cloud-files 5.8.2__tar.gz → 6.0.0__tar.gz

{cloud_files-5.8.2 → cloud_files-6.0.0}/.github/workflows/test-suite.yml

@@ -15,7 +15,7 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        python-version: ["3.9", "3.10", "3.11", "3.12", "3.13"]
+        python-version: ["3.10", "3.11", "3.12", "3.13", "3.14"]
 
     steps:
     - uses: actions/checkout@v2
@@ -25,7 +25,7 @@ jobs:
         python-version: ${{ matrix.python-version }}
     - name: Install dependencies
       run: |
-        python -m pip install --upgrade pip
+        python -m pip install --upgrade pip setuptools wheel
         if [ -f requirements.txt ]; then pip install -e ".[test,monitoring]"; fi
     - name: Test with pytest
       run: |

{cloud_files-5.8.2 → cloud_files-6.0.0}/ChangeLog

@@ -1,6 +1,26 @@
 CHANGES
 =======
 
+6.0.0
+-----
+
+* feat: add file counts to du as -N flag
+
+5.9.0
+-----
+
+* perf: reduce data loading for list\_blogs for GCS
+* perf: memory efficient listing on GCS
+* fix: errant print statement
+* feat: add CloudFile(s).sep
+* fix(https): allow "raw=True" to work
+* fix: don't retry on 403,404 for http head
+* ci: drop py3.9 add py3.14
+* fixtest: try upgrading setuptools
+* feat: add constructor args to CloudFile that are present in CloudFiles
+* perf: faster du using listing (#120)
+* test: change target for nokura
+
 5.8.2
 -----
 

{cloud_files-5.8.2 → cloud_files-6.0.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: cloud-files
-Version: 5.8.2
+Version: 6.0.0
 Summary: Fast access to cloud storage and local FS.
 Home-page: https://github.com/seung-lab/cloud-files/
 Author: William Silversmith

{cloud_files-5.8.2 → cloud_files-6.0.0}/automated_test.py

@@ -757,7 +757,7 @@ def test_to_https_protocol():
   assert pth == "https://s3-hpcrc.rc.princeton.edu/my_bucket/to/heaven"
 
   pth = to_https_protocol("nokura://my_bucket/to/heaven")
-  assert pth == "https://nokura.pni.princeton.edu/my_bucket/to/heaven"
+  assert pth == "https://c10s.pni.princeton.edu/my_bucket/to/heaven"
 
   pth = to_https_protocol("tigerdata://my_bucket/to/heaven")
   assert pth == "https://td.princeton.edu/my_bucket/to/heaven"

{cloud_files-5.8.2 → cloud_files-6.0.0}/cloud_files.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: cloud-files
-Version: 5.8.2
+Version: 6.0.0
 Summary: Fast access to cloud storage and local FS.
 Home-page: https://github.com/seung-lab/cloud-files/
 Author: William Silversmith

cloud_files-6.0.0/cloud_files.egg-info/pbr.json

@@ -0,0 +1 @@
+{"git_version": "38a2b59", "is_release": true}

{cloud_files-5.8.2 → cloud_files-6.0.0}/cloudfiles/cloudfiles.py

@@ -18,6 +18,7 @@ import platform
 import posixpath
 import re
 import shutil
+import threading
 import types
 import time
 
@@ -1007,6 +1008,40 @@ class CloudFiles:
       return results
     return first(results.values())
 
