`dandi.dandiapi`#

This module provides functionality for interacting with a Dandi Archive server via the REST API. Interaction begins with the creation of a DandiAPIClient instance, which can be used to retrieve RemoteDandiset objects (representing Dandisets on the server) and BaseRemoteAsset objects (representing assets without any data associating them with their Dandisets). RemoteDandiset objects can, in turn, be used to retrieve RemoteAsset objects (representing assets associated with Dandisets). Aside from DandiAPIClient, none of these classes should be instantiated directly by the user.

All operations that merely fetch data from the server can be done without authenticating, but any operation that writes, uploads, modifies, or deletes data requires the user to authenticate the DandiAPIClient instance by supplying an API key either when creating the instance or by calling the authenticate() or dandi_authenticate() method.

Example code for printing the metadata of all assets with “two-photon” in their metadata.measurementTechnique[].name for the latest published version of every Dandiset:

import json

from dandi.dandiapi import DandiAPIClient

with DandiAPIClient.for_dandi_instance("dandi") as client:
    for dandiset in client.get_dandisets():
        if dandiset.most_recent_published_version is None:
            continue
        latest_dandiset = dandiset.for_version(dandiset.most_recent_published_version)
        for asset in latest_dandiset.get_assets():
            metadata = asset.get_metadata()
            if any(
                mtt is not None and "two-photon" in mtt.name
                for mtt in (metadata.measurementTechnique or [])
            ):
                print(json.dumps(metadata.json_dict(), indent=4))
                # Can be used to also download the asset:
                # asset.download(pathlib.Path(dandiset.identifier, asset.path))

Example code for accessing asset files as regular Python file objects without downloading their entire content locally. Such file objects could then be passed to functions of pynwb etc.

from dandi.dandiapi import DandiAPIClient

dandiset_id = "000006"  # ephys dataset from the Svoboda Lab
filepath = "sub-anm372795/sub-anm372795_ses-20170718.nwb"  # 450 kB file

with DandiAPIClient() as client:
    asset = client.get_dandiset(dandiset_id, "draft").get_asset_by_path(filepath)
    # https://dandi.readthedocs.io/en/latest/modref/dandiapi.html#dandi.dandiapi.BaseRemoteBlobAsset.as_readable
    # provides file-like object which uses fsspec to provide sparse access to content
    # of the file on S3:
    with asset.as_readable().open() as f:
        print(f.read(4))
        f.seek(100)
        print(f.read(4))

You can see more usages of DANDI API to assist with data streaming at PyNWB: Streaming NWB files.

Client#

class dandi.dandiapi.RESTFullAPIClient(api_url: str, session: requests.Session | None = None, headers: dict | None = None)[source]#

Base class for a JSON-based HTTP(S) client for interacting with a given base API URL.

All request methods can take either an absolute URL or a slash-separated path; in the latter case, the path is appended to the base API URL (separated by a slash) in order to determine the actual URL to make the request of.

RESTFullAPIClient instances are usable as context managers, in which case they will close their associated session on exit.

delete(path: str, **kwargs: Any) → Any[source]#: Convenience method to call request() with the ‘DELETE’ HTTP method.

get(path: str, **kwargs: Any) → Any[source]#: Convenience method to call request() with the ‘GET’ HTTP method.

get_url(path: str) → str[source]#

Append a slash-separated path to the instance’s base URL. The two components are separated by a single slash, removing any excess slashes that would be present after naïve concatenation.

If path is already an absolute URL, it is returned unchanged.

page_size: int | None#: Default number of items to request per page when paginating (None means to use the server’s default)

page_workers: int#: How many pages to fetch at once when parallelizing pagination

paginate(path: str, page_size: int | None = None, params: dict | None = None) → Iterator[source]#

Paginate through the resources at the given path: GET the path, yield the values in the "results" key, and repeat with the URL in the "next" key until it is null.

If the first "next" key is the same as the initially-requested URL but with the page query parameter set to 2, then the remaining pages are fetched concurrently in separate threads, page_workers (default 5) at a time. This behavior requires the initial response to contain a "count" key giving the number of items across all pages.

