Page Menu
Home
Software Heritage
Search
Configure Global Search
Log In
Files
F7123309
D6813.diff
No One
Temporary
Actions
View File
Edit File
Delete File
View Transforms
Subscribe
Mute Notifications
Award Token
Flag For Later
Size
6 KB
Subscribers
None
D6813.diff
View Options
diff --git a/swh/loader/cvs/cvs2gitdump/cvs2gitdump.py b/swh/loader/cvs/cvs2gitdump/cvs2gitdump.py
--- a/swh/loader/cvs/cvs2gitdump/cvs2gitdump.py
+++ b/swh/loader/cvs/cvs2gitdump/cvs2gitdump.py
@@ -652,34 +652,36 @@
# and those lines are followed by content which follows the Log keyword.
# For example, the line:
#
- # $Log$ content which follows
+ # foo $Log$content which follows
#
- # must be expanded like this:
+ # will be expanded like this by CVS:
#
- # $Log: delta,v $
- # Revision 1.2 2021/11/29 14:24:18 stsp
- # log message line 1
- # log message line 2
- # content which follows
+ # foo $Log: delta,v $
+ # foo Revision 1.2 2021/11/29 14:24:18 stsp
+ # foo log message line 1
+ # foo log message line 2
+ # foocontent which follows
+ #
+ # (Side note: Trailing whitespace is stripped from "foo " when
+ # the content which follows gets written to the output file.)
#
# If we did not trim the Log keyword's trailing "$" here then
# the last line would read instead:
#
- # $ content which follows
+ # foo$content which follows
assert(next_match_segment[0] == ord('$'))
next_match_segment = next_match_segment[1:]
- p = prefix
expbuf += filename \
if (expkw & self.RCS_KW_FULLPATH) != 0 \
else os.path.basename(filename)
expbuf += " "
- logbuf = p + (
+ logbuf = prefix + (
'Revision %s %s %s\n' % (
rev[0], time.strftime(
"%Y/%m/%d %H:%M:%S", time.gmtime(rev[1])),
rev[2])).encode('ascii')
for lline in rcs.getlog(rev[0]).splitlines(keepends=True):
- logbuf += p + lline
+ logbuf += prefix + lline
if (expkw & self.RCS_KW_SOURCE) != 0:
expbuf += filename
expbuf += " "
@@ -707,13 +709,10 @@
# the previous keyword and marks the beginning of the next one.
line0 = line0[:-1]
elif logbuf is not None:
- # Trim whitespace from the beginning of text following the Log keyword.
- # But leave a lone trailing empty line as-is. Which seems inconsistent,
- # but testing suggests that this matches CVS's behaviour.
- if len(line) == 1 and line[0] == ord('\n'):
- ret.append(line0 + prefix + line)
- else:
- ret.append(line0 + prefix + line.lstrip())
+ # Trim whitespace from tail of prefix if appending a suffix which
+ # followed the Log keyword on the same line.
+ # Testing suggests that this matches CVS's behaviour.
+ ret.append(line0 + prefix.rstrip() + line)
else:
ret.append(line0 + line)
return b''.join(ret)
diff --git a/swh/loader/cvs/tests/data/rcsbase-log-kw-test-repo.tgz b/swh/loader/cvs/tests/data/rcsbase-log-kw-test-repo.tgz
new file mode 100644
index 0000000000000000000000000000000000000000..0000000000000000000000000000000000000000
GIT binary patch
literal 0
Hc$@<O00001
literal 0
Hc$@<O00001
diff --git a/swh/loader/cvs/tests/test_loader.py b/swh/loader/cvs/tests/test_loader.py
--- a/swh/loader/cvs/tests/test_loader.py
+++ b/swh/loader/cvs/tests/test_loader.py
@@ -995,3 +995,89 @@
}
check_snapshot(GREEK_SNAPSHOT9, loader.storage)
+
+
+RCSBASE_SNAPSHOT = Snapshot(
+ id=hash_to_bytes("2c75041ba8868df04349c1c8f4c29f992967b8aa"),
+ branches={
+ b"HEAD": SnapshotBranch(
+ target=hash_to_bytes("46f076387ff170dc3d4da5e43d953c1fc744c821"),
+ target_type=TargetType.REVISION,
+ )
+ },
+)
+
+
+def test_loader_cvs_expand_log_keyword2(swh_storage, datadir, tmp_path):
+ """Another conversion of RCS history with Log keyword in files"""
+ archive_name = "rcsbase-log-kw-test-repo"
+ archive_path = os.path.join(datadir, f"{archive_name}.tgz")
+ repo_url = prepare_repository_from_archive(archive_path, archive_name, tmp_path)
+ repo_url += "/src" # CVS module name
+
+ loader = CvsLoader(
+ swh_storage, repo_url, cvsroot_path=os.path.join(tmp_path, archive_name)
+ )
+
+ assert loader.load() == {"status": "eventful"}
+
+ assert_last_visit_matches(
+ loader.storage,
+ repo_url,
+ status="full",
+ type="cvs",
+ snapshot=RCSBASE_SNAPSHOT.id,
+ )
+
+ check_snapshot(RCSBASE_SNAPSHOT, loader.storage)
+
+ stats = get_stats(loader.storage)
+ assert stats == {
+ "content": 2,
+ "directory": 3,
+ "origin": 1,
+ "origin_visit": 1,
+ "release": 0,
+ "revision": 3,
+ "skipped_content": 0,
+ "snapshot": 1,
+ }
+
+
+def test_loader_cvs_pserver_expand_log_keyword2(swh_storage, datadir, tmp_path):
+ """Another conversion of RCS history with Log keyword in files"""
+ archive_name = "rcsbase-log-kw-test-repo"
+ archive_path = os.path.join(datadir, f"{archive_name}.tgz")
+ repo_url = prepare_repository_from_archive(archive_path, archive_name, tmp_path)
+ repo_url += "/src" # CVS module name
+
+ # Ask our cvsclient to connect via the 'cvs server' command
+ repo_url = f"fake://{repo_url[7:]}"
+
+ loader = CvsLoader(
+ swh_storage, repo_url, cvsroot_path=os.path.join(tmp_path, archive_name)
+ )
+
+ assert loader.load() == {"status": "eventful"}
+
+ assert_last_visit_matches(
+ loader.storage,
+ repo_url,
+ status="full",
+ type="cvs",
+ snapshot=RCSBASE_SNAPSHOT.id,
+ )
+
+ check_snapshot(RCSBASE_SNAPSHOT, loader.storage)
+
+ stats = get_stats(loader.storage)
+ assert stats == {
+ "content": 2,
+ "directory": 3,
+ "origin": 1,
+ "origin_visit": 1,
+ "release": 0,
+ "revision": 3,
+ "skipped_content": 0,
+ "snapshot": 1,
+ }
File Metadata
Details
Attached
Mime Type
text/plain
Expires
Wed, Dec 18, 4:06 PM (5 h, 3 m ago)
Storage Engine
blob
Storage Format
Raw Data
Storage Handle
3220618
Attached To
D6813: fix Log keyword expansion with trailing whitespace in prefix
Event Timeline
Log In to Comment