Pipeliner

Pipeliner can do several anoying repetitive tasks for you. For example imagine you want to get an email when a certain blog publishes a new article. Or you are loving husband/wife and you want send some nice messages to your wife/husband repetitively. Or imagine you must fill some online forms every month.

Pipeliner runs multiple pipelines. Each pipeline is a list of steps that are executed in a row. Pipelines are defined in configuration file.

Step

Each step gets input data which is modified and then passed to the next step. Step instance is created by using a given class name in class field. Constructor parameters are specified in params field. If a step fails, pipeliner will repeat the step up to 3 times. If step fails after 3 retries, pipeliner will try to run whole pipeline again.

Configuration

Pipeliner uses json configuration which defines pipelines to be run. Each pipeline consists of multiple steps which are performed one by one. Pipelines can be scheduled by providing crontab-like format schedule in schedule field.

Example configuration:

{
  "custom_steps": "../custom_steps/",
  "pipelines": [
    {
      "name": "Notify me when new blog post is published",
      "schedule": "* * * * *",
      "steps": [
        {
          "class": "HttpDownload",
          "params": {
            "url": "http://www.example.com/",
            "headers": {}
          }
        },
        {
          "class": "GetHtmlElementText",
          "params": {
            "element_xpath": "(//*[@class=\"post-title\"])[1]/a"
          }
        },
        {
          "class": "CompareWithPrevious",
          "params": {
            "when_same": {
              "class": "DoNothing"
            },
            "when_different": {
              "class": "SendEmailSsl",
              "params": {
                "smtp_host": "smtp.doe.com",
                "smtp_port": 465,
                "login": "john@doe.com",
                "password": "$3cr37",
                "from_email": "john@doe.com",
                "to_emails": ["sally@doe.com"],
                "subject": "Something important happened!"
              }
            }
          }
        }
      ]
    }
  ]
}

By providing this configuration, pipeliner will download http://www.example.com/. Downloaded content is then passed into the next step which finds html element by (//*[@class=\"post-title\"])[1]/a XPath and passes content of the element into next step where the element content is compared with previous version from previous run. Lastly, email is sent to john@doe.com if the element content is different.

Make your wife/husband happy (or angry)

Example configuration to make your wife/husband happy. Or angry if she/he knows you.

{
  "custom_steps": "../custom_steps/",
  "pipelines": [
    {
      "name": "Make my wife happy",
      "schedule": "*/15 9-14 * * 1-5",
      "steps": [
        {
          "class": "PickRandomText",
          "params": {
            "choices": [
              "I love you my sweetheart",
              "You are the best! :-*",
              "Have a great day honey <3",
              "Hi <3",
              "What would you like me to cook today? :)"
            ]
          }
        },
        {
          "class": "SendMessageFb",
          "params": {
            "login": "john@doe.com",
            "password": "s3cr37pa$$",
            "to_user_name": "Alison Doe"
          }
        }
      ]
    }
  ]
}

How to run

1. Install requirements: pip install -r requirements.txt
2. Create a configuration file
3. Run pipeliner python -m pipeliner /etc/pipeliner/my_config.json

Custom steps

Pipeliner can use custom steps. Just add path to a package with custom steps and use their class names in the configuration file.

{
  "custom_steps": "../custom_steps/",
  "pipelines": [...]
}

Contribution

Feel free to add your step implementations and share with me your use-cases 😜