<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: chenhao5721865</title>
    <description>The latest articles on DEV Community by chenhao5721865 (@chenhao5721865).</description>
    <link>https://dev.to/chenhao5721865</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4172297%2Fe73c3170-7a99-4f28-8e76-28616c3bff32.png</url>
      <title>DEV Community: chenhao5721865</title>
      <link>https://dev.to/chenhao5721865</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/chenhao5721865"/>
    <language>en</language>
    <item>
      <title>I built a free Redis plugin for IntelliJ Community, here's what tripped me up</title>
      <dc:creator>chenhao5721865</dc:creator>
      <pubDate>Fri, 09 Oct 2026 02:13:28 +0000</pubDate>
      <link>https://dev.to/chenhao5721865/i-built-a-free-redis-plugin-for-intellij-community-heres-what-tripped-me-up-4p0o</link>
      <guid>https://dev.to/chenhao5721865/i-built-a-free-redis-plugin-for-intellij-community-heres-what-tripped-me-up-4p0o</guid>
      <description>&lt;p&gt;The official Redis tooling in IntelliJ is Ultimate-only, and I use Community. The third-party plugins on the Marketplace either cost money or fall over in exactly the situations I care about: SSH tunnels, non-UTF-8 data, Remote Development.&lt;/p&gt;

&lt;p&gt;So I wrote my own. It's called Redis Navigator, it's on the JetBrains Marketplace now, and it's completely free. No paid tier, nothing locked.&lt;/p&gt;

&lt;p&gt;Rather than another feature list, I want to write down the bugs that cost me the most time. Maybe it saves someone an evening.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2F9bkqqn0pcrf4q0ezfanr.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2F9bkqqn0pcrf4q0ezfanr.png" alt=" " width="800" height="259"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  1. localhost killed the SSH tunnel, and the error blamed Redis
&lt;/h2&gt;

&lt;p&gt;Jedis kept throwing:&lt;/p&gt;

&lt;p&gt;Unexpected end of stream&lt;/p&gt;

&lt;p&gt;I spent way too long checking Redis config, firewalls, TLS. All fine. The actual problem: JSch passes the target host string to the SSH &lt;em&gt;server&lt;/em&gt; to resolve. My Redis host was set to &lt;code&gt;localhost&lt;/code&gt;, the Windows OpenSSH server resolved that to IPv6 &lt;code&gt;::1&lt;/code&gt;, and my Redis (Memurai) only listens on IPv4. So the tunnel came up and died immediately, and Jedis reported a dead connection in the least helpful way possible.&lt;/p&gt;

&lt;p&gt;The fix is to resolve the host to a literal IP before forwarding:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight java"&gt;&lt;code&gt;&lt;span class="nc"&gt;String&lt;/span&gt; &lt;span class="n"&gt;literal&lt;/span&gt;&lt;span class="o"&gt;;&lt;/span&gt;
&lt;span class="k"&gt;try&lt;/span&gt; &lt;span class="o"&gt;{&lt;/span&gt;
    &lt;span class="n"&gt;literal&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="nc"&gt;InetAddress&lt;/span&gt;&lt;span class="o"&gt;.&lt;/span&gt;&lt;span class="na"&gt;getByName&lt;/span&gt;&lt;span class="o"&gt;(&lt;/span&gt;&lt;span class="n"&gt;host&lt;/span&gt;&lt;span class="o"&gt;).&lt;/span&gt;&lt;span class="na"&gt;getHostAddress&lt;/span&gt;&lt;span class="o"&gt;();&lt;/span&gt;
&lt;span class="o"&gt;}&lt;/span&gt; &lt;span class="k"&gt;catch&lt;/span&gt; &lt;span class="o"&gt;(&lt;/span&gt;&lt;span class="nc"&gt;UnknownHostException&lt;/span&gt; &lt;span class="n"&gt;e&lt;/span&gt;&lt;span class="o"&gt;)&lt;/span&gt; &lt;span class="o"&gt;{&lt;/span&gt;
    &lt;span class="n"&gt;literal&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="n"&gt;host&lt;/span&gt;&lt;span class="o"&gt;;&lt;/span&gt; &lt;span class="c1"&gt;// jump host with its own internal DNS, pass through as-is&lt;/span&gt;
&lt;span class="o"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;If you're debugging tunnel issues: build the channel by hand with &lt;code&gt;ssh -N -L&lt;/code&gt;, push a PING through with a raw socket, and only then test your actual client. Splitting the SSH layer from the Redis layer makes the whole thing obvious.&lt;/p&gt;

&lt;h2&gt;
  
