OpenEuphoria

Re: sequence and strings (newbie help wanted)

• Posted by jc at cknet.net Mar 15, 2001
• 447 views

Thank you very much Derek, a wonderful reply.
I will try to digest this.

Jim Chapman
----- Original Message -----
From: "Derek Parnell" <ddparnell at bigpond.com>
To: "EUforum" <EUforum at topica.com>
Sent: Wednesday, March 14, 2001 10:25 PM
Subject: RE: sequence and strings (newbie help wanted)




Hi there Jim,
here is an alternative program...
------------------
include get.e
sequence cmd
integer vwait
cmd =mmand_line()
for i =to length(cmd) do
    puts(1, cmd[i] & " ")
end for
vwait =it_key()
-------------------

The difference are ...
-- define cmd as a sequence because this is what command_line() returns.
-- define vwait as an integer because this is what wait_key() returns.
-- The command_line() function returns a single sequence, and that sequence
contains a list of sequences - one for each word on the command line plus a
couple of special words.

-- I use a for/end for construct to type out each subsequence contained in
cmd
-- I don't have to define the 'i' because the for automatically does this.
-- The length() function returns the number of sequences inside cmd
-- I use cmd & " " to append a single blank after each typed word.


> I get an error message saying that I have a sequence in a
> character string.

This is because the puts() function expects a special sort of sequence, one
that does NOT have subsequences. The command_line() function returns a
sequence that HAS subsequences, so we have to extract them to use in puts().
That's what the for loop does.

   cmd[1] references the first element in cmd
   cmd[2] references the second element in cmd
   etc ...

> In my ex.err I find:
>
> cmd =7'C',58':',92'\',69'E',85'U',80'P',72'H',79'O',82'R',73'I',
> 65'A',92'\',66'B',73'I',78'N',92'\',101'e',120'x',119'w',46'.',101'e',
> 120'x',101'e'},{67'C',58':',92'\',69'E',85'U',80'P',72'H',79'O',82'R',
> 73'I',65'A',92'\',66'B',73'I',78'N',92'\',102'f',105'i',114'r',115's',
> 116't',46'.',101'e',120'x',119'w'}}

This show the sequence (and subsequences) as lists of numbers enclosed in {}
brackets.

A string is a special sequence type in Euphoria. It is simply a sequence
that has no subsequences (and usually each element is an integer in the
range 0 - 255). Even though its special, there is no special handling by
Euphoria - its only special to the functions which expect it.

Now some general stuff.

A sequence is a list of zero or more objects. An object is either a sequence
or an atom. A atom is a number. There is a special type of atom called an
integer that can only contain a 31-bit whole number. Atoms can hold floating
point type numbers as well.

These are the only data types native to Euphoria. Anything else (such as a
string) is not native and must be handled in your (or someones) Euphoria
code.

Euphoria provides a little bit of help by allowing you to define your own
data types but only allows one method for them, basically a 'is a type of'
method that returns either true or false.

eg.
----------------
type string(object x)
   if atom(x) then
      -- x cannot be an atom
return 0
   end if

   for i = to length(x) do
      -- every element must be an integer
      if not integer(x[i]) then
          return 0
      end if

      -- and have a value from 0 to 255
      if x[1] < 0 or x[i] > 255 then
          return 0
      end if
   end for
   -- if I get to here then 'x' is a true string.
   return 1
end type

string test1, test2
    -- This should work
    test1 =abc"

    -- As should this
    test1 ='a','b','c'}

    -- And this
    test1 =1,2,3}

    -- This should fail with a "type_check failure"
    test2 =1,test1,3}  -- contains a subsequence.

    -- This should fail with a "type_check failure"
    test2 =1, 2.5, 3}  -- contains a non-integer

    -- This should fail with a "type_check failure"
    test2 =1, -2, 3}  -- contains a negative integer

    -- This should fail with a "type_check failure"
    test2 =1, 2, 300}  -- contains a large integer

------------------------

> Also, looking for a word() function, but not finding one in Euphoria. Is
> there a built in function that will parse through command_line() returning
> the Nth word ?

Not really required as command_line() already does this. The first two
elements returned are the interpreter's name that is running and the name of
the euphoria program running.

-----------
cheers
Derek