<feed xmlns='http://www.w3.org/2005/Atom'>
<title>mp3_player/data/catalog.qcat, branch main</title>
<subtitle>Unnamed repository; edit this file 'description' to name the repository.</subtitle>
<id>https://git.nsrddyn.com/mp3_player/atom/data/catalog.qcat?h=main</id>
<link rel='self' href='https://git.nsrddyn.com/mp3_player/atom/data/catalog.qcat?h=main'/>
<link rel='alternate' type='text/html' href='https://git.nsrddyn.com/mp3_player/'/>
<updated>2026-11-15T18:11:31Z</updated>
<entry>
<title>Replace the json catalog with a custom binary format (qcat)</title>
<updated>2026-11-15T18:11:31Z</updated>
<author>
<name>nasr</name>
<email>nsrddyn@gmail.com</email>
</author>
<published>2026-11-15T18:11:31Z</published>
<link rel='alternate' type='text/html' href='https://git.nsrddyn.com/mp3_player/commit/?id=69441c1ffd55f978eb15f81b23ed595e99eca414'/>
<id>urn:sha1:69441c1ffd55f978eb15f81b23ed595e99eca414</id>
<content type='text'>
- catalog.qcat: 32 byte header, fixed size u32 records, one string blob;
  used in place after loading, names point into the blob, no parsing or copies
- integrity: bounds checks + hardware crc32c (arm crc32cx / sse4.2, software
  fallback); corrupted or truncated files are rejected
- 2.6us per load vs 162us for the api json (62x), 17.5KB vs 112KB
- json is only parsed right after downloading from the api, never from disk;
  old json caches are converted once and deleted
- the embedded offline seed is the qcat itself (aligned, read in place)
</content>
</entry>
</feed>
