Skip to content

Latest commit

 

History

History
254 lines (184 loc) · 12.2 KB

RUNNING_AND_DEVELOPING.md

File metadata and controls

254 lines (184 loc) · 12.2 KB

Running and Developing Singer Taps and Targets

Since most Singer Taps and Targets are written in Python, the examples given on this page will all be written in Python.

Running Singer with Python

To run Singer with Python, first make sure Python 3.5.2 is installed on your system or follow these installation instructions for Mac or Ubuntu.

We recommend installing each Tap and Target in a separate Python virtual environment. This will insure that you won't have conflicting dependencies between any Taps and Targets.

Running a Singer Tap

  1. Create and activate a Python 3 virtual environment for the Tap, which we'll call tap-foo. When you run this yourself, change the tap name in the angle brackets < >
python3 -m venv ~/.virtualenvs/<tap-foo>
source ~/.virtualenvs/<tap-foo>/bin/activate
  1. Install the Tap using pip:
pip install <tap-foo>
  1. Edit the Tap's config file (tap_config.json) to include any necessary credentials or parameters.

  2. If the tap supports discovery mode, run it to obtain the catalog:

~/.virtualenvs/<tap-foo>/bin/<tap-foo> --config tap_config.json --discover > catalog.json
  1. Depending on what features the Tap supports, you may need to add metadata in the catalog for stream/field selection or replication-method.

  2. Run the Tap in sync mode:

~/.virtualenvs/<tap-foo>/bin/<tap-foo> --config tap_config.json --catalog catalog.json

The output should consist of SCHEMA, RECORD, STATE, and METRIC messages.

Running a Singer Tap with a Singer Target

To run a Singer Tap with a Singer Target, follow steps 1-5 above to set up the Tap, and then continue with the following steps.

  1. Create and activate a Python 3 virtual environment for the Target, which we'll call target-bar. Again, when you run this yourself, change the target name in the angle brackets < >
python3 -m venv ~/.virtualenvs/<target-bar>
source ~/.virtualenvs/<target-bar>/bin/activate
  1. Install the Target in its virtual environment using pip:
pip install <target-bar>
  1. Edit the Target's config file (target_config.json) to include any necessary credentials or parameters.

  2. Run the Singer Tap, pipe the output to the Singer Target, and save the final state emitted by the target for future runs:

~/.virtualenvs/<tap-foo>/bin/<tap-foo> --config tap_config.json --catalog catalog.json | ~/.virtualenvs/<target-bar>/bin/<target-bar> --config target_config.json >> state.json

tail -1 state.json > state.json.tmp && mv state.json.tmp state.json

Example - Using Singer to populate Google Sheets

The Google Sheets Target can be combined with any Singer Tap to populate a Google Sheet with data. This example will use currency exchange rate data from the Fixer.io Tap. Fixer.io is a free API for current and historical foreign exchange rates published by the European Central Bank.

The steps are:

  1. Activate the Google Sheets API
  2. Configure the Target
  3. Install
  4. Run
  5. Save State (optional)

Step 1 - Activate the Google Sheets API

(originally found in the Google API docs)

  1. Use this wizard to create or select a project in the Google Developers Console and activate the Sheets API. Click Continue, then Go to credentials.

  2. On the Add credentials to your project page, click the Cancel button.

  3. At the top of the page, select the OAuth consent screen tab. Select an Email address, enter a Product name if not already set, and click the Save button.

  4. Select the Credentials tab, click the Create credentials button and select OAuth client ID.

  5. Select the application type Other, enter the name "Singer Sheets Target", and click the Create button.

  6. Click OK to dismiss the resulting dialog.

  7. Click the Download button to the right of the client ID.

  8. Move this file to your working directory and rename it client_secret.json.

Step 2 - Configure the Target

Created a file called config.json in your working directory, following config.sample.json. The required spreadsheet_id parameter is the value between the "/d/" and the "/edit" in the URL of your spreadsheet. For example, consider the following URL that references a Google Sheets spreadsheet:

https://docs.google.com/spreadsheets/d/1qpyC0XzvTcKT6EISywvqESX3A0MwQoFDE8p-Bll4hps/edit#gid=0

The ID of this spreadsheet is 1qpyC0XzvTcKT6EISywvqESX3A0MwQoFDE8p-Bll4hps.

Step 3 - Install

target-gsheet can be run with any Singer Tap to move data from sources like Braintree, Freshdesk and Hubspot to Google Sheets. We'll use the Fixer.io Tap - which pulls currency exchange rate data from a public data set - as an example.

