wandb 0.18.4__py3-none-any.whl → 0.18.6__py3-none-any.whl

wandb/sdk/artifacts/artifact_manifest.py

@@ -48,18 +48,19 @@ class ArtifactManifest:
     def digest(self) -> HexMD5:
         raise NotImplementedError
 
-    def add_entry(self, entry: ArtifactManifestEntry) -> None:
-        if (
-            entry.path in self.entries
-            and entry.digest != self.entries[entry.path].digest
-        ):
-            raise ValueError("Cannot add the same path twice: {}".format(entry.path))
-        self.entries[entry.path] = entry
+    def add_entry(self, entry: ArtifactManifestEntry, overwrite: bool = False) -> None:
+        path = entry.path
+        if not overwrite:
+            prev_entry = self.entries.get(path)
+            if prev_entry and (entry.digest != prev_entry.digest):
+                raise ValueError(f"Cannot add the same path twice: {path!r}")
+        self.entries[path] = entry
 
     def remove_entry(self, entry: ArtifactManifestEntry) -> None:
-        if entry.path not in self.entries:
+        try:
+            del self.entries[entry.path]
+        except LookupError:
             raise FileNotFoundError(f"Cannot remove missing entry: '{entry.path}'")
-        del self.entries[entry.path]
 
     def get_entry_by_path(self, path: str) -> ArtifactManifestEntry | None:
         return self.entries.get(path)

wandb/sdk/artifacts/artifact_manifest_entry.py

@@ -133,7 +133,7 @@ class ArtifactManifestEntry:
     ) -> FilePathStr:
         """Download this artifact entry to the specified root path.
 
-        Arguments:
+        Args:
             root: (str, optional) The root path in which to download this
                 artifact entry. Defaults to the artifact's root.
 

wandb/sdk/artifacts/storage_handlers/azure_handler.py

@@ -28,6 +28,9 @@ class AzureHandler(StorageHandler):
             ".blob.core.windows.net"
         )
 
+    def __init__(self, scheme: str | None = None) -> None:
+        self._cache = get_artifact_file_cache()
+
     def load_path(
         self,
         manifest_entry: ArtifactManifestEntry,
@@ -37,7 +40,7 @@ class AzureHandler(StorageHandler):
         if not local:
             return manifest_entry.ref
 
-        path, hit, cache_open = get_artifact_file_cache().check_etag_obj_path(
+        path, hit, cache_open = self._cache.check_etag_obj_path(
             URIStr(manifest_entry.ref),
             ETag(manifest_entry.digest),
             manifest_entry.size or 0,
@@ -136,10 +139,11 @@ class AzureHandler(StorageHandler):
             name_starts_with=f"{blob_name}/"
         ):
             if len(entries) >= max_objects:
-                raise ValueError(
-                    f"Exceeded {max_objects} objects tracked, pass max_objects to "
-                    f"add_reference"
+                wandb.termwarn(
+                    f"Found more than {max_objects} objects under path, limiting upload "
+                    f"to {max_objects} objects. Increase max_objects to upload more"
                 )
+                break
             if not self._is_directory_stub(blob_properties):
                 suffix = PurePosixPath(blob_properties.name).relative_to(blob_name)
                 entries.append(
@@ -151,6 +155,7 @@ class AzureHandler(StorageHandler):
                         ),
                     )
                 )
+
         return entries
 
     def _get_module(self, name: str) -> ModuleType:

wandb/sdk/artifacts/storage_handlers/gcs_handler.py

@@ -24,8 +24,6 @@ if TYPE_CHECKING:
 class _GCSIsADirectoryError(Exception):
     """Raised when we try to download a GCS folder."""
 
-    pass
-
 
 class GCSHandler(StorageHandler):
     _client: gcs_module.client.Client | None
@@ -169,7 +167,7 @@ class GCSHandler(StorageHandler):
     ) -> ArtifactManifestEntry:
         """Create an ArtifactManifestEntry from a GCS object.
 
