Re: include statement bugs
Patrick Barnes wrote:
----------
>
> On Tue, 19 Oct 2004 22:00:18 +0300, Igor Kachan <kinz at peterlink.ru>
wrote:
> > > Igor, it's obvious that you feel strongly about this issue... why?
> > >
> > > Euphoria as coded assumes that files with the same name are the same
> > > file. As Rob himself has stated, this is a HACK.
> >
> > Ok, it is the very useful hack. Just happy hack.
> >
> > 1. It is very useful to learn Euphoria.
> > Imagine please, you have two different libraries with the same
name,
> > and you need to work with them. It is just very awkwardly and
> > uncomfortably to remember all the time who is who and what is what.
> > The long names for files, the local langs for the names of files,
the
> > case of the letters for the names of files, the descriptions of the
> > files -- all these useful things become just extra feature for
the
> > files with the same name.
> > Anyone will confuse these files, if they have same name.
> > It is very clear for me.
> > And imagine please, you have not 2, but 3 or more same names.
>
> No, that is not the point.
> I'm not going to name two of my libraries the same.
OK, you are not going, but who is going?
> However, a library written by "Alfred" (And don't ask me who Alfred
> is, this is an EXAMPLE) uses a little include called tree.e to
> accomplish its purpose.
OK, I just think that Alfred was clever enough to see the *existing*
libraries *before* his work on *new* one.
If he was not clever enough about this simplest thing -- to see what is
*already done*, I do not think that his library may be too useful for
the existing EU RDS Archive.
The "library routine" is just one of dozen sections of the RDS Archive.
It is too important section to be dirty and ambiguous.
Do you see, I think Alfred is very clever boy, same as you?
If he is not very clever, I am sorry, he needs some good school.
A good school will not teach him on the libraries with the identical
names, I think. But if will, our poor Alfred risks to get crazy.
And our Chris is very-very clever boy, he wanted just replace without
any noise the copyrighted PD standard set of RDS libraries, see,
he wrote: "RDS has established some crappy libraries as standards",
by RDS own hands - force RDS to allow *full path*
instead of the pure library name.
Our Chris is the very-very clever boy, he learned that "crappy Rob's
Euphoria" and now wants to have his own one.
Ok, what a problem, Karl has, Matt has, there is unlimited alphabet
Euphoria.
But if the standard Euphoria is so bad, why he wants his own one
in such unusual manner?
> A library written by "Johnny" uses another include that just happens
> to be called tree.e to accomplish its purpose.
>
> This is a similar problem to the namespacing issues. I *cannot* use
> both "Alfred"'s and "Johhny"'s libraries in my program, because of
> this HACK.
OK, but this is not a real problem at all.
There is no libraries in The Atchive with the identical names and never
will, I think. I think about clever EU boys, remember please.
Just Chris wanted such the feature for his strongly concrete purposes
- first of all to replace the standard copyrighted stuff.
Remember please again, these standard libraries were *stamped* by RDS for a
while. Then Rob deleted these encrypted stamps - you can see one of results
now.
> - It is not *my* responsibility to go and change alfred's or johnny's
> library so that this doesn't happen. Even if I changed the name of
> "tree.e", I'd have to go through an entire library changing all of the
> names over. And now, it wouldn't be compatible with the real version
> of the library. Every time one of them releases a new version, I'd
> have to do it again.
If and if and if and if .....
> - It is not "Alfred"'s or "Johnny"'s responsibility to make sure that
> none of their names are the same as any of the other libraries out
> there. There are thousands of contributions. If I make a new library,
> how far back to I have to search? Do I have to make sure that I don't
> duplicate the file name that a library used more than 10 years ago?
> That's excessive. I'll bet you didn't do that for any libraries
> *you've* released.
If and if and if and if ....
> - If it's the responsibility of anyone, it's Rob.
Yes, Rob hardly knows his responsibility for Euphoria and protects
his own intellectual and material rights as a real expert and excellent
master, I think.
> > 2. The name of a file is its copyright attribute.
