Scenario: BYO encrypted user token
library(shiny) can_decrypt <- gargle::secret_has_key("GMAILR_DEPLOY_DEMO_KEY") knitr::opts_chunk$set( collapse = TRUE, comment = "#>", eval = can_decrypt )
cat("*Decryption key is unavailable, so no code will be evaluated.*")
The following chunk reads the stored token from file, using the environment variable "GMAILR_DEPLOY_DEMO_KEY"
to decrypt it, and tells gmailr to use it.
library(gmailr) try( gm_auth(token = gm_token_read( ".secrets/gmailr-token.rds", key = "GMAILR_DEPLOY_DEMO_KEY" )) )
We wrap that in try()
to guard against the scenario where the token can be decrypted, but then it can't be refreshed.
If auth has been successful, the gm_profile()
call below reveals details about the account associated with the cached token.
If auth has not been successful, gm_profile()
will throw an error.
Other functions that can be helpful for confirming the validity of a re-loaded token are gargle::token_tokeninfo()
and gargle::token_userinfo()
.
This Shiny document sends email from this account:
(whoami <- gm_profile())
Here's the function that takes the inputs for To:
, Subject:
, and Body:
and sends an email when the user clicks "Send mail":
send_email <- function() { gm_mime() |> gm_to(input$field_to) |> gm_from(whoami$emailAddress) |> gm_subject(input$field_subject) |> gm_text_body(input$field_body) |> gm_send_message() }
textInput("field_to", "To:", placeholder = "jane@example.com") textInput("field_subject", "Subject:", placeholder = "fascinating email subject") textAreaInput("field_body", "Body:", placeholder = "riveting, but short, email body") actionButton("btn", "Send mail") observeEvent(input$btn, { output$log_sent_email <- renderPrint({ cat("Sending email...\n") cat(format(send_email())) }) }) verbatimTextOutput("log_sent_email")
renderPrint({ cat( "To: ", input$field_to, "\nSubject: ", input$field_subject, "\n\n", input$field_body, sep = "" ) })
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.