-        Arguments:
+        Args:
             obj: The GCS object
             path: The GCS-style path (e.g.: "gs://bucket/file.txt")
             name: The user assigned name, or None if not specified

wandb/sdk/artifacts/storage_handlers/s3_handler.py

@@ -243,7 +243,7 @@ class S3Handler(StorageHandler):
     ) -> ArtifactManifestEntry:
         """Create an ArtifactManifestEntry from an S3 object.
 
-        Arguments:
+        Args:
             obj: The S3 object
             path: The S3-style path (e.g.: "s3://bucket/file.txt")
             name: The user assigned name, or None if not specified

wandb/sdk/artifacts/storage_handlers/wb_artifact_handler.py

@@ -49,7 +49,7 @@ class WBArtifactHandler(StorageHandler):
 
         Download the referenced artifact; create and return a new symlink to the caller.
 
-        Arguments:
+        Args:
             manifest_entry (ArtifactManifestEntry): The index entry to load
 
         Returns:
@@ -89,7 +89,7 @@ class WBArtifactHandler(StorageHandler):
 
         Recursively resolves the reference until the result is a concrete asset.
 
-        Arguments:
+        Args:
             artifact: The artifact doing the storing path (str): The path to store name
             (str): If specified, the logical name that should map to `path`
 

wandb/sdk/artifacts/storage_handlers/wb_local_artifact_handler.py

@@ -46,7 +46,7 @@ class WBLocalArtifactHandler(StorageHandler):
     ) -> Sequence[ArtifactManifestEntry]:
         """Store the file or directory at the given path within the specified artifact.
 
-        Arguments:
+        Args:
             artifact: The artifact doing the storing
             path (str): The path to store
             name (str): If specified, the logical name that should map to `path`

wandb/sdk/backend/backend.py

@@ -203,7 +203,6 @@ class Backend:
 
     def server_status(self) -> None:
         """Report server status."""
-        pass
 
     def cleanup(self) -> None:
         # TODO: make _done atomic

wandb/sdk/data_types/audio.py

@@ -13,7 +13,7 @@ from .base_types.media import BatchableMedia
 class Audio(BatchableMedia):
     """Wandb class for audio clips.
 
-    Arguments:
+    Args:
         data_or_path: (string or numpy array) A path to an audio file
             or a numpy array of audio data.
         sample_rate: (int) Sample rate, required when passing in raw

wandb/sdk/data_types/base_types/media.py

@@ -24,9 +24,75 @@ if TYPE_CHECKING:  # pragma: no cover
 SYS_PLATFORM = platform.system()
 
