#! /usr/bin/gawk -f # Last edited on 2019-03-29 10:52:33 by jstolfi # Processes HTML indices of papers that replace multitab windows. # Looks for lines of the form "http...", turns them into
  • ..
  • # If the line has "@@", the right half is used as anchor. # Otherwise the href itself is used. # Any other lines are passed through unchanged. /^[ ]*http/ { nf = split($0, fld, /[@][@]/); if (nf == 1) { fld[2] = fld[1]; nf = 2; } if (nf != 2) { printf "** bug «%s»\n", $0 > "/dev/stderr"; print; next; } for (i = 1; i <= nf; i++) { gsub(/^[ ]+/, "", fld[i]); gsub(/[ ]+$/, "", fld[i]); } printf "
  • %s
  • \n", fld[1], fld[2]; next; } // { print; next; }