1. [DOS] Re print sequences
Dan,
There is a potential problem: while {a,b,c} is highly readable, it is not a
legal Euphoria sequence for get(). However {'a','b','c'} is almost as
readable and is a legal sequence for get().
-- Mike Nelson
2. Re: [DOS] Re print sequences
- Posted by Dan B Moyer <DANMOYER at PRODIGY.NET>
Oct 29, 2000
-
Last edited Oct 30, 2000
Mike,
Ok, I'll probably make a differently named variation on the print variation,
so I never use it in a file I might be using a get() on; but his routine
does even better than you might have remembered: it doesn't output either
{a,b,c} or {'a','b','c'} , but rather "abc"! So I changed it to output
{abc}, so I could more easily see the variable's structure, showing
{{abc},{def}}, instead of his {"abc","def"}
But this is a *great* routine, why isn't it included in Euphoria???
Dan
----- Original Message -----
From: "Michael Nelson" <MichaelANelson at WORLDNET.ATT.NET>
To: <EUPHORIA at LISTSERV.MUOHIO.EDU>
Sent: Sunday, October 29, 2000 11:06 PM
Subject: [DOS] Re print sequences
> Dan,
>
> There is a potential problem: while {a,b,c} is highly readable, it is not
a
> legal Euphoria sequence for get(). However {'a','b','c'} is almost as
> readable and is a legal sequence for get().
>
> -- Mike Nelson
3. Re: [DOS] Re print sequences
On Mon, 30 Oct 2000, you wrote:
> Mike,
> [re: Gabriel Boehme's print.e]
> But this is a *great* routine, why isn't it included in Euphoria???
It's also faster, and cuts the size of your data files by 50 - 75%,
depending upon how much of the data is "strings". And the data
is still completely compatible with Euphoria's file input routines.
So, to repeat, why _isn't_ it included in Euphoria??
Irv
4. Re: [DOS] Re print sequences
Irv writes:
>> [re: Gabriel Boehme's print.e]
>
>> But this is a *great* routine, why isn't it included in Euphoria???
>
> It's also faster, and cuts the size of your data files by 50 - 75%,
> depending upon how much of the data is "strings". And the data
> is still completely compatible with Euphoria's file input routines.
>
> So, to repeat, why _isn't_ it included in Euphoria??
If you are serious about saving/restoring large amounts
of Euphoria data, in a compact format, and at high speed, you should
probably be using database.e - the Euphoria Database System.
database.e will be included in Euphoria in the next release.
Whether something is "in" or "out" of Euphoria is a very
subjective decision, and in some ways irrelevant. If you want
print.e to be "in", just add it to your own euphoria\include
directory.
Regards,
Rob Craig
Rapid Deployment Software
http://www.RapidEuphoria.com
5. Re: [DOS] Re print sequences
On Mon, 30 Oct 2000, you wrote:
> Irv writes:
> >> [re: Gabriel Boehme's print.e]
> >
> >> But this is a *great* routine, why isn't it included in Euphoria???
> >
> > It's also faster, and cuts the size of your data files by 50 - 75%,
> > depending upon how much of the data is "strings". And the data
> > is still completely compatible with Euphoria's file input routines.
> >
> > So, to repeat, why _isn't_ it included in Euphoria??
.....
> Whether something is "in" or "out" of Euphoria is a very
> subjective decision, and in some ways irrelevant. If you want
> print.e to be "in", just add it to your own euphoria\include
> directory.
Perhaps there is some logical reason the current method was chosen,
maybe even something not obvious that makes it a better choice for
certain uses. If so, please share.
Regards,
Irv
6. Re: [DOS] Re print sequences
Hi Robert,
Hey now, there's an idea! Can you get Euphoria to support multiple "include"
directories in its search path? Maybe an environment symbol like C uses.
set EULIB=D:\My
----- Original Message -----
From: "Robert Craig" <rds at ATTCANADA.NET>
> just add it to your own euphoria\include directory.
>
------
Derek Parnell
Melbourne, Australia
(Vote [1] The Cheshire Cat for Internet Mascot)
7. Re: [DOS] Re print sequences
- Posted by Robert Craig <rds at ATTCANADA.NET>
Oct 30, 2000
-
Last edited Oct 31, 2000
Derek Parnell writes:
> Hey now, there's an idea! Can you get Euphoria to support
> multiple "include" directories in its search path? Maybe an
> environment symbol like C uses.
This idea has come up before.
I think it's worthy of consideration.
It's on my infamous "list" somewhere.
Regards,
Rob Craig
Rapid Deployment Software
http://www.RapidEuphoria.com
8. Re: [DOS] Re print sequences
- Posted by Robert Craig <rds at ATTCANADA.NET>
Oct 30, 2000
-
Last edited Oct 31, 2000
Irv writes regarding print() vs. Boehme's print.e:
> Perhaps there is some logical reason the current method
> was chosen, maybe even something not obvious that makes
> it a better choice for certain uses. If so, please share.
print() displays any Euphoria object.
It displays all atoms as numbers.
This is the simplest way to display an object.
