#!/bin/sh # prep.sh FIRST LAST — rasterize pages at 130 dpi for reading + tesseract scaffold text cd "$(dirname "$0")/.."; mkdir -p scans work/ocr pdftoppm -jpeg -r 130 -f "$1" -l "$2" "${ROSS_SRC:-source/10731406.pdf}" scans/hi for p in $(seq "$1" "$2"); do f=$(printf 'scans/hi-%03d.jpg' "$p") tesseract "$f" "work/ocr/p$(printf '%04d' "$p")" -l eng --psm 4 >/dev/null 2>&1 done ls scans/hi-*.jpg | tail -n +1 | wc -l