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 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311
|
"""Very low-level nginx config parser based on pyparsing."""
# Forked from https://github.com/fatiherikli/nginxparser (MIT Licensed)
import copy
import logging
import operator
import typing
from typing import Any
from typing import IO
from typing import Iterable
from typing import Iterator
from typing import List
from typing import overload
from typing import SupportsIndex
from typing import Tuple
from typing import Union
from pyparsing import Combine
from pyparsing import Forward
from pyparsing import Group
from pyparsing import Literal
from pyparsing import Optional
from pyparsing import ParseResults
from pyparsing import QuotedString
from pyparsing import Regex
from pyparsing import restOfLine
from pyparsing import stringEnd
from pyparsing import White
from pyparsing import ZeroOrMore
logger = logging.getLogger(__name__)
class RawNginxParser:
# pylint: disable=pointless-statement
"""A class that parses nginx configuration with pyparsing."""
# constants
space = Optional(White(ws=' \t\r\n\u00a0')).leaveWhitespace()
required_space = White(ws=' \t\r\n\u00a0').leaveWhitespace()
left_bracket = Literal("{").suppress()
right_bracket = space + Literal("}").suppress()
semicolon = Literal(";").suppress()
dquoted = QuotedString('"', multiline=True, unquoteResults=False, escChar='\\')
squoted = QuotedString("'", multiline=True, unquoteResults=False, escChar='\\')
quoted = dquoted | squoted
head_tokenchars = Regex(r"(\$\{)|[^{};\s'\"]") # if (last_space)
tail_tokenchars = Regex(r"(\$\{)|[^{;\s]") # else
tokenchars = Combine(head_tokenchars + ZeroOrMore(tail_tokenchars))
paren_quote_extend = Combine(quoted + Literal(')') + ZeroOrMore(tail_tokenchars))
# note: ')' allows extension, but then we fall into else, not last_space.
token = paren_quote_extend | tokenchars | quoted
whitespace_token_group = space + token + ZeroOrMore(required_space + token) + space
assignment = whitespace_token_group + semicolon
comment = space + Literal('#') + restOfLine
block = Forward()
# order matters! see issue 518, and also http { # server { \n}
contents = Group(comment) | Group(block) | Group(assignment)
block_begin = Group(whitespace_token_group)
block_innards = Group(ZeroOrMore(contents) + space).leaveWhitespace()
block << block_begin + left_bracket + block_innards + right_bracket
script = ZeroOrMore(contents) + space + stringEnd
script.parseWithTabs().leaveWhitespace()
def __init__(self, source: str) -> None:
self.source = source
def parse(self) -> ParseResults:
"""Returns the parsed tree."""
return self.script.parseString(self.source)
def as_list(self) -> List[Any]:
"""Returns the parsed tree as a list."""
return self.parse().asList()
class RawNginxDumper:
"""A class that dumps nginx configuration from the provided tree."""
def __init__(self, blocks: List[Any]) -> None:
self.blocks = blocks
def __iter__(self, blocks: typing.Optional[List[Any]] = None) -> Iterator[str]:
"""Iterates the dumped nginx content."""
blocks = blocks or self.blocks
for b0 in blocks:
if isinstance(b0, str):
yield b0
continue
item = copy.deepcopy(b0)
if spacey(item[0]):
yield item.pop(0) # indentation
if not item:
continue
if isinstance(item[0], list): # block
yield "".join(item.pop(0)) + '{'
for parameter in item.pop(0):
yield from self.__iter__([parameter]) # negate "for b0 in blocks"
yield '}'
else: # not a block - list of strings
semicolon = ";"
if isinstance(item[0], str) and item[0].strip() == '#': # comment
semicolon = ""
yield "".join(item) + semicolon
def __str__(self) -> str:
"""Return the parsed block as a string."""
return ''.join(self)
def spacey(x: Any) -> bool:
"""Is x an empty string or whitespace?"""
return (isinstance(x, str) and x.isspace()) or x == ''
class UnspacedList(List[Any]):
"""Wrap a list [of lists], making any whitespace entries magically invisible"""
def __init__(self, list_source: Iterable[Any]) -> None:
# ensure our argument is not a generator, and duplicate any sublists
self.spaced = copy.deepcopy(list(list_source))
self.dirty = False
# Turn self into a version of the source list that has spaces removed
# and all sub-lists also UnspacedList()ed
super().__init__(list_source)
for i, entry in reversed(list(enumerate(self))):
if isinstance(entry, list):
sublist = UnspacedList(entry)
super().__setitem__(i, sublist)
self.spaced[i] = sublist.spaced
elif spacey(entry):
# don't delete comments
if "#" not in self[:i]:
super().__delitem__(i)
@overload
def _coerce(self, inbound: None) -> Tuple[None, None]: ...
@overload
def _coerce(self, inbound: str) -> Tuple[str, str]: ...
