|
1
|
|
|
from coala_utils.string_processing.StringConverter import StringConverter |
|
2
|
|
|
from coala_utils.string_processing import (unescape, convert_to_raw, |
|
3
|
|
|
position_is_escaped, |
|
4
|
|
|
unescaped_rstrip) |
|
5
|
|
|
|
|
6
|
|
|
|
|
7
|
|
|
class LineParser: |
|
8
|
|
|
|
|
9
|
|
|
def __init__(self, |
|
10
|
|
|
key_value_delimiters=('=',), |
|
11
|
|
|
comment_separators=('#',), |
|
12
|
|
|
key_delimiters=(',', ' '), |
|
13
|
|
|
section_name_surroundings=None, |
|
14
|
|
|
section_override_delimiters=(".",)): |
|
15
|
|
|
""" |
|
16
|
|
|
Creates a new line parser. Please note that no delimiter or separator |
|
17
|
|
|
may be an "o" or you may encounter undefined behaviour with the |
|
18
|
|
|
escapes. |
|
19
|
|
|
|
|
20
|
|
|
:param key_value_delimiters: Delimiters that delimit a key from |
|
21
|
|
|
a value |
|
22
|
|
|
:param comment_separators: Used to initiate a comment |
|
23
|
|
|
:param key_delimiters: Delimiters between several keys |
|
24
|
|
|
:param section_name_surroundings: Dictionary, e.g. {"[", "]"} means a |
|
25
|
|
|
section name is surrounded by []. |
|
26
|
|
|
If None, {"[": "]"} is used as |
|
27
|
|
|
default. |
|
28
|
|
|
:param section_override_delimiters: Delimiter for a section override. |
|
29
|
|
|
E.g. "." would mean that |
|
30
|
|
|
section.key is a possible key that |
|
31
|
|
|
puts the key into the section |
|
32
|
|
|
"section" despite of the current |
|
33
|
|
|
section. |
|
34
|
|
|
""" |
|
35
|
|
|
section_name_surroundings = ( |
|
36
|
|
|
{"[": "]"} if section_name_surroundings is None |
|
37
|
|
|
else section_name_surroundings) |
|
38
|
|
|
|
|
39
|
|
|
self.key_value_delimiters = key_value_delimiters |
|
40
|
|
|
self.comment_separators = comment_separators |
|
41
|
|
|
self.key_delimiters = key_delimiters |
|
42
|
|
|
self.section_name_surroundings = section_name_surroundings |
|
43
|
|
|
self.section_override_delimiters = section_override_delimiters |
|
44
|
|
|
|
|
45
|
|
|
def parse(self, line): |
|
46
|
|
|
""" |
|
47
|
|
|
Note that every value in the returned touple *besides the value* is |
|
48
|
|
|
unescaped. This is so since the value is meant to be put into a Setting |
|
49
|
|
|
later thus the escapes may be needed there. |
|
50
|
|
|
|
|
51
|
|
|
:param line: the line to parse |
|
52
|
|
|
:return: section_name (empty string if it's no section name), |
|
53
|
|
|
[(section_override, key), ...], value, comment |
|
54
|
|
|
""" |
|
55
|
|
|
line, comment = self.__separate_by_first_occurrence( |
|
56
|
|
|
line, |
|
57
|
|
|
self.comment_separators) |
|
58
|
|
|
comment = unescape(comment) |
|
59
|
|
|
if line == "": |
|
60
|
|
|
return '', [], '', comment |
|
61
|
|
|
|
|
62
|
|
|
section_name = unescape(self.__get_section_name(line)) |
|
63
|
|
|
if section_name != '': |
|
64
|
|
|
return section_name, [], '', comment |
|
65
|
|
|
|
|
66
|
|
|
# Escapes in value might be needed by the bears |
|
67
|
|
|
keys, value = self.__extract_keys_and_value(line) |
|
68
|
|
|
|
|
69
|
|
|
# Add all the delimiters that stored as tuples |
|
70
|
|
|
all_delimiters = self.key_value_delimiters |
|
71
|
|
|
all_delimiters += self.key_delimiters |
|
72
|
|
|
all_delimiters += self.comment_separators |
|
73
|
|
|
all_delimiters += self.section_override_delimiters |
|
74
|
|
|
all_delimiters = "".join(all_delimiters) |
|
75
|
|
|
|
|
76
|
|
|
# Add all keys and values in section_name_surroundings, which is |
|
77
|
|
|
# stored as a dict |
|
78
|
|
|
