add generic text and comment validator

src/borg/helpers/__init__.py

@@ -26,7 +26,7 @@ from .parseformat import sizeof_fmt, sizeof_fmt_iec, sizeof_fmt_decimal
 from .parseformat import format_line, replace_placeholders, PlaceholderError
 from .parseformat import SortBySpec, NameSpec
 from .parseformat import format_archive, parse_stringified_list, clean_lines
-from .parseformat import Location, location_validator, archivename_validator
+from .parseformat import Location, location_validator, archivename_validator, comment_validator, text_validator
 from .parseformat import BaseFormatter, ArchiveFormatter, ItemFormatter, file_status
 from .parseformat import swidth_slice, ellipsis_truncate
 from .parseformat import BorgJsonEncoder, basic_json_data, json_print, json_dump, prepare_dump_dict

src/borg/helpers/parseformat.py

@@ -573,6 +573,26 @@ def archivename_validator():
     return validator
 
 
+def text_validator(*, name, max_length, invalid_ctrl_chars="\0"):
+    def validator(text):
+        assert isinstance(text, str)
+        if not (len(text) <= max_length):
+            raise argparse.ArgumentTypeError(f'Invalid {name}: "{text}" [length <= {max_length}]')
+        if re.search(f"[{re.escape(invalid_ctrl_chars)}]", text):
+            raise argparse.ArgumentTypeError(f'Invalid {name}: "{text}" [invalid control chars detected]')
+        try:
+            text.encode("utf-8", errors="strict")
+        except UnicodeEncodeError:
+            # looks like text contains surrogate-escapes
+            raise argparse.ArgumentTypeError(f'Invalid {name}: "{text}" [contains non-unicode characters]')
+        return text
+
+    return validator
+
+
+comment_validator = text_validator(name="comment", max_length=10000)
+
+
 class BaseFormatter:
     FIXED_KEYS = {
         # Formatting aids

src/borg/testsuite/helpers.py

@@ -32,7 +32,7 @@ from ..helpers import msgpack
 from ..helpers import yes, TRUISH, FALSISH, DEFAULTISH
 from ..helpers import StableDict, bin_to_hex
 from ..helpers import parse_timestamp, ChunkIteratorFileWrapper, ChunkerParams
-from ..helpers import archivename_validator
+from ..helpers import archivename_validator, text_validator
 from ..helpers import ProgressIndicatorPercent, ProgressIndicatorEndless
 from ..helpers import swidth_slice
 from ..helpers import chunkit
@@ -290,6 +290,29 @@ def test_archivename_invalid(name):
         av(name)
 
 
+@pytest.mark.parametrize("text", ["", "single line", "multi\nline\ncomment"])
+def test_text_ok(text):
+    tv = text_validator(max_length=100, name="name")
+    tv(text)  # must not raise an exception
+
+
+@pytest.mark.parametrize(
+    "text",
+    [
+        "x" * 101,  # too long
+        # invalid chars:
+        "foo\0bar",
+        # contains surrogate-escapes
+        "foo\udc80bar",
+        "foo\udcffbar",
+    ],
+)
+def test_text_invalid(text):
+    tv = text_validator(max_length=100, name="name")
+    with pytest.raises(ArgumentTypeError):
+        tv(text)
+
+
 class FormatTimedeltaTestCase(BaseTestCase):
     def test(self):
         t0 = datetime(2001, 1, 1, 10, 20, 3, 0)