Andrew Fribush

fig. 2.24.1: no one-year pull reaches the 15,000-row export cap; the largest, 1415, is 12,603 lines, and 1,618 dated rows after deduplication

fig. 2.24.1 · no one-year pull reaches the 15,000-row export cap; the largest, 1415, is 12,603 lines, and 1,618 dated rows after deduplication.
NO ONE-YEAR PULL REACHES THE 15,000 CAP; THE LARGEST IS 12,603 LINESONE BAR IS ONE YEAR: GREY, LINES IN ITS EXPORT FILE; INK, ITS ROWS AFTERDEDUPLICATION, DATED ROWS ONLY.5,00010,00015,000-ROW EXPORT CAP: A PULL THAT REACHED IT WOULD BE SPLIT1415 · 12,603 LINES, THE LARGEST PULL1,618 DATED AFTER DEDUP1431 · 10,4011340138014201460

Grey counts every line of the year’s file as the scraper logged it, header and blank footer included, before deduplication across all 121 files and the first capped export. Ink counts the year’s rows in the deduplicated table with a parsed date; 33,580 rows carry no parsed year and are in no bar.

source: chunks/_pull.log lines 2-122 (lines per year file); medieval_soldiers.db, deduplicated dated rows per year, read-only · pullYears.data.ts

View in context: Breaking a 15,000-row cap → · All figures

Andrew Fribush, software and computer engineer in Austin. I build consequential systems where software meets institutions: finance, healthcare, commerce, government data, and the mechanisms that govern AI agents. Taking new engagements: hello@andrewfribush.com.