OLD | NEW |
---|---|
(Empty) | |
1 #!/usr/bin/env python | |
2 # Copyright 2016 The Chromium Authors. All rights reserved. | |
3 # Use of this source code is governed by a BSD-style license that can be | |
4 # found in the LICENSE file. | |
5 | |
6 """This script will search through the target folder specified and try to find | |
7 duplicate includes from h and cc files, and remove them from the cc files. | |
8 | |
9 Usage remove_duplicate_includes.py --dry-run components/foo components/bar | |
10 """ | |
11 | |
12 import argparse; | |
13 import collections; | |
14 import os; | |
15 import re; | |
16 | |
17 parser = argparse.ArgumentParser() | |
18 parser.add_argument('--dry-run', action='store_true', | |
19 help='Does not actually remove lines when specified.') | |
20 parser.add_argument('--span-dirs', action='store_true', | |
21 help='Mapping between h and cc files will not be limited to same folders.') | |
22 parser.add_argument('targets', nargs='+', | |
23 help="Folders to search for duplicate includes in.") | |
24 args = parser.parse_args() | |
25 | |
26 # This could be generlized if desired, and moved to command line arguments. | |
maxbogue
2016/09/30 17:40:31
generalized
skym
2016/09/30 19:18:12
Done.
| |
27 H_FILE_SUFFIX = ".h" | |
28 CC_FILE_SUFFIX = ".cc" | |
29 | |
30 def HasSuffix(file_name, suffix): | |
31 return os.path.splitext(file_name)[1] == suffix | |
32 | |
33 def IsEmpty(line): | |
34 return not line.strip() | |
35 | |
36 # The \s should allow us to ignore any whitespace and only focus on the group | |
37 # captured when comparing between files. | |
38 regex = re.compile('^\s*#include\s+(.*?)\s*$') | |
39 | |
40 # The key here depends on the span-dirs flag, if specified then it will only be | |
maxbogue
2016/09/30 17:40:31
"A map of header files to the includes they contai
skym
2016/09/30 19:18:12
Done.
| |
41 # the file name, and this will allows mapping between files not in the same | |
42 # folder. If this flag not present then full path is used. | |
43 h_file_to_include_set = collections.defaultdict(set) | |
44 | |
45 # Key is always the full path to the cc file. | |
maxbogue
2016/09/30 17:40:31
Explain the purpose of this variable. I think it c
skym
2016/09/30 19:18:12
Done.
| |
46 cc_file_path_set = set() | |
47 | |
48 for relative_root in args.targets: | |
49 absolute_root = os.path.join(os.getcwd(), relative_root) | |
50 for (dir_path, dir_name_list, file_name_list) in os.walk(absolute_root): | |
maxbogue
2016/09/30 17:40:31
I haven't worked with python in a while but I'm pr
skym
2016/09/30 19:18:12
Done.
| |
51 for file_name in file_name_list: | |
52 file_path = os.path.join(dir_path, file_name) | |
53 if HasSuffix(file_name, H_FILE_SUFFIX): | |
54 # Can be either name or path depending on flag. | |
55 file_key = file_name if args.span_dirs else file_path | |
56 with open(file_path) as file_handle: | |
57 for line in file_handle: | |
58 match = regex.search(line) | |
59 if match: | |
60 h_file_to_include_set[file_key].add(match.group(1)) | |
61 elif HasSuffix(file_name, CC_FILE_SUFFIX): | |
62 cc_file_path_set.add(file_path) | |
63 | |
64 for cc_file_path in cc_file_path_set: | |
65 # The lookup must match index method when adding h files, depending on flag. | |
66 cc_file_name = os.path.basename(cc_file_path) | |
67 cc_file_key = cc_file_name if args.span_dirs else cc_file_path | |
68 h_file_key = os.path.splitext(cc_file_key)[0] + H_FILE_SUFFIX | |
maxbogue
2016/09/30 17:40:31
You could strip "_unittest" from the end here if i
skym
2016/09/30 19:18:13
Oooooh, actually maybe I should be looking at the
| |
69 | |
70 if h_file_key in h_file_to_include_set: | |
71 include_set = h_file_to_include_set[h_file_key] | |
72 | |
73 # Read out all the data and reset file position to start overwriting. | |
74 file_handle = open(cc_file_path, "r" if args.dry_run else "r+") | |
75 data = file_handle.readlines() | |
maxbogue
2016/09/30 17:40:31
I'd probably just call this "lines" tbh
skym
2016/09/30 19:18:12
How about line_list?
| |
76 file_handle.seek(0) | |
77 | |
78 # When a section of includes are completely removed, we want to remove the | |
79 # trailing empty as well. | |
80 lastCopiedLineWasEmpty = False | |
81 lastLineWasOmitted = False | |
82 for line in data: | |
83 match = regex.search(line) | |
84 if match is not None and match.group(1) in include_set: | |
85 print "Removed " + match.group(1) + " " + cc_file_name | |
86 lastLineWasOmitted = True | |
87 elif lastCopiedLineWasEmpty and lastLineWasOmitted and IsEmpty(line): | |
88 print "Removed empty line " + cc_file_name | |
89 lastLineWasOmitted = True | |
90 else: | |
91 lastCopiedLineWasEmpty = IsEmpty(line) | |
92 lastLineWasOmitted = False | |
93 if not args.dry_run: | |
94 file_handle.write(line) | |
95 if not args.dry_run: | |
96 file_handle.truncate() | |
97 file_handle.close() | |
OLD | NEW |