Newspapers for reMarkable Tablet

Obviously anything you do with these instructions is at your own risk. I haven't bricked my reMarkable yet, but that doesn't mean it can't happen.

Inspired by a reddit thread about "pdflets" and an example implementation.

Setup

1. Create dummy PDF documents for each newspaper. This should be a unique identifier along with .content and .metadata files. Copy these files into /home/root/.local/share/remarkable/xochitl/ on the reMarkable tablet. Example .content File

   {
       "dummyDocument": false,
       "fileType": "pdf",
       "lineHeight": -1,
       "margins": 100,
       "orientation": "portrait",
       "transform": {
           "m11": 1,
           "m12": 0,
           "m13": 0,
           "m21": 0,
           "m22": 1,
           "m23": 0,
           "m31": 0,
           "m32": 0,
           "m33": 1
       }
   }
   

   Example .metadata File

   {
       "deleted": false,
       "lastModified": "1606059555990",
       "parent": "",
       "pinned": false,
       "type": "DocumentType",
       "version": 1,
       "visibleName": "NY_NYT"
   }
   

2. Create a newspapers.txt file to save the newspaper IDs and their corresponding document UUIDs:

   DC_WP,5b7375ae-532b-4d3b-b12c-faba427bbb95
   NY_NYT,bae562b3-4e7e-477a-98ed-1131580c4b34
   WSJ,24ed4b79-886a-4d89-96c9-7897029c9f5f
   KY_LHL,20954231-5b83-4b71-9a71-b05f3fb00944
   

3. Create a newspapers.sh file to pull the latest newspapers:

   #!/bin/bash
   while IFS= read -r line;
   do
       paper=$(echo $line | cut -d"," -f1)
       uuid=$(echo $line | cut -d"," -f2)
       rm -rf /home/root/.local/share/remarkable/xochitl/uuid.thumbnails
       curl -o /home/root/.local/share/remarkable/xochitl/$uuid.pdf https://cdn.newseum.org/dfp/pdf$(date +%-d)/$paper.pdf
   done < /home/root/.local/share/pdflets/newspaper.txt

IMPORTANT wget on the reMarkable does not work with the Newseum website due to SSL issues. Instead, you will need a static prebuilt curl binary for armv7. You can try the binary from https://github.com/moparisthebest/static-curl, but use at your own risk.

4. Copy newspaper.txt and newspaper.sh into /home/root/.local/share/pdflets/* on the reMarkable tablet
5. Create the systemd timer, newspaper.timer:

   [Unit]
   Description=Pull latest paper every day at 7am
   
   [Timer]
   OnCalendar=*-*-* 7:00:00
   Persistent=true
   
   [Install]
   WantedBy=timers.target
   

Make sure your reMarkable has been configured with the correct timezone! I think mine defaulted to UTC.

rm /etc/localtime
ln -s /usr/share/zoneinfo/America/New_York /etc/localtime
date

6. Create the systemd service, newspaper.service:

[Unit]
Description=Retrieve latest papers

[Service]
Type=oneshot
# run pre script to make sure internet is available
ExecStartPre=/bin/sh -c 'until ping -c1 google.com; do sleep 1; done;'
ExecStart=/home/root/.local/share/pdflets/newspaper.sh

[Install]
WantedBy=multi-user.target

7. Copy newspaper.timer and newspaper.service into /etc/systemd/system/ on the reMarkable tablet.
8. Enable the service: systemctl enable --now newspaper.timer. To also pull the latest papers immediately: systemctl start newspaper.service.