Merge pull request #20 from lulichn/rel/v0.3.0

Ship v0.3.0

CHANGELOG.md

@@ -1,4 +1,4 @@
-0.3.0 (YYYY-MM-DD)
+0.3.0 (2020-03-09)
 ==================
 
 - [Enhancement] Update dependencies
@@ -9,3 +9,41 @@
 - [Enhancement] [#9](https://github.com/lulichn/embulk-input-dynamodb/pull/9) Use TestingEmbulk instead of EmbulkEmbed when testing
 - [Enhancement] [#10](https://github.com/lulichn/embulk-input-dynamodb/pull/10) Reduce test dependencies
 - [Enhancement] [#13](https://github.com/lulichn/embulk-input-dynamodb/pull/13) Use Github Actions instead of CircleCI.
+- [Enhancement] [#15](https://github.com/lulichn/embulk-input-dynamodb/pull/15) Improve development environments
+    - Introduce [scalafmt](https://scalameta.org/scalafmt/) with [spotless](https://github.com/diffplug/spotless)
+    - Fix the format violations
+    - Add [scalafmt](https://scalameta.org/scalafmt/) to CI
+    - Add [CHANGELOG](./CHANGELOG.md)
+    - Add [an example](./example)
+    - Update README about development
+- [Enhancement] [#16](https://github.com/lulichn/embulk-input-dynamodb/pull/16) Cleanup gradle settings
+- [New Feature] [#18](https://github.com/lulichn/embulk-input-dynamodb/pull/18) Introduce new `auth_method`: `"session"`, `"anonymous"`, `"web_identity_token"`, `"default"`.
+    - `"anonymous"`: uses anonymous access. This auth method can access only public files.
+    - `"session"`: uses temporary-generated **access_key_id**, **secret_access_key** and **session_token**.
+    - `"assume_role"`: uses temporary-generated credentials by assuming **role_arn** role.
+    - `"web_identity_token"`: uses temporary-generated credentials by assuming **role_arn** role with web identity.
+    - `"default"`: uses AWS SDK's default strategy to look up available credentials from runtime environment. This method   behaves like the combination of the following methods.
+        1. `"env"`
+        1. `"properties"`
+        1. `"web_identity_token"`
+        1. `"profile"`
+        1. `"instance"`
+- [New Feature] [#18](https://github.com/lulichn/embulk-input-dynamodb/pull/18) Support `http_proxy` option when generating aws credentials.
+- [Enhancement] [#18](https://github.com/lulichn/embulk-input-dynamodb/pull/18) The default value of `auth_method` option become `"default"`. When `access_key_id` and `secret_access_key` options are set, use `"basic"` as `auth_method` for backward compatibility.
+- [Deprecated] [#18](https://github.com/lulichn/embulk-input-dynamodb/pull/18) Make `access_key` and `secret_key` options deprecated. Use `access_key_id` and `secret_access_key` options instead.
+- [Deprecated] [#18](https://github.com/lulichn/embulk-input-dynamodb/pull/18) Make `end_point` option deprecated. Use `endpoint` option instead.
+- [Deprecated] [#19](https://github.com/lulichn/embulk-input-dynamodb/pull/19) The original operation implementation is deprecated, so the below options become deprecated.
+    - **operation**: Use **query** option or **scan** option instead.
+    - **limit**: Use **query.batch_size** option or **query.batch_size** instead.
+    - **scan_limit**: Use **query.batch_size** option or **query.batch_size** instead.
+    - **record_limit**: Use **query.limit** option or **query.limit** instead.
+    - **filters**: Use **query.filter_expression** option or **query.filter_expression** instead.
+- [New Feature] [#19](https://github.com/lulichn/embulk-input-dynamodb/pull/19) Introduce new options **scan**, **query** to support all configurations for Dynamodb Scan/Query Operation API except legacy configurations.
+    - NOTE: This operation stores `null` AttributeValue as `null`, though, in the deprecated operation, `null` is converted arbitrarily. (`string` -> empty string, `long` -> `0`, `double` -> `0.0`, `boolean` -> `false`)
+    - NOTE: This operation stores timestamp values by parsing user-defined format, though the deprecated operation skips storing values when the column type is defined as `timestamp` without any errors.
+    - NOTE: This operation can convert the specific type of the attribute that you specify in **column.attribute_type** to Embulk types, though the deprecated operation can only convert Embulk types that match a particular Dynamodb Attribute type.
+- [Enhancement] [#19](https://github.com/lulichn/embulk-input-dynamodb/pull/19) You can store each dynamodb item as JSON, so **columns** option becomes optional.
+- [Enhancement] [#19](https://github.com/lulichn/embulk-input-dynamodb/pull/19) You can specify the `AttributeValue` type (like `"S"`, `"N"`, `"SS"` and so on) used when converting AttributeValue to Embulk type.
+- [BugFix] [#19](https://github.com/lulichn/embulk-input-dynamodb/pull/19) Avoid `NullPointerException` when Type `N` AttributeValue has `null` in the deprecated operation.
+- [Enhancement] [#19](https://github.com/lulichn/embulk-input-dynamodb/pull/19) Examples work without real Dynamodb.
+- [Enhancement] [#19](https://github.com/lulichn/embulk-input-dynamodb/pull/19) Add more examples.

README.md

@@ -197,7 +197,7 @@ You can see more examples [here](./example).
 $ ./run_dynamodb_local.sh
 $ ./example/prepare_dynamodb_table.sh
 $ ./gradlew classpath
-$ embulk run example/config.yml -Ilib
+$ embulk run example/config-query.yml -Ilib
 ```
 
 ### Run tests

build.gradle

@@ -14,7 +14,7 @@ configurations {
     provided
 }
 
-version = "0.2.0"
+version = "0.3.0"
 
 sourceCompatibility = 1.8
 targetCompatibility = 1.8