Skip to content

Latest commit

 

History

History
114 lines (85 loc) · 4.19 KB

README.md

File metadata and controls

114 lines (85 loc) · 4.19 KB
Raw

Repocket

Repocket is a (very, very raw) tool to manage a local copy of articles from GetPocket. Because, you know, The Cloud is not a reliable storage. Also, grep. Local copies contain a header in plain YAML with the metadata provided by Pocket

Repocket maintains a simple folder structure to represent states. Articles are rendered as text files into two folders for unread and favourite articles. You can move articles to a deleted and archived folders to sync those statuses back to Pocket.

Caveat emptor

  • Local copies are plain text, rendered with w3m, so expect images to be lost. I'm generally OK with this, and the original URLs are still there.
  • This workflow is not meant to be particularly user friendly, it just fits me well. I use the "favourite" state to mark articles that I want to keep a copy for.
  • It works in Linux and OSX, and will very likely not in Windows (although it should be trivial to fix.)

Building

Go >= 1.11.

You MUST have w3m installed. Repocket uses w3m to render the articles into plain text.

Configuration

You must create the config file before running Repocket. It's a very simple yaml stored at ~/.config/repocket/config with these contents:

consumer_key: 85480-9793dd8ed508561cb941d987
favs_dir: <absolute_path_dir_for_favourited_articles>
unread_dir: <absolute_path_dir_for_unread_articles>
deleted_dir: <absolute_path_dir_for_deleted_articles>
archived_dir: <absolute_path_dir_for_archived_articles>
  • The consumer_key indicates the GetPocket Application. You can leave the sample value above (my own app), but if you prefer to use your own just create a new application.

The rest are directories to store articles. All expect an absolute path. Two of these are synced up and downstream:

  • unread_dir is the directory where unread, non archived articles will be downloaded. If a downloaded article is marked as archived in Pocket, then the file will be deleted.
  • favs_dir contains favourited articles. You move articles here from the unread_dir, in the next sync they will be marked as favourite and archived. Articles in your local fav directory are never deleted (even if you unfav them upstream.)

These directories only sync upstream:

  • deleted_dir contains articles that should be deleted in the next run of the delete command. You move articles here when you want to delete them from Pocket.
  • archived_dir contains articles that should be archived in the next run of the archive command. You move articles here when you want to archive them in Pocket, but don't want to keep a local copy.

Files in both deleted_dir and archived_dir are removed after a sync.

Set up

When you first run Repocket, it will authenticate against the Pocket API. It will ask you to browse to a URL where you can grant permissions to read your list n articles. The message looks something like this:

2019/09/09 20:40:12 Browse to this URL, you may ignore errors:
https://getpocket.com/auth/authorize?request_token=62074b8c-ed8a-b5e5-71f3-586bcf&redirect_uri=localhost

Click on the link and accept the authorisation. Once you do this the first time, you simply need to click the link and ignore the browser.

This step will write a new access_token property to your config file so you don't need to auth again.

Synchronizing

Run

GO111MODULE=on go run ./cmd/repocket [favs|delete|archive|unread|sync]

And Repocket will sync the folder associated to the given action.

  • delete: will delete articles added to favs_dir as deleted in Pocket.
  • archive: will delete articles added to favs_dir as deleted in Pocket.
  • favs: will mark articles added to favs_dir as favourited in Pocket.
  • unread: will download all unread articles to unread_dir.
  • sync: will execute the previous three actions, in the same order as shown in this list.

TODO

Some things I'd like to implement:

  • Prepend the source URL to the file.