Parameters:: page_size – If non-None, overrides the client’s page_size attribute for this sequence of pages

patch(path: str, **kwargs: Any) → Any[source]#: Convenience method to call request() with the ‘PATCH’ HTTP method.

post(path: str, **kwargs: Any) → Any[source]#: Convenience method to call request() with the ‘POST’ HTTP method.

put(path: str, **kwargs: Any) → Any[source]#: Convenience method to call request() with the ‘PUT’ HTTP method.

request(method: str, path: str, params: dict | None = None, data: Any = None, files: dict | None = None, json: Any = None, headers: dict | None = None, json_resp: bool = True, retry_statuses: Sequence[int] = (), retry_if: Callable[[requests.Response], Any] | None = None, **kwargs: Any) → Any[source]#

This method looks up the appropriate method, constructs a request URL from the base URL, path, and parameters, and then sends the request. If the method is unknown or if the path is not found, an exception is raised; otherwise, a JSON object is returned with the response.

This is a convenience method to use when making basic requests that do not involve multipart file data that might need to be specially encoded or handled differently.

Parameters:

method (str) – The HTTP method to use in the request (GET, POST, etc.)
path (str) – A string containing the path elements for this request
params (dict) – A dictionary mapping strings to strings, to be used as the key/value pairs in the request parameters.
data – A dictionary, bytes or file-like object to send in the body.
files (dict) – A dictionary of ‘name’ => file-like-objects for multipart encoding upload.
json (dict) – A JSON object to send in the request body.
headers (dict) – If present, a dictionary of headers to encode in the request.
json_resp (bool) –
Whether the response should be parsed as JSON. If False, the raw response object is returned. To get the raw binary content of the response, use the content attribute of the return value, e.g.
```
resp = client.get('my/endpoint', json_resp=False)
print(resp.content)  # Raw binary content
print(resp.headers)  # Dict of headers
```
retry_statuses – a sequence of HTTP response status codes to retry in addition to dandi.consts.RETRY_STATUSES
retry_if – an optional predicate applied to a failed HTTP response to test whether to retry

class dandi.dandiapi.DandiAPIClient(api_url: str | None = None, token: str | None = None, dandi_instance: DandiInstance | None = None)[source]#

Bases: RESTFullAPIClient

A client for interacting with a Dandi Archive server

authenticate(token: str, save_to_keyring: bool = False) → None[source]#

Set the authentication token/API key used by the DandiAPIClient. Before setting the token, a test request to /auth/token is made to check the token’s validity; if it fails, a requests.HTTPError is raised.

If save_to_keyring is true, then (after querying /auth/token but before setting the API key used by the client), the token is saved in the user’s keyring at the same location as used by dandi_authenticate().

Changed in version 0.53.0: save_to_keyring added

check_schema_version(schema_version: str | None = None) → None[source]#

Confirms that the server is using the same version of the Dandi schema as the client. If it is not, a SchemaVersionError is raised.

Parameters:: schema_version – the schema version to confirm that the server uses; if not set, the schema version for the installed dandischema library is used

create_dandiset(name: str, metadata: dict[str, Any], *, embargo: bool = False) → RemoteDandiset[source]#: Creates a Dandiset with the given name & metadata. If embargo is True, the resulting Dandiset will be embargoed.

Changed in version 0.61.0: embargo argument added

dandi_authenticate() → None[source]#: Acquire and set the authentication token/API key used by the DandiAPIClient. If the DANDI_API_KEY environment variable is set, its value is used as the token. Otherwise, the token is looked up in the user’s keyring under the service “dandi-api-INSTANCE_NAME” [1] and username “key”. If no token is found there, the user is prompted for the token, and, if it proves to be valid, it is stored in the user’s keyring.

[1]
E.g., “dandi-api-dandi” for the production server or “dandi-api-dandi-staging” for the staging server

classmethod for_dandi_instance(instance: str | DandiInstance, token: str | None = None, authenticate: bool = False) → DandiAPIClient[source]#: Construct a client instance for the server identified by instance (either the name of a registered Dandi Archive instance or a DandiInstance instance) and an optional authentication token/API key. If no token is supplied and authenticate is true, dandi_authenticate() is called on the instance before returning it.

get_asset(asset_id: str) → BaseRemoteAsset[source]#

Fetch the asset with the given asset ID. If the given asset does not exist, a NotFoundError is raised.

The returned object will not have any information about the Dandiset associated with the asset; for that, the RemoteDandiset.get_asset() method must be used instead.

get_dandiset(dandiset_id: str, version_id: str | None = None, lazy: bool = True) → RemoteDandiset[source]#

Fetches the Dandiset with the given dandiset_id. If version_id is not specified, the RemoteDandiset’s version is set to the most recent published version if there is one, otherwise to the draft version.

If lazy is true, no requests are actually made until any data is requested from the RemoteDandiset.

Returns a generator of all Dandisets on the server. For each Dandiset, the RemoteDandiset’s version is set to the most recent published version if there is one, otherwise to the draft version.

