Sure - if you "manually" copy one string to another, obviously
your second string needs storage space. However, how often
is this done in the code? I can't image the domain name string
needs to be copied too often (you would only need to do this to
make a temporary modification to the string ...). If we have a few
more bytes times a few temporary copies, this is still not much
extra RAM; without knowing how often you make temporary
copies of the domain name string, I would _guess_ that even
*doubling* it's size would add perhaps 200 _extra_ bytes (?).
This is certainly worth doing.
As for the issue of "typing" long URL's, I must say that very
rarely do I type *any* URL's. Normally, I click on a link from
another web page, a web search engine page, or my "hotlist".
So, being too lazy to type 40 characters *just isn't relevant*.
I will type a URL when I need to, but in fact I rarely need to
do any typing, anyway.
Michael has indicated that this issue will be addressed using
dynamic allocation, for the Linux version of Arachne only.
Fine, use whatever method you like for the Linux version, but
please don't ignore the problem with the DOS version. The
person who pointed this problem out (complained ;-) to me
is a DOS Arachne user, not a Linux user. When I checked
my own favourite URL's, I found that several had fairly long
domain names, with nearly 40 characters, so I have simply
been lucky so far not to have exceeded this Arachne limit
myself. Clearly, others have not been so lucky ...
Joe.
> -----Original Message-----
> From: Bernie [SMTP:[EMAIL PROTECTED]]
> Sent: Wednesday, 15 November 2000 23:18
> To: '[EMAIL PROTECTED]'
> Subject: RE: Outstanding bugs
>
> Joe wrote:
> > Firstly, strings in C (AFAIK) are always passed by reference
> > (ie. using pointers), not by value (not an option in C, I think),
> > so as this string gets "passed" from one routine to another,
> > it doesn't consume the extra bytes each time.
>
> If we do like this:
>
> void func(char *s1)
> {
> ...
> strcpy(s1, s2);
> ...
> }
>
> we will use more memory.
>
> > > I wouldn't go there, and I think I'm speaking for the majority of
> users.
> > >
> > [da Silva, Joe]
> >
> > This is the part that I *really* don't understand! Arachne users
> > have no control over domain names - the web site owners do.
>
> Exactly, so it was a poor choice since it's:
>
> 1. long to type
> 2. hard to remember (not always true, but most of the time -
> http://ge.astrid.lindgren.nobelpriset.nu/ is easy to remember (a Swedish
> sentance).
> 3. Rather unusable in advertisments
>
> Therefor the chance that some web site owner chooses such a long name is
> very low since most users will not type such a long address.
> //Bernie
> http://bernie.arachne.cz/
