Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 13 additions & 1 deletion sdks/python/apache_beam/yaml/yaml_provider.py
Original file line number Diff line number Diff line change
Expand Up @@ -1057,7 +1057,7 @@ def normalize_package(package):
# Ignore the exact path by which this package was referenced,
# but do create a new environment if it changed.
with open(package, 'rb') as fin:
return os.path.basename(package) + '-' + hashlib.file_digest(
return os.path.basename(package) + '-' + _file_digest(
fin, 'sha256').hexdigest()
else:
# Assume urls and pypi identifiers are immutable.
Expand Down Expand Up @@ -1363,3 +1363,15 @@ def standard_providers():
io_providers(),
load_providers(
os.path.join(os.path.dirname(__file__), 'standard_providers.yaml')))


def _file_digest(fileobj, digest):
if hasattr(hashlib, 'file_digest'): # Python 3.11+
return hashlib.file_digest(fileobj, digest)
else:
hasher = hashlib.new(digest)
data = fileobj.read(1 << 20)
while data:
hasher.update(data)
data = fileobj.read(1 << 20)
return hasher
42 changes: 42 additions & 0 deletions sdks/python/apache_beam/yaml/yaml_provider_unit_test.py
Original file line number Diff line number Diff line change
Expand Up @@ -261,6 +261,48 @@ def test_recursive(self):
assert_that(result | beam.Map(lambda x: x.value), equal_to([120]))


class PythonProviderDepsTest(unittest.TestCase):
def test_env_package_sensitive(self):
self.assertNotEqual(
yaml_provider.PypiExpansionService._key('base', ['pkg1']),
yaml_provider.PypiExpansionService._key('base', ['pkg2']))

def test_env_base_sensitive(self):
self.assertNotEqual(
yaml_provider.PypiExpansionService._key('base1', ['pkg']),
yaml_provider.PypiExpansionService._key('base2', ['pkg']))

def test_env_order_invariant(self):
self.assertEqual(
yaml_provider.PypiExpansionService._key('base', ['pkg1', 'pkg2']),
yaml_provider.PypiExpansionService._key('base', ['pkg2', 'pkg1']))

def test_env_path_invariant(self):
with tempfile.TemporaryDirectory() as tmpdir:
os.mkdir(os.path.join(tmpdir, 'a'))
pkgA = os.path.join(tmpdir, 'a', 'pkg.tgz')
os.mkdir(os.path.join(tmpdir, 'b'))
pkgB = os.path.join(tmpdir, 'b', 'pkg.tgz')
with open(pkgA, 'w') as fout:
fout.write('content')
with open(pkgB, 'w') as fout:
fout.write('content')
self.assertEqual(
yaml_provider.PypiExpansionService._key('base', [pkgA]),
yaml_provider.PypiExpansionService._key('base', [pkgB]))

def test_env_content_sensitive(self):
with tempfile.TemporaryDirectory() as tmpdir:
pkg = os.path.join(tmpdir, 'pkg.tgz')
with open(pkg, 'w') as fout:
fout.write('content')
before = yaml_provider.PypiExpansionService._key('base', [pkg])
with open(pkg, 'w') as fout:
fout.write('new content')
after = yaml_provider.PypiExpansionService._key('base', [pkg])
self.assertNotEqual(before, after)


if __name__ == '__main__':
logging.getLogger().setLevel(logging.INFO)
unittest.main()
Loading