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
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
|
import dataclasses
import itertools
import sys
from typing import Optional, TYPE_CHECKING, Iterable
if TYPE_CHECKING:
from typing import Self
from .parsing import Deb822Element
_DATA_CLASS_OPTIONAL_ARGS = {}
if sys.version_info >= (3, 10):
# The `slots` feature greatly reduces the memory usage by avoiding the `__dict__`
# instance. But at the end of the day, performance is "nice to have" for this
# feature and all current consumers are at Python 3.12 (except the CI tests...)
_DATA_CLASS_OPTIONAL_ARGS["slots"] = True
@dataclasses.dataclass(frozen=True, **_DATA_CLASS_OPTIONAL_ARGS)
class Position:
"""Describes a "cursor" position inside a file
It consists of a line position (0-based line number) and a cursor position. This is modelled
after the "Position" in Language Server Protocol (LSP).
"""
line_position: int
"""Describes the line position as a 0-based line number
See line_number if you want a human-readable line number
"""
cursor_position: int
"""Describes a cursor position ("between two characters") or a character offset.
When this value is 0, the position is at the start of a line. When it is 1, then
the position is between the first and the second character (etc.).
"""
@property
def line_number(self) -> int:
"""The line number as human would count it"""
return self.line_position + 1
def relative_to(self, new_base: "Position") -> "Position":
"""Offsets the position relative to another position
This is useful to avoid the `position_in_file()` method by caching where
the parents position and then for its children you use `range_in_parent()`
plus `relative_to()` to rebase the range.
>>> parent: Locatable = ... # doctest: +SKIP
>>> children: Iterable[Locatable] = ... # doctest: +SKIP
>>> # This will expensive
>>> parent_pos = parent.position_in_file( # doctest: +SKIP
... skip_leading_comments=False
... )
>>> for child in children: # doctest: +SKIP
... child_pos = child.position_in_parent()
... # Avoid a position_in_file() for each child
... child_pos_in_file = child_pos.relative_to(parent_pos)
... ... # Use the child_pos_in_file for something
:param new_base: The position that should have been the origin rather than
(0, 0).
:returns: The range offset relative to the base position.
"""
if self.line_position == 0 and self.cursor_position == 0:
return new_base
if new_base.line_position == 0 and new_base.cursor_position == 0:
return self
if self.line_position == 0:
line_number = new_base.line_position
line_char_offset = new_base.cursor_position + self.cursor_position
else:
line_number = self.line_position + new_base.line_position
line_char_offset = self.cursor_position
return Position(
line_number,
line_char_offset,
)
@dataclasses.dataclass(frozen=True, **_DATA_CLASS_OPTIONAL_ARGS)
class Range:
"""Describes a range inside a file
This can be useful to describe things like "from line 4, cursor position 2
to line 7 to cursor position 10". When describing a full line including the
newline, use line N, cursor position 0 to line N+1. cursor position 0.
It is also used to denote the size of objects (in that case, the start position
is set to START_POSITION as a convention if the precise location is not
specified).
This is modelled after the "Range" in Language Server Protocol (LSP).
"""
start_pos: Position
end_pos: Position
@property
def start_line_position(self) -> int:
"""Describes the start line position as a 0-based line number
See start_line_number if you want a human-readable line number
"""
return self.start_pos.line_position
@property
def start_cursor_position(self) -> int:
"""Describes the starting cursor position
When this value is 0, the position is at the start of a line. When it is 1, then
the position is between the first and the second character (etc.).
"""
return self.start_pos.cursor_position
@property
def start_line_number(self) -> int:
"""The start line number as human would count it"""
return self.start_pos.line_number
@property
def end_line_position(self) -> int:
"""Describes the end line position as a 0-based line number
See end_line_number if you want a human-readable line number
"""
return self.end_pos.line_position
@property
def end_line_number(self) -> int:
"""The end line number as human would count it"""
return self.end_pos.line_number
@property
def end_cursor_position(self) -> int:
"""Describes the end cursor position
When this value is 0, the position is at the start of a line. When it is 1, then
the position is between the first and the second character (etc.).
