verify-commits.py raw
1 #!/usr/bin/env python3
2 # Copyright (c) 2018-2022 The Limenka developers
3 # Distributed under the MIT software license, see the accompanying
4 # file COPYING or http://www.opensource.org/licenses/mit-license.php.
5 """Verify commits against a trusted keys list."""
6 import argparse
7 import hashlib
8 import logging
9 import os
10 import subprocess
11 import sys
12 import time
13
14 GIT = os.getenv('GIT', 'git')
15
16 def tree_sha512sum(commit='HEAD'):
17 """Calculate the Tree-sha512 for the commit.
18
19 This is copied from github-merge.py. See https://github.com/limenka/limenka-maintainer-tools."""
20
21 # request metadata for entire tree, recursively
22 files = []
23 blob_by_name = {}
24 for line in subprocess.check_output([GIT, 'ls-tree', '--full-tree', '-r', commit]).splitlines():
25 name_sep = line.index(b'\t')
26 metadata = line[:name_sep].split() # perms, 'blob', blobid
27 assert metadata[1] == b'blob'
28 name = line[name_sep + 1:]
29 files.append(name)
30 blob_by_name[name] = metadata[2]
31
32 files.sort()
33 # open connection to git-cat-file in batch mode to request data for all blobs
34 # this is much faster than launching it per file
35 p = subprocess.Popen([GIT, 'cat-file', '--batch'], stdout=subprocess.PIPE, stdin=subprocess.PIPE)
36 overall = hashlib.sha512()
37 for f in files:
38 blob = blob_by_name[f]
39 # request blob
40 p.stdin.write(blob + b'\n')
41 p.stdin.flush()
42 # read header: blob, "blob", size
43 reply = p.stdout.readline().split()
44 assert reply[0] == blob and reply[1] == b'blob'
45 size = int(reply[2])
46 # hash the blob data
47 intern = hashlib.sha512()
48 ptr = 0
49 while ptr < size:
50 bs = min(65536, size - ptr)
51 piece = p.stdout.read(bs)
52 if len(piece) == bs:
53 intern.update(piece)
54 else:
55 raise IOError('Premature EOF reading git cat-file output')
56 ptr += bs
57 dig = intern.hexdigest()
58 assert p.stdout.read(1) == b'\n' # ignore LF that follows blob data
59 # update overall hash with file hash
60 overall.update(dig.encode("utf-8"))
61 overall.update(" ".encode("utf-8"))
62 overall.update(f)
63 overall.update("\n".encode("utf-8"))
64 p.stdin.close()
65 if p.wait():
66 raise IOError('Non-zero return value executing git cat-file')
67 return overall.hexdigest()
68
69 def main():
70
71 # Enable debug logging if running in CI
72 if 'CI' in os.environ and os.environ['CI'].lower() == "true":
73 logging.getLogger().setLevel(logging.DEBUG)
74
75 # Parse arguments
76 parser = argparse.ArgumentParser(usage='%(prog)s [options] [commit id]')
77 parser.add_argument('--disable-tree-check', action='store_false', dest='verify_tree', help='disable SHA-512 tree check')
78 parser.add_argument('--clean-merge', type=float, dest='clean_merge', default=float('inf'), help='Only check clean merge after <NUMBER> days ago (default: %(default)s)', metavar='NUMBER')
79 parser.add_argument('commit', nargs='?', default='HEAD', help='Check clean merge up to commit <commit>')
80 args = parser.parse_args()
81
82 # get directory of this program and read data files
83 dirname = os.path.dirname(os.path.abspath(__file__))
84 print("Using verify-commits data from " + dirname)
85 with open(dirname + "/trusted-git-root", "r", encoding="utf8") as f:
86 verified_root = f.read().splitlines()[0]
87 with open(dirname + "/trusted-sha512-root-commit", "r", encoding="utf8") as f:
88 verified_sha512_root = f.read().splitlines()[0]
89 with open(dirname + "/allow-revsig-commits", "r", encoding="utf8") as f:
90 revsig_allowed = f.read().splitlines()
91 with open(dirname + "/allow-unclean-merge-commits", "r", encoding="utf8") as f:
92 unclean_merge_allowed = f.read().splitlines()
93 with open(dirname + "/allow-incorrect-sha512-commits", "r", encoding="utf8") as f:
94 incorrect_sha512_allowed = f.read().splitlines()
95 with open(dirname + "/trusted-keys", "r", encoding="utf8") as f:
96 trusted_keys = f.read().splitlines()
97
98 # Set commit and variables
99 current_commit = args.commit
100 if ' ' in current_commit:
101 print("Commit must not contain spaces", file=sys.stderr)
102 sys.exit(1)
103 verify_tree = args.verify_tree
104 no_sha1 = True
105 prev_commit = ""
106 initial_commit = current_commit
107
108 # Iterate through commits
109 while True:
110
111 # Log a message to prevent Travis from timing out
112 logging.debug("verify-commits: [in-progress] processing commit {}".format(current_commit[:8]))
113
114 if current_commit == verified_root:
115 print('There is a valid path from "{}" to {} where all commits are signed!'.format(initial_commit, verified_root))
116 sys.exit(0)
117 else:
118 # Make sure this commit isn't older than trusted roots
119 check_root_older_res = subprocess.run([GIT, "merge-base", "--is-ancestor", verified_root, current_commit])
120 if check_root_older_res.returncode != 0:
121 print(f"\"{current_commit}\" predates the trusted root, stopping!")
