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
|
"""File wrangling."""
import fnmatch, os, os.path, re, sys
import ntpath, posixpath
from coverage.backward import to_string, open_python_source
from coverage.misc import CoverageException, join_regex
class FileLocator(object):
"""Understand how filenames work."""
def __init__(self):
# The absolute path to our current directory.
self.relative_dir = os.path.normcase(abs_file(os.curdir) + os.sep)
# Cache of results of calling the canonical_filename() method, to
# avoid duplicating work.
self.canonical_filename_cache = {}
def relative_filename(self, filename):
"""Return the relative form of `filename`.
The filename will be relative to the current directory when the
`FileLocator` was constructed.
"""
fnorm = os.path.normcase(filename)
if fnorm.startswith(self.relative_dir):
filename = filename[len(self.relative_dir):]
return filename
def canonical_filename(self, filename):
"""Return a canonical filename for `filename`.
An absolute path with no redundant components and normalized case.
"""
if filename not in self.canonical_filename_cache:
if not os.path.isabs(filename):
for path in [os.curdir] + sys.path:
if path is None:
continue
f = os.path.join(path, filename)
if os.path.exists(f):
filename = f
break
cf = abs_file(filename)
self.canonical_filename_cache[filename] = cf
return self.canonical_filename_cache[filename]
def get_python_source(filename):
"""Return the source code, as a string."""
if os.path.exists(filename):
# A regular text file: open it.
with open_python_source(filename) as f:
return f.read()
# Maybe it's in a zip file?
source = get_zip_data(filename)
if source is not None:
return source
# Couldn't find source.
raise CoverageException(
"No source for code: '%s'." % filename
)
def get_zip_data(filename):
"""Get data from `filename` if it is a zip file path.
Returns the string data read from the zip file, or None if no zip file
could be found or `filename` isn't in it. The data returned will be
an empty string if the file is empty.
"""
import zipimport
markers = ['.zip'+os.sep, '.egg'+os.sep]
for marker in markers:
if marker in filename:
parts = filename.split(marker)
try:
zi = zipimport.zipimporter(parts[0]+marker[:-1])
except zipimport.ZipImportError:
continue
try:
data = zi.get_data(parts[1])
except IOError:
continue
return to_string(data)
return None
if sys.platform == 'win32':
def actual_path(path):
"""Get the actual path of `path`, including the correct case."""
if path in actual_path.cache:
return actual_path.cache[path]
head, tail = os.path.split(path)
if not tail:
actpath = head
elif not head:
actpath = tail
else:
head = actual_path(head)
if head in actual_path.list_cache:
files = actual_path.list_cache[head]
else:
try:
files = os.listdir(head)
except OSError:
files = []
actual_path.list_cache[head] = files
normtail = os.path.normcase(tail)
for f in files:
if os.path.normcase(f) == normtail:
tail = f
break
actpath = os.path.join(head, tail)
actual_path.cache[path] = actpath
return actpath
actual_path.cache = {}
actual_path.list_cache = {}
else:
def actual_path(filename):
"""The actual path for non-Windows platforms."""
return filename
def abs_file(filename):
"""Return the absolute normalized form of `filename`."""
path = os.path.expandvars(os.path.expanduser(filename))
path = os.path.abspath(os.path.realpath(path))
path = actual_path(path)
return path
def isabs_anywhere(filename):
"""Is `filename` an absolute path on any OS?"""
return ntpath.isabs(filename) or posixpath.isabs(filename)
def prep_patterns(patterns):
"""Prepare the file patterns for use in a `FnmatchMatcher`.
If a pattern starts with a wildcard, it is used as a pattern
as-is. If it does not start with a wildcard, then it is made
absolute with the current directory.
If `patterns` is None, an empty list is returned.
"""
prepped = []
for p in patterns or []:
if p.startswith(("*", "?")):
prepped.append(p)
else:
prepped.append(abs_file(p))
return prepped
class TreeMatcher(object):
"""A matcher for files in a tree."""
def __init__(self, directories):
self.dirs = list(directories)
def __repr__(self):
return "<TreeMatcher %r>" % self.dirs
def info(self):
"""A list of strings for displaying when dumping state."""
return self.dirs
def match(self, fpath):
"""Does `fpath` indicate a file in one of our trees?"""
for d in self.dirs:
if fpath.startswith(d):
if fpath == d:
# This is the same file!
return True
if fpath[len(d)] == os.sep:
# This is a file in the directory
return True
return False
class ModuleMatcher(object):
"""A matcher for modules in a tree."""
def __init__(self, module_names):
self.modules = list(module_names)
def __repr__(self):
return "<ModuleMatcher %r>" % (self.modules)
def info(self):
"""A list of strings for displaying when dumping state."""
return self.modules
def match(self, module_name):
"""Does `module_name` indicate a module in one of our packages?"""