 
+def check_windows_valid_filename(path: Union[int, str]) -> bool:
+    r"""Verify that the given path does not contain any invalid characters for a Windows filename.
+
+    Windows filenames cannot contain the following characters:
+    < > : " \ / | ? *
+
+    For more details, refer to the official documentation:
+    https://learn.microsoft.com/en-us/windows/win32/fileio/naming-a-file#naming-conventions
+
+    Args:
+        path: The file path to check, which can be either an integer or a string.
+
+    Returns:
+        bool: True if the path does not contain any invalid characters, False otherwise.
+    """
+    return not bool(re.search(r'[<>:"\\?*]', path))  # type: ignore
+
+
 def _wb_filename(
     key: Union[str, int], step: Union[str, int], id: Union[str, int], extension: str
 ) -> str:
+    r"""Generates a safe filename/path for storing media files, using the provided key, step, and id.
+
+    The filename is made safe by:
+    1. Removing any leading slashes to prevent writing to absolute paths
+    2. Replacing '.' and '..' with underscores to prevent directory traversal attacks
+
+    If the key contains slashes (e.g. 'images/cats/fluffy.jpg'), subdirectories will be created:
+        media/
+          images/
+            cats/
+              fluffy.jpg_step_id.ext
+
+    Args:
+        key: Name/path for the media file
+        step: Training step number
+        id: Unique identifier
+        extension: File extension (e.g. '.jpg', '.mp3')
+
+    Returns:
+        A sanitized filename string in the format: key_step_id.extension
+
+    Raises:
+        ValueError: If running on Windows and the key contains invalid filename characters
+                   (\\, :, *, ?, ", <, >, |)
+    """
+    if SYS_PLATFORM == "Windows" and not check_windows_valid_filename(key):
+        raise ValueError(
+            f'Media {key} is invalid. Please remove invalid filename characters (\\, :, *, ?, ", <, >, |)'
+        )
+
+    # On Windows, convert forward slashes to backslashes.
+    # This ensures that the key is a valid filename on Windows.
+    if SYS_PLATFORM == "Windows":
+        key = str(key).replace("/", os.sep)
+
+    # Avoid writing to absolute paths by striping any leading slashes.
+    # The key has already been validated for windows operating systems in util.check_windows_valid_filename
+    # This ensures the key does not contain invalid characters for windows, such as '\' or ':'.
+    # So we can check only for '/' in the key.
+    key = str(key).lstrip(os.sep)
+
+    # Avoid directory traversal by replacing dots with underscores.
+    keys = key.split(os.sep)
+    keys = [k.replace(".", "_") if k in (os.curdir, os.pardir) else k for k in keys]
+
+    # Recombine the key into a relative path.
+    key = os.sep.join(keys)
+
     return f"{str(key)}_{str(step)}_{str(id)}{extension}"
 
 
@@ -100,11 +166,6 @@ class Media(WBValue):
         """
         assert self.file_is_set(), "bind_to_run called before _set_file"
 
-        if SYS_PLATFORM == "Windows" and not util.check_windows_valid_filename(key):
-            raise ValueError(
-                f"Media {key} is invalid. Please remove invalid filename characters"
-            )
-
         # The following two assertions are guaranteed to pass
         # by definition file_is_set, but are needed for
         # mypy to understand that these are strings below.

wandb/sdk/data_types/bokeh.py

@@ -13,7 +13,7 @@ from .base_types.media import Media
 class Bokeh(Media):
     """Wandb class for Bokeh plots.
 
-    Arguments:
+    Args:
         val: Bokeh plot
     """
 

wandb/sdk/data_types/helper_types/bounding_boxes_2d.py

@@ -16,7 +16,7 @@ if TYPE_CHECKING:  # pragma: no cover
 class BoundingBoxes2D(JSONMetadata):
     """Format images with 2D bounding box overlays for logging to W&B.
 
-    Arguments:
+    Args:
         val: (dictionary) A dictionary of the following form:
             box_data: (list of dictionaries) One dictionary for each bounding box, containing:
                 position: (dictionary) the position and size of the bounding box, in one of two formats

wandb/sdk/data_types/helper_types/image_mask.py

@@ -18,7 +18,7 @@ if TYPE_CHECKING:  # pragma: no cover
 class ImageMask(Media):
     """Format image masks or overlays for logging to W&B.
 
-    Arguments:
+    Args:
         val: (dictionary)
             One of these two keys to represent the image:
                 mask_data : (2D numpy array) The mask containing an integer class label
@@ -117,7 +117,7 @@ class ImageMask(Media):
     def __init__(self, val: dict, key: str) -> None:
         """Initialize an ImageMask object.
 
-        Arguments:
+        Args:
             val: (dictionary) One of these two keys to represent the image:
                 mask_data : (2D numpy array) The mask containing an integer class label
                     for each pixel in the image

wandb/sdk/data_types/histogram.py

@@ -33,7 +33,7 @@ class Histogram(WBValue):
         wandb.Histogram(np_histogram=hist)
         ```
 
-    Arguments:
+    Args:
         sequence: (array_like) input data for histogram
         np_histogram: (numpy histogram) alternative input of a precomputed histogram
         num_bins: (int) Number of bins for the histogram.  The default number of bins

wandb/sdk/data_types/html.py

@@ -18,7 +18,7 @@ if TYPE_CHECKING:  # pragma: no cover
 class Html(BatchableMedia):
     """Wandb class for arbitrary html.
 