"""
return self.end_pos.cursor_position
@property
def line_count(self) -> int:
"""The number of lines (newlines) spanned by this range.
Will be zero when the range fits inside one line.
"""
return self.end_line_position - self.start_line_position
@classmethod
def between(cls, a: Position, b: Position) -> "Self":
"""Computes the range between two positions
Unlike the constructor, this will always create a "positive" range.
That is, the "earliest" position will always be the start position
regardless of the order they were passed to `between`. When using
the Range constructor, you have freedom to do "inverse" ranges
in case that is ever useful
"""
if a.line_position > b.line_position or (
a.line_position == b.line_position and a.cursor_position > b.cursor_position
):
# Order swap, so `a` is always the earliest position
a, b = b, a
return cls(
a,
b,
)
def relative_to(self, new_base: Position) -> "Range":
"""Offsets the range relative to another position
This is useful to avoid the `position_in_file()` method by caching where
the parents position and then for its children you use `range_in_parent()`
plus `relative_to()` to rebase the range.
>>> parent: Locatable = ... # doctest: +SKIP
>>> children: Iterable[Locatable] = ... # doctest: +SKIP
>>> # This will expensive
>>> parent_pos = parent.position_in_file( # doctest: +SKIP
... skip_leading_comments=False
... )
>>> for child in children: # doctest: +SKIP
... child_range = child.range_in_parent()
... # Avoid a position_in_file() for each child
... child_range_in_file = child_range.relative_to(parent_pos)
... ... # Use the child_range_in_file for something
:param new_base: The position that should have been the origin rather than
(0, 0).
:returns: The range offset relative to the base position.
"""
if new_base == START_POSITION:
return self
return Range(
self.start_pos.relative_to(new_base),
self.end_pos.relative_to(new_base),
)
def as_size(self) -> "Range":
"""Reduces the range to a "size"
The returned range will always have its start position to (0, 0) and
its end position shifted accordingly if it was not already based at
(0, 0).
The original range is not mutated and, if it is already at (0, 0), the
method will just return it as-is.
"""
if self.start_pos == START_POSITION:
return self
line_count = self.line_count
if line_count:
new_end_cursor_position = self.end_cursor_position
else:
delta = self.end_cursor_position - self.start_cursor_position
new_end_cursor_position = delta
return Range(
START_POSITION,
Position(
line_count,
new_end_cursor_position,
),
)
@classmethod
def from_position_and_size(cls, base: Position, size: "Range") -> "Self":
"""Compute a range from a position and the size of another range
This provides you with a range starting at the base position that has
the same effective span as the size parameter.
:param base: The desired starting position
:param size: A range, which will be used as a size (that is, it will
be reduced to a size via the `as_size()` method) for the resulting
range
:returns: A range at the provided base position that has the size of
the provided range.
"""
line_position = base.line_position
cursor_position = base.cursor_position
size_rebased = size.as_size()
lines = size_rebased.line_count
if lines:
line_position += lines
cursor_position = size_rebased.end_cursor_position
else:
delta = (
size_rebased.end_cursor_position - size_rebased.start_cursor_position
)
cursor_position += delta
return cls(
base,
Position(
line_position,
cursor_position,
),
)
@classmethod
def from_position_and_sizes(
cls, base: Position, sizes: Iterable["Range"]
) -> "Self":
"""Compute a range from a position and the size of number of ranges
:param base: The desired starting position
:param sizes: All the ranges that combined makes up the size of the
desired position. Note that order can affect the end result. Particularly
the end character offset gets reset everytime a size spans a line.
:returns: A range at the provided base position that has the size of
the provided range.