+  def subtree_size(self, prefix:GetPathType = "") -> dict[str,int]:
+    """High performance size calculation for directory trees."""
+    prefix, return_multiple = toiter(prefix, is_iter=True)
+    total_bytes = 0
+    total_files = 0
+
+    total = totalfn(prefix, None)
+
+    lock = threading.Lock()
+
+    def size_thunk(prefix):
+      nonlocal total_bytes
+      nonlocal total_files
+      nonlocal lock
+
+      with self._get_connection() as conn:
+        subtree_files, subtree_bytes = conn.subtree_size(prefix)
+        with lock:
+          total_files += subtree_files
+          total_bytes += subtree_bytes
+    
+    schedule_jobs(
+      fns=( partial(size_thunk, path) for path in prefix ),
+      concurrency=self.num_threads,
+      progress=self.progress,
+      green=self.green,
+      total=total,
+    )
+
+    return {
+      "N": total_files,
+      "num_bytes": total_bytes,
+    }
+
   @parallelize(desc="Delete")
   def delete(
     self, paths:GetPathType, total:Optional[int] = None, 
@@ -1666,6 +1701,12 @@ class CloudFiles:
       return os.path.join(*paths)
     return posixpath.join(*paths)
 
+  @property
+  def sep(self) -> str:
+    if self._path.protocol == "file":
+      return os.sep
+    return posixpath.sep
+
   def dirname(self, path:str) -> str:
     if self._path.protocol == "file":
       return os.path.dirname(path)
@@ -1706,11 +1747,17 @@ class CloudFiles:
 
 class CloudFile:
   def __init__(
-    self, path:str, cache_meta:bool = False, 
+    self,
+    path:str,
+    cache_meta:bool = False, 
     secrets:SecretsType = None,
     composite_upload_threshold:int = int(1e8),
     locking:bool = True,
     lock_dir:Optional[str] = None,
+    endpoint:Optional[str] = None, 
+    no_sign_request:bool = False,
+    request_payer:Optional[str] = None,
+    use_https:bool = False,
   ):
     path = paths.normalize(path)
     self.cf = CloudFiles(
@@ -1719,6 +1766,10 @@ class CloudFile:
       composite_upload_threshold=composite_upload_threshold,
       locking=locking,
       lock_dir=lock_dir,
+      use_https=use_https,
+      endpoint=endpoint,
+      request_payer=request_payer,
+      no_sign_request=no_sign_request,
     )
     self.filename = paths.basename(path)
     
@@ -1726,6 +1777,10 @@ class CloudFile:
     self._size:Optional[int] = None
     self._head = None
 
+  @property
+  def sep(self) -> str:
+    return self.cf.sep
+
   @property
   def protocol(self):
     return self.cf.protocol

{cloud_files-5.8.2 → cloud_files-6.0.0}/cloudfiles/interfaces.py

@@ -48,6 +48,7 @@ MEM_POOL = None
 
 S3_ACLS = {
   "tigerdata": "private",
+  "nokura": "public-read",
 }
 
 S3ConnectionPoolParams = namedtuple('S3ConnectionPoolParams', 'service bucket_name request_payer')
@@ -303,6 +304,22 @@ class FileInterface(StorageInterface):
 
     return self.io_with_lock(do_size, path, exclusive=False)
 
+  def subtree_size(self, prefix:str = "") -> tuple[int,int]:
+    total_bytes = 0
+    total_files = 0
+
+    subdir = self.get_path_to_file("")
+    if prefix:
+      subdir = os.path.join(subdir, os.path.dirname(prefix))
+
+    for root, dirs, files in os.walk(subdir):
+      for f in files:
+          path = os.path.join(root, f)
+          total_files += 1
+          total_bytes += os.path.getsize(path)
+
+    return (total_files, total_bytes)
+
   def exists(self, file_path):
     path = self.get_path_to_file(file_path)
     def do_exists():
@@ -580,8 +597,7 @@ class MemoryInterface(StorageInterface):
 
     Returns: iterator
     """
-    layer_path = self.get_path_to_file("")        
-    path = os.path.join(layer_path, prefix) + '*'
+    layer_path = self.get_path_to_file("")
 
     remove = layer_path
     if len(remove) and remove[-1] != '/':
@@ -615,6 +631,23 @@ class MemoryInterface(StorageInterface):
     filenames.sort()
     return iter(filenames)
 
