Summary claims 143 ✅ / 0 divergence / 0 partial, which is misleading: many ✅ rows embed documented rsync divergences, and some (e.g. -r, -b, -L shorts) are not parsed at all.
The §8 row labeled -M, --preserve conflates two flags: -M is --remote-option; --preserve is FastSync-only (-p+-t).
§15 "1 GB per connection" is wrong: MAX_CONNECTION_MEMORY is 256 MiB.
--rsync-path is claimed to cross the wire; fastsync_server_path is client-only.
-p row should state the group/other-write (and file setuid/setgid/sticky) masking divergence.
-D is ✅ but implies the --specials socket case.
Fake-super row text elsewhere says "no replay" while replay exists.
Fix: correct these statements and reclassify rows that are not true parity.
Audit findings (documentation only).
- Summary claims 143 ✅ / 0 divergence / 0 partial, which is misleading: many ✅ rows embed documented rsync divergences, and some (e.g. `-r`, `-b`, `-L` shorts) are not parsed at all.
- The §8 row labeled `-M, --preserve` conflates two flags: `-M` is `--remote-option`; `--preserve` is FastSync-only (`-p`+`-t`).
- §15 "1 GB per connection" is wrong: `MAX_CONNECTION_MEMORY` is 256 MiB.
- `--rsync-path` is claimed to cross the wire; `fastsync_server_path` is client-only.
- `-p` row should state the group/other-write (and file setuid/setgid/sticky) masking divergence.
- `-D` is ✅ but implies the `--specials` socket case.
- Fake-super row text elsewhere says "no replay" while replay exists.
Fix: correct these statements and reclassify rows that are not true parity.
Fixed. The summary tally is now accurate (117 ✅ / 13 ⚠️ / 27 ❌), -M/--preserve are separate rows, §15 says 256 MiB, --rsync-path is client-only, the -p masking divergence is stated, -D sockets are covered, fake-super replay is described, -F and -i were reclassified ✅ → ⚠️, and stale --bwlimit/munge/checksum bullets were corrected. Closing as completed.
Fixed. The summary tally is now accurate (117 ✅ / 13 ⚠️ / 27 ❌), `-M`/`--preserve` are separate rows, §15 says 256 MiB, `--rsync-path` is client-only, the `-p` masking divergence is stated, `-D` sockets are covered, fake-super replay is described, `-F` and `-i` were reclassified ✅ → ⚠️, and stale `--bwlimit`/munge/checksum bullets were corrected. Closing as completed.
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Audit findings (documentation only).
-r,-b,-Lshorts) are not parsed at all.-M, --preserveconflates two flags:-Mis--remote-option;--preserveis FastSync-only (-p+-t).MAX_CONNECTION_MEMORYis 256 MiB.--rsync-pathis claimed to cross the wire;fastsync_server_pathis client-only.-prow should state the group/other-write (and file setuid/setgid/sticky) masking divergence.-Dis ✅ but implies the--specialssocket case.Fix: correct these statements and reclassify rows that are not true parity.
Fixed. The summary tally is now accurate (117 ✅ / 13 ⚠️ / 27 ❌),
-M/--preserveare separate rows, §15 says 256 MiB,--rsync-pathis client-only, the-pmasking divergence is stated,-Dsockets are covered, fake-super replay is described,-Fand-iwere reclassified ✅ → ⚠️, and stale--bwlimit/munge/checksum bullets were corrected. Closing as completed.