Re: '\n' output in floats
On Tue, 5 Dec 2000, Allan Rae wrote: > I thought I'd fixed this but it seems to have crept back in again -- or I > didn't completely fix it last time. > > I'll look at this later in the week -- it's a matter of one too many '\n' > after a figure/table in a float when you multiple figures in a float (and > a manually inserted \subfigure). Thanks to the now mostly working minipage and ERT insets this is no longer a problem although it does require existing documents to be slightly tweaked when coonverted/used by 1.2.0cvs. The main thing is to ensure that there are no line breaks in what you are doing, ie. all the figures and manually inserted ERT _must_ be on one line -- otherwise LaTeX will give an error message about end of paragraph before end of subfloat. At present there is also a minor annoyance of paragraphs in minipages not having separate attributes from their parent paragraph. I'll take a look at removing the couple of lines of comments in the sources that were related to the previous '\n' handling. Allan. (ARRae)
Re: Updated bug list
Perhaps this should be split into two errors: - Insert a figure float into an empty document; insert a minipage into the float; insert a paragraph (a few words) into the minipage; then set lines above and below the paragraph within the minipage -> the lines are printed above and below, i.e. outside, the minipage! (More terribly, the complete layout settings seem to be identical for the minipage and its paragraph; it seems like "minipage" is considered a property of paragraphs, not as an environment A drawing error and a paragraph property error. - Insert a figure float into an empty document; insert a minipage into the float; insert a paragraph (a few words) into the minipage; then set lines above and below the paragraph within the minipage -> the lines are printed above and below, i.e. outside, the minipage! - Paragraph attributes of a paragraph inside a minipage are those of the surrounding paragraph. These should be separate. I'm trying to get a couple of figures aligned in multiple adjacent minipages and noticed that despite setting the paragraph attributes of the paragraphs within the minipages the changes weren't visible. One good thing though I've noticed is that with a bit of care it's possible to stack multiple figures and tables in a figure float and use \subfigure _without_ having to fix the '\n' problem I reported some time ago (and which doesn't appear in Michael's bug list). Allan. (ARRae)
Re: math_mathmlstream.C
On Wed, Nov 07, 2001 at 09:08:21PM -0400, Garst R. Reese wrote: > #include "support/LOstream.h" > + std::vecto Both are fixed now Andre' -- André Pönitz .. [EMAIL PROTECTED]
Re: CVS Update: lyx-devel
Andre Poenitz a écrit :
> On Wed, Nov 07, 2001 at 02:13:42PM +, Jose Abilio Oliveira Matos wrote:
> > > The part does not look really MathMLish, does it?
> >
> > Nope. That is the _alt_ernative text version of the equation, like
> > in html where for the img tag you have the alt attribute for those
> > browsers who don't show graphics (lynx, w3m, ...)
> >
> > > Maybe I should re-formulate question:
> > > In an ideal world, how should LyX's docbook export of math stuff look like?
> >
> > Use the MathML, of course. ;-)
>
> Something like this:
>
>
> \frac{1}{2}
> 12
>
>
> ?
>
> > No I don't, but I will look for it, now I'm leaving for my classes.
> >
> > A quick search in google show this site that has in the end a list
> > of elements for MathML 1 (representation):
> >
> > http://tech.irt.org/articles/js081/
>
> That's a start...
You can either look at:
http://www.w3.org/Math/testsuite/download.html (there an example for each tag)
or at:
http://www.zvon.org/download2.php/MathML?title=MathML+Reference
(start with Output/index.html. It lists all the tags, gives examples, and links to
the reference where the element is talked about).
BG
csv2lyx to the contribs?
On Wed, 7 Nov 2001, Guenter Milde wrote: > By the way, how about uploading the csv2lyx script to the contribs on > the main lyx server? Why not? If you think it's a good idea, let me know. I send you the last version. Maybe I must do other things, like a README file or so. Changin the version nummer is a good idea too (0.0.6fix1 was a joke :-)
[PATCH] more minibuffer
this fixes two crashes (I think - JMarc does it look right to you ?)
