|
139
|
1 |
|
self.__dict__ = self |
|
140
|
|
|
|
|
141
|
1 |
|
@classmethod |
|
142
|
|
|
def create_default(cls): |
|
143
|
|
|
"""Create an uninitialized object with keys as attributes.""" |
|
144
|
1 |
|
if cls is AttributeDictionary: |
|
145
|
1 |
|
msg = "AttributeDictionary class must be subclassed to use" |
|
146
|
1 |
|
raise NotImplementedError(msg) |
|
147
|
|
|
|
|
148
|
1 |
|
obj = cls.__new__(cls) |
|
149
|
1 |
|
obj.__dict__ = obj |
|
150
|
1 |
|
return obj |
|
151
|
|
|
|
|
152
|
|
|
|
|
153
|
1 |
|
class SortedList(List): |
|
154
|
|
|
"""List converter that is sorted on disk.""" |
|
155
|
|
|
|
|
156
|
1 |
|
@classmethod |
|
157
|
|
|
def create_default(cls): |
|
158
|
|
|
"""Create an uninitialized object.""" |
|
159
|
1 |
|
if cls is SortedList: |
|
160
|
1 |
|
msg = "SortedList class must be subclassed to use" |
|
161
|
1 |
|
raise NotImplementedError(msg) |
|
162
|
1 |
|
if not cls.item_type: |
|
163
|
1 |
|
msg = "SortedList subclass must specify item type" |
|
164
|
1 |
|
raise NotImplementedError(msg) |
|
165
|
|
|
|
|
166
|
1 |
|
return cls.__new__(cls) |
|
167
|
|
|
|
|
168
|
1 |
|
@classmethod |
|
169
|
|
|
def to_data(cls, obj): |
|
170
|
|
|
"""Convert all attribute values for optimal dumping to YAML.""" |
|
171
|
1 |
|
value = cls.to_value(obj) |
|
172
|
|
|
|
|
173
|
1 |
|
data = [] |
|
174
|
|
|
|
|
175
|
1 |
|
for item in sorted(value): |
|
176
|
1 |
|
data.append(cls.item_type.to_data(item)) # pylint: disable=no-member |
|
177
|
|
|
|
|
178
|
|
|
return data |
|
179
|
|
|
|
jacebrowning /
yorm