Changed in version 0.61.0: draft, embargoed, empty, mine, order, and search parameters added

Parameters:

draft – If true, Dandisets that have only draft versions (i.e., that haven’t yet been published) will be included in the results (default true)
embargoed – If true, embargoed Dandisets will be included in the results (default false)
empty – If true, empty Dandisets will be included in the results (default true)
mine – If true, only Dandisets owned by the authenticated user will be retrieved (default false)
order – The field to sort the results by. The accepted field names are "id", "name", "modified", and "size". Prepend a hyphen to the field name to reverse the sort order.
search – A search string to filter the returned Dandisets by. The string is searched for in the metadata of Dandiset versions.

page_size: int | None#: Default number of items to request per page when paginating (None means to use the server’s default)

page_workers: int#: How many pages to fetch at once when parallelizing pagination

Dandisets#

class dandi.dandiapi.RemoteDandiset[source]#

Representation of a Dandiset (as of a certain version) retrieved from the API.

Stringifying a RemoteDandiset returns a string of the form "server_id:dandiset_id/version_id".

This class should not be instantiated by end-users directly. Instead, instances should be retrieved from the appropriate attributes & methods of DandiAPIClient and RemoteDandiset.

property api_path: str#: The path (relative to the base endpoint for a Dandi Archive API) at which API requests for interacting with the Dandiset itself are made

property api_url: str#: The URL at which API requests for interacting with the Dandiset itself are made

client: DandiAPIClient#: The DandiAPIClient instance that returned this RemoteDandiset and which the latter will use for API requests

property contact_person: str#: The name of the registered contact person for the Dandiset

property created: datetime#: The timestamp at which the Dandiset was created

delete() → None[source]#: Delete the Dandiset from the server. Any further access of the instance’s data attributes afterwards will result in a NotFoundError.

download_directory(assets_dirpath: str, dirpath: str | Path, chunk_size: int = 8388608) → None[source]#: Download all assets under the virtual directory assets_dirpath to the directory dirpath. Downloads are synchronous.

property draft_version: Version#: The draft version of the Dandiset

property embargo_status: EmbargoStatus#: The current embargo status for the Dandiset

for_version(version_id: str | Version) → RemoteDandiset[source]#: Returns a copy of the RemoteDandiset with the version attribute set to the given Version object or the Version with the given version ID. If a version ID given and the version does not exist, a NotFoundError is raised.

classmethod from_data(client: DandiAPIClient, data: dict[str, Any]) → RemoteDandiset[source]#

Construct a RemoteDandiset instance from a DandiAPIClient and a dict of raw string fields in the same format as returned by the API. If the "most_recent_published_version" field is set, that is used as the Dandiset’s version; otherwise, "draft_version" is used.

This is a low-level method that non-developers would normally only use when acquiring data using means outside of this library.

get_asset(asset_id: str) → RemoteAsset[source]#: Fetch the asset in this version of the Dandiset with the given asset ID. If the given asset does not exist, a NotFoundError is raised.

get_asset_by_path(path: str) → RemoteAsset[source]#: Fetch the asset in this version of the Dandiset whose path equals path. If the given asset does not exist, a NotFoundError is raised.

get_assets(order: str | None = None) → Iterator[RemoteAsset][source]#

Returns an iterator of all assets in this version of the Dandiset.

Assets can be sorted by a given field by passing the name of that field as the order parameter. The accepted field names are "created", "modified", and "path". Prepend a hyphen to the field name to reverse the sort order.

get_assets_by_glob(pattern: str, order: str | None = None) → Iterator[RemoteAsset][source]#

New in version 0.44.0.

Returns an iterator of all assets in this version of the Dandiset whose path attributes match the glob pattern pattern

get_assets_with_path_prefix(path: str, order: str | None = None) → Iterator[RemoteAsset][source]#

Returns an iterator of all assets in this version of the Dandiset whose path attributes start with path

get_metadata() → Dandiset[source]#: Fetch the metadata for this version of the Dandiset as a dandischema.models.Dandiset instance

Note

Only published Dandiset versions can be expected to have valid metadata. Consider using get_raw_metadata() instead in order to fetch unstructured, possibly-invalid metadata.

get_raw_metadata() → dict[str, Any][source]#: Fetch the metadata for this version of the Dandiset as an unprocessed dict

get_version(version_id: str) → VersionInfo[source]#: Get information about a given version of the Dandiset. If the given version does not exist, a NotFoundError is raised.

