liquidiert/deta_dart

Dart package for easy usage of Deta Base, Drive and Auth

Warning

Deta is dead :( Thus this project is archived.

Deta Cloud for Flutter

deta_dart is meant for easy usage of Deta Base, Drive and Auth. All rigths of the beformentioned belong to Deta.

Getting started

Simply import:

import 'package:deta_dart/deta_dart.dart';

and intialize with:

// deta base
DetaBase.instance.init("deta_project_id", "deta_project_key");
// deta drive
DetaDrive.instance.init("deta_project_id", "deta_project_key");

Although it is possible to init as often as you want I recommend to do this once at program start.

Usage

Base

There are multiple options to access a Base instance. The simplest version is to use the default methods:

• addItem
• addMultipleItems
• getItem
• updateItem
• queryItem
• deleteitem

those all expect at least the Deta Base path you want to edit. addItem also expects the object to post for example. The default methods allow flexible path usage but automatically parse your object to fullfill the object format that Deta expects. So there is no need to format those yourself.

Example:

await DetaBase.instance.addItem(
        "/stats/items", {"key": "I'm_a_key", "wins": 0, "losses": 0});

There are also generic versions of the default methods provided. In order to use those DetaBase has to additionally be intialized with a map of types and factory functions (btw I'm using the json_serializable package here):

final factories = <Type, Function>{
    UserStats: (json) => UserStats.fromJson(json)
};

DetaBase.instance.initGenericFactories(factories);

Once that's done you can call the generic versions like the following:

final resp = await DetaBase.instance
    .getGenericItem<UserStats>("/stats/items/$uid");

which will return an instance of UserStats. If you wonder what UserStats consists of, see here. Following generic methods are available:

• addGenericItem
• addMultipleGenericItems
• updateGenericItem
• getGenericItem
• querySingleGenericItem
• queryGenericItems

For those who want the outmost flexibility there is also a "raw" version of modifying calls:

• addRawItem
• addMultipleRawItems
• queryRaw

Those calls can be used if you want to alter the Deta format or if sudden changes of the format are made and this package isn't update soon enough. Though keep in mind that you have to fullfill all Deta format requirements.

Example:

await DetaBase.instance.addRawItem(
        "/stats/items", {"items": {"key": "I'm_a_key", "wins": 0, "losses": 0}});

Finally for the shy ones there are "safe" calls available:

• addItemSafe
• addMultipleItemsSafe
• updateItemSafe
• getItemSafe
• queryItemSafe
• deleteItemSafe

Those calls make sure that:

1. The Deta URL structure is fullfilled
2. The Deta object structure is fullfilled

To use them you just have to provide your Deta Base name (name only; no slashes) and the object / key (depending on the call).

Example:

await DetaBase.instance.addItemSafe(
        "stats", {"key": "I'm_a_key", "wins": 0, "losses": 0});

Drive

The Deta Drive api is built for easy usage and error reduction. E.g. Deta only allows uploads of files smaller than 10MB with their direct upload call. In order to fullfill this requirement the uploadSmallFile method throws an error once the file exceeds 10MB. On the other hand, if you use the chunkedUpload method, a file that is smaller than 5MB (which is the minimal size for a chunk) automatically switches to the uploadSmallFile method. It also warns you btw, indicating that you probably should add a size check before calling the method.
Long story short, after intialization following methods are available:

• uploadSmallFile
• chunkedUpload -> which is a Stream; more on that later
• downloadFile
• listFiles
• listFilesStream
• deleteFiles

uploadSmallFile should be used for files that are smaller than 10MB, it throws an ArgumentError exception if the file is larger than that.

chunkedUpload is meant for files larger than 10MB as mentioned before this method is a Stream. So you can listen for changes and probably display a loading bar. If the file is smaller than 10MB only the result of the uploadSmallFile will be returned otherwise intermediate results are a JSON object of the upload chunked part call response and the last yield is the response of the end chunked upload call.
If something failes during the upload process the chunked upoad will automatically be aborted.

From the remaining calls only listFiles and listFilesStream are special. Both return instances of FileList which is a helper class.
listFilesStream is also a stream, which polls for changes every five seconds. You can change that behavior via the timeOut parameter.

Example (simple CRD):

// upload a file < 10MB
final imgBytes =
        await File.fromUri(Uri.parse("./test/testSrc/Cat03.jpg")).readAsBytes();
final resp =
    await DetaDrive.instance.uploadSmallFile("test", "test.jpg", imgBytes);

// download a file
final img = await DetaDrive.instance.downloadFile("test", "test.jpg");

// delete files -> the name list can contain up to 1000 names
final nameList = ["test.jpg"];
final delResp = await DetaDrive.instance.deleteFiles("test", nameList);

Example (chunked C):

final bytes =
        await File.fromUri(Uri.parse("./test/testSrc/test.pdf")).readAsBytes();
// .last is the last stream response; in this case the response from the end chunked upload call
// if the provided bytes are smaller than 5MB uploadSmallFile will be used instead; also a warning is logged
final resp = await DetaDrive.instance
    .chunkedUpload("test", "test.pdf", bytes)
    .last;