Merge feat/p7-times: Phase 7 Wave D (real dir/symlink time preservation making -O/-J meaningful; secluded-args -> Impossible/Divergence; PROTOCOL 2.17.0)

This commit is contained in:
2026-09-12 11:22:23 +02:00
25 changed files with 1000 additions and 77 deletions
+81 -9
View File
@@ -127,6 +127,11 @@ static bool prepare_scanner(const Config* config, int num_threads, PreparedScann
options->excluded_paths = NULL;
options->excluded_mutex = NULL;
options->hardlinks = NULL;
/* P7 Wave D: capture source directory times whenever metadata rides the
wire. Whether they are APPLIED is decided receiver-side (-O skips). */
options->capture_dir_times = config->use_metadata;
options->dir_entries = NULL;
options->dir_entries_mutex = NULL;
if (config->preserve_hard_links) {
out->hardlinks = hardlink_table_create();
if (!out->hardlinks) {
@@ -1115,14 +1120,51 @@ static bool send_file_direct(File* file, int fd, bool use_metadata, int compress
}
/* Transmit one explicit directory entry (--dirs): a STATUS_MKDIR frame whose
payload is only the destination path. The receiver validates the path and
creates the directory under the receive root. */
static bool send_directory_entry(Client* client, File* file) {
payload is the destination path and, when metadata is negotiated, the
directory's metadata frame. The receiver validates the path, creates the
directory under the receive root, and (metadata case) defers applying its
times to the end of the transfer so -O/--omit-dir-times is honored. */
static bool send_directory_entry(const Client* client, File* file, const Config* config) {
if (!file || !file_wire_path(file))
return false;
if (!send_status(client->file_descriptor, STATUS_MKDIR))
if (!send_status(client->file_descriptor, STATUS_MKDIR) ||
!send_wire_str(client->file_descriptor, file_wire_path(file)))
return false;
return send_wire_str(client->file_descriptor, file_wire_path(file));
return !config->use_metadata || metadata_send(client->file_descriptor, file->metadata);
}
/* P7 Wave D: transmit every captured source directory's metadata in terminal
STATUS_DIR_TIMES frames (count, then (path, metadata) pairs) after all file
data and the optional delete manifest. The receiver applies them at the END
of its own transfer (after deletion and --delay-updates publication) so a
directory's mtime is not clobbered by writing its children. A non-metadata
transfer (or an empty set) sends nothing, keeping the stream byte-identical.
The receiver rejects a frame whose count exceeds MAX_MANIFEST_ENTRIES, so a
huge tree is CHUNKED into repeated frames of at most that many entries each
(the receiver's loop handles repeated STATUS_DIR_TIMES frames). Every frame
stays within the receiver's bound, and a frame that would exceed it is never
emitted. */
static bool send_dir_times(const Client* client, const Config* config, ArrayList* dir_entries) {
if (!client || !config || !config->use_metadata || !dir_entries || dir_entries->size == 0)
return true;
int fd = client->file_descriptor;
int index = 0;
while (index < dir_entries->size) {
int remaining = dir_entries->size - index;
int chunk = remaining > MAX_MANIFEST_ENTRIES ? MAX_MANIFEST_ENTRIES : remaining;
if (!send_status(fd, STATUS_DIR_TIMES) || !send_int(fd, chunk))
return false;
for (int i = 0; i < chunk; i++) {
File* file = (File*)dir_entries->items[index + i];
if (!file || !file_wire_path(file))
return false;
if (!send_wire_str(fd, file_wire_path(file)) || !metadata_send(fd, file->metadata))
return false;
}
index += chunk;
}
return true;
}
/* Transmit one symlink entry: a STATUS_SYMLINK frame carrying the destination
@@ -1313,10 +1355,10 @@ static int send_chunk_with_removal(Client* client, Chunk* chunk, Config* config,
if (f == NULL)
continue;
if (f->is_dir) {
/* Explicit directory entry (--dirs): a MKDIR frame carrying only the
destination path. Directories have no source to remove and no
incremental check. */
if (!send_directory_entry(client, f))
/* Explicit directory entry (--dirs): a MKDIR frame carrying the
destination path (and metadata when negotiated). Directories have no
source to remove and no incremental check. */
if (!send_directory_entry(client, f, config))
return -1;
change_emit_dir_sent(config, f);
continue;
@@ -1501,6 +1543,13 @@ static int send_chunks_multithreaded(void* pipeline_context) {
if (send_delete_manifest(client->file_descriptor, NULL, NULL, context->missing_args) != 0)