-    Arguments:
+    Args:
         data: (string or io object) HTML to display in wandb
         inject: (boolean) Add a stylesheet to the HTML object.  If set
             to False the HTML will pass through unchanged.

wandb/sdk/data_types/image.py

@@ -65,7 +65,7 @@ def _server_accepts_artifact_path() -> bool:
 class Image(BatchableMedia):
     """Format images for logging to W&B.
 
-    Arguments:
+    Args:
         data_or_path: (numpy array, string, io) Accepts numpy array of
             image data, or a PIL image. The class attempts to infer
             the data format and converts it.

wandb/sdk/data_types/molecule.py

@@ -25,7 +25,7 @@ if TYPE_CHECKING:  # pragma: no cover
 class Molecule(BatchableMedia):
     """Wandb class for 3D Molecular data.
 
-    Arguments:
+    Args:
         data_or_path: (string, io)
             Molecule can be initialized from a file name or an io object.
         caption: (string)
@@ -106,7 +106,7 @@ class Molecule(BatchableMedia):
     ) -> "Molecule":
         """Convert RDKit-supported file/object types to wandb.Molecule.
 
-        Arguments:
+        Args:
             data_or_path: (string, rdkit.Chem.rdchem.Mol)
                 Molecule can be initialized from a file name or an rdkit.Chem.rdchem.Mol object.
             caption: (string)
@@ -173,7 +173,7 @@ class Molecule(BatchableMedia):
     ) -> "Molecule":
         """Convert SMILES string to wandb.Molecule.
 
-        Arguments:
+        Args:
             data: (string)
                 SMILES string.
             caption: (string)

wandb/sdk/data_types/object_3d.py

@@ -190,7 +190,7 @@ def box3d(
 class Object3D(BatchableMedia):
     """Wandb class for 3D point clouds.
 
-    Arguments:
+    Args:
         data_or_path: (numpy array, string, io)
             Object3D can be initialized from a file or a numpy array.
 
@@ -341,7 +341,7 @@ class Object3D(BatchableMedia):
     ) -> "Object3D":
         """Initializes Object3D from a file or stream.
 
-        Arguments:
+        Args:
             data_or_path (Union["TextIO", str]): A path to a file or a `TextIO` stream.
             file_type (str): Specifies the data format passed to `data_or_path`. Required when `data_or_path` is a
                 `TextIO` stream. This parameter is ignored if a file path is provided. The type is taken from the file extension.
@@ -356,7 +356,7 @@ class Object3D(BatchableMedia):
     def from_numpy(cls, data: "np.ndarray") -> "Object3D":
         """Initializes Object3D from a numpy array.
 
-        Arguments:
+        Args:
             data (numpy array): Each entry in the array will
                 represent one point in the point cloud.
 
@@ -394,7 +394,7 @@ class Object3D(BatchableMedia):
     ) -> "Object3D":
         """Initializes Object3D from a python object.
 
-        Arguments:
+        Args:
             points (Sequence["Point"]): The points in the point cloud.
             boxes (Sequence["Box3D"]): 3D bounding boxes for labeling the point cloud. Boxes
             are displayed in point cloud visualizations.

wandb/sdk/data_types/plotly.py

@@ -33,7 +33,7 @@ if TYPE_CHECKING:  # pragma: no cover
 class Plotly(Media):
     """Wandb class for plotly plots.
 