"""
line_position = base.line_position
cursor_position = base.cursor_position
for size in sizes:
size_rebased = size.as_size()
lines = size_rebased.line_count
if lines:
line_position += lines
cursor_position = size_rebased.end_cursor_position
else:
delta = (
size_rebased.end_cursor_position
- size_rebased.start_cursor_position
)
cursor_position += delta
return cls(
base,
Position(
line_position,
cursor_position,
),
)
START_POSITION = Position(0, 0)
SECOND_CHAR_POS = Position(0, 1)
SECOND_LINE_POS = Position(1, 0)
ONE_CHAR_RANGE = Range.between(START_POSITION, SECOND_CHAR_POS)
ONE_LINE_RANGE = Range.between(START_POSITION, SECOND_LINE_POS)
class Locatable:
__slots__ = ()
@property
def parent_element(self):
# type: () -> Optional[Deb822Element]
raise NotImplementedError
def position_in_parent(self, *, skip_leading_comments: bool = True) -> Position:
"""The start position of this token/element inside its parent
This is operation is generally linear to the number of "parts" (elements/tokens)
inside the parent.
:param skip_leading_comments: If True, then if any leading comment that
that can be skipped will be excluded in the position of this locatable.
This is useful if you want the position "semantic" content of a field
without also highlighting a leading comment. Remember to align this
parameter with the `size` call, so the range does not "overshoot"
into the next element (or falls short and only covers part of an
element). Note that this option can only be used to filter out leading
comments when the comments are a subset of the element. It has no
effect on elements that are entirely made of comments.
"""
# pylint: disable=unused-argument
# Note: The base class makes no assumptions about what tokens can be skipped,
# therefore, skip_leading_comments is unused here. However, I do not want the
# API to differ between elements and tokens.
parent = self.parent_element
if parent is None:
raise TypeError(
"Cannot determine the position since the object is detached"
)
relevant_parts = itertools.takewhile(
lambda x: x is not self, parent.iter_parts()
)
span = Range.from_position_and_sizes(
START_POSITION,
(x.size(skip_leading_comments=False) for x in relevant_parts),
)
return span.end_pos
def range_in_parent(self, *, skip_leading_comments: bool = True) -> Range:
"""The range of this token/element inside its parent
This is operation is generally linear to the number of "parts" (elements/tokens)
inside the parent.
:param skip_leading_comments: If True, then if any leading comment that
that can be skipped will be excluded in the position of this locatable.
This is useful if you want the position "semantic" content of a field
without also highlighting a leading comment. Remember to align this
parameter with the `size` call, so the range does not "overshoot"
into the next element (or falls short and only covers part of an
element). Note that this option can only be used to filter out leading
comments when the comments are a subset of the element. It has no
effect on elements that are entirely made of comments.
"""
pos = self.position_in_parent(skip_leading_comments=skip_leading_comments)
return Range.from_position_and_size(
pos, self.size(skip_leading_comments=skip_leading_comments)
)
def position_in_file(self, *, skip_leading_comments: bool = True) -> Position:
"""The start position of this token/element in this file
This is an *expensive* operation and in many cases have to traverse
the entire file structure to answer the query. Consider whether
you can maintain the parent's position and then use
`position_in_parent()` combined with
`child_position.relative_to(parent_position)`
:param skip_leading_comments: If True, then if any leading comment that
that can be skipped will be excluded in the position of this locatable.
This is useful if you want the position "semantic" content of a field
without also highlighting a leading comment. Remember to align this
parameter with the `size` call, so the range does not "overshoot"
into the next element (or falls short and only covers part of an
element). Note that this option can only be used to filter out leading
comments when the comments are a subset of the element. It has no
effect on elements that are entirely made of comments.
"""
position = self.position_in_parent(
skip_leading_comments=skip_leading_comments,
)
parent = self.parent_element
if parent is not None:
parent_position = parent.position_in_file(skip_leading_comments=False)
position = position.relative_to(parent_position)
return position
def size(self, *, skip_leading_comments: bool = True) -> Range:
"""Describe the objects size as a continuous range
:param skip_leading_comments: If True, then if any leading comment that
that can be skipped will be excluded in the position of this locatable.
This is useful if you want the position "semantic" content of a field
without also highlighting a leading comment. Remember to align this
parameter with the `position_in_file` or `position_in_parent` call,
so the range does not "overshoot" into the next element (or falls
short and only covers part of an element). Note that this option can
only be used to filter out leading comments when the comments are a
subset of the element. It has no effect on elements that are entirely
made of comments.
"""
raise NotImplementedError
|