<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Idempotency on 每日拍拍</title>
    <link>https://dailypypy.org/tags/idempotency/</link>
    <description>Recent content in Idempotency on 每日拍拍</description>
    <generator>Hugo -- gohugo.io</generator>
    <language>zh-tw</language>
    <copyright>© 2026 每日拍拍</copyright>
    <lastBuildDate>Fri, 09 Oct 2026 10:57:00 +0800</lastBuildDate><atom:link href="https://dailypypy.org/tags/idempotency/index.xml" rel="self" type="application/rss+xml" />
    <follow_challenge>
      <feedId>155076163427069952</feedId>
      <userId>154825760438254592</userId>
    </follow_challenge>
    
    
    <item>
      <title>DuckDB MERGE 實戰：增量匯入、去重與可重跑 ETL</title>
      <link>https://dailypypy.org/learn/duckdb-merge-idempotent-etl/</link>
      <pubDate>Fri, 09 Oct 2026 10:57:00 +0800</pubDate>
      
      <guid>https://dailypypy.org/learn/duckdb-merge-idempotent-etl/</guid>
      <description>&lt;!---
1440x768
prompt: masterpiece, best quality, highres, clean anime illustration, japanese anime style, soft shading, flat color design, 1girl, black hair, green eyes, white off-shoulder shirt, black short skirt, slight dutch angle, upper body, standing with one hand on chin thinking and the other arm folded, focused confident smile, looking at viewer, pastel apricot and powder blue background, subtle translucent data rows converging into one orderly stack without symbols, logos, markings, numbers, letters, or text, neat composition, detailed eyes, cute and smart vibe, minimal background, polished illustration, no text
negative prompt: worst quality, bad eye, bad hand, extra limbs, manga, multiple views, monochrome, text, signature
dedup note: Planned 2026-10-09 topic `duckdb-merge-idempotent-etl` passed a strict 185-post inventory and current-source scan. ../python-duckdb/ teaches general local CSV/Parquet querying; ../duckdb-window-functions/ teaches analytical ranking and frames; ../duckdb-asof-joins-time-series/ teaches nearest-time joins; ../duckdb-remote-parquet-s3/ teaches object-storage scans and pushdown. This focused follow-up instead centers MERGE INTO, deterministic source deduplication, incremental batches, insert/update/delete rules, transaction-scoped audit counts, rerun safety, and loss-resistant tests.
source check: Reviewed against the current official DuckDB MERGE INTO, Python API, transaction, and INSERT documentation on 2026-10-09. The article targets DuckDB 1.5.5 and uses explicit source normalization, deterministic tie-breaking, RETURNING merge_action, and a single transaction for data plus audit metadata.
---&gt;
&lt;p&gt;





&lt;figure&gt;
    &lt;img class=&#34;my-0 rounded-md&#34; loading=&#34;lazy&#34; alt=&#34;featured&#34; src=&#34;./featured.png&#34; /&gt;

  
&lt;/figure&gt;
&lt;/p&gt;</description>
      <media:content xmlns:media="http://search.yahoo.com/mrss/" url="https://dailypypy.org/learn/duckdb-merge-idempotent-etl/featured.png" />
    </item>
    
  </channel>
</rss>
