diff --git a/swh/model/identifiers.py b/swh/model/identifiers.py
index c226dfef9deac339dff590211947174e2226d317..90f3db0ae46af7850ad5030b8cd5fead4c53f0e1 100644
--- a/swh/model/identifiers.py
+++ b/swh/model/identifiers.py
@@ -246,6 +246,53 @@ def format_author(author):
def revision_identifier(revision):
"""Return the intrinsic identifier for a revision.
+
+ The fields used for the revision identifier computation are:
+ - directory
+ - parents
+ - author
+ - author_date
+ - committer
+ - committer_date
+ - metadata -> extra_headers
+ - message
+
+ A revision's identifier is the 'git'-checksum of a commit manifest
+ constructed as follows (newlines are a single ASCII newline character):
+
+ ```
+ tree <directory identifier>
+ [for each parent in parents]
+ parent <parent identifier>
+ [end for each parents]
+ author <author> <author_date>
+ committer <committer> <committer_date>
+ [for each key, value in extra_headers]
+ <key> <encoded value>
+ [end for each extra_headers]
+
+ <message>
+ ```
+
+ The directory identifier is the ascii representation of its hexadecimal
+ encoding.
+
+ Author and committer are formatted with the `format_author` function.
+ Dates are formatted with the `format_date_offset` function.
+
+ Extra headers are an ordered list of [key, value] pairs. Keys are strings
+ and get encoded to utf-8 for identifier computation. Values are either byte
+ strings, unicode strings (that get encoded to utf-8), or integers (that get
+ encoded to their utf-8 decimal representation).
+
+ Multiline extra header values are escaped by indenting the continuation
+ lines with one ascii space.
+
+ The headers are separated from the commit message with an empty line.
+
+ The checksum of the full manifest is computed using the 'commit' git object
+ type.
+
"""
components = [
b'tree ', identifier_to_str(revision['directory']).encode(), b'\n',
@@ -263,29 +310,25 @@ def revision_identifier(revision):
b' ', format_date_offset(revision['committer_date']), b'\n',
])
+ # Handle extra headers
metadata = revision.get('metadata', {})
- if 'gpgsig' in metadata:
- gpgsig = metadata['gpgsig']
- if isinstance(gpgsig, str):
- gpgsig = gpgsig.encode('utf-8')
- components.extend([b'gpgsig', b' ', gpgsig, b'\n'])
-
- if 'extra_headers' in metadata:
- headers = metadata['extra_headers']
- keys = list(headers.keys())
- keys.sort()
- for header_key in keys:
- val = headers[header_key]
- if isinstance(val, int):
- val = str(val).encode('utf-8')
- if isinstance(val, str):
- val = val.encode('utf-8')
- if isinstance(header_key, str):
- key = header_key.encode('utf-8')
- else:
- key = header_key
-
- components.extend([key, b' ', val, b'\n'])
+ for key, value in metadata.get('extra_headers', []):
+
+ # Integer values: decimal representation
+ if isinstance(value, int):
+ value = str(value).encode('utf-8')
+
+ # Unicode string values: utf-8 encoding
+ if isinstance(value, str):
+ value = value.encode('utf-8')
+
+ # multi-line values: indent continuation lines
+ if b'\n' in value:
+ value_chunks = value.split(b'\n')
+ value = b'\n '.join(value_chunks)
+
+ # encode the key to utf-8
+ components.extend([key.encode('utf-8'), b' ', value, b'\n'])
components.extend([b'\n', revision['message']])
diff --git a/swh/model/tests/test_identifiers.py b/swh/model/tests/test_identifiers.py
index 5993313fc30e7470a8be3629c13d1d219097056c..834b923b27db32a711f0c5a54d7230181d49b33d 100644
--- a/swh/model/tests/test_identifiers.py
+++ b/swh/model/tests/test_identifiers.py
@@ -220,22 +220,22 @@ class RevisionIdentifier(unittest.TestCase):
linus_tz = datetime.timezone(datetime.timedelta(minutes=-420))
linus_gpgsig = b'''\
-----BEGIN PGP SIGNATURE-----
- Version: GnuPG v1.4.13 (Darwin)
-\x20
- iQIcBAABAgAGBQJVJcYsAAoJEBiY3kIkQRNJVAUQAJ8/XQIfMqqC5oYeEFfHOPYZ
- L7qy46bXHVBa9Qd8zAJ2Dou3IbI2ZoF6/Et89K/UggOycMlt5FKV/9toWyuZv4Po
- L682wonoxX99qvVTHo6+wtnmYO7+G0f82h+qHMErxjP+I6gzRNBvRr+SfY7VlGdK
- wikMKOMWC5smrScSHITnOq1Ews5pe3N7qDYMzK0XVZmgDoaem4RSWMJs4My/qVLN
- e0CqYWq2A22GX7sXl6pjneJYQvcAXUX+CAzp24QnPSb+Q22Guj91TcxLFcHCTDdn
- qgqMsEyMiisoglwrCbO+D+1xq9mjN9tNFWP66SQ48mrrHYTBV5sz9eJyDfroJaLP
- CWgbDTgq6GzRMehHT3hXfYS5NNatjnhkNISXR7pnVP/obIi/vpWh5ll6Gd8q26z+
- a/O41UzOaLTeNI365MWT4/cnXohVLRG7iVJbAbCxoQmEgsYMRc/pBAzWJtLfcB2G
- jdTswYL6+MUdL8sB9pZ82D+BP/YAdHe69CyTu1lk9RT2pYtI/kkfjHubXBCYEJSG
- +VGllBbYG6idQJpyrOYNRJyrDi9yvDJ2W+S0iQrlZrxzGBVGTB/y65S8C+2WTBcE
- lf1Qb5GDsQrZWgD+jtWTywOYHtCBwyCKSAXxSARMbNPeak9WPlcW/Jmu+fUcMe2x
- dg1KdHOa34shrKDaOVzW
- =od6m
- -----END PGP SIGNATURE-----'''
+Version: GnuPG v1.4.13 (Darwin)
+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+=od6m
+-----END PGP SIGNATURE-----'''
self.revision = {
'id': 'bc0195aad0daa2ad5b0d76cce22b167bc3435590',
@@ -307,10 +307,10 @@ class RevisionIdentifier(unittest.TestCase):
tzinfo=linus_tz),
'message': b'Linux 4.2-rc2\n',
'metadata': {
- 'extra_headers': {
- 'svn-revision': 10,
- 'svn-repo-uuid': '046f1af7-66c2-d61b-5410-ce57b7db7bff',
- }
+ 'extra_headers': [
+ ['svn-repo-uuid', '046f1af7-66c2-d61b-5410-ce57b7db7bff'],
+ ['svn-revision', 10],
+ ]
}
}
@@ -336,7 +336,9 @@ class RevisionIdentifier(unittest.TestCase):
'offset': 480,
},
'metadata': {
- 'gpgsig': linus_gpgsig,
+ 'extra_headers': [
+ ['gpgsig', linus_gpgsig],
+ ],
},
'message': b'''Merge branch 'master' of git://github.com/alexhenrie/git-po