OpenEuphoria

2. Re: Basic Euphoria Sequences

• Posted by don cole <doncole at pacbell.net> Feb 10, 2006
• 454 views

Some Slacker wrote:
> 
> Ok, I know, I'm a moron, but I really need a helping hand making
> new sequences, specifically...
> the number of elements in each sequence is a set of 14 (I think)
> sequences 2 of these sequences are themselves sequences but of 
> an unknown length perhaps one element but maybe more, and I need
> each sequence to be able to be looked at like sequence[1][3] for 
> obviously the third element of the first record... I'm not really 
> sure how to set this sequence in motion
> }}}
<eucode>
> NewRecord=repeat(0,14)
> -- this will make a 14 element sequence of 0's I know that, but how
> -- then do I make say elements 5 and 6 themselves sequences (with their
> -- own list of records) sequence[1][5][1..or more]
> </eucode>
{{{

> 
> I know this is a basic question, and I probably already know the 
> answer, but until I see it...   heh.
> 
> Thanks in advance,
> Some Slacker


I'm not sure what your trying to do but I would think:

</eucode>
{{{

--Writing: 
         seq[1][1]=atom_data
         seq[1][5]=seq_data

--Reading:

   for x=1 to 14 do
     if atom(seq[1][x]) then
        treat_as_atom()
    elsif sequence(seq[1][x]) then
        treat_as_sequence()
    end for

--or

   for x=1 to 14 do
     if length(seq[1][x])=1 then
        treat_as_atom()
    elsif length(seq[1][x])>1 then
        treat_as_sequence()
    end for

Hope this helps,

Don Cole
 A Bug is an un-documented feature.
A Feature is a documented Bug.

4. Re: Basic Euphoria Sequences

• Posted by don cole <doncole at pacbell.net> Feb 10, 2006
• 467 views

Some Slacker wrote:
> 
> don cole wrote:
> > I'm not sure what your trying to do but I would think:
> > 
> > </eucode>
{{{

> > --Writing: 
> >          seq[1][1]=atom_data
> >          seq[1][5]=seq_data
> > 
> > --Reading:
> > 
> >    for x=1 to 14 do
> >      if atom(seq[1][x]) then
> >         treat_as_atom()
> >     elsif sequence(seq[1][x]) then
> >         treat_as_sequence()
> >     end for
> > 
> > --or
> > 
> >    for x=1 to 14 do
> >      if length(seq[1][x])=1 then
> >         treat_as_atom()
> >     elsif length(seq[1][x])>1 then
> >         treat_as_sequence()
> >     end for
> > 
> > Hope this helps,
> > 
> > Don Cole
> 
> This is somewhat helpful, but I'm still unsure how to best start 
> this sequence. To be more specific the 14 elements will all be sequences
> for things like timestamps, strings, numbers, more strings, more numbers
> etc.
> 
> now for some of the strings i need to be able to seperate out the strings
> to be a list(for a listview, or a combo box for example) so that some
> of the elements of the sequence are single strings or just atoms, or 
> formatted for time or currency etc. and some of the elements are themselves
> their own list of strings...
> 
> if the data could look something like...
> 
> {{{time},{time2},{time3},{time4},{string},{string2},{atom},{atom (formatted
>   for currency)}, {atom (again currency)}, {string3} etc. }}
> 
> now what if the strings are lists for a combo box, can i just add elements
> to each of these strings?
> 
>
> {{{time},{time2},{time3},{time4},{{string1},{string1a}},{{string2},{string2a}},
> {atom},{atom (formatted for currency)}, {atom (again currency)}, 
> {string3} etc. }}
> 
> I hope this makes my problem just a bit clearer, I know I'm unclear, 
> but I don't know what it is I want, I just know I want it!!! LOL!
> 
> thanks again, 
> Some (Serious) Slacker

Hello Some Slacker,

I have similar thing I am doing to keep track of Movie inventory.

rec[1]=movie_num
rec[2]-title
rec[11]=history
rec[11][1]=date
rec[11][2]=item--purchased ,sold,rented,traded etc.
rec[11]3]=amount
--as you can see record 11 could be of any length. A movie could be bought,
--sold, bought back again ,rented 3 times and then sold again. A long history.

  The best way to keep track of everything is with an index.
  For my index I use movie number.
  Maybe you could use something like transaction number.
  Make that rec[1].

--add a new movie
database=append(database,rec)
index=append(index,rec[1])
--find a movie
a=find(movie_num,index)
title=database[a][2]

So on and so forth.
There are A lot of database programs in the Archives you could look over.
But for your purposes it would be best to construct your own.

Starting off with transaction num being rec[1]
and building on that.

Don Cole
 A Bug is an un-documented feature.
A Feature is a documented Bug.

