fix: report partial Syncthing transfer progress

src/archive_clients/syncthing.py

@@ -139,19 +139,14 @@ class SyncthingTransferObserver:
 
     def status(self) -> SyncthingTransferStatus:
         published = load_published_transfer(self.local_job_directory)
-        total = 0
-        for entry in published.manifest.files:
-            total += _verified_job_file_size(
-                self.local_job_directory,
-                entry.payload_relative_path,
-                entry.logical_bytes,
-            )
-        for artifact in published.manifest.artifacts:
-            total += _verified_job_file_size(
-                self.local_job_directory,
-                artifact.payload_relative_path,
-                artifact.logical_bytes,
-            )
+        declared_files = [
+            (entry.payload_relative_path, entry.logical_bytes)
+            for entry in published.manifest.files
+        ] + [
+            (artifact.payload_relative_path, artifact.logical_bytes)
+            for artifact in published.manifest.artifacts
+        ]
+        total = sum(size for _, size in declared_files)
 
         completion = self.transport.get_json(
             "/rest/db/completion?"
@@ -178,11 +173,36 @@ class SyncthingTransferObserver:
             name for name in needed_names
             if name == self.job_relative_path or name.startswith(prefix)
         }
-        fraction = float(raw_fraction) / 100
-        complete = fraction == 1 and not relevant
+        complete = not relevant
+        if complete:
+            for relative_path, expected_bytes in declared_files:
+                _verified_job_file_size(
+                    self.local_job_directory,
+                    relative_path,
+                    expected_bytes,
+                )
+            completed_bytes = total
+        else:
+            observed_bytes = sum(
+                _received_job_file_bytes(
+                    self.local_job_directory,
+                    relative_path,
+                    expected_bytes,
+                )
+                for relative_path, expected_bytes in declared_files
+            )
+            observed_fraction = observed_bytes / total if total else 1.0
+            # Folder completion remains useful when Syncthing has already
+            # atomically published a file but still reports it in its need
+            # queue; the allocated-byte estimate is needed for sparse temp
+            # files that are pre-sized before their blocks arrive.
+            fraction = min(observed_fraction, float(raw_fraction) / 100)
+            completed_bytes = int(total * fraction)
+        if complete:
+            fraction = 1.0
         return SyncthingTransferStatus(
             fraction,
-            total if complete else int(total * fraction),
+            completed_bytes,
             total,
             complete,
             len(relevant),
@@ -472,6 +492,42 @@ def _verified_job_file_size(
     return metadata.st_size
 
 
+def _received_job_file_bytes(
+    job_directory: Path,
+    relative_path: str,
+    expected_bytes: int,
+) -> int:
+    """Return a conservative receive estimate for one job-owned file.
+
+    Syncthing writes incomplete files as ``.syncthing.<name>.tmp`` and may
+    pre-size that sparse temporary to its final logical length. Allocated
+    blocks, rather than ``st_size``, therefore provide the useful progress
+    signal until the final atomic rename occurs.
+    """
+
+    relative = PurePosixPath(relative_path)
+    current = job_directory
+    for component in relative.parts[:-1]:
+        current = current / component
+    final = current / relative.name
+    try:
+        metadata = final.lstat()
+    except FileNotFoundError:
+        metadata = None
+    if metadata is not None:
+        if stat.S_ISREG(metadata.st_mode) and metadata.st_size == expected_bytes:
+            return expected_bytes
+        return 0
+    temporary = current / f".syncthing.{relative.name}.tmp"
+    try:
+        temporary_metadata = temporary.lstat()
+    except FileNotFoundError:
+        return 0
+    if not stat.S_ISREG(temporary_metadata.st_mode):
+        return 0
+    return min(expected_bytes, temporary_metadata.st_blocks * 512)
+
+
 def _needed_names(value: dict[str, Any]) -> set[str]:
     result: set[str] = set()
     for key in ("progress", "queued", "rest"):