Mandrill Template Support

[john-griffin]

Adds the ability to use Mandrill's built in template support. This means template rendering can be fully offloaded to Mandrill. This is done by adding a template function to MandrillHelper that inserts template config into the private params. These params are detected when trying to deliver the email which switches over the API path and inserts params into the request.

Here is an example of how it can be used below. Please note the only new public api is Bamboo.MandrillHelper.template the rest of the config can be done using existing functions.

new_email
|> to("[email protected]")
|> from("[email protected]")
|> Bamboo.MandrillHelper.template("invoice")
|> Bamboo.MandrillHelper.put_param("merge_language", "handlebars")
|> Bamboo.MandrillHelper.put_param("global_merge_vars", [ %{ "name": "reference_number", "content": "ABC001" }])
|> Mailer.deliver_now

Tests are included and this is working against a production Mandrill account. Regular message sending via the API is unaffected by this change.

[paulcsmith]

What do you think of calling this maybe_put_template_params. That way it's clear that it sometimes won't add the params

[paulcsmith]

Left a couple comments. This is really cool! Thanks for submitting the PR

[john-griffin]

@paulcsmith thanks for the awesome feedback! I've incorporated those changes and rebased.

[paulcsmith]

@john-griffin This looks fantastic! Thanks for the PR

[maxim]

Hey folks, there are currently no docs anywhere that provide the snippet at the top of this issue, and if you aren't familiar with all the details of Mandrill, you can spend a bunch of time trying to figure out why the stuff you pass into template_content doesn't get interpolated (hint: it's because you can only use handlebars interpolation with merge_vars and global_merge_vars, but not with template_content, so trying to use template function alone won't work). Just a tiny suggestion that perhaps the snippet at the top of this issue should be added as an example to the function's doc.

[rootkc]

@maxim exactly that just happened to me. Thanks for this comment 😄

[germsvel]

@rootkc if you're up for it, a pull-request to improve the function's documentation would be very welcome!