Add support for --wrap option

bagit.py

@@ -14,6 +14,7 @@
 import signal
 import sys
 import tempfile
+import textwrap
 import unicodedata
 import warnings
 from collections import defaultdict
@@ -137,7 +138,8 @@ def find_locale_dir():
 
 
 def make_bag(
- bag_dir, bag_info=None, processes=1, checksums=None, checksum=None, encoding="utf-8"
+ bag_dir, bag_info=None, processes=1, checksums=None, checksum=None,
+ encoding="utf-8", tag_wrap_column=0
 ):
  """
  Convert a given directory into a bag. You can pass in arbitrary
@@ -256,7 +258,7 @@ def make_bag(
  )
 
  bag_info["Payload-Oxum"] = "%s.%s" % (total_bytes, total_files)
- _make_tag_file("bag-info.txt", bag_info)
+ _make_tag_file("bag-info.txt", bag_info, tag_wrap_column)
 
  for c in checksums:
  _make_tagmanifest_file(c, bag_dir, encoding="utf-8")
@@ -450,7 +452,7 @@ def payload_entries(self):
  if key.startswith("data" + os.sep)
  )
 
- def save(self, processes=1, manifests=False):
+ def save(self, processes=1, manifests=False, tag_wrap_column=0):
  """
  save will persist any changes that have been made to the bag
  metadata (self.info).
@@ -463,6 +465,11 @@ def save(self, processes=1, manifests=False):
 
  If you want to control the number of processes that are used when
  recalculating checksums use the processes parameter.
+
+ If you want long tag values to be wrapped by breaking long strings at
+ whitespace characters, set tag_wrap_column to a value greater than 0.
+ An integer value greater than 0 causes line-wrapping to be performed on
+ a best-effort basis to limit line lengths to the given value.
  """
  # Error checking
  if not self.path:
@@ -514,7 +521,7 @@ def save(self, processes=1, manifests=False):
  LOGGER.info(_("Updating Payload-Oxum in %s"), self.tag_file_name)
  self.info["Payload-Oxum"] = "%s.%s" % (total_bytes, total_files)
 
- _make_tag_file(self.tag_file_name, self.info)
+ _make_tag_file(self.tag_file_name, self.info, tag_wrap_column)
 
  # Update tag-manifest for changes to manifest & bag-info files
  for alg in self.algorithms:
@@ -1219,19 +1226,47 @@ def _parse_tags(tag_file):
  yield (tag_name, tag_value.strip())
 
 
-def _make_tag_file(bag_info_path, bag_info):
+def _make_tag_file(bag_info_path, bag_info, tag_wrap_column):
  headers = sorted(bag_info.keys())
  with open_text_file(bag_info_path, "w") as f:
  for h in headers:
  values = bag_info[h]
  if not isinstance(values, list):
  values = [values]
  for txt in values:
- # strip CR, LF and CRLF so they don't mess up the tag file
- txt = re.sub(r"\n|\r|(\r\n)", "", force_unicode(txt))
+ txt = _format_tag_value(h, txt, tag_wrap_column)
  f.write("%s: %s\n" % (h, txt))
 
 
+def _format_tag_value(header, txt, tag_wrap_column):
+ txt = force_unicode(txt)
+ if tag_wrap_column > 1:
+ # Account for colon & space written after the property name.
+ prop_width = len(header) + 2
+ # If the wrap column falls in the middle of the first word of the
+ # value, start on the next line. This avoids a very long first line.
+ first_break = prop_width + len(txt.split(None, 1)[0])
+ if tag_wrap_column <= first_break:
+ # Start value on the next line.
+ txt = '\n'.join(textwrap.wrap(txt, initial_indent='\n ',
+ width=tag_wrap_column,
+ break_long_words=False,
+ break_on_hyphens=False,
+ subsequent_indent=' '))
+ else:
+ # Account for tag name by temporarily adding a leading space
+ # before calling wrap(), then removing the space later below.
+ txt = '\n'.join(textwrap.wrap(txt, initial_indent=' '*prop_width,
+ width=tag_wrap_column,
+ break_long_words=False,
+ break_on_hyphens=False,
+ subsequent_indent=' '))
+ txt = txt[prop_width:]
+ else:
+ txt = re.sub(r"\n|\r|(\r\n)", "", txt)
+ return txt
+
+
 def make_manifests(data_dir, processes, algorithms=DEFAULT_CHECKSUMS, encoding="utf-8"):
  LOGGER.info(
  _("Using %(process_count)d processes to generate manifests: %(algorithms)s"),
@@ -1499,6 +1534,17 @@ def _make_parser():
  " without performing checksum validation to detect corruption."
  ),
  )
+ parser.add_argument(
+ "--wrap-tags",
+ type=int,
+ dest="tag_wrap_column",
+ default=0,
+ help=_(
+ "Limit line lengths in the tag file (bag-info.txt) by"
+ " wrapping long values and indenting subsequent lines with a"
+ " space character. (Default: don't.)"
+ ),
+ )
 
  checksum_args = parser.add_argument_group(
  _("Checksum Algorithms"),
@@ -1596,6 +1642,7 @@ def main():
  bag_info=args.bag_info,
  processes=args.processes,
  checksums=args.checksums,
+ tag_wrap_column=args.tag_wrap_column,
  )
  except Exception as exc:
  LOGGER.error(

test.py

@@ -115,6 +115,30 @@ def test_validate_fast(self):
  os.remove(j(self.tmpdir, "data", "loc", "2478433644_2839c5e8b8_o_d.jpg"))
  self.assertRaises(bagit.BagValidationError, self.validate, bag, fast=True)
 
+ def test_validate_wrapped_tag_lines(self):
+ info = {"External-Description":
+ ('BagIt is a set of hierarchical file layout conventions'
+ ' designed to support storage and transfer of arbitrary'
+ ' digital content. A bag consists of a directory containing'
+ ' the payload files and other accompanying metadata files'
+ ' known as "tag" files.')}
+ bag = bagit.make_bag(self.tmpdir, bag_info=info, tag_wrap_column=79)
+ self.assertEqual(self.validate(bag, fast=True), True)
+ os.remove(j(self.tmpdir, "data", "loc", "2478433644_2839c5e8b8_o_d.jpg"))
+ self.assertRaises(bagit.BagValidationError, self.validate, bag, fast=True)
+
+ def test_validate_wrapped_tag_lines_short(self):
+ info = {"External-Description":
+ ('BagIt is a set of hierarchical file layout conventions'
+ ' designed to support storage and transfer of arbitrary'
+ ' digital content. A bag consists of a directory containing'
+ ' the payload files and other accompanying metadata files'
+ ' known as "tag" files.')}
+ bag = bagit.make_bag(self.tmpdir, bag_info=info, tag_wrap_column=18)
+ self.assertEqual(self.validate(bag, fast=True), True)
+ os.remove(j(self.tmpdir, "data", "loc", "2478433644_2839c5e8b8_o_d.jpg"))
+ self.assertRaises(bagit.BagValidationError, self.validate, bag, fast=True)
+
  def test_validate_completeness(self):
  bag = bagit.make_bag(self.tmpdir)
  old_path = j(self.tmpdir, "data", "README")