Echo JS 0.11.0

I tend to favor url query-string parameters.  This does *NOT* require a server round trip to capture/change.  This is what the history api is for and how modern SPA should work.  Especially for effectively complex query forms.

While it's nice to see these changes and have them working, with a relatively clean DX, I've mostly been using the Tanstack router lately and pretty happy with it... I'm not sure why I'd go back.

It's security through obscurity.. and in a browser context it has only ever served to make some individuals feel safer.  In meaningful terms it does very little and in general, minification alone does almost as good a job.

My only questions are what features of this rely specifically on Node.js that prevent use with Deno, Bun or Browsers?

As well as, the suggestion that with the TS source, it would be trivial to also publish to JSR as well as NPM.

And this is a big part of why I haven't started publishing the MS-SQL client I had been working on, AI assisted...  Even though I have some pretty thorough tests for general functionality across node, deno and bun, I just don't quite trust it yet... let alone I still have a couple refactors I want to do around transactions and the documentation project.

It's hard to actually bulletproof test something that people might rely on.  And my client is mostly a wrapper around ODBC at this point.  I'm planning to do a strip-down version for general ODBC that isn't MS-SQL specific.

https://github.com/tsdrivers/mssql

Harder still are features that I want to target that only work for MS SQL Server on Windows with a Windows client (FILESTREAM).

The main advantages that I wanted are around query ergonomics...

https://tsdrivers.github.io/mssql/guide/queries.html

Just curious, is this completely, or mostly AI generated?

Not that I've never developed something offline/private then done a single public commit before, it's just a little sus these days.

Interesting, and haven't really dug into it.... I've done similar before using other JSON diff libraries.  Also, unsure how this fits in with using the browser's built in history API.

Interesting...  Would be cool to see it adapted for MUI and Mantine component libraries.

Given other existing clients, I'm curious as to the "why" on creating/using this one vs. some of the others.  I am a bit familiar with shortcomings I've experienced from other clients.

Also, a nice to have addition would be an implementation of request/response channels (ie: distributed rpc).  Of course on the latter part, I've mostly moved to using Redis for coordinating that type of work... despite SPOF issues.

Nice improvements on the description and readme, etc... the positioning of the tool is much clearer now.