Add phase field to file_status JSON output with correct lifecycle emission

Author: YK-03

src/borg/archive.py

@@ -1408,7 +1408,13 @@ def process_pipe(self, *, path, cache, fd, mode, user=None, group=None):
             item.uid = uid
         if gid is not None:
             item.gid = gid
-        self.process_file_chunks(item, cache, self.stats, self.show_progress, backup_io_iter(self.chunker.chunkify(fd)))
+        self.print_file_status(None, path, phase="start")
+        try:
+            self.process_file_chunks(
+                item, cache, self.stats, self.show_progress, backup_io_iter(self.chunker.chunkify(fd))
+            )
+        finally:
+            self.print_file_status(None, path, phase="end")
         item.get_size(memorize=True)
         self.stats.nfiles += 1
         self.add_item(item, stats=self.stats)
@@ -1475,17 +1481,21 @@ def process_file(self, *, path, parent_fd, name, st, cache, flags=flags_normal,
                     # Only chunkify the file if needed
                     changed_while_backup = False
                     if "chunks" not in item:
-                        start_reading = time.time_ns()
-                        with backup_io("read"):
-                            self.process_file_chunks(
-                                item,
-                                cache,
-                                self.stats,
-                                self.show_progress,
-                                backup_io_iter(self.chunker.chunkify(None, fd)),
-                            )
-                            self.stats.chunking_time = self.chunker.chunking_time
-                        end_reading = time.time_ns()
+                        self.print_file_status(None, path, phase="start")
+                        try:
+                            start_reading = time.time_ns()
+                            with backup_io("read"):
+                                self.process_file_chunks(
+                                    item,
+                                    cache,
+                                    self.stats,
+                                    self.show_progress,
+                                    backup_io_iter(self.chunker.chunkify(None, fd)),
+                                )
+                                self.stats.chunking_time = self.chunker.chunking_time
+                            end_reading = time.time_ns()
+                        finally:
+                            self.print_file_status(None, path, phase="end")
                         with backup_io("fstat2"):
                             st2 = os.fstat(fd)
                         if self.files_changed == "disabled" or is_special_file:

src/borg/archiver/__init__.py

@@ -156,33 +156,31 @@ def print_warning_instance(self, warning):
         self.print_warning(msg, *args, wc=wc, wt="curly", msgid=msgid)
 
     def print_file_status(self, status, path, *, phase=None, error=None):
-        # START lifecycle event (JSON only)
+        # START lifecycle event
         if self.output_list and self.log_json and phase == "start" and status is None:
             json_data = {"type": "file_status", "phase": "start"}
             json_data |= text_to_json("path", path)
             if error is not None:
                 json_data["error"] = error
             print(json.dumps(json_data), file=sys.stderr)
             return
-        # END lifecycle event (JSON only)
+
+        # END lifecycle event
         if self.output_list and self.log_json and phase == "end" and status is None:
             json_data = {"type": "file_status", "phase": "end"}
             json_data |= text_to_json("path", path)
             if error is not None:
                 json_data["error"] = error
             print(json.dumps(json_data), file=sys.stderr)
             return
-        # regular status event (A, M, U, -, d, s, etc.)
+
+        # if we get called with status == None, the final file status was already printed
         if self.output_list and status is not None and (self.output_filter is None or status in self.output_filter):
             if self.log_json:
                 json_data = {"type": "file_status", "status": status}
                 json_data |= text_to_json("path", path)
-                if phase is not None:
-                    json_data["phase"] = phase
-                if error is not None:
-                    json_data["error"] = error
                 print(json.dumps(json_data), file=sys.stderr)
-            elif not self.log_json and status is not None:
+            else:
                 logging.getLogger("borg.output.list").info("%1s %s", status, remove_surrogates(path))
 
     def preprocess_args(self, args):