alias.md

1. Aliases can be defined inside a pipeline as follows:

• In Xml:

  <aliases>
      <alias name="file" type="com.qwshen.etl.source.FileReader" />
      <alias name="flat" type="com.qwshen.etl.source.FlatReader" />
      <alias name="setting" type="com.qwshen.etl.common.SparkConfActor" />
      <alias name="sql" type="com.qwshen.etl.transform.SqlTransformer" />
  </aliases>

• In Json

  "aliases": [
    {
      "name": "file",
      "type": "com.qwshen.etl.source.FileReader"
    },
    {
      "name": "flat",
      "type": "com.qwshen.etl.source.FlatReader"
    },
    {
      "name": "setting",
      "type": "com.qwshen.etl.setting.SparkConfSetter"
    },
    {
      "name": "sql",
      "type": "com.qwshen.etl.transform.SqlTransformer"
    }
  ]

• In Yaml

  aliases:
    - name: file
      type: com.qwshen.etl.source.FileReader
    - name: flat
      type: com.qwshen.etl.source.FlatReader
    - name: setting
      type: com.qwshen.etl.common.SparkConfActor
    - name: sql
      type: com.qwshen.etl.transform.SqlTransformer

2. Besides, aliases can be defined in a centralized file:

• In Xml. The following content is defined in a separated file - alias.xml

  <aliases>
      <alias name="file" type="com.qwshen.etl.source.FileReader" />
      <alias name="flat" type="com.qwshen.etl.source.FlatReader" />
      <alias name="setting" type="com.qwshen.etl.common.SparkConfActor" />
      <alias name="sql" type="com.qwshen.etl.transform.SqlTransformer" />
  </aliases>

  Then in a pipeline, include above aliases as follows:

  <aliases include="./miscellaneous/alias.xml" />

• In Json. The following aliases are defined in a separated file - alias.json

  {
    "aliases": [
      {
        "name": "file",
        "type": "com.qwshen.etl.source.FileReader"
      },
      {
        "name": "flat",
        "type": "com.qwshen.etl.source.FlatReader"
      },
      {
        "name": "setting",
        "type": "com.qwshen.etl.setting.SparkConfSetter"
      },
      {
        "name": "sql",
        "type": "com.qwshen.etl.transform.SqlTransformer"
      }
    ]
  }

  Then in a pipeline, include above aliases as follows:

  "aliases": {
    "include": "./miscellaneous/alias.json"
  }

• In Yaml. The following aliases are defined in a separated file - alias.yaml

  aliases:
    - name: file
      type: com.qwshen.etl.source.FileReader
    - name: flat
      type: com.qwshen.etl.source.FlatReader
    - name: setting
      type: com.qwshen.etl.common.SparkConfActor
    - name: sql
      type: com.qwshen.etl.transform.SqlTransformer

  Then in a pipeline, include above aliases as follows:

  aliases:
    include: ./miscellaneous/alias.yaml

3. The mixed mode - for the common aliases that are used across multiple pipelines, the aliases are better defined in a separated file, and include it in each pipeline. However, one pipeline may have a particular alias.

• In Xml. The common aliases are defined in a separated file - alias.xml

  <aliases>
      <alias name="file" type="com.qwshen.etl.source.FileReader" />
      <alias name="flat" type="com.qwshen.etl.source.FlatReader" />
      <alias name="sql" type="com.qwshen.etl.transform.SqlTransformer" />
  </aliases>

  Then in a pipeline, include above aliases with one additional alias as follows:

  <aliases include="./miscellaneous/alias.xml">
    <alias name="setting" type="com.qwshen.etl.common.SparkConfActor" />
  </aliases>

• In Json. The common aliases are defined in a separated file - alias.json

  {
    "aliases": [
      {
        "name": "file",
        "type": "com.qwshen.etl.source.FileReader"
      },
      {
        "name": "flat",
        "type": "com.qwshen.etl.source.FlatReader"
      },
      {
        "name": "sql",
        "type": "com.qwshen.etl.transform.SqlTransformer"
      }
    ]
  }

  Then in a pipeline, include above aliases with one additional alias as follows:

  "aliases": [
    {
      "include": "./miscellaneous/alias.json"
    },
    {
      "name": "setting",
      "type": "com.qwshen.etl.setting.SparkConfSetter"
    }
  ]

• In Yaml. The common aliases are defined in a separated file - alias.yaml

  aliases:
    - name: file
      type: com.qwshen.etl.source.FileReader
    - name: flat
      type: com.qwshen.etl.source.FlatReader
    - name: sql
      type: com.qwshen.etl.transform.SqlTransformer

Then in a pipeline, include above aliases with one additional alias as follows:

aliases:
  - include: ./miscellaneous/alias.yaml
  - name: setting
    type: com.qwshen.etl.common.SparkConfActor