Changed in version 0.49.0: This method now returns a VersionInfo instance instead of just a Version.

get_versions(order: str | None = None) → Iterator[Version][source]#

Returns an iterator of all available Versions for the Dandiset

Versions can be sorted by a given field by passing the name of that field as the order parameter. Currently, the only accepted field name is "created". Prepend a hyphen to the field name to reverse the sort order.

identifier: str#: The Dandiset identifier

iter_upload_raw_asset(filepath: str | Path, asset_metadata: dict[str, Any], jobs: int | None = None, replace_asset: RemoteAsset | None = None) → Iterator[dict][source]#

Upload the file at filepath with metadata asset_metadata to this version of the Dandiset, returning a generator of status dicts.

Deprecated since version 0.36.0: Use the iter_upload() method of LocalAsset instances instead

Parameters:

filepath (str or PathLike) – the path to the local file to upload
asset_metadata (dict) – Metadata for the uploaded asset file. Must include a “path” field giving the forward-slash-separated path at which the uploaded file will be placed on the server.
jobs (int) – Number of threads to use for uploading; defaults to 5
replace_asset (RemoteAsset) – If set, replace the given asset, which must have the same path as the new asset

Returns:

A generator of dicts containing at least a "status" key. Upon successful upload, the last dict will have a status of "done" and an "asset" key containing the resulting RemoteAsset.

json_dict() → dict[str, Any][source]#: Convert to a JSONable dict, omitting the client attribute and using the same field names as in the API

property modified: datetime#: The timestamp at which the Dandiset was last modified

property most_recent_published_version: Version | None#: The most recent published (non-draft) version of the Dandiset, or None if no versions have been published

publish(max_time: float = 120) → RemoteDandiset[source]#

Publish the draft version of the Dandiset and wait at most max_time seconds for the publication operation to complete. If the operation does not complete in time, a ValueError is raised.

Returns a copy of the RemoteDandiset with the version attribute set to the new published Version.

refresh() → None[source]#: Update the RemoteDandiset in-place with the latest data from the server. The RemoteDandiset continues to have the same version as before, but the cached version data is internally cleared and may be different upon subsequent access.

set_metadata(metadata: Dandiset) → None[source]#: Set the metadata for this version of the Dandiset to the given value

set_raw_metadata(metadata: dict[str, Any]) → None[source]#: Set the metadata for this version of the Dandiset to the given value

upload_raw_asset(filepath: str | Path, asset_metadata: dict[str, Any], jobs: int | None = None, replace_asset: RemoteAsset | None = None) → RemoteAsset[source]#

Upload the file at filepath with metadata asset_metadata to this version of the Dandiset and return the resulting asset. Blocks until the upload is complete.

Deprecated since version 0.36.0: Use the upload() method of LocalAsset instances instead

Parameters:

filepath (str or PathLike) – the path to the local file to upload
asset_metadata (dict) – Metadata for the uploaded asset file. Must include a “path” field giving the forward-slash-separated path at which the uploaded file will be placed on the server.
jobs (int) – Number of threads to use for uploading; defaults to 5
replace_asset (RemoteAsset) – If set, replace the given asset, which must have the same path as the new asset

property version: Version#: The version in question of the Dandiset

property version_api_path: str#: The path (relative to the base endpoint for a Dandi Archive API) at which API requests for interacting with the version in question of the Dandiset are made

property version_api_url: str#: The URL at which API requests for interacting with the version in question of the Dandiset are made

property version_id: str#: The identifier for the Dandiset version

wait_until_valid(max_time: float = 120) → None[source]#: Wait at most max_time seconds for the Dandiset to be valid for publication. If the Dandiset does not become valid in time, a ValueError is raised.

class dandi.dandiapi.Version[source]#

The version information for a Dandiset retrieved from the API.

Stringifying a Version returns its identifier.

This class should not be instantiated by end-users directly. Instead, instances should be retrieved from the appropriate attributes & methods of RemoteDandiset.

asset_count: int#: The number of assets in the version

created: datetime#: The timestamp at which the version was created

identifier: str#: The version identifier

json_dict() → dict[str, Any]#: Convert to a JSONable dict, omitting the client attribute and using the same field names as in the API

model_computed_fields: ClassVar[dict[str, ComputedFieldInfo]] = {}#: A dictionary of computed field names and their corresponding ComputedFieldInfo objects.

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True, 'populate_by_name': True}#: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

model_fields: ClassVar[dict[str, FieldInfo]] = {'asset_count': FieldInfo(annotation=int, required=True), 'created': FieldInfo(annotation=datetime, required=True), 'identifier': FieldInfo(annotation=str, required=True, alias='version', alias_priority=2), 'modified': FieldInfo(annotation=datetime, required=True), 'name': FieldInfo(annotation=str, required=True), 'size': FieldInfo(annotation=int, required=True), 'status': FieldInfo(annotation=VersionStatus, required=True)}#

Metadata about the fields defined on the model, mapping of field names to [FieldInfo][pydantic.fields.FieldInfo].