+  def subtree_size(self, prefix:str = "") -> tuple[int,int]:
+    layer_path = self.get_path_to_file("")        
+
+    remove = layer_path
+    if len(remove) and remove[-1] != '/':
+      remove += '/'
+
+    total_bytes = 0
+    total_files = 0
+    for filename, binary in self._data.items():
+      f_prefix = f.removeprefix(remove)[:len(prefix)]
+      if f_prefix == prefix:
+        total_bytes += len(binary)
+        total_files += 1
+
+    return (total_files, total_bytes)
+
 class GoogleCloudStorageInterface(StorageInterface):
   exists_batch_size = Batch._MAX_BATCH_SIZE
   delete_batch_size = Batch._MAX_BATCH_SIZE
@@ -816,6 +849,8 @@ class GoogleCloudStorageInterface(StorageInterface):
     blobs = self._bucket.list_blobs(
       prefix=path, 
       delimiter=delimiter,
+      page_size=2500,
+      fields="items(name),nextPageToken",
     )
 
     for page in blobs.pages:
@@ -835,6 +870,26 @@ class GoogleCloudStorageInterface(StorageInterface):
           yield filename
 
 
+  @retry
+  def subtree_size(self, prefix:str = "") -> tuple[int,int]:
+    layer_path = self.get_path_to_file("")        
+    path = posixpath.join(layer_path, prefix)
+
+    blobs = self._bucket.list_blobs(
+      prefix=path,
+      page_size=5000,
+      fields="items(name,size),nextPageToken",
+    )
+
+    total_bytes = 0
+    total_files = 0
+    for page in blobs.pages:
+      for blob in page:
+        total_bytes += blob.size
+        total_files += 1
+
+    return (total_files, total_bytes)
+
   def release_connection(self):
     global GC_POOL
     with GCS_BUCKET_POOL_LOCK:
@@ -882,6 +937,8 @@ class HttpInterface(StorageInterface):
     key = self.get_path_to_file(file_path)
     headers = self.default_headers()
     with self.session.head(key, headers=headers) as resp:
+      if resp.status_code in (404, 403):
+        return None
       resp.raise_for_status()
       return resp.headers
 
@@ -889,6 +946,9 @@ class HttpInterface(StorageInterface):
     headers = self.head(file_path)
     return int(headers["Content-Length"])
 
+  def subtree_size(self, prefix:str = "") -> tuple[int,int]:
+    raise NotImplementedError()
+
   @retry
   def get_file(self, file_path, start=None, end=None, part_size=None):
     key = self.get_path_to_file(file_path)
@@ -899,24 +959,20 @@ class HttpInterface(StorageInterface):
       end = int(end - 1) if end is not None else ''
       headers["Range"] = f"bytes={start}-{end}"
     
-    resp = self.session.get(key, headers=headers)
-    
-    if resp.status_code in (404, 403):
-      return (None, None, None, None)
-    resp.close()
-    resp.raise_for_status()
+    with self.session.get(key, headers=headers, stream=True) as resp:    
+      if resp.status_code in (404, 403):
+        return (None, None, None, None)
+      resp.raise_for_status()
+      resp.raw.decode_content = False
+      content = resp.raw.read()
+      content_encoding = resp.headers.get('Content-Encoding', None)  
 
     # Don't check MD5 for http because the etag can come in many
     # forms from either GCS, S3 or another service entirely. We
     # probably won't figure out how to decode it right.
     # etag = resp.headers.get('etag', None)
-    content_encoding = resp.headers.get('Content-Encoding', None)
-
-    # requests automatically decodes these
-    if content_encoding in (None, '', 'gzip', 'deflate', 'br'):
-      content_encoding = None
     
-    return (resp.content, content_encoding, None, None)
+    return (content, content_encoding, None, None)
 
   @retry
   def save_file(self, src, dest, resumable) -> tuple[bool, int]:
@@ -1017,7 +1073,6 @@ class HttpInterface(StorageInterface):
         )
 
       for res in results.get("items", []):
-        print(res["name"])
         yield res["name"].removeprefix(strip)
       
       token = results.get("nextPageToken", None)
@@ -1490,6 +1545,47 @@ class S3Interface(StorageInterface):
       for filename in iterate(resp):
         yield filename
 
