Tm-http-request from one tiddlyhost to another - possible?

[twMat]

Is it possible to tm-http-request from one tiddlyhost wiki to another tiddlyhost wiki? Has anyone even done it?

…and could an arbitrary tiddlyhost wiki be targeted, i.e without requirements (e.g special installed plugins)?

[linonetwo]

I use tm-http-request recently Use http request message action widget for LLM AI, anyone tried this before?

I think it can fetch whole wiki HTML, but parse the fetched result will be hard. Because you will need to parse script tag that contains the tiddlers json store. I’m not sure we have tool to do this in tw.

I will be a lot easier if you are fetching a nodejs based wiki.

[yedhukrishna]

@linonetwo does it have API authentication?

[saqimtiaz]

A search in the forum for tm-http-request brings up a fair few relevant threads:

• How delay an action? - #5 by saqimtiaz
• Can TiddlyWiki read external tiddlers from URL parameter? - #3 by Springer
• More tm-http-request examples - Discussion - Talk TW (tiddlywiki.org)

[linonetwo]

You can attach bearer token. At least hat works for AI requests.

[twMat]

Guys - thank you - but I should have emphasized more clearly that I’m asking specifically about Tiddlyhost. (I do say it in the headline but I understand if it was missed, my bad)

However, I should clarify one aspect though; Can one tiddlyhosted wiki send a tm-http-request to an arbitrary other tiddlyhosted wiki? I.e without requirements on the targeted wiki.

[I’m editing the OP to reflect these things]

…

@linonetwo regarding there being a tool to “parse the fetched result” - @inmysocks (and I) did create TWederation which uses single file wikis (and works) so maybe this does what you are thinking of.

[saqimtiaz]

@twMat

Guys - thank you - but I should have emphasized more clearly that I’m asking specifically about Tiddlyhost. (I do say it in the headline but I understand if it was missed, my bad)

The first two links I posted are entirely relevant and even have special handling to use the tiddler.json endpoint for TiddlyHost wikis instead of fetching the entire wiki.

See also this demo of those techniques by Springer: My ExternalContent, once loaded, keeps getting saved - Discussion - Talk TW (tiddlywiki.org)

[saqimtiaz]

@linonetwo

I think it can fetch whole wiki HTML, but parse the fetched result will be hard. Because you will need to parse script tag that contains the tiddlers json store. I’m not sure we have tool to do this in tw.

The deserialize[] operator handles this use case: https://tiddlywiki.com/#deserialize%20Operator

[btheado]

@saqimtiaz

• More tm-http-request examples - Discussion - Talk TW (tiddlywiki.org)

Most of those examples are now in the official documentation at https://tiddlywiki.com/#WidgetMessage%3A%20tm-http-request%20Examples

[Springer]

@twMat

without requirements on the targeted wiki.

No special plugins are required. I suspect — but I haven’t tested — that the remote wiki may need to be public (rather than another one of your private wikis). I just tested whether the “allow site to load in iframe” setting needs to be on; apparently not.