-    Arguments:
+    Args:
         val: matplotlib or plotly figure
     """
 

wandb/sdk/data_types/saved_model.py

@@ -232,7 +232,6 @@ class _SavedModel(WBValue, Generic[SavedModelObjType]):
         except Exception as e:
             if DEBUG_MODE:
                 print(f"{cls}._maybe_init({obj_or_path}) failed: {e}")
-            pass
 
         for child_cls in cls.__subclasses__():
             maybe_instance = child_cls._maybe_init(obj_or_path, **kwargs)

wandb/sdk/data_types/table.py

@@ -191,7 +191,7 @@ class Table(Media):
     This class is the primary class used to generate the Table Visualizer
     in the UI: https://docs.wandb.ai/guides/data-vis/tables.
 
-    Arguments:
+    Args:
         columns: (List[str]) Names of the columns in the table.
             Defaults to ["Input", "Output", "Expected"].
         data: (List[List[any]]) 2D row-oriented array of values.
@@ -318,7 +318,7 @@ class Table(Media):
         This can be one of the normal python classes, an internal W&B type, or an
         example object, like an instance of wandb.Image or wandb.Classes.
 
-        Arguments:
+        Args:
             col_name: (str) - The name of the column to cast.
             dtype: (class, wandb.wandb_sdk.interface._dtypes.Type, any) - The target dtype.
             optional: (bool) - If the column should allow Nones.
@@ -704,7 +704,7 @@ class Table(Media):
         If the state has been updated since the last update, wraps the data
         appropriately in the Key classes.
 
-        Arguments:
+        Args:
             force_last: (bool) Wraps the last column of data even if there
                 are no key updates.
         """
@@ -748,7 +748,7 @@ class Table(Media):
     def _apply_key_updates(self, only_last=False):
         """Appropriately wraps the underlying data in special Key classes.
 
-        Arguments:
+        Args:
             only_last: only apply the updates to the last row (used for performance when
             the caller knows that the only new data is the last row and no updates were
             applied to the column types)
@@ -797,7 +797,7 @@ class Table(Media):
     def add_column(self, name, data, optional=False):
         """Adds a column of data to the table.
 
-        Arguments:
+        Args:
             name: (str) - the unique name of the column
             data: (list | np.array) - a column of homogeneous data
             optional: (bool) - if null-like values are permitted
@@ -838,7 +838,7 @@ class Table(Media):
     def get_column(self, name, convert_to=None):
         """Retrieves a column from the table and optionally converts it to a NumPy object.
 
-        Arguments:
+        Args:
             name: (str) - the name of the column
             convert_to: (str, optional)
                 - "numpy": will convert the underlying data to numpy object
@@ -1015,7 +1015,7 @@ class PartitionedTable(Media):
 class JoinedTable(Media):
     """Join two tables for visualization in the Artifact UI.
 
-    Arguments:
+    Args:
         table1 (str, wandb.Table, ArtifactManifestEntry):
             the path to a wandb.Table in an artifact, the table object, or ArtifactManifestEntry
         table2 (str, wandb.Table):

wandb/sdk/data_types/trace_tree.py

@@ -80,7 +80,7 @@ class Span:
 class WBTraceTree(Media):
     """Media object for trace tree data.
 
-    Arguments:
+    Args:
         root_span (Span): The root span of the trace tree.
         model_dict (dict, optional): A dictionary containing the model dump.
             NOTE: model_dict is a completely-user-defined dict. The UI will render

wandb/sdk/data_types/video.py

@@ -49,7 +49,7 @@ def write_gif_with_image_io(
 class Video(BatchableMedia):
     """Format a video for logging to W&B.
 
-    Arguments:
+    Args:
         data_or_path: (numpy array, string, io)
             Video can be initialized with a path to a file or an io object.
             The format must be "gif", "mp4", "webm" or "ogg".
@@ -103,10 +103,11 @@ class Video(BatchableMedia):
             )
 
         if isinstance(data_or_path, (BytesIO, str)) and fps:
-            wandb.termwarn(
+            msg = (
                 "`fps` argument does not affect the frame rate of the video "
-                + "when providing a file path or raw bytes."
+                "when providing a file path or raw bytes."
             )
+            wandb.termwarn(msg)
 
         if isinstance(data_or_path, BytesIO):
             filename = os.path.join(

wandb/sdk/interface/router.py

@@ -25,8 +25,6 @@ logger = logging.getLogger("wandb")
 class MessageRouterClosedError(Exception):
     """Router has been closed."""
 