+  def subtree_size(self, prefix:str = "") -> tuple[int,int]:
+    layer_path = self.get_path_to_file("")        
+    path = posixpath.join(layer_path, prefix)
+
+    @retry
+    def s3lst(path, continuation_token=None):
+      kwargs = {
+        'Bucket': self._path.bucket,
+        'Prefix': path,
+        **self._additional_attrs
+      }
+
+      if continuation_token:
+        kwargs['ContinuationToken'] = continuation_token
+
+      return self._conn.list_objects_v2(**kwargs)
+
+    resp = s3lst(path)
+    
+    def iterate(resp):
+      if 'Contents' not in resp.keys():
+        resp['Contents'] = []
+
+      for item in resp['Contents']:
+        yield item.get('Size', 0)
+
+    total_bytes = 0
+    total_files = 0
+    for num_bytes in iterate(resp):
+      total_files += 1
+      total_bytes += num_bytes
+
+    while resp['IsTruncated'] and resp['NextContinuationToken']:
+      resp = s3lst(path, resp['NextContinuationToken'])
+
+      for num_bytes in iterate(resp):
+        total_files += 1
+        total_bytes += num_bytes
+
+    return (total_files, total_bytes)
+
   def release_connection(self):
     global S3_POOL
     service = self._path.alias or 's3'

{cloud_files-5.8.2 → cloud_files-6.0.0}/cloudfiles_cli/cloudfiles_cli.py

@@ -802,14 +802,22 @@ def __rm(cloudpath, progress, paths):
 @click.option('-c', '--grand-total', is_flag=True, default=False, help="Sum a grand total of all inputs.")
 @click.option('-s', '--summarize', is_flag=True, default=False, help="Sum a total for each input argument.")
 @click.option('-h', '--human-readable', is_flag=True, default=False, help='"Human-readable" output. Use unit suffixes: Bytes, KiB, MiB, GiB, TiB, PiB, and EiB.')
-def du(paths, grand_total, summarize, human_readable):
+@click.option('-N', '--count-files', is_flag=True, default=False, help='Also report the number of files.')
+def du(paths, grand_total, summarize, human_readable, count_files):
   """Display disk usage statistics."""
   results = []
+
+  list_data = False
+
   for path in paths:
     npath = normalize_path(path)
     if ispathdir(path):
       cf = CloudFiles(npath)
-      results.append(cf.size(cf.list()))
+      if summarize:
+        results.append(cf.subtree_size())
+      else:
+        list_data = True
+        results.append(cf.size(cf.list()))
     else:
       cf = CloudFiles(os.path.dirname(npath))
       sz = cf.size(os.path.basename(npath))
@@ -838,8 +846,15 @@ def du(paths, grand_total, summarize, human_readable):
       return f"{(val / 2**60):.2f} EiB"
 
   summary = {}
+  num_files = 0
   for path, res in zip(paths, results):
-    summary[path] = sum(res.values())
+    if list_data:
+      summary[path] = sum(res.values())
+      num_files += len(res)
+    else:
+      summary[path] = res["num_bytes"]
+      num_files += res["N"]
+
     if summarize:
       print(f"{SI(summary[path])}\t{path}")
 
@@ -849,7 +864,10 @@ def du(paths, grand_total, summarize, human_readable):
         print(f"{SI(size)}\t{pth}")
 
   if grand_total:
-    print(f"{SI(sum(summary.values()))}\ttotal") 
+    print(f"{SI(sum(summary.values()))}\tbytes total")
+
+  if count_files:
+    print(f"{num_files}\tfiles total") 
 
 @main.command()
 @click.argument('paths', nargs=-1)

cloud_files-5.8.2/cloud_files.egg-info/pbr.json

@@ -1 +0,0 @@
-{"git_version": "99528f8", "is_release": true}