<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Kotlin on Tushar Choudhary</title><link>https://tushar-c23.github.io/tags/kotlin/</link><description>Recent content in Kotlin on Tushar Choudhary</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Tue, 15 Sep 2026 00:30:00 +0530</lastBuildDate><atom:link href="https://tushar-c23.github.io/tags/kotlin/index.xml" rel="self" type="application/rss+xml"/><item><title>Three stores, two cursors: where an SSE consumer should resume</title><link>https://tushar-c23.github.io/posts/sse-cursor-resume/</link><pubDate>Tue, 15 Sep 2026 00:30:00 +0530</pubDate><guid>https://tushar-c23.github.io/posts/sse-cursor-resume/</guid><description>&lt;p&gt;Picture a service consuming events over Server-Sent Events: a handful of long-lived streams, one owning pod each, running all day. For a consumer like this the interesting code isn&amp;rsquo;t the parsing. It&amp;rsquo;s what happens when the stream breaks, because it will, and the only question that matters then is &lt;em&gt;where do you start reading again&lt;/em&gt;.&lt;/p&gt;
&lt;p&gt;Every SSE guide answers that the same way: &lt;code&gt;Last-Event-ID&lt;/code&gt;. Keep the id of the last event you saw, send it on reconnect, the server resumes after it. That&amp;rsquo;s the protocol&amp;rsquo;s answer and it&amp;rsquo;s a good one.&lt;/p&gt;</description></item></channel></rss>