Hi Örjan,
not a beginners's task, I'd say, and less a problem of
Word as a problem of logic. The best forum maybe
microsoft.public.word.vba.general, not because of
Word programming, but because of the linguistic expertise there.
First you would need an exhaustive list of all tags.
Which reveals a first problem, as unfortunately e.g. gender "<GEN>"
has different meanings, depending on the preceding language.
From that follows, that the data do not really fit into
a table, as they are not 2-dimensional.
Let's say the expression1 is equal "Abwinkelung".
Then there is English(expression1) and
possibly Spanish(expression1), but there is, too,
Gender(Spanish(expression1)), which makes it 3-dimensional.
For a 2-dimensional represention you would need a list of
all possible genders in all possible languages.
And "<USE>" maybe dependant of the gender of the language
of the expression...
Then there is the complication, that the records may contain
different numbers of fields. Pretty bad. Must be possible
to overcome, but would be very complicated, see below "recursion".
As a start, if you really want to go through that,
I'd fill up all records with all possible fields,
which would determine the number columns.
Difficult enough.
Very off topic, but a hobby of mine:
Otherwise, you'll run into problems with recursion
similar to the "travelling salesman" problem.
Theoretically solvable, but only if there is no time limit.
In that context, I've learned a new expression:
computation time in CPU-years.
Greetings from Bavaria, Germany
Helmut Weber, MVP
"red.sys" & chr(64) & "t-online.de"
Word XP, Win 98
http://word.mvps.org/
?rjan Skogl?sa - 26 Jan 2005 08:18 GMT
Hi Helmut,
thanks for your input. I see it was more difficult than I thought.
As for now, I "solved" it by discarding every information but the
direct term pair. (I deleted everything but the first german and the
first english pair and used that as a simple term listing.)
Next time, according to your advice I first decide whether I have the
time or not.
If yes, I´d go for that exhaustive 2-dimensional list of tags (perhaps
manually collected) and then compare every tag to that list until I
get a match, insert the tag's value in the corresponding table column
and then match the next tag to the _next_ item in the list. That
should work, shouldn´t it? (At least if the the tags always come in
the same order.) But probably it would be better to do this in Excel?
Well, I´ll leave this for now.
Schöne Grüße aus Schonen.
Örjan
>Hi Örjan,
>
[quoted text clipped - 40 lines]
>Word XP, Win 98
>http://word.mvps.org/