The problems came from the French accents and the German umlauts. I changed the XML
charset, so this should be fine now. New version is under CVS.

Hi Stefan,

The parsing now seems OK, IE downloads the XML and the RSS reader of Avant 
Browser 
accepts the feed. 
I still have a problem however : below the header, only the first 3 items 
are displayed and it seems only partially.
I included the HTML result so you can "visualize" what's going on (this may 
also be on the avant browser side !).

Cheers,
Marc


> The problems came from the French accents and the German umlauts. I 
changed the XML
> charset, so this should be fine now. New version is under CVS.

> I included the HTML result so you can "visualize" what's going on (this may 
> also be on the avant browser side !).

I believe it is on the avant browser side. Below is the same shown in the
SharpReader browser, where everything is displayed correctly.

When compiling elogd on AMD64 (in 64 bit mode) there are many 
warnings like these:

src/regex.c:3769: warning: cast from pointer to integer of different size
src/regex.c:3769: warning: cast from pointer to integer of different size
src/regex.c:3775: warning: cast to pointer from integer of different size

The reason is the (int) cast, which (I think) is not necessary.

> When compiling elogd on AMD64 (in 64 bit mode) there are many 
> warnings like these:
> 
> src/regex.c:3769: warning: cast from pointer to integer of different size
> src/regex.c:3769: warning: cast from pointer to integer of different size
> src/regex.c:3775: warning: cast to pointer from integer of different size
> 
> The reason is the (int) cast, which (I think) is not necessary.

Unfortunately I don't have a 64-bit compiler to try. Can you please check if
elogd.c has similar warnings or only regex.c? The ones in elogd.c I can fix,
but regex.c is straight from the GNU C library, so we would have to wait intil
they fix it. Or can you get it from the 64-bit C library? The point is that I
need the source code, so that I can compile it under Windows as well. Can you
try to compile elogd without regex.c, like 

gcc -g -O -W -Wall -o elogd src/elogd.c

If that works fine, we don't have to compile regex.c under linux. But I guess
it's still needed fot Solaris and other non-GNU Unix brands.

> > When compiling elogd on AMD64 (in 64 bit mode) there are many 
> > warnings like these:
> > 
> > src/regex.c:3769: warning: cast from pointer to integer of different size
> > src/regex.c:3769: warning: cast from pointer to integer of different size
> > src/regex.c:3775: warning: cast to pointer from integer of different size
> > 
> > The reason is the (int) cast, which (I think) is not necessary.
> 
> Unfortunately I don't have a 64-bit compiler to try. Can you please check if
> elogd.c has similar warnings or only regex.c? 

  elogd.c shows the same warnings.

> The ones in elogd.c I can fix,
> but regex.c is straight from the GNU C library, so we would have to wait intil
> they fix it. Or can you get it from the 64-bit C library? The point is that I
> need the source code, so that I can compile it under Windows as well. Can you
> try to compile elogd without regex.c, like 
> 
> gcc -g -O -W -Wall -o elogd src/elogd.c

  This works OK.
 

After THREE tries to enter a long detailed list of questions, all have been 
replaced by just one:  why do ELOG textareas and entry forms lose all data 
if the browser goes back/forward?  Or if a submission causes an error?  
Other forms in other applications don't have this kind of data loss.

Submitting this now before I lose it again.  Will submit rest of questions 
as separate entries.

Is there any way to search for all attributes _except_ a certain value?