This replaces Model.__fields__ from Pydantic V1.

modified: datetime#: The timestamp at which the version was last modified

name: str#: The name of the version

size: int#: The total size in bytes of all assets in the version

status: VersionStatus#

class dandi.dandiapi.VersionInfo[source]#

Bases: Version

New in version 0.49.0.

Version information for a Dandiset, including information about validation errors

asset_validation_errors: List[RemoteAssetValidationError]#

model_computed_fields: ClassVar[dict[str, ComputedFieldInfo]] = {}#: A dictionary of computed field names and their corresponding ComputedFieldInfo objects.

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True, 'populate_by_name': True}#: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

model_fields: ClassVar[dict[str, FieldInfo]] = {'asset_count': FieldInfo(annotation=int, required=True), 'asset_validation_errors': FieldInfo(annotation=List[dandi.dandiapi.RemoteAssetValidationError], required=True), 'created': FieldInfo(annotation=datetime, required=True), 'identifier': FieldInfo(annotation=str, required=True, alias='version', alias_priority=2), 'modified': FieldInfo(annotation=datetime, required=True), 'name': FieldInfo(annotation=str, required=True), 'size': FieldInfo(annotation=int, required=True), 'status': FieldInfo(annotation=VersionStatus, required=True), 'version_validation_errors': FieldInfo(annotation=List[dandi.dandiapi.RemoteValidationError], required=True)}#

Metadata about the fields defined on the model, mapping of field names to [FieldInfo][pydantic.fields.FieldInfo].

This replaces Model.__fields__ from Pydantic V1.

version_validation_errors: List[RemoteValidationError]#

class dandi.dandiapi.RemoteValidationError[source]#

New in version 0.49.0.

Validation error record obtained from a server. Not to be confused with dandi.validate_types.ValidationResult, which provides richer representation of validation errors.

field: str#

json_dict() → dict[str, Any]#: Convert to a JSONable dict, omitting the client attribute and using the same field names as in the API

message: str#

model_computed_fields: ClassVar[dict[str, ComputedFieldInfo]] = {}#: A dictionary of computed field names and their corresponding ComputedFieldInfo objects.

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True, 'populate_by_name': True}#: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

model_fields: ClassVar[dict[str, FieldInfo]] = {'field': FieldInfo(annotation=str, required=True), 'message': FieldInfo(annotation=str, required=True)}#

Metadata about the fields defined on the model, mapping of field names to [FieldInfo][pydantic.fields.FieldInfo].

This replaces Model.__fields__ from Pydantic V1.

Assets#

class dandi.dandiapi.BaseRemoteAsset[source]#

Representation of an asset retrieved from the API without associated Dandiset information.

This is an abstract class; its concrete subclasses are BaseRemoteBlobAsset (for assets backed by blobs) and BaseRemoteZarrAsset (for assets backed by Zarrs).

Stringifying a BaseRemoteAsset returns a string of the form "server_id:asset/asset_id".

This class should not be instantiated by end-users directly. Instead, instances should be retrieved from the appropriate methods of DandiAPIClient.

property api_path: str#: The path (relative to the base endpoint for a Dandi Archive API) at which API requests for interacting with the asset itself are made

property api_url: str#: The URL at which API requests for interacting with the asset itself are made

abstract property asset_type: AssetType#: New in version 0.36.0.

The type of the asset’s underlying data

property base_download_url: str#: The URL from which the asset can be downloaded, sans any Dandiset identifiers (cf. RemoteAsset.download_url)

client: DandiAPIClient#: The DandiAPIClient instance that returned this BaseRemoteAsset and which the latter will use for API requests

created: datetime#: The date at which the asset was created

property digest_type: DigestType#: New in version 0.36.0.

The primary digest algorithm used by Dandi Archive for the asset, determined based on its underlying data: dandi-etag for blob resources, dandi-zarr-checksum for Zarr resources

download(filepath: str | Path, chunk_size: int = 8388608) → None[source]#

Download the asset to filepath. Blocks until the download is complete.

Raises:: ValueError – if the asset is not backed by a blob

classmethod from_base_data(client: DandiAPIClient, data: dict[str, Any], metadata: dict[str, Any] | None = None) → BaseRemoteAsset[source]#

Construct a BaseRemoteAsset instance from a DandiAPIClient, a dict of raw data in the same format as returned by the API’s pagination endpoints, and optional raw asset metadata.

