tags: low-level infrastructure for archive tags

Read or modify this set, only add validated str to it:

Archive.tags: Optional[set[str]]

borg info [--json] <archive> displays a list of comma-separated archive tags (currently always empty).

src/borg/archive.py

@@ -473,6 +473,7 @@ def __init__(
         self.name = name  # overwritten later with name from archive metadata
         self.name_in_manifest = name  # can differ from .name later (if borg check fixed duplicate archive names)
         self.comment = None
+        self.tags = None
         self.numeric_ids = numeric_ids
         self.noatime = noatime
         self.noctime = noctime
@@ -495,6 +496,7 @@ def __init__(
         self.create = create
         if self.create:
             self.items_buffer = CacheChunkBuffer(self.cache, self.key, self.stats)
+            self.tags = set()
         else:
             if name_is_id:
                 # we also go over the manifest here to avoid quick&dirty deleted archives
@@ -521,6 +523,7 @@ def load(self, id):
         self.metadata = self._load_meta(self.id)
         self.name = self.metadata.name
         self.comment = self.metadata.get("comment", "")
+        self.tags = set(self.metadata.get("tags", []))
 
     @property
     def ts(self):
@@ -573,6 +576,7 @@ def info(self):
                     "hostname": self.metadata.hostname,
                     "username": self.metadata.username,
                     "comment": self.metadata.get("comment", ""),
+                    "tags": sorted(self.tags),
                     "chunker_params": self.metadata.get("chunker_params", ""),
                 }
             )
@@ -632,6 +636,7 @@ def save(self, name=None, comment=None, timestamp=None, stats=None, additional_m
             "version": 2,
             "name": name,
             "comment": comment or "",
+            "tags": list(sorted(self.tags)),
             "item_ptrs": item_ptrs,  # see #1473
             "command_line": join_cmd(sys.argv),
             "hostname": hostname,

src/borg/archiver/info_cmd.py

@@ -32,6 +32,7 @@ def do_info(self, args, repository, manifest, cache):
                 output_data.append(info)
             else:
                 info["duration"] = format_timedelta(timedelta(seconds=info["duration"]))
+                info["tags"] = ",".join(info["tags"])
                 print(
                     textwrap.dedent(
                         """
@@ -40,6 +41,7 @@ def do_info(self, args, repository, manifest, cache):
                 Comment: {comment}
                 Hostname: {hostname}
                 Username: {username}
+                Tags: {tags}
                 Time (start): {start}
                 Time (end): {end}
                 Duration: {duration}

src/borg/constants.py

@@ -12,6 +12,7 @@
 # this set must be kept complete, otherwise rebuild_manifest might malfunction:
 # fmt: off
 ARCHIVE_KEYS = frozenset(['version', 'name', 'hostname', 'username', 'time', 'time_end',
+                          'tags',  # v2+ archives
                           'items',  # legacy v1 archives
                           'item_ptrs',  # v2+ archives
                           'comment', 'chunker_params',

src/borg/item.pyx

@@ -511,6 +511,7 @@ cdef class ArchiveItem(PropDict):
     time = PropDictProperty(str)
     time_end = PropDictProperty(str)
     comment = PropDictProperty(str, 'surrogate-escaped str')
+    tags = PropDictProperty(list)  # list of s-e-str
     chunker_params = PropDictProperty(tuple)
     recreate_cmdline = PropDictProperty(list)  # legacy, list of s-e-str
     recreate_command_line = PropDictProperty(str, 'surrogate-escaped str')

src/borg/testsuite/archiver/info_cmd.py

@@ -32,6 +32,7 @@ def test_info_json(archivers, request):
     assert isinstance(archive["command_line"], str)
     assert isinstance(archive["duration"], float)
     assert len(archive["id"]) == 64
+    assert archive["tags"] == []
     assert "stats" in archive
     checkts(archive["start"])
     checkts(archive["end"])