# Makefile to build the file word_map.tbl
# that maps English words to Vietnamese words.
# Last edited on 2026-07-23 10:32:19 by stolfi

OUTTBL := word_map.tbl
MAKEFILE := word_map.make

all: ${OUTTBL}

WORKBIN := ${STOLFIHOME}/projects/voynich/work
BANKBIN := ${STOLFIHOME}/projects/voynich/work

# We obtain the list of English words and symbols from the plaintext
# itself, using the same sample-functions that we used for the plaintext
# sample (engl/wow).

ENGLLANG := engl
ENGLBOOK := wow
ENGLSMP := ${ENGLLANG}/${ENGLBOOK}
ENGLSRC := org/main.wds
ENGLAWK := ../../${ENGLSMP}/sample_fns.gawk
ENGLTBL := ../../${ENGLSMP}/word_map.tbl
ENGLDIC := /tmp/.engl.dic

${ENGLDIC}: ${ENGLSRC} ${MAKEFILE} \
            ${WORKBIN}/wds_to_tlw.gawk \
	    ${ENGLAWK} ${ENGLTBL}
	cat ${ENGLSRC} \
	  | ${WORKBIN}/wds_to_tlw.gawk \
              -f ${ENGLAWK} \
              -v table=${ENGLTBL} \
	      -v lang="${ENGLLANG}" -v book=${ENGLBOOK} -v sec="tot.1" \
	  | gawk '($$1 ~ /[as]/) { print $$3; }' \
          | sort | uniq -c | expand \
          | sort -b -k1,1nr \
          | gawk '/./ { print $$2; }' \
          > ${ENGLDIC}
	wc -l ${ENGLDIC} 1>&2
	head -n 200 ${ENGLDIC} | ${WORKBIN}/format-words-filled 1>&2

# We obtain the list of Vietnamese words (no symbols) from the
# Vietnamese Pentateuch sample (viet/ptt), using the same
# sample-functions that we used for that sample, with compounding so
# that it has about the same number of distinct words as the English
# sample:

VIETLANG := viet
VIETBOOK := ptt
VIETSMP := ${VIETLANG}/${VIETBOOK}
VIETSRC := ../../${VIETSMP}/org/main.wds
VIETAWK := ../../${VIETSMP}/sample_fns.gawk
VIETTBL := ../../${VIETSMP}/word_map.tbl
VIETDIC := /tmp/.viet.dic

${VIETDIC}: ${VIETSRC} ${MAKEFILE} \
            ${WORKBIN}/wds_to_tlw.gawk \
	    ${VIETAWK} ${TBL} \
            enlarge-lexicon
	cat ${VIETSRC} \
	  | ${WORKBIN}/wds_to_tlw.gawk \
              -f ${VIETAWK} \
              -v table=${VIETTBL} \
	      -v lang="${VIETLANG}" -v book=${VIETBOOK} -v sec="tot.1" \
	  | gawk '($$1 == "a") { print $$3; }' \
          | enlarge-lexicon -v num=`cat ${ENGLDIC} | wc -l` \
          | sort | uniq -c | expand \
          | sort -b -k1,1nr \
          | gawk '/./ { print $$2; }' \
          > ${VIETDIC}
	wc -l ${VIETDIC} 1>&2
	head -n 200 ${VIETDIC} | ${WORKBIN}/format-words-filled 1>&2

# Then we feed the two lists to the script make-word-subst-table

${OUTTBL}: ${ENGLSRC} ${VIETSRC} ${MAKEFILE} \
            ${ENGLDIC} \
            ${VIETDIC} \
            make-word-subst-table.gawk
	make-word-subst-table.gawk \
            -v old=${ENGLDIC} \
            -v new=${VIETDIC} \
            -v order=2 \
          > ${OUTTBL}