<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Parser on ./Code</title><link>https://blog.ouankou.com/tags/parser/</link><description>Recent content in Parser on ./Code</description><generator>Hugo</generator><language>en-US</language><copyright>© Anjia Wang</copyright><lastBuildDate>Wed, 15 Apr 2026 16:01:55 -0700</lastBuildDate><atom:link href="https://blog.ouankou.com/tags/parser/index.xml" rel="self" type="application/rss+xml"/><item><title>How `ompparser` Turns OpenMP Pragmas Into `OpenMPIR` In REX</title><link>https://blog.ouankou.com/2026/04/02/how-ompparser-turns-openmp-pragmas-into-openmpir-in-rex/</link><pubDate>Thu, 02 Apr 2026 00:00:00 +0000</pubDate><guid>https://blog.ouankou.com/2026/04/02/how-ompparser-turns-openmp-pragmas-into-openmpir-in-rex/</guid><description>&lt;p&gt;The previous posts in this series covered two frontend boundaries around this stage:&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;how REX preserves OpenMP pragmas instead of letting Clang swallow them,&lt;/li&gt;
&lt;li&gt;and how Stage 2 converts &lt;code&gt;OpenMPIR&lt;/code&gt; into &lt;code&gt;SgOmp*&lt;/code&gt; nodes in the Sage AST.&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;What was still missing was the parser itself.&lt;/p&gt;
&lt;p&gt;That missing step matters because it is where directive text stops being “a pragma string we carried forward” and becomes a compiler data structure that later stages can reason about.&lt;/p&gt;</description></item></channel></rss>