  
  2. Editing a String removed its TTL
&lt;/h2&gt;

&lt;p&gt;One of those Redis semantics things: bare &lt;code&gt;SET&lt;/code&gt; clears the TTL. I knew this. I still shipped the bug and only caught it in testing.&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight java"&gt;&lt;code&gt;&lt;span class="n"&gt;jedis&lt;/span&gt;&lt;span class="o"&gt;.&lt;/span&gt;&lt;span class="na"&gt;set&lt;/span&gt;&lt;span class="o"&gt;(&lt;/span&gt;&lt;span class="n"&gt;key&lt;/span&gt;&lt;span class="o"&gt;,&lt;/span&gt; &lt;span class="n"&gt;value&lt;/span&gt;&lt;span class="o"&gt;,&lt;/span&gt; &lt;span class="nc"&gt;SetParams&lt;/span&gt;&lt;span class="o"&gt;.&lt;/span&gt;&lt;span class="na"&gt;setParams&lt;/span&gt;&lt;span class="o"&gt;().&lt;/span&gt;&lt;span class="na"&gt;keepttl&lt;/span&gt;&lt;span class="o"&gt;());&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;h2&gt;
  
  
  3. Pattern search has to drain the SCAN cursor
&lt;/h2&gt;

&lt;p&gt;SCAN is cursor-based and one page is just a partial result. If your pattern search stops after one page, you're silently missing keys. It needs to loop until the cursor wraps back to 0, with some cap in case a pattern goes wild (mine bails at 50k keys). Scroll pagination is only appropriate for plain &lt;code&gt;*&lt;/code&gt; browsing.&lt;/p&gt;

&lt;p&gt;Two Jedis 6 things that got me: &lt;code&gt;Pipeline.sync()&lt;/code&gt; returns void, so batch results need &lt;code&gt;syncAndReturnAll()&lt;/code&gt;. And &lt;code&gt;sscan&lt;/code&gt;/&lt;code&gt;hscan&lt;/code&gt; want the cursor as &lt;code&gt;byte[]&lt;/code&gt; while &lt;code&gt;ScanResult.getCursor()&lt;/code&gt; hands you back a &lt;code&gt;String&lt;/code&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  4. A mojibake bug that was actually a feature request
&lt;/h2&gt;

&lt;p&gt;Testing with Chinese data, everything showed up as garbage. The plugin was fine. The data was fine, technically. I had loaded my test data through a GBK console, so the GBK bytes went straight into Redis and my UTF-8 decoding did exactly what it should: display nonsense.&lt;/p&gt;

&lt;p&gt;Then it clicked: some companies really do have GBK-era data sitting in production Redis. So per-connection charset (UTF-8 / GBK / auto-detect) went from nice-to-have to core feature. GBK data displays correctly and editing in GBK mode round-trips without corrupting anything.&lt;/p&gt;

&lt;p&gt;Practical takeaway: load Chinese test data through a socket or script, never through a console. The other takeaway is to look at your bugs and ask whether they're someone else's daily reality.&lt;/p&gt;

&lt;h2&gt;
  
  
  5. Split Mode testing, blocked by someone else's bug
&lt;/h2&gt;

&lt;p&gt;Remote Development support matters to me, so I tested with &lt;code&gt;runIdeSplitMode&lt;/code&gt; from the IntelliJ Platform Gradle Plugin. The client just hung on "Connecting to remote host..." forever.&lt;/p&gt;

&lt;p&gt;Turned out the plugin had a bug with spaces in the project path: when the backend restarts itself, it goes through a native restarter that drops the Gradle-provided sandbox config, and the new process never starts. My project lives in &lt;code&gt;C:\Project Exploration\...&lt;/code&gt;. Direct hit.&lt;/p&gt;

&lt;p&gt;Workaround was dumb but effective: copy the source to a path without spaces. I filed &lt;a href="https://github.com/JetBrains/intellij-platform-gradle-plugin/issues/2252" rel="noopener noreferrer"&gt;an issue upstream&lt;/a&gt; and JetBrains fixed it within a week, relaunching the backend via exit code instead of the restarter. Once that ships in a release I can delete the workaround.&lt;/p&gt;

&lt;p&gt;The good news is everything in the plugin works in the JetBrains Client: tool window, JSON preview, editing, SSH tunnels, all verified in Split Mode.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fjz8lkahjlcthgbc5yzky.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fjz8lkahjlcthgbc5yzky.png" alt=" " width="800" height="598"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;If this sounds useful, the plugin is here:&lt;/p&gt;

&lt;p&gt;&lt;a href="https://plugins.jetbrains.com/plugin/34398-redis-navigator" rel="noopener noreferrer"&gt;Redis Navigator on JetBrains Marketplace&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Bugs and feature requests go to &lt;a href="https://github.com/chenhao5721865/redis-navigator-issues" rel="noopener noreferrer"&gt;GitHub issues&lt;/a&gt;. It's early days, I read everything and I respond fast.&lt;/p&gt;

</description>
      <category>redis</category>
      <category>java</category>
      <category>intellij</category>
      <category>devtools</category>
    </item>
  </channel>
</rss>
