|
| 1 | +# SPDX-FileCopyrightText: 2022 Eva Herrada |
| 2 | +# |
| 3 | +# SPDX-License-Identifier: MIT |
| 4 | + |
| 5 | +""" |
| 6 | +
|
| 7 | +file-compare.py |
| 8 | +=============== |
| 9 | +
|
| 10 | +Functionality to compare a file across all Adafruit CircuitPython repos |
| 11 | +and output the text of the files along with which and how many repos use that |
| 12 | +exact file text. |
| 13 | +
|
| 14 | +* Author(s): Eva Herrada |
| 15 | +
|
| 16 | +""" |
| 17 | +import argparse |
| 18 | +from typing import Optional |
| 19 | + |
| 20 | +import requests |
| 21 | +from requests.structures import CaseInsensitiveDict |
| 22 | + |
| 23 | +from adabot.lib.common_funcs import list_repos |
| 24 | + |
| 25 | + |
| 26 | +def compare(git_file: str, token: Optional[str] = None) -> list: |
| 27 | + """Uses requests to compare files across the adafruit org |
| 28 | +
|
| 29 | + .. note:: |
| 30 | +
|
| 31 | + The GitHub API token is not necessary as long as all repos |
| 32 | + being accessed are public. However: it does make things easier |
| 33 | + as you won't get rate-limited quite as often |
| 34 | +
|
| 35 | + :param str git_file: The file to compare |
| 36 | + :param str|None token: The (optional but recommended) github API token |
| 37 | + :return: A list containing all the unique file texts, sorted from most to |
| 38 | + least common along with the repos that have that exact file text. |
| 39 | + :rtype: list |
| 40 | + """ |
| 41 | + |
| 42 | + files = {} |
| 43 | + |
| 44 | + all_repos = list_repos() |
| 45 | + print("Got Repos List") |
| 46 | + print(f"Repos found: {len(all_repos)}") |
| 47 | + |
| 48 | + for repo in all_repos: |
| 49 | + name = repo["name"] |
| 50 | + url = f"https://raw.githubusercontent.com/adafruit/{name}/main/{git_file}" |
| 51 | + |
| 52 | + if token: |
| 53 | + # If repo is private - we need to add a token in header: |
| 54 | + headers = CaseInsensitiveDict() |
| 55 | + headers["Authorization"] = f"token {token}" |
| 56 | + |
| 57 | + resp = requests.get(url, headers=headers) |
| 58 | + else: |
| 59 | + resp = requests.get(url) |
| 60 | + |
| 61 | + if resp.status_code != 200: |
| 62 | + print(name) |
| 63 | + print(resp.status_code) |
| 64 | + if resp.text not in files: |
| 65 | + files[resp.text] = [1, [repo["html_url"]]] |
| 66 | + else: |
| 67 | + files[resp.text][0] = files[resp.text][0] + 1 |
| 68 | + files[resp.text][1].append(repo["html_url"]) |
| 69 | + |
| 70 | + top = 0 |
| 71 | + sort = [] |
| 72 | + for text, repos in files.items(): |
| 73 | + if repos[0] >= top: |
| 74 | + sort.insert(0, [repos[0], text, repos[1]]) |
| 75 | + top = repos[0] |
| 76 | + else: |
| 77 | + for i, val in enumerate(sort): |
| 78 | + if val[0] <= repos[0]: |
| 79 | + sort.insert(i, [repos[0], text, repos[1]]) |
| 80 | + break |
| 81 | + |
| 82 | + return sort |
| 83 | + |
| 84 | + |
| 85 | +if __name__ == "__main__": |
| 86 | + |
| 87 | + parser = argparse.ArgumentParser( |
| 88 | + description="Compare files across the adafruit CircuitPython repos", |
| 89 | + ) |
| 90 | + parser.add_argument( |
| 91 | + "gh_token", |
| 92 | + metavar="GH_TOKEN", |
| 93 | + type=str, |
| 94 | + help="GitHub token with proper scopes", |
| 95 | + ) |
| 96 | + |
| 97 | + parser.add_argument( |
| 98 | + "--file", |
| 99 | + metavar="<FILE>", |
| 100 | + type=str, |
| 101 | + dest="file", |
| 102 | + required=True, |
| 103 | + help="File to compare", |
| 104 | + ) |
| 105 | + |
| 106 | + parser.add_argument( |
| 107 | + "-o", |
| 108 | + metavar="<OUTFILE>", |
| 109 | + type=str, |
| 110 | + dest="outfile", |
| 111 | + default=None, |
| 112 | + help="File to send output to", |
| 113 | + ) |
| 114 | + |
| 115 | + args = parser.parse_args() |
| 116 | + |
| 117 | + results = compare(args.file, args.gh_token) |
| 118 | + |
| 119 | + for index, value in enumerate(results): |
| 120 | + print(f"##### {index+1}/{len(results)} #####") |
| 121 | + print(value[0]) |
| 122 | + print("START OF FILE") |
| 123 | + print(value[1]) |
| 124 | + print("END OF FILE") |
| 125 | + print(value[2]) |
| 126 | + print() |
| 127 | + if args.outfile: |
| 128 | + with open(args.outfile, "w") as F: |
| 129 | + for index, value in enumerate(results): |
| 130 | + F.write(f"##### {index+1}/{len(results)} #####\n") |
| 131 | + F.write(f"{value[0]}\n") |
| 132 | + F.write("START OF FILE\n") |
| 133 | + F.write(f"{value[1]}\n") |
| 134 | + F.write("END OF FILE\n") |
| 135 | + for r in value[2]: |
| 136 | + F.write(r + "\n") |
| 137 | + F.write("\n") |
0 commit comments