OpenEuphoria

Re: Giving Euphoria more of an appeal to others?

• Posted by Larry Miller <larrymiller at sasktel.n??> Apr 17, 2008
• 580 views

Jeremy Cowgar wrote:
> 
> Euphoria is a great language, however, there are a few things that I think are
> preventing it from gaining even more popularity. I am curious as to the users
> opinions and feedback.
> 
> I'll focus on #1 as I can help there.
> 
> 1. A standard library. One of the reasons Python is so popular is because when
> you download and install it, you have just about everything you need (they
> take
> it a bit to extremes though, I think). Now, on the contrary, with Euphoria,
> how do I add two dates? How can I trim the whitespace off the end or beginning
> of a sequence acting as a string? How do I split a string by spaces? How do
> I create a hash table? etc... All of those can be done by downloads found in
> "The Archive" but which one? Then find it. Then use one "split" function and
> then one of the other libs you use requires a different "split" function,
> etc...
> Then, when wrapping libraries it seems everyone has their own functions for
> peeking at strings, assigning and checking functions, etc... 
> 
> CK Lester already started on such a library. Submissions to The Archive seem
> to show that such a thing is needed. There are others who have attempted at
> creating unofficial standard libs. Euphoria has itself an include directory
> with euphoria functions. 
> 
> I am wondering, why not start building an official standard library that is
> distributed with Euphoria? Not additions in C, but such helpers placed in
> EUPHORIA/include
> ?? CK Lester has started on such a library.
> 
> I would be willing to start organizing such an addition by standardizing
> naming,
> documenting functions in the standard .htx format and importantly, providing
> unit tests for the functions and, in fact, functions that are already in
> existence
> in Euphoria.
> 
> I would start on String functions, then Date and Time functions, the Hash
> functions
> and move on from there. It would take a bit of work, but a solid week of work
> on it, I think we would all be amazed at what could be done.
> 
> It would be a *fantastic* addition to 3.2.
> 
> 2. The website is very old in appearance. It's very functional, which is of
> course good but it needs a face lift. I invited a co-worker to visit
> rapideuphoria.com
> and check out Euphoria. They saw the website, the flashing banner "Simpler
> than
> basic, more powerful than C++, etc..." and thought it was a toy from 1980.
> Seriously,
> the claim of more powerful than C++ is pretty subjective and hard to
> substantiate.
> 
> 3. Get Euphoria into the major Linux distributions. Debian, Ubuntu, Fedora,
> GenToo, FreeBSD, OpenSUSE, Mandrivia and others.
> 
> These are just some thoughts. Please comment, especially on #1. If the
> community
> likes the idea, I would get started right away.
> 
> --
> Jeremy Cowgar
> <a href="http://jeremy.cowgar.com">http://jeremy.cowgar.com</a>


I am in full agreement that a standard set of libraries is needed. This has been
attempted before but got bogged down with endless discussions and little was
produced. While most, if not all, of the desired functions are already in the
archive this is hardly an ideal situation. One of the biggest problems is the
complete lack of consistency. Some are well documented while others have
virtually none. There is a wide variety in code quality. Some libraries are
written by professional programmers while others are by new users. Many have had
little testing done.

The complete standard library need not be included in the distribution. The
website should have a prominent link to them and it should be clearly indicated
that they have official sanction. Any user could be reasonably certain that they
followed a defined standard and had been properly tested. This would be
particularly useful to a new user.

I really hope that this comes about. This needs to be controlled by one
individual or a small group. We must avoid the endless discussions about the
names of individual functions and what functions are to be included in what
files, etc. If this occurs nothing will ever be decided and nothing accomplished.

Larry Miller