一键导入
dash_data_layer
dash_data_layer 收录了来自 craiglabenz 的 25 个 skills,并提供仓库级职业覆盖和站内 skill 详情页。
这个仓库中的 skills
Understand how sources declare operation capabilities via supportedOperations and how SourceList respects them.
FirestoreAdminSource connects google_cloud_firestore for server-side Dart applications to a SourceList.
FirestoreSource seamlessly connects Cloud Firestore streams to a SourceList.
Perform untyped Firestore document merges using the raw method on Firestore sources.
Cached data is mapped uniquely per request hash (RequestDetails with specific filter and pagination), supporting a powerful request-based caching behavior.
The SourceList class manages the delegation of read/write attempts to various configured Sources, with a read-thru cache system prioritizing local Sources.
ConnectivityPlusStream provides a stream of network statuses based on connectivity_plus so DefaultRetryPolicy can intelligently manage failed Operations.
HiveSource is a LocalSource implementation that writes to Hive boxes.
Remote sources are the API-specific bindings for pkg:data_layer.
Watching realtime data amounts to calling `watch`, `watchByIds`, or `watchList` on a Repository whose SourceList contains a `WatchableSource`.
Authoring a new local source allows different storage mechanisms to participate in the data layer caching architecture.
Proxy Sources offer a build-a-bear type solution for irregular or JSON RPC APIs.
Forces reads to go to the server when locally cached data is suspected of being stale
Automatically invalidate local caches after a specified time to live (ttl).
Ejecting all data from local sources is as simple as calling .clear()
Repositories are simple wrappers around SourceLists plus custom business logic.
Bindings define the serialization and request networking footprint for your specific data model.
Use a MessageRepository to define a special DTO class for your writes.
There are distinct usage scenarios where mapping data asynchronously from a server securely strictly introduces UI latency against pre-populated ma...
Filters communicate intent to limit to read a specific subset of data. They are never executed locally and rely on "remote" sources correctly applying the filter to their network request.
Paginating reads relies on passing a Pagination object to a RequestDetails and having a remote Source which correctly applies its values to the network request.
Reading data involves calling getById, getByIds, or getItems on a Repository.
SourceLists can be configured to store failed Operations for later retry if a `RetryPolicy` is supplied.
SourceLists should be defined with Sources sorted by immediacy; with on-device Sources appearing first and remote, API-based Sources appearing last. Often, 1 local Source and 1 remote Source is all that will appear in a SourceList.
Writing data involves calling setItem and, optionally, setItems on a Repository.