nx_corpus_measure.nx
buildroot/runtime/nx_corpus_measure.nx
about
nx_corpus_measure.nx -- data-driven MEASURED index over a REAL crawled corpus (the honest "bigger index" step at
laptop scale; the NAS multiplies it). Reads a manifest of natively-crawled pages (nishi_fetch, off-WSL), indexes all
via nx_search_inverted (bounded footprint), MEASURES the real index (docs / unique-terms / postings bytes), and runs
a RANKED multi-term query -> top-k docs by term-match score. Add a manifest line = a bigger index, zero rebuild.
license_tier: ORIGINAL
dependencies 1 imports · 0 importers
imports: nx_search_inverted.nx
imported by: nobody (leaf or entry point)
call flow from main pre-order; caps 40 nodes / depth 6 declared; ↻ = already shown
structs
| none |
consts
| 7 | const CM_MAGIC_131072: i64 = 131072 |
| 8 | const CM_MAGIC_8388608: i64 = 8388608 |
| 10 | const CM_MAXDOC: i64 = 512 |
functions
| 12 | func gw(s: *u8) -> i64 { var n: i64=0; while s[n]!=(0 as u8){n=n+1} sys_write(1,s,n); return 0 } called by 1: main |
| 13 | func gn(v: i64) -> i64 { if v==0 { sys_write(1,"0" as *u8,1); return 0 } var m: i64=v; if m<0 { sys_write(1,"-" as *u8,1); m=0-m } let t: *u8=sys_mmap(24); var k: i64=0; while m>0 { t[k]=(48+(m%10)) as u8; m=m/10; k=k+1 } let o: *u8=sys_mmap(24); var w: i64=0; var q: i64=k-1; while q>=0 { o[w]=t[q]; w=w+1; q=q-1 } sys_write(1,o,w); return 0 } called by 1: main |
| 14 | func cm_slen(s: *u8) -> i64 { var n: i64=0; while s[n]!=(0 as u8){n=n+1} return n } |
| 17 | func cm_has(idx: *NxInvIndex, term: *u8, tn: i64, docid: i64) -> i64 |
| 26 | func main() -> i64 |