goto send_fail;
}
/* P7 Wave D: transmit the captured directory times last. The scanner thread
(and all parallel workers) has been joined before scanner_done was set, so
the list is complete and race-free; on an early stop the list may be
incomplete and is deliberately not sent. */
if (!context->scan_stopped_early &&
!send_dir_times(client, context->config, context->dir_entries))
goto send_fail;
bool ok = finalize_transfer(client, context->config, context->remove_source_files);
if (!ok && context->config->use_delete)
log_message(LOG_LEVEL_ERROR,
@@ -1542,6 +1591,10 @@ static int scan_directory_multithreaded(void* pipeline_context) {
return thrd_error;
}
prepared.options.stop_condition = &context->stop_condition;
/* P7 Wave D: the recursive scan feeds the shared directory-time list; the
parallel workers append under the context's dedicated mutex. */
prepared.options.dir_entries = context->dir_entries;
prepared.options.dir_entries_mutex = &context->dir_entries_mutex;
/* The keep-set manifest for the late modes is built from this data pass, so
the parallel scanner records the protected excluded prefixes here. The
early modes already transmitted the pre-scan keep-set and its protected
@@ -1843,6 +1896,9 @@ int send_files(Config* config) {
DirectoryScanner* scanner = NULL;
ArrayList* manifest = NULL;
ArrayList* remove_sources = NULL;
/* P7 Wave D: captured source directory times, transmitted in trailing
STATUS_DIR_TIMES frame(s) (only when metadata rides the wire). */
ArrayList* dir_entries = NULL;
/* Protected excluded prefixes (delete-excluded default protection). */
ArrayList* excluded = NULL;
bool delete_early = config->use_delete && config_delete_timing_early(config);
@@ -1855,6 +1911,11 @@ int send_files(Config* config) {
receive_daemon_motd(client, config);
if (!prepare_scanner(config, 0, &prepared))
goto send_fail;
if (config->use_metadata) {
dir_entries = array_list_create(file_destroy);
if (!dir_entries)
goto send_fail;
}
if (config->remove_source_files)
remove_sources = array_list_create(source_file_destroy);
if (config->remove_source_files && !remove_sources)
@@ -1919,6 +1980,10 @@ int send_files(Config* config) {
StopCondition stop = stop_condition_make(config->stop_after_mins > 0, config->stop_after_mins,
config->stop_at_set, config->stop_at, now_mono);
prepared.options.stop_condition = &stop;
/* The early-delete pre-scan above already ran; only the data pass should feed
the directory-time list (otherwise every directory would be captured
twice). */
prepared.options.dir_entries = dir_entries;
scanner = directory_scanner_create_with_options(config->send_directory, &prepared.options);
if (!scanner)
goto send_fail;
@@ -2037,6 +2102,11 @@ int send_files(Config* config) {
}
}
}
/* P7 Wave D: every directory has now been traversed (or the scan stopped
early), so transmit the captured directory times last. The receiver defers
applying them until after its own deletion/publication phase. */
if (!send_dir_times(client, config, dir_entries))
goto send_fail;
bool ok = finalize_transfer(client, config, remove_sources);
if (!ok && config->use_delete)
log_message(LOG_LEVEL_ERROR,
@@ -2078,6 +2148,8 @@ send_fail:
array_list_delete(missing_args);
if (remove_sources)
array_list_delete(remove_sources);
if (dir_entries)
array_list_delete(dir_entries);
if (scanner)
directory_scanner_destroy(scanner);
prepared_scanner_destroy(&prepared);
+83
View File
@@ -488,6 +488,9 @@ DirectoryScanner* directory_scanner_create_with_options(const char* root_directo
scanner->hardlinks = options->hardlinks;
scanner->prune_empty_dirs = options->prune_empty_dirs;
scanner->stop_condition = options->stop_condition;
scanner->capture_dir_times = options->capture_dir_times;
scanner->dir_entries = options->dir_entries;
scanner->dir_entries_mutex = options->dir_entries_mutex;
scanner->dirs_root_emitted = false;
scanner->list_index = 0;
scanner->dirs_batch = NULL;
@@ -595,6 +598,63 @@ static Chunk* chunk_data_to_chunk(ArrayList* chunk_data) {
return chunk;
}
/* P7 Wave D: append one traversed source directory's captured metadata to the
* shared pending-directory-time list. The File carries no payload; only the
* wire path (absolute fs path normally, the bare relative path under
* -R + --files-from) and its metadata are used, and the sender transmits them