@overload
def _coerce(self, inbound: List[Any]) -> Tuple["UnspacedList", List[Any]]: ...
def _coerce(self, inbound: Any) -> Tuple[Any, Any]:
"""
Coerce some inbound object to be appropriately usable in this object
:param inbound: string or None or list or UnspacedList
:returns: (coerced UnspacedList or string or None, spaced equivalent)
:rtype: tuple
"""
if not isinstance(inbound, list): # str or None
return inbound, inbound
else:
if not hasattr(inbound, "spaced"):
inbound = UnspacedList(inbound)
return inbound, inbound.spaced
def insert(self, i: "SupportsIndex", x: Any) -> None:
"""Insert object before index."""
idx = operator.index(i)
item, spaced_item = self._coerce(x)
slicepos = self._spaced_position(idx) if idx < len(self) else len(self.spaced)
self.spaced.insert(slicepos, spaced_item)
if not spacey(item):
super().insert(idx, item)
self.dirty = True
def append(self, x: Any) -> None:
"""Append object to the end of the list."""
item, spaced_item = self._coerce(x)
self.spaced.append(spaced_item)
if not spacey(item):
super().append(item)
self.dirty = True
def extend(self, x: Any) -> None:
"""Extend list by appending elements from the iterable."""
item, spaced_item = self._coerce(x)
self.spaced.extend(spaced_item)
super().extend(item)
self.dirty = True
def __add__(self, other: List[Any]) -> "UnspacedList":
new_list = copy.deepcopy(self)
new_list.extend(other)
new_list.dirty = True
return new_list
def pop(self, *args: Any, **kwargs: Any) -> None:
"""Function pop() is not implemented for UnspacedList"""
raise NotImplementedError("UnspacedList.pop() not yet implemented")
def remove(self, *args: Any, **kwargs: Any) -> None:
"""Function remove() is not implemented for UnspacedList"""
raise NotImplementedError("UnspacedList.remove() not yet implemented")
def reverse(self) -> None:
"""Function reverse() is not implemented for UnspacedList"""
raise NotImplementedError("UnspacedList.reverse() not yet implemented")
def sort(self, *_args: Any, **_kwargs: Any) -> None:
"""Function sort() is not implemented for UnspacedList"""
raise NotImplementedError("UnspacedList.sort() not yet implemented")
def __setslice__(self, *args: Any, **kwargs: Any) -> None:
raise NotImplementedError("Slice operations on UnspacedLists not yet implemented")
def __setitem__(self, i: Union["SupportsIndex", slice], value: Any) -> None:
if isinstance(i, slice):
raise NotImplementedError("Slice operations on UnspacedLists not yet implemented")
item, spaced_item = self._coerce(value)
self.spaced.__setitem__(self._spaced_position(i), spaced_item)
if not spacey(item):
super().__setitem__(i, item)
self.dirty = True
def __delitem__(self, i: Union["SupportsIndex", slice]) -> None:
if isinstance(i, slice):
raise NotImplementedError("Slice operations on UnspacedLists not yet implemented")
self.spaced.__delitem__(self._spaced_position(i))
super().__delitem__(i)
self.dirty = True
def __deepcopy__(self, memo: Any) -> "UnspacedList":
new_spaced = copy.deepcopy(self.spaced, memo=memo)
new_list = UnspacedList(new_spaced)
new_list.dirty = self.dirty
return new_list
def is_dirty(self) -> bool:
"""Recurse through the parse tree to figure out if any sublists are dirty"""
if self.dirty:
return True
return any((isinstance(x, UnspacedList) and x.is_dirty() for x in self))
def _spaced_position(self, idx: "SupportsIndex") -> int:
"""Convert from indexes in the unspaced list to positions in the spaced one"""
int_idx = operator.index(idx)
pos = spaces = 0
# Normalize indexes like list[-1] etc, and save the result
if int_idx < 0:
int_idx = len(self) + int_idx
if not 0 <= int_idx < len(self):
raise IndexError("list index out of range")
int_idx0 = int_idx
# Count the number of spaces in the spaced list before int_idx in the unspaced one
while int_idx != -1:
if spacey(self.spaced[pos]):
spaces += 1
else:
int_idx -= 1
pos += 1
return int_idx0 + spaces
# Shortcut functions to respect Python's serialization interface
# (like pyyaml, picker or json)
def loads(source: str) -> UnspacedList:
"""Parses from a string.
:param str source: The string to parse
:returns: The parsed tree
:rtype: list
"""
return UnspacedList(RawNginxParser(source).as_list())
def load(file_: IO[Any]) -> UnspacedList:
"""Parses from a file.
:param file file_: The file to parse
:returns: The parsed tree
:rtype: list
"""
return loads(file_.read())
def dumps(blocks: UnspacedList) -> str:
"""Dump to a Unicode string.
:param UnspacedList blocks: The parsed tree
:rtype: str
"""
return str(RawNginxDumper(blocks.spaced))
def dump(blocks: UnspacedList, file_: IO[Any]) -> None:
"""Dump to a file.
:param UnspacedList blocks: The parsed tree
:param IO[Any] file_: The file stream to dump to. It must be opened with
Unicode encoding.
:rtype: None
"""
file_.write(dumps(blocks))
|