This is a low-level method that non-developers would normally only use when acquiring data using means outside of this library.

get_content_url(regex: str = '.*', follow_redirects: bool | int = False, strip_query: bool = False) → str[source]#

Returns a URL for downloading the asset, found by inspecting the metadata; specifically, returns the first contentUrl that matches regex. Raises NotFoundError if the metadata does not contain a matching URL.

If follow_redirects is True, a HEAD request is made to resolve any & all redirects before returning the URL. If follow_redirects is an integer, at most that many redirects are followed.

If strip_query is true, any query parameters are removed from the final URL before returning it.

get_digest() → Digest[source]#: New in version 0.36.0: Replaces the previous version of get_digest(), now renamed to get_raw_digest()

Retrieves the DANDI etag digest of the appropriate type for the asset: a dandi-etag digest for blob resources or a dandi-zarr-checksum for Zarr resources

get_download_file_iter(chunk_size: int = 8388608) → Callable[[int], Iterator[bytes]][source]#

Returns a function that when called (optionally with an offset into the asset to start downloading at) returns a generator of chunks of the asset.

Raises:: ValueError – if the asset is not backed by a blob

get_metadata() → Asset[source]#: Fetch the metadata for the asset as a dandischema.models.Asset instance

Note

Only assets in published Dandiset versions can be expected to have valid metadata. Consider using get_raw_metadata() instead in order to fetch unstructured, possibly-invalid metadata.

get_raw_digest(digest_type: str | models.DigestType | None = None) → str[source]#

Retrieves the value of the given type of digest from the asset’s metadata. Raises NotFoundError if there is no entry for the given digest type.

If no digest type is specified, the same type as used by get_digest() is returned.

Changed in version 0.36.0: Renamed from get_digest() to get_raw_digest()

get_raw_metadata() → dict[str, Any][source]#: Fetch the metadata for the asset as an unprocessed dict

identifier: str#: The asset identifier

json_dict() → dict[str, Any]#: Convert to a JSONable dict, omitting the client attribute and using the same field names as in the API

model_computed_fields: ClassVar[dict[str, ComputedFieldInfo]] = {}#: A dictionary of computed field names and their corresponding ComputedFieldInfo objects.

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True, 'populate_by_name': True}#: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

Metadata about the fields defined on the model, mapping of field names to [FieldInfo][pydantic.fields.FieldInfo].

This replaces Model.__fields__ from Pydantic V1.

model_post_init(__context: Any) → None#

This function is meant to behave like a BaseModel method to initialise private attributes.

It takes context as an argument since that’s what pydantic-core passes when calling it.

Args:: self: The BaseModel instance. __context: The context.

modified: datetime#: The date at which the asset was last modified

path: str#: The asset’s (forward-slash-separated) path

size: int#: The size of the asset in bytes

class dandi.dandiapi.BaseRemoteBlobAsset[source]#

Bases: BaseRemoteAsset

New in version 0.36.0.

A BaseRemoteAsset whose actual data is a blob resource

as_readable() → RemoteReadableAsset[source]#: New in version 0.50.0.

Returns a Readable instance that can be used to obtain a file-like object for reading bytes directly from the asset on the server

property asset_type: AssetType#: New in version 0.36.0.

The type of the asset’s underlying data

blob: str#: The ID of the underlying blob resource

model_computed_fields: ClassVar[dict[str, ComputedFieldInfo]] = {}#: A dictionary of computed field names and their corresponding ComputedFieldInfo objects.

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True, 'populate_by_name': True}#: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

model_fields: ClassVar[dict[str, FieldInfo]] = {'blob': FieldInfo(annotation=str, required=True), 'client': FieldInfo(annotation=DandiAPIClient, required=True, exclude=True), 'created': FieldInfo(annotation=datetime, required=True), 'identifier': FieldInfo(annotation=str, required=True, alias='asset_id', alias_priority=2), 'modified': FieldInfo(annotation=datetime, required=True), 'path': FieldInfo(annotation=str, required=True), 'size': FieldInfo(annotation=int, required=True)}#

Metadata about the fields defined on the model, mapping of field names to [FieldInfo][pydantic.fields.FieldInfo].

This replaces Model.__fields__ from Pydantic V1.

class dandi.dandiapi.AssetType(value)[source]#

New in version 0.36.0.