* in trailing STATUS_DIR_TIMES frame(s). `mutex` (optional) serializes the
* append for the parallel scanner's shared workers. An unstattable or
* non-directory path is silently skipped (the transfer is unaffected); an
* allocation failure is fatal and reported to the caller. */
static bool scanner_capture_dir_time(ArrayList* dir_entries, mtx_t* mutex, const char* root_path,
const char* fs_path, bool relative_mode, bool preserve_atimes,
bool preserve_crtimes) {
if (!dir_entries || !root_path || !fs_path)
return true;
struct stat st;
if (stat(fs_path, &st) != 0 || !S_ISDIR(st.st_mode))
return true;
char* rel = scanner_path_relative(root_path, fs_path);
if (!rel)
return true;
if (relative_mode && rel[0] == '\0') {
/* -R + --files-from: the transfer root itself has no bare relative wire
path (matches the -R scan, which never emits the root). */
free(rel);
return true;
}
File* file = file_create(fs_path);
if (!file) {
free(rel);
return false;
}
file->is_dir = true;
file->metadata = file_metadata_create(fs_path, &st, preserve_atimes, preserve_crtimes);
if (!file->metadata) {
free(rel);
file_destroy(file);
return false;
}
if (relative_mode) {
file->send_path = rel;
rel = NULL;
}
free(rel);
bool added;
if (mutex) {
mtx_lock(mutex);
added = array_list_add(dir_entries, file);
mtx_unlock(mutex);
} else {
added = array_list_add(dir_entries, file);
}
if (!added) {
file_destroy(file);
return false;
}
return true;
}
/* Open the next queued directory and set up its filter context. Returns 1 when
a directory is open, 0 when the queue is exhausted, and -1 on a fatal error.
A directory that cannot be opened is an I/O error: it is recorded on the
@@ -661,6 +721,17 @@ static int open_next_directory(DirectoryScanner* scanner) {
scanner->current_path = NULL;
return -1;
}
if (scanner->capture_dir_times &&
!scanner_capture_dir_time(scanner->dir_entries, scanner->dir_entries_mutex,
scanner->root_path, scanner->current_path, scanner->relative_mode,
scanner->preserve_atimes, scanner->preserve_crtimes)) {
closedir(scanner->current_dir);
scanner->current_dir = NULL;
free(scanner->current_path);
scanner->current_path = NULL;
scanner->failed = true;
return -1;
}
return 1;
}
return 0;
@@ -1584,6 +1655,18 @@ ParallelScanner* parallel_scanner_create_with_options(const char* root_directory
parallel_scanner_destroy(ps);
return NULL;
}
/* P7 Wave D: the parallel scanner never runs a DirectoryScanner over the
transfer root itself (it hands the root's immediate subdirectories to
workers), so capture the root's directory time here. */
if (options->capture_dir_times &&
!scanner_capture_dir_time(options->dir_entries, options->dir_entries_mutex, root_directory,
root_directory, options->relative && options->file_list != NULL,
options->preserve_atimes, options->preserve_crtimes)) {
array_list_delete(root_files);
array_list_delete(subdirs);
parallel_scanner_destroy(ps);
return NULL;
}
unsigned long long cs = options->chunk_size > 0 ? options->chunk_size : DESIRED_CHUNK_SIZE;
ps->initial_chunk = batch_files(root_files, cs, ps->result_queue, &ps->failed);
+16
View File
@@ -98,6 +98,18 @@ typedef struct {
* (without marking the scan as failed), so a busy scan itself stops early.
* Client-only, never serialized to the wire. */
const StopCondition* stop_condition;
/* P7 Wave D (protocol 2.17.0): directory-time capture sink. When
* `capture_dir_times` is true the recursive scan appends one is_dir File
* (with metadata, no payload) per source directory it traverses to
* `dir_entries`, so the sender can transmit trailing STATUS_DIR_TIMES
* frame(s) and the receiver can apply directory mtimes AFTER all children
* are written. `dir_entries_mutex` (optional) guards the list
* for the parallel scanner's shared worker threads; the caller owns both.
* The --dirs generator does not use this (its directory entries carry their
* metadata inline through STATUS_MKDIR). */
bool capture_dir_times;
ArrayList* dir_entries;
mtx_t* dir_entries_mutex;
} ScannerOptions;
/* Internal per-scanner filter state. FilterNode chains represent the ordered
@@ -173,6 +185,10 @@ typedef struct {
HardLinkTable* hardlinks;
/* Phase 6: sender stop deadline (from ScannerOptions). */
const StopCondition* stop_condition;
/* P7 Wave D directory-time capture (see ScannerOptions). */
bool capture_dir_times;
ArrayList* dir_entries;
mtx_t* dir_entries_mutex;
} DirectoryScanner;
typedef struct {