support platforms with no os.link (#4903)

support platforms with no os.link, fixes #4901

if we don't have os.link, we just extract another copy instead of making a hardlink.

for that to work, we need to have (and keep) the chunks list in hardlink_masters.

src/borg/archive.py

@@ -48,6 +48,7 @@ from .remote import cache_if_remote
 from .repository import Repository, LIST_SCAN_LIMIT
 
 has_lchmod = hasattr(os, 'lchmod')
+has_link = hasattr(os, 'link')
 
 
 class Statistics:
@@ -635,7 +636,7 @@ Utilization of max. archive size: {csize_max:.0%}
         if 'source' in item:
             source = os.path.join(dest, *item.source.split(os.sep)[stripped_components:])
             chunks, link_target = hardlink_masters.get(item.source, (None, source))
-            if link_target:
+            if link_target and has_link:
                 # Hard link was extracted previously, just link
                 with backup_io('link'):
                     os.link(link_target, path)
@@ -645,8 +646,14 @@ Utilization of max. archive size: {csize_max:.0%}
                 item.chunks = chunks
         yield hardlink_set
         if not hardlink_set and hardlink_masters:
-            # Update master entry with extracted item path, so that following hardlinks don't extract twice.
-            hardlink_masters[item.get('source') or original_path] = (None, path)
+            if has_link:
+                # Update master entry with extracted item path, so that following hardlinks don't extract twice.
+                # We have hardlinking support, so we will hardlink not extract.
+                hardlink_masters[item.get('source') or original_path] = (None, path)
+            else:
+                # Broken platform with no hardlinking support.
+                # In this case, we *want* to extract twice, because there is no other way.
+                pass
 
     def extract_item(self, item, restore_attrs=True, dry_run=False, stdout=False, sparse=False,
                      hardlink_masters=None, stripped_components=0, original_path=None, pi=None):

src/borg/archiver.py

@@ -39,6 +39,7 @@ try:
     from .archive import Archive, ArchiveChecker, ArchiveRecreater, Statistics, is_special
     from .archive import BackupError, BackupOSError, backup_io, OsOpen, stat_update_check
     from .archive import FilesystemObjectProcessors, MetadataCollector, ChunksProcessor
+    from .archive import has_link
     from .cache import Cache, assert_secure, SecurityManager
     from .constants import *  # NOQA
     from .compress import CompressionSpec
@@ -768,11 +769,20 @@ class Archiver:
         strip_components = args.strip_components
         dirs = []
         partial_extract = not matcher.empty() or strip_components
-        hardlink_masters = {} if partial_extract else None
+        hardlink_masters = {} if partial_extract or not has_link else None
 
         def peek_and_store_hardlink_masters(item, matched):
-            if (partial_extract and not matched and hardlinkable(item.mode) and
-                    item.get('hardlink_master', True) and 'source' not in item):
+            # not has_link:
+            # OS does not have hardlink capability thus we need to remember the chunks so that
+            # we can extract all hardlinks as separate normal (not-hardlinked) files instead.
+            #
+            # partial_extract and not matched and hardlinkable:
+            # we do not extract the very first hardlink, so we need to remember the chunks
+            # in hardlinks_master, so we can use them when we extract some 2nd+ hardlink item
+            # that has no chunks list.
+            if ((not has_link or (partial_extract and not matched and hardlinkable(item.mode)))
+                and
+                (item.get('hardlink_master', True) and 'source' not in item)):
                 hardlink_masters[item.get('path')] = (item.get('chunks'), None)
 
         filter = self.build_filter(matcher, peek_and_store_hardlink_masters, strip_components)

src/borg/repository.py

@@ -295,14 +295,18 @@ class Repository:
             secure_erase(old_config_path)
 
         if os.path.isfile(config_path):
+            link_error_msg = ("Failed to securely erase old repository config file (hardlinks not supported>). "
+                              "Old repokey data, if any, might persist on physical storage.")
             try:
                 os.link(config_path, old_config_path)
             except OSError as e:
                 if e.errno in (errno.EMLINK, errno.ENOSYS, errno.EPERM, errno.EACCES, errno.ENOTSUP):
-                    logger.warning("Failed to securely erase old repository config file (hardlinks not supported>). "
-                                   "Old repokey data, if any, might persist on physical storage.")
+                    logger.warning(link_error_msg)
                 else:
                     raise
+            except AttributeError:
+                # some python ports have no os.link, see #4901
+                logger.warning(link_error_msg)
 
         with SaveFile(config_path) as fd:
             config.write(fd)

src/borg/testsuite/__init__.py

@@ -79,6 +79,10 @@ def are_symlinks_supported():
 
 @functools.lru_cache()
 def are_hardlinks_supported():
+    if not hasattr(os, 'link'):
+        # some pythons do not have os.link
+        return False
+
     with unopened_tempfile() as file1path, unopened_tempfile() as file2path:
         open(file1path, 'w').close()
         try:

src/borg/testsuite/archiver.py

@@ -1025,7 +1025,11 @@ class ArchiverTestCase(ArchiverTestCaseBase):
         self.create_regular_file('cache2/%s' % CACHE_TAG_NAME,
                                  contents=b'invalid signature')
         os.mkdir('input/cache3')
-        os.link('input/cache1/%s' % CACHE_TAG_NAME, 'input/cache3/%s' % CACHE_TAG_NAME)
+        if are_hardlinks_supported():
+            os.link('input/cache1/%s' % CACHE_TAG_NAME, 'input/cache3/%s' % CACHE_TAG_NAME)
+        else:
+            self.create_regular_file('cache3/%s' % CACHE_TAG_NAME,
+                                     contents=CACHE_TAG_CONTENTS + b' extra stuff')
 
     def test_create_stdin(self):
         self.cmd('init', '--encryption=repokey', self.repository_location)
@@ -2205,8 +2209,9 @@ class ArchiverTestCase(ArchiverTestCaseBase):
                 assert sti1.st_atime == sto1.st_atime
             assert sti1.st_ctime == sto1.st_ctime
             assert sti1.st_mtime == sto1.st_mtime
-            # note: there is another hardlink to this, see below
-            assert sti1.st_nlink == sto1.st_nlink == 2
+            if are_hardlinks_supported():
+                # note: there is another hardlink to this, see below
+                assert sti1.st_nlink == sto1.st_nlink == 2
             # read
             with open(in_fn, 'rb') as in_f, open(out_fn, 'rb') as out_f:
                 assert in_f.read() == out_f.read()

src/borg/testsuite/upgrader.py

@@ -141,6 +141,7 @@ def test_keys(attic_repo, attic_key_file):
     assert key_valid(keyfile_path)
 
 
+@pytest.mark.skipif(not are_hardlinks_supported(), reason='hardlinks not supported')
 def test_convert_all(attic_repo, attic_key_file, inplace):
     """test all conversion steps
 
@@ -166,7 +167,7 @@ def test_convert_all(attic_repo, attic_key_file, inplace):
     with AtticRepositoryUpgrader(repo_path, create=False) as repository:
         # replicate command dispatch, partly
         os.umask(UMASK_DEFAULT)
-        backup = repository.upgrade(dryrun=False, inplace=inplace)
+        backup = repository.upgrade(dryrun=False, inplace=inplace)  # note: uses hardlinks internally
         if inplace:
             assert backup is None
             assert first_inode(repository.path) == orig_inode