- craft a prettyLink for
- TwGeoSpatial or TwGeoSpatial
- https://tiddlywiki5-git-geospatial-plugin-jermolene.vercel.app/plugins/tiddlywiki/geospatial/index.html
- linkTo XmmxH
2. TiddlyWiki HTTP Endpoints
The implementation in TiddlyWiki 5 is simplified, without full support for “bags” and “recipes”. The API docs are here:
The API that Eric is using to push data is PUT /recipes/default/tiddlers/{title}:
3. “MultiWikiServer” Pull Request
* fully supporting the “bags and recipes” model to allow data to be partitioned on the server
* improved synchronisation with instant updates
* improved handling of large attachments
* switching from storing data as individual files to using a SQLite database
These changes are the culmination of many years thinking and planning. A big part of the motivation for this work is to be able to undertake the kinds of business opportunities that Hans is working towards.
4. TiddlyWiki as an Azure Web App
There’s very little to it:
* The file package.json is boilerplate that pulls in TiddlyWiki as a dependency
* The folder wiki/ is a standard TiddlyWiki wiki containing the initial tiddlers that the app starts with
* The file `./github/workflows/main_jermotest1.yml` is a GitHub Action that automatically redeploys the app every time there is a commit to this repository. This file was automatically generated by Azure when I set up the Web App, pointing it at this repo
The great thing is that no modifications were required to get TiddlyWiki running on Azure. The capability I mentioned whereby files in the home directory are preserved across restarts gives us a very simple, robust way to deploy as many apps as we need.
5. TiddlyWiki Authentication
TiddlyWIki 5 supports anonymous access and basic authentication. For DataFix projects we’re expecting to use Microsoft’s federated identity provider which enables us to use any private or public Active Directory for authentication. The authentication is done by a proxy layer that is invisible to TiddlyWiki; we just get a trusted header containing the username with every request. I’ve tested it with a dummy Azure-based Active Directory.
6. TiddlyWiki Import Options
TiddlyWIki supports a lot of different ways to get data into it:
* In the browser, interactive users will typically just use drag and drop to import data. Anything can be imported as a tiddler: images, text, movies
* In the browser, wikitext code can pull data over HTTP and import it as tiddlers. For example, the Geospatial plugin includes reusable code to pull images from Flickr
* On the server, data can be imported from the file system with the “load” command and over HTTP via the “fetch” commands
* On the server, there is an HTTP API for external apps to save and retrieve data
* On the server, there is a JavaScript API for custom modules to use to import data
/.