6. Re: Basic Euphoria Sequences

• Posted by ags <eu at 531pi.co.nz> Feb 10, 2006
• 461 views

Some Slacker wrote:
> 
> Ok, I know, I'm a moron, but I really need a helping hand making
> new sequences, specifically...
> the number of elements in each sequence is a set of 14 (I think)
> sequences 2 of these sequences are themselves sequences but of 
> an unknown length perhaps one element but maybe more, and I need
> each sequence to be able to be looked at like sequence[1][3] for 
> obviously the third element of the first record... I'm not really 
> sure how to set this sequence in motion
> }}}
<eucode>
> NewRecord=repeat(0,14)
> -- this will make a 14 element sequence of 0's I know that, but how
> -- then do I make say elements 5 and 6 themselves sequences (with their
> -- own list of records) sequence[1][5][1..or more]
> </eucode>
{{{


This is not a specific answer (as sequences themselves are so versatile) but
some practices I use.  They may or may not suit you.

If I'm using sequences as records, I always define constants for the 'fields'
eg:

sequence records

constant REC_ID = 1,         -- atom
         REC_NAME = 2,       -- string sequence
         REC_DATE = 3,       -- DateTime sequence structure
         REC_DIRENT = 4,     -- output of dir(...)
         REC_PATH = 5        -- string sequence

I then immediately manually define a 'blank' record.  This allows you to have
certain defaults if you wish not to fill out all of them at the time of record
creation.  It also gives you a visual clue of what your records look like:

constant blank_record = {-1, "", {}, {}, ""}

When creating a new record I fill in a copy of the blank record:

sequence newrec
newrec = blank_record
newrec[REC_ID] = get_next_id() --  as an example
newrec[REC_NAME] = file_name
newrec[REC_DATE] = nowDateTime()
newrec[REC_DIRENT] = dir(current_path & "\\" & file_name)
newrec[REC_PATH] = current_path

Then I append the new record to the sequence containing the records:

records = append(records, newrec)

I find this the most reliable in terms of readability (you can see what
is being assigned to which field of a record), flexibility (you can assign
to newrec in any order) and expandability (by changing the REC_* constants
and the blank_record you can alter the structure of the records with minimum
recoding).  Imagine you assign the records like:

sequence newrec
newrec = {get_next_id(),file_name,nowDateTime(),
            dir(current_path & "\\" & file_name), current_path)

And then you realise that storing file_name is not necessary.  You can 
leave it in and ignore it at the expense of memory/complexity, or you can 
modify the constants, the blank_record and any part of the code that refers 
to [REC_NAME] (which is easy to search, rather than going cross-eyed looking
at sequence curly brackets).

In general terms in answer to your question, if you want record[4] to be a
sequence, or sequence of sequences then you just assign a sequence to that
position, eg record[4] = {1, 2, 3, 4} but I would create a function to return
the sequence structure you are after.

Hope this helps
Gary

8. Re: Basic Euphoria Sequences

• Posted by Some Slacker <some_slacker at yahoo.com> Feb 10, 2006
• 474 views

don cole wrote:
> 
> ags wrote:
> 
> > 
> > This is not a specific answer (as sequences themselves are so versatile) but
> > some practices I use.  They may or may not suit you.
> > 
> > If I'm using sequences as records, I always define constants for the
> > 'fields'
> > eg:
> > }}}
<eucode>
> > sequence records
> > 
> > constant REC_ID = 1,         -- atom
> >          REC_NAME = 2,       -- string sequence
> >          REC_DATE = 3,       -- DateTime sequence structure
> >          REC_DIRENT = 4,     -- output of dir(...)
> >          REC_PATH = 5        -- string sequence
> > 
> > </eucode>
{{{

> > 
> > I then immediately manually define a 'blank' record.  This allows you to
> > have
> > certain defaults if you wish not to fill out all of them at the time of
> > record
> > creation.  It also gives you a visual clue of what your records look like:
> > }}}
<eucode>
> > constant blank_record = {-1, "", {}, {}, ""}
> > </eucode>
{{{

> > 
> > When creating a new record I fill in a copy of the blank record:
> > }}}
<eucode>
> > sequence newrec 
> > newrec = blank_record
> > newrec[REC_ID] = get_next_id() --  as an example
> > newrec[REC_NAME] = file_name
> > newrec[REC_DATE] = nowDateTime()
> > newrec[REC_DIRENT] = dir(current_path & "\\" & file_name)
> > newrec[REC_PATH] = current_path
> > </eucode>
{{{

> > Then I append the new record to the sequence containing the records:
> > }}}
<eucode>
> > records = append(records, newrec)
> 
> 
> This I highly approve of. 
> 
> Don Cole
> 
> >
Yeah, I hate to say it but for this code...
"I'm Lovin' It!"