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
|
# Licensed under the Apache License: http://www.apache.org/licenses/LICENSE-2.0
# For details: https://github.com/nedbat/coveragepy/blob/master/NOTICE.txt
"""Better tokenizing for coverage.py."""
from __future__ import annotations
import ast
import io
import keyword
import re
import sys
import token
import tokenize
from typing import Iterable, List, Optional, Set, Tuple
from coverage import env
from coverage.types import TLineNo, TSourceTokenLines
TokenInfos = Iterable[tokenize.TokenInfo]
def _phys_tokens(toks: TokenInfos) -> TokenInfos:
"""Return all physical tokens, even line continuations.
tokenize.generate_tokens() doesn't return a token for the backslash that
continues lines. This wrapper provides those tokens so that we can
re-create a faithful representation of the original source.
Returns the same values as generate_tokens()
"""
last_line: Optional[str] = None
last_lineno = -1
last_ttext: str = ""
for ttype, ttext, (slineno, scol), (elineno, ecol), ltext in toks:
if last_lineno != elineno:
if last_line and last_line.endswith("\\\n"):
# We are at the beginning of a new line, and the last line
# ended with a backslash. We probably have to inject a
# backslash token into the stream. Unfortunately, there's more
# to figure out. This code::
#
# usage = """\
# HEY THERE
# """
#
# triggers this condition, but the token text is::
#
# '"""\\\nHEY THERE\n"""'
#
# so we need to figure out if the backslash is already in the
# string token or not.
inject_backslash = True
if last_ttext.endswith("\\"):
inject_backslash = False
elif ttype == token.STRING:
if "\n" in ttext and ttext.split("\n", 1)[0][-1] == "\\":
# It's a multi-line string and the first line ends with
# a backslash, so we don't need to inject another.
inject_backslash = False
if inject_backslash:
# Figure out what column the backslash is in.
ccol = len(last_line.split("\n")[-2]) - 1
# Yield the token, with a fake token type.
yield tokenize.TokenInfo(
99999, "\\\n",
(slineno, ccol), (slineno, ccol+2),
last_line
)
last_line = ltext
if ttype not in (tokenize.NEWLINE, tokenize.NL):
last_ttext = ttext
yield tokenize.TokenInfo(ttype, ttext, (slineno, scol), (elineno, ecol), ltext)
last_lineno = elineno
class MatchCaseFinder(ast.NodeVisitor):
"""Helper for finding match/case lines."""
def __init__(self, source: str) -> None:
# This will be the set of line numbers that start match or case statements.
self.match_case_lines: Set[TLineNo] = set()
self.visit(ast.parse(source))
if sys.version_info >= (3, 10):
def visit_Match(self, node: ast.Match) -> None:
"""Invoked by ast.NodeVisitor.visit"""
self.match_case_lines.add(node.lineno)
for case in node.cases:
self.match_case_lines.add(case.pattern.lineno)
self.generic_visit(node)
def source_token_lines(source: str) -> TSourceTokenLines:
"""Generate a series of lines, one for each line in `source`.
Each line is a list of pairs, each pair is a token::
[('key', 'def'), ('ws', ' '), ('nam', 'hello'), ('op', '('), ... ]
Each pair has a token class, and the token text.
If you concatenate all the token texts, and then join them with newlines,
you should have your original `source` back, with two differences:
trailing white space is not preserved, and a final line with no newline
is indistinguishable from a final line with a newline.
"""
ws_tokens = {token.INDENT, token.DEDENT, token.NEWLINE, tokenize.NL}
line: List[Tuple[str, str]] = []
col = 0
source = source.expandtabs(8).replace("\r\n", "\n")
tokgen = generate_tokens(source)
if env.PYBEHAVIOR.soft_keywords:
match_case_lines = MatchCaseFinder(source).match_case_lines
for ttype, ttext, (sline, scol), (_, ecol), _ in _phys_tokens(tokgen):
mark_start = True
for part in re.split("(\n)", ttext):
if part == "\n":
yield line
line = []
col = 0
mark_end = False
elif part == "":
mark_end = False
elif ttype in ws_tokens:
mark_end = False
else:
if mark_start and scol > col:
line.append(("ws", " " * (scol - col)))
mark_start = False
tok_class = tokenize.tok_name.get(ttype, "xx").lower()[:3]
if ttype == token.NAME:
if keyword.iskeyword(ttext):
# Hard keywords are always keywords.
tok_class = "key"
elif sys.version_info >= (3, 10): # PYVERSIONS
# Need the version_info check to keep mypy from borking
# on issoftkeyword here.
if env.PYBEHAVIOR.soft_keywords and keyword.issoftkeyword(ttext):
# Soft keywords appear at the start of the line,
# on lines that start match or case statements.
if len(line) == 0:
is_start_of_line = True
elif (len(line) == 1) and line[0][0] == "ws":
is_start_of_line = True
else:
is_start_of_line = False
if is_start_of_line and sline in match_case_lines:
tok_class = "key"
line.append((tok_class, part))
mark_end = True
scol = 0
if mark_end:
col = ecol
if line:
yield line
class CachedTokenizer:
"""A one-element cache around tokenize.generate_tokens.
When reporting, coverage.py tokenizes files twice, once to find the
structure of the file, and once to syntax-color it. Tokenizing is
expensive, and easily cached.
This is a one-element cache so that our twice-in-a-row tokenizing doesn't
actually tokenize twice.
"""
def __init__(self) -> None:
self.last_text: Optional[str] = None
self.last_tokens: List[tokenize.TokenInfo] = []
def generate_tokens(self, text: str) -> TokenInfos:
"""A stand-in for `tokenize.generate_tokens`."""
if text != self.last_text:
self.last_text = text
readline = io.StringIO(text).readline
try:
self.last_tokens = list(tokenize.generate_tokens(readline))
except:
self.last_text = None
raise
return self.last_tokens
# Create our generate_tokens cache as a callable replacement function.
generate_tokens = CachedTokenizer().generate_tokens
def source_encoding(source: bytes) -> str:
"""Determine the encoding for `source`, according to PEP 263.
`source` is a byte string: the text of the program.
Returns a string, the name of the encoding.
"""
readline = iter(source.splitlines(True)).__next__
return tokenize.detect_encoding(readline)[0]
|