all_delimiters += "".join(self.section_name_surroundings.keys()) |
|
79
|
|
|
all_delimiters += "".join(self.section_name_surroundings.values()) |
|
80
|
|
|
|
|
81
|
|
|
value = convert_to_raw(value, all_delimiters) |
|
82
|
|
|
|
|
83
|
|
|
key_touples = [] |
|
84
|
|
|
for key in keys: |
|
85
|
|
|
key = convert_to_raw(key, all_delimiters) |
|
86
|
|
|
section, key = self.__separate_by_first_occurrence( |
|
87
|
|
|
key, |
|
88
|
|
|
self.section_override_delimiters, |
|
89
|
|
|
True, |
|
90
|
|
|
True) |
|
91
|
|
|
key_touples.append((unescape(section), unescape(key))) |
|
92
|
|
|
|
|
93
|
|
|
return '', key_touples, value, comment |
|
94
|
|
|
|
|
95
|
|
|
@staticmethod |
|
96
|
|
|
def __separate_by_first_occurrence(string, |
|
97
|
|
|
delimiters, |
|
98
|
|
|
strip_delim=False, |
|
99
|
|
|
return_second_part_nonempty=False): |
|
100
|
|
|
""" |
|
101
|
|
|
Separates a string by the first of all given delimiters. Any whitespace |
|
102
|
|
|
characters will be stripped away from the parts. |
|
103
|
|
|
|
|
104
|
|
|
:param string: The string to separate. |
|
105
|
|
|
:param delimiters: The delimiters. |
|
106
|
|
|
:param strip_delim: Strips the delimiter from the |
|
107
|
|
|
result if true. |
|
108
|
|
|
:param return_second_part_nonempty: If no delimiter is found and this |
|
109
|
|
|
is true the contents of the string |
|
110
|
|
|
will be returned in the second part |
|
111
|
|
|
of the touple instead of the first |
|
112
|
|
|
one. |
|
113
|
|
|
:return: (first_part, second_part) |
|
114
|
|
|
""" |
|
115
|
|
|
temp_string = string.replace("\\\\", "oo") |
|
116
|
|
|
i = temp_string.find("\\") |
|
117
|
|
|
while i != -1: |
|
118
|
|
|
temp_string = temp_string[:i] + "oo" + temp_string[i+2:] |
|
119
|
|
|
i = temp_string.find("\\", i+2) |
|
120
|
|
|
|
|
121
|
|
|
delim_pos = len(string) |
|
122
|
|
|
used_delim = "" |
|
123
|
|
|
for delim in delimiters: |
|
124
|
|
|
pos = temp_string.find(delim) |
|
125
|
|
|
if 0 <= pos < delim_pos: |
|
126
|
|
|
delim_pos = pos |
|
127
|
|
|
used_delim = delim |
|
128
|
|
|
|
|
129
|
|
|
if return_second_part_nonempty and delim_pos == len(string): |
|
130
|
|
|
return "", string.strip(" \n") |
|
131
|
|
|
|
|
132
|
|
|
first_part = string[:delim_pos] |
|
133
|
|
|
second_part = string[delim_pos + ( |
|
134
|
|
|
len(used_delim) if strip_delim else 0):] |
|
135
|
|
|
|
|
136
|
|
|
if not position_is_escaped(second_part, len(second_part) - 1): |
|
137
|
|
|
first_part = unescaped_rstrip(first_part) |
|
138
|
|
|
second_part = unescaped_rstrip(second_part) |
|
139
|
|
|
|
|
140
|
|
|
return (first_part.lstrip().rstrip("\n"), |
|
141
|
|
|
second_part.lstrip().rstrip("\n")) |
|
142
|
|
|
|
|
143
|
|
|
def __get_section_name(self, line): |
|
144
|
|
|
for begin, end in self.section_name_surroundings.items(): |
|
145
|
|
|
if (line[0:len(begin)] == begin and |
|
146
|
|
|
line[len(line) - len(end):len(line)] == end): |
|
147
|
|
|
return line[len(begin):len(line) - len(end)].strip(" \n") |
|
148
|
|
|
|
|
149
|
|
|
return '' |
|
150
|
|
|
|
|
151
|
|
|
def __extract_keys_and_value(self, line): |
|
152
|
|
|
key_part, value = self.__separate_by_first_occurrence( |
|
153
|
|
|
line, |
|
154
|
|
|
self.key_value_delimiters, |
|
155
|
|
|
True, |
|
156
|
|
|
True) |
|
157
|
|
|
keys = list(StringConverter( |
|
158
|
|
|
key_part, |
|
159
|
|
|
list_delimiters=self.key_delimiters).__iter__( |
|
160
|
|
|
remove_backslashes=False)) |
|
161
|
|
|
|
|
162
|
|
|
return keys, value |
|
163
|
|
|
|
coala-analyzer /
coala