An enum for the different kinds of resources that an asset’s actual data can be

BLOB = 1#

ZARR = 2#

class dandi.dandiapi.RemoteAsset[source]#

Bases: BaseRemoteAsset

Subclass of BaseRemoteAsset that includes information about the Dandiset to which the asset belongs.

This is an abstract class; its concrete subclasses are RemoteBlobAsset (for assets backed by blobs) and RemoteZarrAsset (for assets backed by Zarrs).

This class should not be instantiated by end-users directly. Instead, instances should be retrieved from the appropriate attributes & methods of RemoteDandiset.

property api_path: str#: The path (relative to the base endpoint for a Dandi Archive API) at which API requests for interacting with the asset itself are made

property api_url: str#: The URL at which API requests for interacting with the asset itself are made

dandiset_id: str#: The identifier for the Dandiset to which the asset belongs

delete() → None[source]#: Delete the asset

property download_url: str#: The URL from which the asset can be downloaded, including Dandiset identifiers (cf. BaseRemoteAsset.base_download_url)

classmethod from_data(dandiset: RemoteDandiset, data: dict[str, Any], metadata: dict[str, Any] | None = None) → RemoteAsset[source]#

Construct a RemoteAsset instance from a RemoteDandiset, a dict of raw data in the same format as returned by the API’s pagination endpoints, and optional raw asset metadata.

This is a low-level method that non-developers would normally only use when acquiring data using means outside of this library.

model_computed_fields: ClassVar[dict[str, ComputedFieldInfo]] = {}#: A dictionary of computed field names and their corresponding ComputedFieldInfo objects.

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True, 'populate_by_name': True}#: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

model_fields: ClassVar[dict[str, FieldInfo]] = {'client': FieldInfo(annotation=DandiAPIClient, required=True, exclude=True), 'created': FieldInfo(annotation=datetime, required=True), 'dandiset_id': FieldInfo(annotation=str, required=True, exclude=True), 'identifier': FieldInfo(annotation=str, required=True, alias='asset_id', alias_priority=2), 'modified': FieldInfo(annotation=datetime, required=True), 'path': FieldInfo(annotation=str, required=True), 'size': FieldInfo(annotation=int, required=True), 'version_id': FieldInfo(annotation=str, required=True, exclude=True)}#

Metadata about the fields defined on the model, mapping of field names to [FieldInfo][pydantic.fields.FieldInfo].

This replaces Model.__fields__ from Pydantic V1.

rename(dest: str) → None[source]#: New in version 0.41.0.

Change the path of the asset on the server to the given value and update the RemoteAsset in place. If another asset already exists at the given path, a requests.HTTPError is raised.

set_metadata(metadata: Asset) → None[source]#: Set the metadata for the asset to the given value and update the RemoteAsset in place.

abstract set_raw_metadata(metadata: dict[str, Any]) → None[source]#: Set the metadata for the asset on the server to the given value and update the RemoteAsset in place.

version_id: str#: The identifier for the version of the Dandiset to which the asset belongs

class dandi.dandiapi.RemoteBlobAsset[source]#

Bases: RemoteAsset, BaseRemoteBlobAsset

New in version 0.36.0.

A RemoteAsset whose actual data is a blob resource

client: DandiAPIClient#: The DandiAPIClient instance that returned this BaseRemoteAsset and which the latter will use for API requests

created: datetime#: The date at which the asset was created

dandiset_id: str#: The identifier for the Dandiset to which the asset belongs

identifier: str#: The asset identifier

model_computed_fields: ClassVar[dict[str, ComputedFieldInfo]] = {}#: A dictionary of computed field names and their corresponding ComputedFieldInfo objects.

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True, 'populate_by_name': True}#: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

model_fields: ClassVar[dict[str, FieldInfo]] = {'blob': FieldInfo(annotation=str, required=True), 'client': FieldInfo(annotation=DandiAPIClient, required=True, exclude=True), 'created': FieldInfo(annotation=datetime, required=True), 'dandiset_id': FieldInfo(annotation=str, required=True, exclude=True), 'identifier': FieldInfo(annotation=str, required=True, alias='asset_id', alias_priority=2), 'modified': FieldInfo(annotation=datetime, required=True), 'path': FieldInfo(annotation=str, required=True), 'size': FieldInfo(annotation=int, required=True), 'version_id': FieldInfo(annotation=str, required=True, exclude=True)}#

Metadata about the fields defined on the model, mapping of field names to [FieldInfo][pydantic.fields.FieldInfo].

