Buildcache creation/extraction use temp tarfile to preserve hardlinks during copy to/from prefix. (#15003)

* Buildcache creation change the way prefix is copied to workdir.
* install_tree copies hardlinked files
* tarfile creates hardlinked files on extraction.
* create a temporary tarfile from prefix and extract it to workdir

* Use temp tarfile to move workdir to prefix to preserve hardlinks instead of copying

2 files changed, 30 insertions, 26 deletions

diff --git a/lib/spack/spack/binary_distribution.py b/lib/spack/spack/binary_distribution.py
index 515a6166d2..49f8416d6f 100644
--- a/lib/spack/spack/binary_distribution.py
+++ b/lib/spack/spack/binary_distribution.py
@@ -18,7 +18,7 @@ import json
 from six.moves.urllib.error import URLError
 
 import llnl.util.tty as tty
-from llnl.util.filesystem import mkdirp, install_tree
+from llnl.util.filesystem import mkdirp
 
 import spack.cmd
 import spack.config as config
@@ -308,7 +308,7 @@ def build_tarball(spec, outdir, force=False, rel=False, unsigned=False,
     tmpdir = tempfile.mkdtemp()
     cache_prefix = build_cache_prefix(tmpdir)
 
-    tarfile_name = tarball_name(spec, '.tar.gz')
+    tarfile_name = tarball_name(spec, '.tar.bz2')
     tarfile_dir = os.path.join(cache_prefix, tarball_directory_name(spec))
     tarfile_path = os.path.join(tarfile_dir, tarfile_name)
     spackfile_path = os.path.join(
@@ -342,8 +342,18 @@ def build_tarball(spec, outdir, force=False, rel=False, unsigned=False,
             raise NoOverwriteException(url_util.format(remote_specfile_path))
 
     # make a copy of the install directory to work with
-    workdir = os.path.join(tempfile.mkdtemp(), os.path.basename(spec.prefix))
-    install_tree(spec.prefix, workdir, symlinks=True)
+    workdir = os.path.join(tmpdir, os.path.basename(spec.prefix))
+    # install_tree copies hardlinks
+    # create a temporary tarfile from prefix and exract it to workdir
+    # tarfile preserves hardlinks
+    temp_tarfile_name = tarball_name(spec, '.tar')
+    temp_tarfile_path = os.path.join(tarfile_dir, temp_tarfile_name)
+    with closing(tarfile.open(temp_tarfile_path, 'w')) as tar:
+        tar.add(name='%s' % spec.prefix,
+                arcname='.')
+    with closing(tarfile.open(temp_tarfile_path, 'r')) as tar:
+        tar.extractall(workdir)
+    os.remove(temp_tarfile_path)
 
     # create info for later relocation and create tar
     write_buildinfo_file(spec.prefix, workdir, rel=rel)
@@ -368,7 +378,7 @@ def build_tarball(spec, outdir, force=False, rel=False, unsigned=False,
             tty.die(e)
 