These commands will install tap-fixerio and target-gsheet with pip in their own virtual environments:

# Install tap-fixerio in its own virtualenv
python3 -m venv ~/.virtualenvs/tap-fixerio
source ~/.virtualenvs/tap-fixerio/bin/activate
pip install tap-fixerio
deactivate

# Install target-gsheet in its own virtualenv
python3 -m venv ~/.virtualenvs/target-gsheet
source ~/.virtualenvs/target-gsheet/bin/activate
pip install target-gsheet
deactivate

Step 4 - Run

This command will pipe the output of tap-fixerio to target-gsheet, using the configuration file created in Step 2:

~/.virtualenvs/tap-fixerio/bin/tap-fixerio | ~/.virtualenvs/target-gsheet/bin/target-gsheet -c config.json

target-gsheet will attempt to open a new window or tab in your default browser to perform authentication. If this fails, copy the URL from the console and manually open it in your browser.

If you are not already logged into your Google account, you will be prompted to log in. If you are logged into multiple Google accounts, you will be asked to select one account to use for the authorization. Click the Accept button to allow target-gsheet to access your Google Sheet. You can close the tab after the signup flow is complete.

Each stream generated by the Tap will be written to a different sheet in your Google Sheet. For the Fixer.io Tap you'll see a single sheet named exchange_rate.

Step 5 - Save State (optional)

Taps like the Fixer.io Tap can also accept a --state argument that, if present, points to a file containing the last persisted State value. This enables Taps to work incrementally - the State checkpoints the last value that was handled by the Target, and the next time the Tap is run it should pick up from that point.

To run the Fixer.io Tap incrementally, point it to a State file and capture the persister's stdout like this:

~/.virtualenvs/tap-fixerio/bin/tap-fixerio | ~/.virtualenvs/target-gsheet/bin/target-gsheet -c config.json >> state.json

tail -1 state.json > state.json.tmp && mv state.json.tmp state.json
(rinse and repeat)

Developing a Tap

If you can't find an existing Tap for your data source, then it's time to build your own.

Topics:

Hello, world

A Tap is just a program, written in any language, that outputs data to stdout according to the [Singer spec]. In fact, your first Tap can be written from the command line, without any programming at all:

printf '{"type":"SCHEMA", "stream":"hello","key_properties":[],"schema":{"type":"object", "properties":{"value":{"type":"string"}}}}\n{"type":"RECORD","stream":"hello","schema":"hello","record":{"value":"world"}}\n'

This writes the datapoint {"value":"world"} to the hello stream along with a schema indicating that value is a string. That data can be piped into any Target, like the [Google Sheets Target], over stdin:

printf '{"type":"SCHEMA", "stream":"hello","key_properties":[],"schema":{"type":"object", "properties":{"value":{"type":"string"}}}}\n{"type":"RECORD","stream":"hello","schema":"hello","record":{"value":"world"}}\n' | target-gsheet -c config.json

A Python Tap

To move beyond Hello, world you'll need a real programming language. Although any language will do, we have built a Python library to help you get up and running quickly.

Let's write a Tap called tap_ip.py that retrieves the current IP using icanhazip.com, and writes that data with a timestamp.

First, install the Singer helper library with pip:

pip install singer-python

Then, open up a new file called tap_ip.py in your favorite editor.

import singer
import urllib.request
from datetime import datetime, timezone

We'll use the datetime module to get the current timestamp, the singer module to write data to stdout in the correct format, and the urllib.request module to make a request to icanhazip.com.

now = datetime.now(timezone.utc).isoformat()
schema = {
    'properties':   {
        'ip': {'type': 'string'},
        'timestamp': {'type': 'string', 'format': 'date-time'},
    },
}

This sets up some of the data we'll need - the current time, and the schema of the data we'll be writing to the stream formatted as a JSON Schema.

with urllib.request.urlopen('http://icanhazip.com') as response:
    ip = response.read().decode('utf-8').strip()
    singer.write_schema('my_ip', schema, 'timestamp')
    singer.write_records('my_ip', [{'timestamp': now, 'ip': ip}])

Finally, we make the HTTP request, parse the response, and then make two calls to the singer library:

  • singer.write_schema which writes the schema of the my_ip stream and defines its primary key
  • singer.write_records to write a record to that stream

We can send this data to Google Sheets by running our new Tap with the Google Sheets Target:

python tap_ip.py | target-gsheet -c config.json

Tap Template

The python-based singer-tap-template provides some skeleton code for a new tap, so you can spend less time on setup and more time coding!

Developing a Target

A Singer Target should read lines from stdin and process Schema, Record, and State Messages.

Guidelines

  • Schema validation should be performed to ensure that the records being written for a stream comform to the schema defined for that stream.
  • Write State messages to stdout once all data that appeared in the stream before the State message has been processed by the Target. Note that although the State message is sent into the target, in most cases the Target's process won't actually store it anywhere or do anything with it other than repeat it back to stdout.
  • Targets should be able to handle records with nested fields. Depending on the type of Target, the record may need to be flattened (de-nested). See target-csv's flatten() function for an example.

Target Template

The python-based singer-target-template provides some skeleton code for a new target, so you can spend less time on setup and more time coding!