122 sys.exit(0)
123
124 if verify_tree:
125 if current_commit == verified_sha512_root:
126 print("All Tree-SHA512s matched up to {}".format(verified_sha512_root), file=sys.stderr)
127 verify_tree = False
128 no_sha1 = False
129 else:
130 # Skip the tree check if we are older than the trusted root
131 check_root_older_res = subprocess.run([GIT, "merge-base", "--is-ancestor", verified_sha512_root, current_commit])
132 if check_root_older_res.returncode != 0:
133 print(f"\"{current_commit}\" predates the trusted SHA512 root, disabling tree verification.")
134 verify_tree = False
135 no_sha1 = False
136
137
138 os.environ['LIMENKA_VERIFY_COMMITS_ALLOW_SHA1'] = "0" if no_sha1 else "1"
139 allow_revsig = current_commit in revsig_allowed
140
141 # Check that the commit (and parents) was signed with a trusted key
142 valid_sig = False
143 verify_res = subprocess.run([GIT, '-c', 'gpg.program={}/gpg.sh'.format(dirname), 'verify-commit', "--raw", current_commit], capture_output=True)
144 for line in verify_res.stderr.decode().splitlines():
145 if line.startswith("[GNUPG:] VALIDSIG "):
146 key = line.split(" ")[-1]
147 valid_sig = key in trusted_keys
148 elif (line.startswith("[GNUPG:] REVKEYSIG ") or line.startswith("[GNUPG:] EXPKEYSIG ")) and not allow_revsig:
149 valid_sig = False
150 break
151 if not valid_sig:
152 if prev_commit != "":
153 print("No parent of {} was signed with a trusted key!".format(prev_commit), file=sys.stderr)
154 print("Parents are:", file=sys.stderr)
155 parents = subprocess.check_output([GIT, 'show', '-s', '--format=format:%P', prev_commit]).decode('utf8').splitlines()[0].split(' ')
156 for parent in parents:
157 subprocess.call([GIT, 'show', '-s', parent], stdout=sys.stderr)
158 else:
159 print("{} was not signed with a trusted key!".format(current_commit), file=sys.stderr)
160 sys.exit(1)
161
162 # Check the Tree-SHA512
163 if (verify_tree or prev_commit == "") and current_commit not in incorrect_sha512_allowed:
164 tree_hash = tree_sha512sum(current_commit)
165 if ("Tree-SHA512: {}".format(tree_hash)) not in subprocess.check_output([GIT, 'show', '-s', '--format=format:%B', current_commit]).decode('utf8').splitlines():
166 print("Tree-SHA512 did not match for commit " + current_commit, file=sys.stderr)
167 sys.exit(1)
168
169 # Merge commits should only have two parents
170 parents = subprocess.check_output([GIT, 'show', '-s', '--format=format:%P', current_commit]).decode('utf8').splitlines()[0].split(' ')
171 if len(parents) > 2:
172 print("Commit {} is an octopus merge".format(current_commit), file=sys.stderr)
173 sys.exit(1)
174
175 # Check that the merge commit is clean
176 commit_time = int(subprocess.check_output([GIT, 'show', '-s', '--format=format:%ct', current_commit]).decode('utf8').splitlines()[0])
177 check_merge = commit_time > time.time() - args.clean_merge * 24 * 60 * 60 # Only check commits in clean_merge days
178 allow_unclean = current_commit in unclean_merge_allowed
179 if len(parents) == 2 and check_merge and not allow_unclean:
180 current_tree = subprocess.check_output([GIT, 'show', '--format=%T', current_commit]).decode('utf8').splitlines()[0]
181
182 # This merge-tree functionality requires git >= 2.38. The
183 # --write-tree option was added in order to opt-in to the new
184 # behavior. Older versions of git will not recognize the option and
185 # will instead exit with code 128.
186 try:
187 recreated_tree = subprocess.check_output([GIT, "merge-tree", "--write-tree", parents[0], parents[1]]).decode('utf8').splitlines()[0]
188 except subprocess.CalledProcessError as e:
189 if e.returncode == 128:
190 print("git v2.38+ is required for this functionality.", file=sys.stderr)
191 sys.exit(1)
192 else:
193 raise e
194
195 if current_tree != recreated_tree:
196 print("Merge commit {} is not clean".format(current_commit), file=sys.stderr)
197 subprocess.call([GIT, 'diff', recreated_tree, current_tree])
198 sys.exit(1)
199
200 prev_commit = current_commit
201 current_commit = parents[0]
202
203 if __name__ == '__main__':
204 main()
205