-    pass
-
 
 class MessageFutureObject(MessageFuture):
     def __init__(self) -> None:

wandb/sdk/internal/datastore.py

@@ -277,7 +277,7 @@ class DataStore:
     def write(self, obj: "Record") -> Tuple[int, int, int]:
         """Write a protocol buffer.
 
-        Arguments:
+        Args:
             obj: Protocol buffer to write.
 
         Returns:

wandb/sdk/internal/file_pusher.py

@@ -129,7 +129,7 @@ class FilePusher:
     def file_changed(self, save_name: LogicalPath, path: str, copy: bool = True):
         """Tell the file pusher that a file's changed and should be uploaded.
 
-        Arguments:
+        Args:
             save_name: string logical location of the file relative to the run
                 directory.
             path: actual string path of the file to upload on the filesystem.

wandb/sdk/internal/file_stream.py

@@ -575,7 +575,7 @@ class FileStreamApi:
     def push(self, filename: str, data: str) -> None:
         """Push a chunk of a file to the streaming endpoint.
 
-        Arguments:
+        Args:
             filename: Name of file to append to.
             data: Text to append to the file.
         """
@@ -584,7 +584,7 @@ class FileStreamApi:
     def push_success(self, artifact_id: str, save_name: str) -> None:
         """Notification that a file upload has been successfully completed.
 
-        Arguments:
+        Args:
             artifact_id: ID of artifact
             save_name: saved name of the uploaded file
         """
@@ -595,7 +595,7 @@ class FileStreamApi:
 
         Anything pushed after finish will be dropped.
 
-        Arguments:
+        Args:
             exitcode: The exitcode of the watched process.
         """
         logger.info("file stream finish called")
@@ -620,7 +620,7 @@ def request_with_retry(
 ) -> Union["requests.Response", "requests.RequestException"]:
     """Perform a requests http call, retrying with exponential backoff.
 
-    Arguments:
+    Args:
         func:        An http-requesting function to call, like requests.post
         max_retries: Maximum retries before giving up.
                      By default, we retry 30 times in ~2 hours before dropping the chunk

wandb/sdk/internal/handler.py

@@ -21,6 +21,7 @@ from typing import (
     cast,
 )
 
+from wandb.errors.links import url_registry
 from wandb.proto.wandb_internal_pb2 import (
     HistoryRecord,
     InternalMessages,
@@ -37,7 +38,7 @@ from wandb.proto.wandb_internal_pb2 import (
 )
 
 from ..interface.interface_queue import InterfaceQueue
-from ..lib import handler_util, proto_util, tracelog, wburls
+from ..lib import handler_util, proto_util, tracelog
 from . import context, sample, tb_watcher
 from .settings_static import SettingsStatic
 from .system.system_monitor import SystemMonitor
@@ -578,7 +579,7 @@ class HandleManager:
                 if not self._dropped_history:
                     message = (
                         "Step only supports monotonically increasing values, use define_metric to set a custom x "
-                        f"axis. For details see: {wburls.wburls.get('wandb_define_metric')}"
+                        f"axis. For details see: {url_registry.url('define-metric')}"
                     )
                     self._internal_messages.warning.append(message)
                     self._dropped_history = True

wandb/sdk/internal/internal.py

@@ -55,7 +55,7 @@ def wandb_internal(
 
     Read from record queue and dispatch work to various threads.
 
-    Arguments:
+    Args:
         settings: settings object
         record_q: records to be handled
         result_q: for sending results back