> > Authors love the names of their files, routines, constants,
> > variables. I have investigated the big EU libs - win32lib,
w32engin,
> > some others. 11000 names. The name conflicts are very-very rare.
>
> Really? So, whom does 'ini.e' belong to? Who owns the name 'struct.e'?
> win32lib is known, and noone will else will use that name. However,
> many names are generic, and if the programmers think logically when
> they write a library, utility files are quite likely to exist with the
> same name.
Yes, really, the copyrighted stuff *must* to have its own name.
And if you do see the copyrighted existing 'struct.e' and want your
own similar library, the best way is to use the *new* good name to avoid
any confusion first of all.
Remember please Tommy's efforts around his new " Win4Eu".
> > It is the very coommon thing about intellectual property.
>
> A NAME is not intellectual property. It never is, if you had any idea
> about legal mumbo jumbo.
> A NAME can be a trademark. However, you cannot trademark generic
> terms. Noone can trademark the name "math.e", because it is a generic
> term.
The name and product are indivisible.
And there are the protected names - trademarks.
If you care, you can register the name of your program, but it is the very
expensive thing.
No one care here, as I can see, we just respect the rights on the name
without registration. The case with Chris is single and first as far as I
know.
> > 3. RDS's 'hack' is very good in those regards.
> > There are no any scandals in our community about the
> > intellectual properties - it is just impossible thing,
>
> Again see above, you are mistaken in your idea of 'intellectual property'
No, I am not mistaken, I respect the right of an author on the name of his
product independently if this name is *registered and protected* by the law
or I just see this name in The Archive.
There is no difference for me - author is author. If he was the very first
- he is the very first foreve.
> Igor, you AGAIN pushed your ideals, but did not address any of the
> points I raised. I can and will not take you for anything other than a
> Troll unless you can answer them:
>
> >Patrick Barnes wrote:
> >I believe that Euphoria should NOT assume that 2 files in different
> >locations are the same file. This is because:
> >1. It produces unexpected, undocumented behaviour.
The behaviour of Euphoria interpereter is strongly documented - filename of
library. Full path and filename are different things.
> >2. It is illogical. There is no supporting argument for keeping it as-is
My arguments see please above and below.
> >3. It makes using third-party libraries more difficult.
No, there is no such the third-party libraries at all.
You have nothing to use this way.
> As well, fixing this HACK would ensure that these problems as
> described above do not occur. Igor, can you give me even a single
> example where remedying it would cause problems?
No one is going to create the libraries with the identical names now.
Clever boys do not want, the very-very clever boy wants, but can not.
I do not think Rob wants to implement this strange useless feature
in the standard EU.
Let us wait and see some forked examples.
> Would correcting the behaviour break libraries? NO!
Useless and bad feature.
Irv did not find the libs with the identical names in other
languages for now.
Let us see the reason if he finds one.
And his "perfect solution" is like to prison or such.
> Would it produce undocumented behaviour? NO!
The Euphoria documentation is clear, interpreter has no this bug,
there is no this problem.
What do you want, MrTrick?
What about fish & chips? 
> And Rob, you have this annoying habit of being a wallflower whenever a
> discussion comes up about the language itself. I know you are proud of
> Euphoria, and you have every right to be, but don't pretend that it's
> perfect, and that every action you have ever made regarding the
> language is correct. Please weigh in on this issue. Do *you* have any
> opinion regarding why it shouldn't be changed?
Let us respect the Rob's intellectual rights on his Euphoria.
He is the author of that excellent language.
Life is life, no?
I just do not like if some too clever boy returns from the vacation and
says - all around is shit, crap, shitty crap, crappy shit, I'll replace
all that just now with my own big one, but only under YOURS NAME,
hurry up Rob, implement this for me just now, I love your Euphoria
more than you do love, stupid Rob.
And this too clever boy just forgot his naive questions to this
list years ago.
Good behaviour of grateful and thankful apprentice, no?
Again: What about fish & chips, MrTrick?
Regards,
Igor Kachan
kinz at peterlink.ru
|
Not Categorized, Please Help
|
|