     # create compressed tarball of the install prefix
-    with closing(tarfile.open(tarfile_path, 'w:gz')) as tar:
+    with closing(tarfile.open(tarfile_path, 'w:bz2')) as tar:
         tar.add(name='%s' % workdir,
                 arcname='%s' % os.path.basename(spec.prefix))
     # remove copy of install directory
@@ -407,8 +417,8 @@ def build_tarball(spec, outdir, force=False, rel=False, unsigned=False,
         sign_tarball(key, force, specfile_path)
     # put tarball, spec and signature files in .spack archive
     with closing(tarfile.open(spackfile_path, 'w')) as tar:
-        tar.add(name='%s' % tarfile_path, arcname='%s' % tarfile_name)
-        tar.add(name='%s' % specfile_path, arcname='%s' % specfile_name)
+        tar.add(name=tarfile_path, arcname='%s' % tarfile_name)
+        tar.add(name=specfile_path, arcname='%s' % specfile_name)
         if not unsigned:
             tar.add(name='%s.asc' % specfile_path,
                     arcname='%s.asc' % specfile_name)
@@ -579,7 +589,7 @@ def extract_tarball(spec, filename, allow_root=False, unsigned=False,
     stagepath = os.path.dirname(filename)
     spackfile_name = tarball_name(spec, '.spack')
     spackfile_path = os.path.join(stagepath, spackfile_name)
-    tarfile_name = tarball_name(spec, '.tar.gz')
+    tarfile_name = tarball_name(spec, '.tar.bz2')
     tarfile_path = os.path.join(tmpdir, tarfile_name)
     specfile_name = tarball_name(spec, '.spec.yaml')
     specfile_path = os.path.join(tmpdir, specfile_name)
@@ -638,7 +648,17 @@ def extract_tarball(spec, filename, allow_root=False, unsigned=False,
     # so the pathname should be the same now that the directory layout
     # is confirmed
     workdir = os.path.join(tmpdir, os.path.basename(spec.prefix))
-    install_tree(workdir, spec.prefix, symlinks=True)
+    # install_tree copies hardlinks
+    # create a temporary tarfile from prefix and exract it to workdir
+    # tarfile preserves hardlinks
+    temp_tarfile_name = tarball_name(spec, '.tar')
+    temp_tarfile_path = os.path.join(tmpdir, temp_tarfile_name)
+    with closing(tarfile.open(temp_tarfile_path, 'w')) as tar:
+        tar.add(name='%s' % workdir,
+                arcname='.')
+    with closing(tarfile.open(temp_tarfile_path, 'r')) as tar:
+        tar.extractall(spec.prefix)
+    os.remove(temp_tarfile_path)
 
     # cleanup
     os.remove(tarfile_path)
diff --git a/lib/spack/spack/test/packaging.py b/lib/spack/spack/test/packaging.py
index edad8e29fa..8f11b8ec99 100644
--- a/lib/spack/spack/test/packaging.py
+++ b/lib/spack/spack/test/packaging.py
@@ -107,11 +107,6 @@ echo $PATH"""
         buildcache.buildcache(parser, args)
 
         files = os.listdir(spec.prefix)
-        assert 'link_to_dummy.txt' in files
-        assert 'dummy.txt' in files
-        assert os.path.realpath(
-            os.path.join(spec.prefix, 'link_to_dummy.txt')
-        ) == os.path.realpath(os.path.join(spec.prefix, 'dummy.txt'))
 
         # create build cache with relative path and signing
         args = parser.parse_args(
@@ -129,13 +124,6 @@ echo $PATH"""
         args = parser.parse_args(['install', '-f', str(pkghash)])
         buildcache.buildcache(parser, args)
 
-        files = os.listdir(spec.prefix)
-        assert 'link_to_dummy.txt' in files
-        assert 'dummy.txt' in files
-        assert os.path.realpath(
-            os.path.join(spec.prefix, 'link_to_dummy.txt')
-        ) == os.path.realpath(os.path.join(spec.prefix, 'dummy.txt'))
-
     else:
         # create build cache without signing
         args = parser.parse_args(
@@ -152,10 +140,6 @@ echo $PATH"""
         files = os.listdir(spec.prefix)
         assert 'link_to_dummy.txt' in files
         assert 'dummy.txt' in files
-        assert os.path.realpath(
-            os.path.join(spec.prefix, 'link_to_dummy.txt')
-        ) == os.path.realpath(os.path.join(spec.prefix, 'dummy.txt'))
-
         # test overwrite install without verification
         args = parser.parse_args(['install', '-f', '-u', str(pkghash)])
         buildcache.buildcache(parser, args)
@@ -242,7 +226,7 @@ def test_relocate_links(tmpdir):
         old_src = os.path.join(old_dir, filename)
         os.symlink(old_src, filename)
         filenames = [filename]
-        new_dir = '/opt/rh/devtoolset/'
+        new_dir = '/opt/rh/devtoolset'
         relocate_links(filenames, old_dir, new_dir)
         assert os.path.realpath(filename) == os.path.join(new_dir, filename)