I did some programming, since I find this an interesting problem.
Words.dbf contains a memo field, "words", with the words which may have repetitions.
The result is copied into a table "Words2".
In the resulting tables, the words are sorted alphabetically; this is more or less automatic, because of the SELECT...DISTINCT. (On the other hand, since I didn't explicitly sort, there is no guarantee that the result will always sort correctly.)
close database all
erase words2.dbf
create table words2 (words M)
use words in 0
select words
scan
m.words = words
m.words = strtran(words, " ", chr(13) + chr(10))
alines(awords, m.words)
dimension awords(alen(awords), 1)
create cursor temp (word C(100))
append from array awords fields word
select distinct word from temp into cursor temp
select words2
append blank
for i = 1 to reccount("temp")
goto (i) in temp
replace words with alltrim(temp.word) + space(1) additive
next
replace words with alltrim(words)
endscan
Difference in opinions hath cost many millions of lives: for instance, whether flesh be bread, or bread be flesh; whether whistling be a vice or a virtue; whether it be better to kiss a post, or throw it into the fire... (from Gulliver's Travels)
