Using OpenTodo › ReferenceVersion 0.1.0

Capturing pages

Many tasks start as "deal with this page later". OpenTodo can receive a page from three places, and all of them open the same capture screen at /share on your own server:

The capture screen proposes:

• Title: the page title, or the link's host name when there is no title. The title field is a regular quick-add line, so typing tomorrow #work p1 into it sets the date, project and priority.
• Description: the selected or shared text, followed by the link.
• Project: the Inbox by default; pick any project.

Press Enter to save or Esc to cancel. Nothing is saved until you confirm. Saving creates an ordinary task on the device first and syncs it like any other quick-add task, so it works with no network and reaches your other devices when the network returns.

When the capture screen was opened as a popup (bookmarklet, extension) or in a window opened just for the share, it closes itself after Save or Esc. Otherwise it takes you to the Inbox or to the project you picked.

Share sheet (installed app)#

The web app manifest declares a Web Share Target (GET /share with title, text and url). The browser adds OpenTodo to the share menu once the app is installed:

Some apps put the link inside the shared text instead of the link field. The capture screen then takes the last http(s) link from the text and moves it to the end of the description.

After an update that adds or changes the share target, the browser may need the app to be opened once (or reinstalled) before the share menu entry appears.

Bookmarklet#

Settings → Capture shows an Add to OpenTodo button generated for your server's address. Drag it to the bookmarks bar (or use Copy bookmarklet and paste the code as the address of a new bookmark). Clicking it on any page opens a small popup with the capture screen, prefilled with the page's address, title and the text you selected.

The bookmarklet contains no password or token. It only opens your server's /share page in a popup, where your normal sign-in applies. Its source, with your server's address in place of https://todo.example.com:

javascript:(function () {
  var e = encodeURIComponent
  var s = String(window.getSelection ? getSelection() : '').slice(0, 16384)
  var u = "https://todo.example.com/share" + '?url=' + e(location.href) + '&title=' + e(document.title) + (s ? '&text=' + e(s) : '')
  if (!window.open(u, 'opentodo-capture', 'popup,width=480,height=640'))
    alert('Allow pop-ups for this site to capture it to OpenTodo.')
})()

If nothing happens, the browser blocked the popup: allow pop-ups for the site you are on. Some sites block bookmarklets with their own security policy; the extension works there.

Browser extension#

extension/ holds a small Manifest V3 extension for Chromium-based browsers (Chrome, Edge, Brave, Vivaldi) and Firefox 121 or later. It adds:

• a toolbar button Add to OpenTodo,
• a context-menu entry Add to OpenTodo on pages and selections,
• a keyboard shortcut, Alt+Shift+O by default (change it at chrome://extensions/shortcuts or in Firefox's Manage Extension Shortcuts).

Each opens your server's capture screen in a popup window with the current tab's address, title and selection.

Install (it is not published in the browser stores yet):

• Chromium: download opentodo-browser-extension.zip from a release (or run make extension), unzip it, open chrome://extensions, turn on Developer mode, Load unpacked, pick the unzipped folder.
• Firefox: open about:debugging#/runtime/this-firefox, Load Temporary Add-on…, pick manifest.json from the unzipped folder. Temporary add-ons are removed when Firefox restarts; a signed build is a release task.

First use: if no server is set, the extension opens its options page and asks for your server's address (for example https://todo.example.com). Once it is saved, the capture you started continues. Change it later in the extension's options.

Privacy: the extension stores only the server address (in the browser's extension sync storage). It holds no token or password, has no host permissions and no content scripts, and makes no network requests itself; the only traffic is the capture popup you open, which loads your own server. Permissions: activeTab (read the current tab's address, title and selection when you click), contextMenus, scripting (read the selection), storage (the server address).

Signed out#

If the device is not signed in, the capture screen first shows the sign-in form with a note, then continues to the capture screen with the shared page intact. Until you save, the shared title, text and link stay in the page's memory only: they are removed from the address bar and are not written to storage or sent to the server (the server's request log records paths, never query strings). Reloading the sign-in page therefore loses the shared content; share it again.

Limits and safety#

• The title is cut to 500 characters, the text to 16 KB with a [truncated] note.
• A link is kept only if it starts with http:// or https://; anything else (javascript:, data:, file: …) is dropped and the rest of the share is shown normally.
• Shared content is shown as plain text, never as HTML, and the capture screen never opens the shared link.
• There is no auto-save: a link to /share cannot create a task without you pressing Save.
• Files and images cannot be shared yet; the app is offered in the share menu for text and links only.

Edit this page on GitHub