if not module_name:
return False
for m in self.modules:
if module_name.startswith(m):
if module_name == m:
return True
if module_name[len(m)] == '.':
# This is a module in the package
return True
return False
class FnmatchMatcher(object):
"""A matcher for files by filename pattern."""
def __init__(self, pats):
self.pats = pats[:]
# fnmatch is platform-specific. On Windows, it does the Windows thing
# of treating / and \ as equivalent. But on other platforms, we need to
# take care of that ourselves.
fnpats = (fnmatch.translate(p) for p in pats)
fnpats = (p.replace(r"\/", r"[\\/]") for p in fnpats)
if sys.platform == 'win32':
# Windows is also case-insensitive. BTW: the regex docs say that
# flags like (?i) have to be at the beginning, but fnmatch puts
# them at the end, and having two there seems to work fine.
fnpats = (p + "(?i)" for p in fnpats)
self.re = re.compile(join_regex(fnpats))
def __repr__(self):
return "<FnmatchMatcher %r>" % self.pats
def info(self):
"""A list of strings for displaying when dumping state."""
return self.pats
def match(self, fpath):
"""Does `fpath` match one of our filename patterns?"""
return self.re.match(fpath) is not None
def sep(s):
"""Find the path separator used in this string, or os.sep if none."""
sep_match = re.search(r"[\\/]", s)
if sep_match:
the_sep = sep_match.group(0)
else:
the_sep = os.sep
return the_sep
class PathAliases(object):
"""A collection of aliases for paths.
When combining data files from remote machines, often the paths to source
code are different, for example, due to OS differences, or because of
serialized checkouts on continuous integration machines.
A `PathAliases` object tracks a list of pattern/result pairs, and can
map a path through those aliases to produce a unified path.
`locator` is a FileLocator that is used to canonicalize the results.
"""
def __init__(self, locator=None):
self.aliases = []
self.locator = locator
def add(self, pattern, result):
"""Add the `pattern`/`result` pair to the list of aliases.
`pattern` is an `fnmatch`-style pattern. `result` is a simple
string. When mapping paths, if a path starts with a match against
`pattern`, then that match is replaced with `result`. This models
isomorphic source trees being rooted at different places on two
different machines.
`pattern` can't end with a wildcard component, since that would
match an entire tree, and not just its root.
"""
# The pattern can't end with a wildcard component.
pattern = pattern.rstrip(r"\/")
if pattern.endswith("*"):
raise CoverageException("Pattern must not end with wildcards.")
pattern_sep = sep(pattern)
# The pattern is meant to match a filepath. Let's make it absolute
# unless it already is, or is meant to match any prefix.
if not pattern.startswith('*') and not isabs_anywhere(pattern):
pattern = abs_file(pattern)
pattern += pattern_sep
# Make a regex from the pattern. fnmatch always adds a \Z to
# match the whole string, which we don't want.
regex_pat = fnmatch.translate(pattern).replace(r'\Z(', '(')
# We want */a/b.py to match on Windows too, so change slash to match
# either separator.
regex_pat = regex_pat.replace(r"\/", r"[\\/]")
# We want case-insensitive matching, so add that flag.
regex = re.compile(r"(?i)" + regex_pat)
# Normalize the result: it must end with a path separator.
result_sep = sep(result)
result = result.rstrip(r"\/") + result_sep
self.aliases.append((regex, result, pattern_sep, result_sep))
def map(self, path):
"""Map `path` through the aliases.
`path` is checked against all of the patterns. The first pattern to
match is used to replace the root of the path with the result root.
Only one pattern is ever used. If no patterns match, `path` is
returned unchanged.
The separator style in the result is made to match that of the result
in the alias.
"""
for regex, result, pattern_sep, result_sep in self.aliases:
m = regex.match(path)
if m:
new = path.replace(m.group(0), result)
if pattern_sep != result_sep:
new = new.replace(pattern_sep, result_sep)
if self.locator:
new = self.locator.canonical_filename(new)
return new
return path
def find_python_files(dirname):
"""Yield all of the importable Python files in `dirname`, recursively.
To be importable, the files have to be in a directory with a __init__.py,
except for `dirname` itself, which isn't required to have one. The
assumption is that `dirname` was specified directly, so the user knows
best, but sub-directories are checked for a __init__.py to be sure we only
find the importable files.
"""
for i, (dirpath, dirnames, filenames) in enumerate(os.walk(dirname)):
if i > 0 and '__init__.py' not in filenames:
# If a directory doesn't have __init__.py, then it isn't
# importable and neither are its files
del dirnames[:]
continue
for filename in filenames:
# We're only interested in files that look like reasonable Python
# files: Must end with .py or .pyw, and must not have certain funny
# characters that probably mean they are editor junk.
if re.match(r"^[^.#~!$@%^&*()+=,]+\.pyw?$", filename):
yield os.path.join(dirpath, filename)
|