From e0dbae331f22c8bb149c87ab960ea4fc66af29ed Mon Sep 17 00:00:00 2001
From: Nicolas Dandrimont <nicolas@dandrimont.eu>
Date: Wed, 15 Feb 2017 17:45:59 +0100
Subject: [PATCH] identifiers: properly escape newlines in author
specifications
Found by investigating T75
---
swh/model/identifiers.py | 22 ++++++++++++++-------
swh/model/tests/test_identifiers.py | 30 +++++++++++++++++++++++++++++
2 files changed, 45 insertions(+), 7 deletions(-)
diff --git a/swh/model/identifiers.py b/swh/model/identifiers.py
index c53513ae..4eb2b9d7 100644
--- a/swh/model/identifiers.py
+++ b/swh/model/identifiers.py
@@ -114,6 +114,18 @@ def _perms_to_bytes(perms):
return oc.encode('ascii')
+def escape_newlines(snippet):
+ """Escape the newlines present in snippet according to git rules.
+
+ New lines in git manifests are escaped by indenting the next line by one
+ space."""
+
+ if b'\n' in snippet:
+ return b'\n '.join(snippet.split(b'\n'))
+ else:
+ return snippet
+
+
def directory_identifier(directory):
"""Return the intrinsic identifier for a directory.
@@ -347,7 +359,7 @@ def format_author_line(header, author, date_offset):
"""
- ret = [header.encode(), b' ', format_author(author)]
+ ret = [header.encode(), b' ', escape_newlines(format_author(author))]
date_offset = normalize_timestamp(date_offset)
@@ -443,13 +455,9 @@ def revision_identifier(revision):
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([key.encode('utf-8'), b' ',
+ escape_newlines(value), b'\n'])
if revision['message'] is not None:
components.extend([b'\n', revision['message']])
diff --git a/swh/model/tests/test_identifiers.py b/swh/model/tests/test_identifiers.py
index 16a34bb9..21552950 100644
--- a/swh/model/tests/test_identifiers.py
+++ b/swh/model/tests/test_identifiers.py
@@ -615,6 +615,29 @@ o6X/3T+vm8K3bf3driRr34c=
'message': b'tagging version 20081029\n\nr56558\n',
}
+ self.release_newline_in_author = {
+ 'author': {
+ 'email': b'esycat@gmail.com',
+ 'fullname': b'Eugene Janusov\n<esycat@gmail.com>',
+ 'name': b'Eugene Janusov\n',
+ },
+ 'date': {
+ 'negative_utc': None,
+ 'offset': 600,
+ 'timestamp': {
+ 'microseconds': 0,
+ 'seconds': 1377480558,
+ },
+ },
+ 'id': b'\\\x98\xf5Y\xd04\x16-\xe2->\xbe\xb9T3\xe6\xf8\x88R1',
+ 'message': b'Release of v0.3.2.',
+ 'name': b'0.3.2',
+ 'synthetic': False,
+ 'target': (b'\xc0j\xa3\xd9;x\xa2\x86\\I5\x17'
+ b'\x000\xf8\xc2\xd79o\xd3'),
+ 'target_type': 'revision',
+ }
+
@istest
def release_identifier(self):
self.assertEqual(
@@ -649,3 +672,10 @@ o6X/3T+vm8K3bf3driRr34c=
identifiers.release_identifier(self.release_negative_utc),
identifiers.identifier_to_str(self.release_negative_utc['id'])
)
+
+ @istest
+ def release_identifier_newline_in_author(self):
+ self.assertEqual(
+ identifiers.release_identifier(self.release_newline_in_author),
+ identifiers.identifier_to_str(self.release_newline_in_author['id'])
+ )
--
GitLab