def _SyncFilesToCloud(input_api, output_api):
"""Searches for .sha1 files and uploads them to Cloud Storage.
It validates all the hashes and skips upload if not necessary.
"""
cloud_storage = LoadSupport(input_api)
# Look in both buckets, in case the user uploaded the file manually. But this
# script focuses on WPR archives, so it only uploads to the internal bucket.
hashes_in_cloud_storage = cloud_storage.List(cloud_storage.INTERNAL_BUCKET)
hashes_in_cloud_storage += cloud_storage.List(cloud_storage.PUBLIC_BUCKET)
results = []
for affected_file in input_api.AffectedFiles(include_deletes=False):
hash_path = affected_file.AbsoluteLocalPath()
file_path, extension = os.path.splitext(hash_path)
if extension != '.sha1':
continue
with open(hash_path, 'rb') as f:
file_hash = f.read(1024).rstrip()
if file_hash in hashes_in_cloud_storage:
results.append(
output_api.PresubmitNotifyResult(
'File already in Cloud Storage, skipping upload: %s' %
hash_path))
continue
if not re.match('^([A-Za-z0-9]{40})$', file_hash):
results.append(
output_api.PresubmitError(
'Hash file does not contain a valid SHA-1 hash: %s' %
hash_path))
continue
if not os.path.exists(file_path):
results.append(
output_api.PresubmitError(
'Hash file exists, but file not found: %s' % hash_path))
continue
if cloud_storage.GetHash(file_path) != file_hash:
results.append(
output_api.PresubmitError(
'Hash file does not match file\'s actual hash: %s' %
hash_path))
continue
try:
cloud_storage.Insert(cloud_storage.INTERNAL_BUCKET, file_hash,
file_path)
results.append(
output_api.PresubmitNotifyResult(
'Uploaded file to Cloud Storage: %s' % hash_path))
except cloud_storage.CloudStorageError, e:
results.append(
output_api.PresubmitError(
'Unable to upload to Cloud Storage: %s\n\n%s' %
(hash_path, e)))
def _SyncFilesToCloud(input_api, output_api):
"""Searches for .sha1 files and uploads them to Cloud Storage.
It validates all the hashes and skips upload if not necessary.
"""
# Because this script will be called from a magic PRESUBMIT demon,
# avoid angering it; don't pollute its sys.path.
old_sys_path = sys.path
try:
sys.path = [os.path.join(os.pardir, os.pardir, 'telemetry')] + sys.path
from telemetry.page import cloud_storage
finally:
sys.path = old_sys_path
hashes_in_cloud_storage = cloud_storage.List(cloud_storage.DEFAULT_BUCKET)
results = []
for hash_path in input_api.AbsoluteLocalPaths():
file_path, extension = os.path.splitext(hash_path)
if extension != '.sha1':
continue
with open(hash_path, 'rb') as f:
file_hash = f.read(1024).rstrip()
if file_hash in hashes_in_cloud_storage:
results.append(
output_api.PresubmitNotifyResult(
'File already in Cloud Storage, skipping upload: %s' %
hash_path))
continue
if not re.match('^([A-Za-z0-9]{40})$', file_hash):
results.append(
output_api.PresubmitError(
'Hash file does not contain a valid SHA-1 hash: %s' %
hash_path))
continue
if not os.path.exists(file_path):
results.append(
output_api.PresubmitError(
'Hash file exists, but file not found: %s' % hash_path))
continue
if cloud_storage.GetHash(file_path) != file_hash:
results.append(
output_api.PresubmitError(
'Hash file does not match file\'s actual hash: %s' %
hash_path))
continue
try:
cloud_storage.Insert(cloud_storage.DEFAULT_BUCKET, file_hash,
file_path)
except cloud_storage.CloudStorageError:
results.append(
output_api.PresubmitError(
'Unable to upload to Cloud Storage: %s' % hash_path))
return results
def AddRecordedPages(self, urls):
(target_wpr_file, target_wpr_file_path) = self._NextWprFileName()
for url in urls:
self._SetWprFileForPage(url, target_wpr_file)
shutil.move(self.temp_target_wpr_file_path, target_wpr_file_path)
# Update the hash file.
with open(target_wpr_file_path + '.sha1', 'wb') as f:
f.write(cloud_storage.GetHash(target_wpr_file_path))
f.flush()
self._WriteToFile()
self._DeleteAbandonedWprFiles()
def _SyncFilesToCloud(input_api, output_api):
"""Searches for .sha1 files and uploads them to Cloud Storage.
It validates all the hashes and skips upload if not necessary.
"""
cloud_storage = LoadSupport(input_api)
results = []
for hash_path, file_hash in _GetFilesNotInCloud(input_api):
file_path, _ = os.path.splitext(hash_path)
if not re.match('^([A-Za-z0-9]{40})$', file_hash):
results.append(output_api.PresubmitError(
'Hash file does not contain a valid SHA-1 hash: %s' % hash_path))
continue
if not os.path.exists(file_path):
results.append(output_api.PresubmitError(
'Hash file exists, but file not found: %s' % hash_path))
continue
if cloud_storage.GetHash(file_path) != file_hash:
results.append(output_api.PresubmitError(
'Hash file does not match file\'s actual hash: %s' % hash_path))
continue
try:
bucket_input = raw_input('Uploading to Cloud Storage: %s\n'
'Is this file [P]ublic or Google-[i]nternal?'
% file_path).lower()
if 'public'.startswith(bucket_input):
bucket = cloud_storage.PUBLIC_BUCKET
elif ('internal'.startswith(bucket_input) or
'google-internal'.startswith(bucket_input)):
bucket = cloud_storage.INTERNAL_BUCKET
else:
results.append(output_api.PresubmitError(
'Response was neither "public" nor "internal": %s' % bucket_input))
return results
cloud_storage.Insert(bucket, file_hash, file_path)
results.append(output_api.PresubmitNotifyResult(
'Uploaded file to Cloud Storage: %s' % file_path))
except cloud_storage.CloudStorageError, e:
results.append(output_api.PresubmitError(
'Unable to upload to Cloud Storage: %s\n\n%s' % (file_path, e)))
def assertCorrectHashFile(self, file_path):
self.assertTrue(os.path.exists(file_path + '.sha1'))
with open(file_path + '.sha1', 'rb') as f:
self.assertEquals(cloud_storage.GetHash(file_path), f.read())
def _UpdateHashFile(file_path):
with open(file_path + '.sha1', 'wb') as f:
f.write(cloud_storage.GetHash(file_path))
f.flush()