# Last edited on 2002-01-16 02:59:10 by stolfi
# Sampling functions for latn/ptt
# The "Pentateuch" in Latin (Vulgate), all lowercase.
# To be included in select-evt-lines, fix-raw-words, select-gud-bad-words

function select_evt_line(subsec,chapter,unit,linenum)
{
  # Make each book into a subsection; 
  # Consider only running text (unit type "P"):
  if (unit !~ /^[P]/) { return 0; }
  if (subsec == "gen.1")
    { return (chapter ~ /^[a]/); }
  else if (subsec == "exo.1")
    { return (chapter ~ /^[b]/); }
  else if (subsec == "lev.1")
    { return (chapter ~ /^[c]/); }
  else if (subsec == "num.1")
    { return (chapter ~ /^[d]/); }
  else if (subsec == "deu.1")
    { return (chapter ~ /^[e]/); }
  else  
    { arg_error(("bad subsection \"" subsec "\"")); }
}

function fix_raw_word(word)
{
  # Map upper case to lower case (just in case).
  word = tolower(word);
  return word;
}

function define_patterns()
{
  # No patterns needed 
}

function is_good_word(word)
{ 
  # Accept only lowercase alpha, but no "w" or "j".
  # The text uses neither hyphen nor apostrophe. 
  return (word ~ /^[a-ik-vx-z]+$/);
}