At 15:38 Uhr -0500 02.11.1999, [EMAIL PROTECTED] wrote:
>1. Are there really big advantages to switching from the old item x
>of line y approach?
Yes, it is faster and you dont have to care about commas or returns
being in your values.
>What happens if you put an array into a field?
Do not do that... you can put elements of an array anywhere, but not
entire arrays. A way to store/build arrays into/from text data is on
the feature request list.
>Can you put a delimiter into an array?
Yes.
>What exactly is an *associative* array as opposed to a plain old array?
Just that you can use other things than numbers as indexes in your array.
>2. I scripted the following to strip out duplicate lines from a list
>field (that users can put lines into, and enter duplicates by
>mistake). It is my first attempt at using lineOffset and seems to
>work OK.
>on mouseUp
> put field runlist into runlister
> repeat with i = 1 to the number of lines of runlister
> repeat until it =0
> get lineOffset (line i of runlister, runlister, i)
> if it is not 0 then delete line it+i of runlister
> end repeat
> put empty into it
> end repeat
> put runlister into field runlist
>end mouseUp
You did not set the wholeMatches to true, so if one line is "old" and
another is "cold", one will be deleted!
>It seems hellish quick to me, but is it efficient?
I would do it this way:
on mouseUp
put empty into outlist
repeat for each line it in fld "runlist"
if it is not among the lines of outlist then
put it&return after outlist
end if
end repeat
put outlist into fld "runlist"
end mouseUp
I guess this is even faster.
>Does it speed things up to put the field into a variable?
Yes. Putting stuff into a field needs reformatting of the field,
which takes time (but not so much in 2.3 any more, thanx Scott!!!
:-))))
>3. ...
>
>However, I would also like to say that part of the reluctance to
>report 'bugs' is that half the time I suspect it is me doing
>something wrong, and anyway there are lots of smarter guys on the
>list who will report it if it is a bug.
It is too far fetched say that everything that you do wrong is a bug
in the software, because it is to complex to do it right. But take
comfort in thinking a bit like that. xTalk is supposed to be the
programming language "for the rest of us"... that is "non
professional programmers". I am a professional programmer (which does
not make me one of those "smarter guys", does it?), but I like
programming in xTalk because, in order to have readable code, I don't
have to write more comments than code like I have to in C++, Java, or
any other language.
>We psychologists have a term for this, but sadly it has slipped my mind.
inferiority complex ;-)))
Regards
R�diger
--------------------------------------------------------------------
| Ruediger zu Dohna GINIT GmbH 0721-96681-63 [EMAIL PROTECTED] |
| PGP-Fingerprint: F1 BF 9D 95 57 26 48 42 FE F8 E8 02 41 1A EE 3E |
--------------------------------------------------------------------