Boehme's print.e looks for sequences containing
atoms that are all in the range 32 to 255, and it displays
those sequences as strings (whether they are meant to be
strings or not).
Boehme's method is more compact than print(),
which is useful if you are writing out large
amounts of data (as I suspect you are Irv).
EDS uses a more sophisticated approach to making
the data compact, one which has the added benefit
of not losing accuracy when floating-point numbers are
written, and read back in.
print() does not exist just to supply
the input to get(). print() can be used by itself simply
to display objects. There is no way that I would change
print() to use Boehme's method, and with the advent of EDS,
I don't think Boehme's print() is useful enough to enough people
to justify adding it to the standard includes.
Regards,
Rob Craig
Rapid Deployment Software
http://www.RapidEuphoria.com
9. Re: [DOS] Re print sequences
----- Original Message -----
From: "Robert Craig" <rds at ATTCANADA.NET>
To: <EUPHORIA at LISTSERV.MUOHIO.EDU>
Sent: Monday, October 30, 2000 6:53 PM
Subject: Re: [DOS] Re print sequences
> Irv writes regarding print() vs. Boehme's print.e:
> > Perhaps there is some logical reason the current method
> > was chosen, maybe even something not obvious that makes
> > it a better choice for certain uses. If so, please share.
>
> print() displays any Euphoria object.
> It displays all atoms as numbers.
> This is the simplest way to display an object.
But it is *not* the most *helpful*!! If a "newbie" or otherwise inexpert
programmer is trying to use "print" to attempt to debug an improperly
constructed sequence involving textual items, seeing all those numbers does
*not* help them to see what the actual (& possibly complex) structure *is*,
as opposed to what the programmer *wanted* it to be. Seeing *text* in the
output sequence *does* help see the structure, because then the programmer
can compare recognizable, understandable things to what they *intended* to
be there.
>
> Boehme's print.e looks for sequences containing
> atoms that are all in the range 32 to 255, and it displays
> those sequences as strings (whether they are meant to be
> strings or not).
Is that a problem??
>
> Boehme's method is more compact than print(),
> which is useful if you are writing out large
> amounts of data (as I suspect you are Irv).
> EDS uses a more sophisticated approach to making
> the data compact, one which has the added benefit
> of not losing accuracy when floating-point numbers are
> written, and read back in.
Well *I* wasn't trying to write out large amounts of data, just trying to
debug a malformed sequence structure, and the regular "print" did *not* help
me, and Gabriel's *did*, and I think it would be potentially helpful to
others in a similar circumstance too.
>
> print() does not exist just to supply
> the input to get(). print() can be used by itself simply
> to display objects.
But for *text* based sequence structures, Gabriel's method displays them
*more usefully*! I could *not* have debugged a recent mal-formed text
sequence (ie, inadvertently constructed *way* different than I intended)
using the *regular* "print", because it was too complex to make any sense of
it with *all those numbers*; with the *text* output from Gabriel's method I
could see almost *instantly* what was wrong. (I did temporarily change his
"quote" output around text "back" to "curly braces", so I could see all the
sequences even more clearly.)
There is no way that I would change
> print() to use Boehme's method, and with the advent of EDS,
> I don't think Boehme's print() is useful enough to enough people
> to justify adding it to the standard includes.
>
That's too bad, because I would think many beginners might struggle
unnecessarily trying to figure out why their not enormous data structures
weren't working, when an improved "print" would solve that problem for them
almost immediately. And to "include" it they would need to *know* about it,
which they might very well *not*. So having it *in* the language would be
"user-friendly", rather than: "futz around with what you're trying to do
for a while, get frustrated, ask people on the list if there's a better way,
find out there *is*, go get it, "include" it, *then* use it. I think having
the improved "print" *in* the language would be *way* better than that.
And speaking of "beginners", I think a *lot* (?) of people have been often
confused about how exactly to get something to "print" out to the screen;
possibly having some previous familiarity with "print" as a command to do
this, they try it in Euphoria and just get a bunch of *numbers*. If an
alteration of Gabriel's version (with print of a single sequence to the
screen being output without quotes) were *standard*, it would have the
*dual* advantage of being more *familiar* (than "puts"), *and* it would help
in the case when what they are trying to print out (unintentionally) has
sequences inside sequences: instead of erroring out like "puts" does,
Gabriel's print would just *print* it. It would probably be *wrong*, but at
least by having it there to *see*, the programmer could have an idea what
they *should* have specified to print.
I found Gabriel's "print" to be *very* helpful; if it were a part of
Euphoria, I really think others would find it to be useful too. If it's not
in it, & they don't *know* about it, then they can't use it.
Respectfully,
Dan Moyer
P.S. I was making a parser to abstract all the Euphoria topics, routine
names, and routine *syntax* out of the library document, so I could include
them in a variation of Judith's improvement of David's IDE, so users could
more easily find & paste the Euphoria routines into the code editor, just by
clicking on their names in a list & then clicking on "paste". Using
Gabriel's "print" let me debug the structure of my data sequence so it would
be correct & let the parser function right.
> Regards,
> Rob Craig
> Rapid Deployment Software
> http://www.RapidEuphoria.com
