tap-datadog

This tap datadog was created by Degreed as a datadog to be used for extracting data via Meltano into defined targets

tap-datadog

These are the steps required for using this repo as a 'datadog' for a Meltano extractor. Note: we will use tap-datadog as the example throughout the process.

1. Being aware of case sensitivity, replace the following throughout the repo:

• tap-datadog >tap-datadog
• tap_datadog > tap_datadog
• TapDatadogStream > TapDatadogStream (inside streams.py)

2. Update the following folders/files to:

• tap_datadog > tap_datadog
• tap-datadog.sh > tap-datadog.sh

3. Inside streams.py update TapDatadogStream with the authentication used for the tap-datadog api calls. Note: all streams in streams.py work as a heirarchy further down. i.e. you can replace the http headers in another stream

4. Using Events(TapDatadogStream) as an example, build your first stream to be synced. There are comments to help identify what values to use

For setting the records_jsonpath value in the stream, you can use a tool likle postman to make a sample call and view the response json. After identifying what keys and values you need to extract, you will need to narrow down the json path. This is a helpful site that you can paste the response text in and help locate the correct path to use. In this example, we want to only extract the id and type values inside data:

{
    "meta": {
        "page": {
            "after": "293048209rudjkfjdsf"
        }
    },
    "data": [
        {
            "type": "error",
            "id": "234234324324234"
        },
        {
            "type": "log",
            "id": "2342123123"
        },
        {
            "type": "log",
            "id": "09823044ugkdf"
        }
    ],
    "links": {
        "next": "https://api.datadoghq.com/api/v2/logs/events?..."
    }
}

Using the link above and entering the value $.data[*], the correct fields are now displaying, confirming that is the correct path:

[
  {
    "type": "error",
    "id": "234234324324234"
  },
  {
    "type": "log",
    "id": "2342123123"
  },
  {
    "type": "log",
    "id": "09823044ugkdf"
  }
]

For the schema, you can create the .json file and place it in the schemas/ folder, or you can create the schema on the fly using the eample in the Events stream

• Option 1: Adding the events.json file to the schemas/ folder:

{
        "type": "object",
        "properties": {
                "id": {
                        "type": "string"
                },
                "type": {
                        "type": "string"
                }
        }
}

• Option 2: Defining schema using hte PropertiesList in the stream class:

schema = th.PropertiesList(
        th.Property("id", th.NumberType),
        th.Property("name", th.StringType),
    ).to_dict()

5. In tap.py add each stream added in streams.py to STREAM_TYPES and define the configuration required:

    config_jsonschema = th.PropertiesList(
        th.Property("api_token", th.StringType, required=False, description="api token for Basic auth"),
        th.Property("start_date", th.StringType, required=False, description="start date for sync"),
    ).to_dict()

6. After updating those components and confirming all references to datadog or Datadog have been updated, you can test the tap locally.

Testing locally

To test locally, pipx poetry

pipx install poetry

Install poetry for the package

poetry install

To confirm everything is setup properly, run the following:

poetry run tap-datadog --help

To run the tap locally outside of Meltano and view the response in a text file, run the following:

poetry run tap-datadog > output.txt 

A full list of supported settings and capabilities is available by running: tap-datadog --about

Config Guide

To test locally, create a config.json with required config values in your tap_datadog folder (i.e. tap_datadog/config.json)

{
  "api_key": "$DD_API_KEY",
  "app_key": "$DD_APP_KEY",
  "start_date": "2022-10-05T00:00:00Z"
}

note: It is critical that you delete the config.json before pushing to github. You do not want to expose an api key or token

Add to Meltano

The provided meltano.yml provides the correct setup for the tap to be installed in the data-houston repo.

At this point you should move all your updated tap files into its own tap-datadog github repo. You also want to make sure you update in the setup.py the url of the repo for you tap.

Update the following in meltano within the data-houston repo with the new tap-datadog credentials/configuration.

plugins:
  extractors:
  - name: tap-datadog
    namespace: tap_datadog
    pip_url: git+https://github.com/degreed-data-engineering/tap-datadog
    capabilities:
    - state
    - catalog
    - discover
    config:
      api_key: $DD_API_KEY
      app_key: $DD_APP_KEY
      start_date: '2022-10-05T00:00:00Z'

To test in data-houston, run the following:

1. make meltano - spins up meltano
2. meltano install extractor tap-datadog - installs the tap
3. meltano invoke tap-datadog --discover > catalog.json - tests the catalog/discovery
4. meltano invoke tap-datadog > output.txt - runs tap with .txt output in meltano/degreed/

That should be it! Feel free to contribute to the tap to help add functionality for any future sources

Singer SDK Dev Guide

See the dev guide for more instructions on how to use the Singer SDK to develop your own taps and targets.