| Devon Loehr | 3b59be9 | 2025-03-10 14:16:16 | [diff] [blame] | 1 | # Copyright 2025 The Chromium Authors |
| 2 | # Use of this source code is governed by a BSD-style license that can be |
| 3 | # found in the LICENSE file. |
| 4 | |
| 5 | import os |
| 6 | import sys |
| 7 | |
| 8 | ### |
| 9 | # The ASTRewriter plugin emits substitution directives independently for each |
| 10 | # TU. This means there will be several duplicates (e.g. in headers that are |
| 11 | # included in multiple source files). The format used for paths is also |
| 12 | # inconsistent. |
| 13 | # |
| 14 | # This is a general-purpose post-processing script that can deduplicate, |
| 15 | # filter edits based on path, add user headers if not already present, etc. |
| 16 | # It also adds the begin/end tags when writing out the edits. |
| 17 | # |
| 18 | # usage: `python3 dedup.py directives.txt` |
| 19 | ### |
| 20 | |
| 21 | ### Configurable options |
| 22 | # List of headers to add to each modified file |
| 23 | headers_to_add = ["base/strings/to_string.h"] |
| 24 | # List of paths we do/don't want to replace in |
| 25 | paths_to_exclude = ["third_party"] |
| 26 | paths_to_include = ["/components/", "/content/", "/chrome/"] |
| 27 | |
| 28 | |
| 29 | # Paths we don't want to process |
| 30 | def filter_path(path): |
| 31 | """ |
| 32 | Examine a path and return true if we want to filter it out, |
| 33 | e.g. because it's in third_party. Feel free to customize the logic. |
| 34 | """ |
| 35 | if (any(exclude in path for exclude in paths_to_exclude)): |
| 36 | return True |
| 37 | |
| 38 | if (not any(include in path for include in paths_to_include)): |
| 39 | return True |
| 40 | |
| 41 | return False |
| 42 | |
| 43 | |
| 44 | ### Actual work |
| 45 | def ProcessFile(filename, deduped_contents, unique_paths): |
| 46 | """ Read every replacement in a file, normalizing paths and removing |
| 47 | duplicates, as well as any paths we choose to filter out. Keep track |
| 48 | of all unique paths we see so we know which files to add headers to. |
| 49 | |
| 50 | filename: the name of the file to be processed |
| 51 | deduped_contents: the set of replacements we've already processed |
| 52 | unique_paths: the set of unique replacement paths we've seen. |
| 53 | """ |
| 54 | with open(filename) as f: |
| 55 | for line in f.readlines(): |
| 56 | parts = line.split(":::") |
| 57 | if len(parts) < 2: |
| 58 | print("Skipping unexpected line: ", line) |
| 59 | continue |
| 60 | path = os.path.normpath(parts[1]) |
| 61 | if filter_path(path): |
| 62 | continue |
| 63 | |
| 64 | if path not in unique_paths: |
| 65 | unique_paths.add(path) |
| 66 | |
| 67 | parts[1] = path |
| 68 | new_line = ":::".join(parts) |
| 69 | if new_line not in deduped_contents: |
| 70 | deduped_contents.add(new_line) |
| 71 | |
| 72 | |
| 73 | def DedupFiles(filenames): |
| 74 | deduped_contents = set() |
| 75 | unique_paths = set() |
| 76 | |
| 77 | for file in filenames: |
| 78 | ProcessFile(file, deduped_contents, unique_paths) |
| 79 | |
| 80 | # This may not be necessary if the tool already emits these directives, |
| 81 | # but sometimes that may be inconvenient. |
| 82 | for path in unique_paths: |
| 83 | for header in headers_to_add: |
| 84 | deduped_contents.add( |
| 85 | f"include-user-header:::{path}:::-1:::-1:::{header}\n") |
| 86 | |
| 87 | output_file = "deduped.txt" |
| 88 | WriteFile(output_file, sorted(deduped_contents)) |
| 89 | |
| 90 | |
| 91 | def WriteFile(outfile, lines): |
| 92 | with open(outfile, "w") as f: |
| 93 | f.write("==== BEGIN EDITS ====\n") |
| 94 | f.write("".join(lines)) |
| 95 | f.write("==== END EDITS ====\n") |
| 96 | |
| 97 | |
| 98 | if __name__ == "__main__": |
| 99 | DedupFiles(sys.argv[1:]) |