This replaces Model.__fields__ from Pydantic V1.

modified: datetime#: The date at which the asset was last modified

path: str#: The asset’s (forward-slash-separated) path

set_raw_metadata(metadata: dict[str, Any]) → None[source]#: Set the metadata for the asset on the server to the given value and update the RemoteBlobAsset in place.

size: int#: The size of the asset in bytes

version_id: str#: The identifier for the version of the Dandiset to which the asset belongs

Zarr Assets#

class dandi.dandiapi.BaseRemoteZarrAsset[source]#

Bases: BaseRemoteAsset

New in version 0.36.0.

A BaseRemoteAsset whose actual data is a Zarr resource

property asset_type: AssetType#: New in version 0.36.0.

The type of the asset’s underlying data

get_entry_by_path(path: str) → RemoteZarrEntry[source]#: Fetch the entry in this Zarr whose path equals path. If the given entry does not exist, a NotFoundError is raised.

iterfiles(prefix: str | None = None) → Iterator[RemoteZarrEntry][source]#: Returns a generator of all RemoteZarrEntrys within the Zarr, optionally limited to those whose path starts with the given prefix

model_computed_fields: ClassVar[dict[str, ComputedFieldInfo]] = {}#: A dictionary of computed field names and their corresponding ComputedFieldInfo objects.

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True, 'populate_by_name': True}#: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

model_fields: ClassVar[dict[str, FieldInfo]] = {'client': FieldInfo(annotation=DandiAPIClient, required=True, exclude=True), 'created': FieldInfo(annotation=datetime, required=True), 'identifier': FieldInfo(annotation=str, required=True, alias='asset_id', alias_priority=2), 'modified': FieldInfo(annotation=datetime, required=True), 'path': FieldInfo(annotation=str, required=True), 'size': FieldInfo(annotation=int, required=True), 'zarr': FieldInfo(annotation=str, required=True)}#

Metadata about the fields defined on the model, mapping of field names to [FieldInfo][pydantic.fields.FieldInfo].

This replaces Model.__fields__ from Pydantic V1.

rmfiles(files: Iterable[RemoteZarrEntry], reingest: bool = True) → None[source]#

Delete one or more files from the Zarr.

If reingest is true, after performing the deletion, the client triggers a recalculation of the Zarr’s checksum and waits for it to complete.

zarr: str#: The ID of the underlying Zarr resource

class dandi.dandiapi.RemoteZarrAsset[source]#

Bases: RemoteAsset, BaseRemoteZarrAsset

New in version 0.36.0.

A RemoteAsset whose actual data is a Zarr resource

client: DandiAPIClient#: The DandiAPIClient instance that returned this BaseRemoteAsset and which the latter will use for API requests

created: datetime#: The date at which the asset was created

dandiset_id: str#: The identifier for the Dandiset to which the asset belongs

identifier: str#: The asset identifier

model_computed_fields: ClassVar[dict[str, ComputedFieldInfo]] = {}#: A dictionary of computed field names and their corresponding ComputedFieldInfo objects.

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True, 'populate_by_name': True}#: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

Metadata about the fields defined on the model, mapping of field names to [FieldInfo][pydantic.fields.FieldInfo].

This replaces Model.__fields__ from Pydantic V1.

modified: datetime#: The date at which the asset was last modified

path: str#: The asset’s (forward-slash-separated) path

set_raw_metadata(metadata: dict[str, Any]) → None[source]#: Set the metadata for the asset on the server to the given value and update the RemoteZarrAsset in place.

size: int#: The size of the asset in bytes

version_id: str#: The identifier for the version of the Dandiset to which the asset belongs

class dandi.dandiapi.RemoteZarrEntry[source]#

Bases: object

New in version 0.36.0.

A file within a RemoteZarrAsset

Changed in version 0.48.0:

No longer represents directories
No longer implements BasePath

client: DandiAPIClient#: The DandiAPIClient instance used for API requests

digest: Digest#: The entry’s digest

get_download_file_iter(chunk_size: int = 8388608) → Callable[[int], Iterator[bytes]][source]#: Returns a function that when called (optionally with an offset into the file to start downloading at) returns a generator of chunks of the file

match(pattern: str) → bool[source]#: Tests whether the path matches the given glob pattern

modified: datetime#: The timestamp at which the entry was last modified

property name: str#: The basename of the path object

parts: tuple[str, ...]#: The path components of the entry

size: int#: The entry’s size in bytes

property stem: str#: The basename without its final file extension, if any

property suffix: str#: The final file extension of the basename, if any

property suffixes: list[str]#: A list of the basename’s file extensions

zarr_id: str#: The ID of the Zarr backing the asset

dandi.dandiapi#

Client#

Dandisets#

Assets#

Zarr Assets#

`dandi.dandiapi`#