1. Weird error messages & more
Robert, could you in future versions document every error message possible,
since I have no clue how to fix errors such as:
-- "unable to write error message file (ex.err)"
-- "current directory is not available"
etc. And please incorporate dir95.e into the a new release as well,
short-names/long-names, etc. are just two messy.
Why should I need to check the OS (ex/exw), just because I want to have the
SHORT name or the LONG name of a file.
The one gives me this, the other that, and using dir95.e I end up using
interrupts .. hmm, that's weird those aren't implemented
in the Exw interpreter, why ?
While we at fixes/suggestions ... here's again a suggestion I made some time
ago...
The 'each' keyword, executes the current statement, for every element of the
sequence that follows.
Example:
printf (1, "The next number is %2d", {each s})
Is equilivent to the current:
for index = 1 to length(s) do
printf (1, "The next number is %2d", {s[index]})
end for
But for readability, maintainability, speed, etc. the first is much cleaner.
And in complex loops, this can make it much more readable.
puts (each fhandle_list, each text_image)
Will output to each filehandle in fhandle_list, every line in text_image.
It is equilivent (some one tell me how to spell this word, my spell checker
claims it doesn't exist) to..
for index1 = 1 to length(fhandle_list) do
for index2 = 1 to length(text_image) do
puts (fhandle_list[index1], text_image[index2])
end for
end for
Or look at this:
s = repeat (0, 10)
each s = getc (fhandle)
Equilivent to:
s = repeat (0,10)
for index = 1 to length(s) do
s[index] = getc (fhandle)
end for
But the real advantage comes when used in conjunction with functions....
Introduce the new keyword 'reply' ...
It means the same as return, except unlike return after processing the value,
the function might continue at that point, if
another value if desired.
Look at the new count function ()
function count ()
integer index
index = 1
while 1 do
reply index
index += 1
end while
end function
s = repeat (0, 10)
each s = count ()
-- s now contains: { 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 }
-- If the above could also have been written as:
s = each count ()
-- it would end up in an infinite loop.
integer i
i = count ()
-- i is now set to 1
For example, what if the built-in find () and match () were written this way
(they would still be backwards
compatible), however for those programs that need it they could use a list
of all places found, rather that just the first
place
it is found.
Example:
sequence s
s = "Hello world, what's up ?"
printf (1, "A space is found at position: %d", {each find(' ', s)})
And like many others, I'm still awaiting the concatonated assignment...
{ succes_value, return_value } = get (fhandle)
Etc.
Ralf Nieuwenhuijsen
.... Mailto://nieuwen at xs4all.nl
.... Http://www.xs4all.nl/~nieuwen
.... Uin://9389920
2. Re: Weird error messages & more
- Posted by Robert Craig <rds at ATTCANADA.NET>
Apr 12, 1999
-
Last edited Apr 13, 1999
Ralf writes:
< a bunch of suggestions>
Thanks. I'll add them to the wish list.
> "unable to write error message file (ex.err)"
Maybe ex.err existed already and was read-only, or
maybe Euphoria was trying to store ex.err on a floppy that
had write protection.
Regards,
Rob Craig
Rapid Deployment Software
http://members.aol.com/FilesEu/
3. Re: Weird error messages & more
Ralf Nieuwenhuijsen wrote:
>While we at fixes/suggestions ... here's again a suggestion I made
>some time ago...
>
>The 'each' keyword, executes the current statement, for every element
>of the sequence that follows.
>
>Example:
>
> printf (1, "The next number is %2d", {each s})
>
>Is equilivent to the current:
>
> for index = 1 to length(s) do
> printf (1, "The next number is %2d", {s[index]})
> end for
>
>But for readability, maintainability, speed, etc. the first is much
>cleaner.
I disagree. There's nothing wrong with the readability, maintainability,
etc. of the second example. In fact, the first example could be *horribly*
confusing, since the loop operation -- instead of being obvious -- is
"hidden" in that little "each" keyword, lurking way out there on the end.
[Not to mention the serious logistical problems the interpreter would have
to face, trying to figure out where the programmer wants the "each" loop to
apply to within multiple nested operations. Are you trying to give Rob
nightmares? :) ]
>And in complex loops, this can make it much more readable.
>
> puts (each fhandle_list, each text_image)
>
>Will output to each filehandle in fhandle_list, every line in
>text_image.
>It is equilivent (some one tell me how to spell this word, my spell
>checker claims it doesn't exist) to..
>
> for index1 = 1 to length(fhandle_list) do
> for index2 = 1 to length(text_image) do
> puts (fhandle_list[index1], text_image[index2])
> end for
> end for
Again, I find example 2 far more "readable" than example 1, since you can
clearly see what's going on. Yes, you have to type a little more for example
2, but the benefits of the obvious loops far outweigh any possible
disadvantages resulting from the extra typing, IMO.
[BTW, the word you're looking for is "equivalent." :) ]
>Or look at this:
>
> s = repeat (0, 10)
> each s = getc (fhandle)
>
>Equilivent to:
>
> s = repeat (0,10)
> for index = 1 to length(s) do
> s[index] = getc (fhandle)
> end for
In Euphoria 2.1, you can include get.e and code this:
s = get_bytes(fhandle, 10)
Which is much cleaner than either of the above examples.
>But the real advantage comes when used in conjunction with
>functions....
>Introduce the new keyword 'reply' ...
>It means the same as return, except unlike return after processing
>the value, the function might continue at that point, if
>another value if desired.
>
>Look at the new count function ()
>
> function count ()
> integer index
> index = 1
> while 1 do
> reply index
> index += 1
> end while
> end function
>
> s = repeat (0, 10)
> each s = count ()
> -- s now contains: { 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 }
> -- If the above could also have been written as:
> s = each count ()
> -- it would end up in an infinite loop.
Why not just code it like this:
function count(integer n)
sequence s
s = repeat(0, n)
for i = 1 to n do
s[i] = i
end for
return s
end function
s = count(10)
This is far more readable and maintainable than the "each"/"reply" stuff
would be.
>For example, what if the built-in find () and match () were written
>this way (they would still be backwards compatible), however for
>those programs that need it they could use a list of all places
>found, rather that just the first place it is found.
>
> Example:
>
> sequence s
> s = "Hello world, what's up ?"
>
> printf (1, "A space is found at position: %d", {each find(' ', s)})
You'll get an infinite loop, since "find(' ', s)" will always return 6, the
position of the first space in s.
Again, why not just make use of how Euphoria actually works, and code it
like this:
function find_all(object value, sequence s)
object not_value
sequence list
integer f
not_value = not value
list = {}
f = find(value, s)
while f do
list = append(list, f) -- append to position list
s[f] = not_value -- so we don't find this one again
f = find(value, s) -- find the next one
end while
return list
end function
sequence s, pos
s = "Hello world, what's up ?"
pos = find_all(' ', s)
for i = 1 to length(pos) do
printf(1, "A space is found at position: %d\n", pos[i])
end for
It's far more obvious what's going on this way.
>And like many others, I'm still awaiting the concatonated
>assignment...
>
> { succes_value, return_value } = get (fhandle)
I sort of agree with you here, but what happens if the function returns an
atom value instead of a sequence? Or a length-3 sequence instead of
length-2? Type-check failure, I suppose. I agree that this would greatly
help to simplify some code, and actually might *not* require the Euphoria
interpreter to be completely rewritten. :)
Be seeing you,
Gabriel Boehme
P.S. Thanks, by the way, for forcing me to come up with the find_all()
function! :)
-------
Freedom's just another word for not caring about the quality of your work.
from a "Dilbert" cartoon by Scott Adams
-------
