Search Read Messages

[timothymamo]

I'm getting started with gmail-processor and testing out a simple script I'd like to implement.

I'm trying to do a simple search for read messages with an attachment between specific dates. I then want to label these processed messages, with gmail-processed so that the script doesn't process them again. Below is the config I'm using:

function run() {
  const config = {
    description:
      "Use this to get started with Gmail Processor. It stores an attachment to Google Drive.",
    settings: {
      markProcessedMethod: "mark-read",
    },
    global: {
      thread: {
        match: {
          query:
            'has:attachment -in:trash -in:drafts -in:spam -label:gmail-processed is:read after:${date.now:date:startOfQuarter-1d:yyyy-MM-dd} before:${date.now:date:endOfQuarter+1d:yyyy-MM-dd} subject:"^*word*$"',
        },
      },
    },
    threads: [
      {
        match: {
          query: 'to:[email protected] is:read',
        },
        messages: [
          {
            match: {
              query: 'is:read',
            },
            attachments: [
              {
                match: {
                  name: "^.*\.pdf$",
                },
                actions: [
                  {
                    name: "attachment.store",
                    args: {
                      location:
                        "/GmailProcessor-Tests/e2e/simple/${message.date:date::yyyy-MM-dd}/${message.subject}-${attachment.name}",
                      conflictStrategy: "keep",
                    },
                  },
                ],
              },
            ],
          },
        ],
      },
    ],
  }

  return GmailProcessorLib.run(config, "dry-run")
}

After running the script, the process does find the email I'm looking for but outputs the following DEBUG message:

1:28:46 PM Info 2024-07-31T11:28:46.931Z TRACE [ThreadProcessor.processEntity()] Processing thread id '123...456' (lastMessageDate:2024-07-31T08:26:36.000Z, subject:'word') started ...
1:28:46 PM Info 2024-07-31T11:28:46.932Z TRACE [MessageProcessor.processConfigs()] Processing message configs started ...
1:28:46 PM Info 2024-07-31T11:28:46.934Z TRACE [MessageProcessor.processConfig()] Processing message config '0' started ...
1:28:47 PM Info 2024-07-31T11:28:47.047Z DEBUG NO MATCH: message is read
1:28:47 PM Info 2024-07-31T11:28:47.053Z INFO Skipping non-matching message id '123...456' (date:'2024-07-31T08:26:36.000Z',  subject:'word', from:"email>).

Meaning the subsequent actions are not performed.

I've tested the above, without the is:read option, by setting the email to being unread and works perfectly, both in dry-run and in safe-mode.

What am I missing? Is this possible?

P.S. For further context I want to run this script every quarter, only, for accounting purposes, meaning I'll be reading the message when I receive it (mainly to check if Its an invoice I need to pay or has been paid).

[ahochsteger]

@timothymamo the reason is that the config uses mark-read as the markProcessedMethod which implies that certain matches are automatically added to the configuration, since this is normally what a typical user might expect.

You may overcome this limitation if you use custom instead, but then you have to add the action to mark processed messages yourself.

See here for the documentation of the available options:
https://ahochsteger.github.io/gmail-processor/docs/reference/enum-types#markprocessedmethod

[timothymamo]

Ahhhh thanks for the quick reply @ahochsteger