thanks
john
--
"This bulletin discusses three security vulnerabilities that are unrelated
except in the sense that both affect ISA Server 2000"
- Microsoft Product Security
Index: src/ChangeLog
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/ChangeLog,v
retrieving revision 1.384
diff -u -r1.384 ChangeLog
--- src/ChangeLog 2001/11/05 17:07:22 1.384
+++ src/ChangeLog 2001/11/07 17:17:17
@@ -1,3 +1,8 @@
+2001-11-07 John Levon <[EMAIL PROTECTED]>
+
+ * minibuffer.h:
+ * minibuffer.C: fix crash, improve drop-down completion
+
2001-11-03 John Levon <[EMAIL PROTECTED]>
* minibuffer.h:
Index: src/minibuffer.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/minibuffer.C,v
retrieving revision 1.42
diff -u -r1.42 minibuffer.C
--- src/minibuffer.C2001/11/05 17:07:22 1.42
+++ src/minibuffer.C2001/11/07 17:17:23
@@ -70,7 +70,7 @@
void MiniBuffer::dd_init()
{
dropdown_ = new DropDown(owner_, the_buffer);
- dropdown_->result.connect(slot(this, &MiniBuffer::set_input));
+ dropdown_->result.connect(slot(this, &MiniBuffer::set_complete_input));
}
@@ -153,14 +153,15 @@
// Perfect match
string const tmp =
comp[0] + _(" [sole completion]");
- stored_set(comp[0]);
+ stored_set(comp[0] + " ");
set_input(tmp);
} else {
// More that one match
// Find maximal avaliable prefix
string const tmp = comp[0];
string test(input);
- test += tmp[test.length()];
+ if (tmp.length() > test.length())
+ test += tmp[test.length()];
while (test.length() < tmp.length()) {
vector vtmp;
lyx::copy_if(comp.begin(),
@@ -378,6 +379,13 @@
}
+void MiniBuffer::set_complete_input(string const & str)
+{
+ if (!str.empty())
+ set_input(str);
+}
+
+
void MiniBuffer::set_input(string const & str)
{
fl_set_input(the_buffer, str.c_str());
Index: src/minibuffer.h
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/minibuffer.h,v
retrieving revision 1.20
diff -u -r1.20 minibuffer.h
--- src/minibuffer.h2001/11/05 17:07:22 1.20
+++ src/minibuffer.h2001/11/07 17:17:23
@@ -74,7 +74,9 @@
void stored_slot();
///
void stored_set(string const &);
- ///
+ /// set the minibuffer content if str non-empty
+ void set_complete_input(string const &);
+ /// set the minibuffer content
void set_input(string const &);
///
void init();
Index: src/frontends/xforms/ChangeLog
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/ChangeLog,v
retrieving revision 1.193
diff -u -r1.193 ChangeLog
--- src/frontends/xforms/ChangeLog 2001/11/05 17:07:23 1.193
+++ src/frontends/xforms/ChangeLog 2001/11/07 17:17:34
@@ -1,3 +1,7 @@
+2001-11-07 John Levon <[EMAIL PROTECTED]>
+
+ * DropDown.C: fix crash, improve behaviour a bit
+
2001-11-03 John Levon <[EMAIL PROTECTED]>
* Makefile.am:
Index: src/frontends/xforms/DropDown.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/DropDown.C,v
retrieving revision 1.1
diff -u -r1.1 DropDown.C
--- src/frontends/xforms/DropDown.C 2001/11/05 17:07:23 1.1
+++ src/frontends/xforms/DropDown.C 2001/11/07 17:17:34
@@ -100,6 +100,7 @@
fl_hide_form(form_);
return 1;
}
+ XUngrabPointer(fl_get_display(), CurrentTime);
} else if (xev->type == KeyPress) {
char s_r[10]; s_r[9] = '\0';
KeySym keysym_return;
@@ -114,6 +115,18 @@
case XK_Return:
completed();
return 1;
+ case XK_Escape:
+ fl_select_browser_line(browser_, 0);
+ completed();
+ return 1;
+ default:
+ // FIXME: if someone has a got a way to
+ // convince the event to fall back to the
+
Re: Updated bug list
> "Juergen" == Juergen Spitzmueller <[EMAIL PROTECTED]> writes: Juergen> We could of course use geometry for the AMS-classes and other Juergen> classes who don't support landscape natively. I haven't Juergen> checked, but it seems to me that ejour belongs to these ones, Juergen> too. Are we supposed to do this? What do the LaTeX gurus say? I'm not a LaTeX guru, but we could have yet another tag in the layout file to declare the options it know about (or a tag explicitely for landscape). JMarc
[PATCH] cleanups
several cleanups. Lars, I added ::WORKAREA instead of removing the debug. Can someone apply my lyxserver patch too please ? thanks john -- "This bulletin discusses three security vulnerabilities that are unrelated except in the sense that both affect ISA Server 2000" - Microsoft Product Security cleanups.diff.gz Description: application/gunzip
Re: 1.1.6fix3 pictures no longer display ( eps ghostscript preview )
On Wed, Nov 07, 2001 at 03:08:27PM +0100, Morten S. Nielsen wrote: > Sorry if this is a bit outdated, but this bug still seems to exist. > > In figinset.C there seems to be a problem with execlp if one uses > ghostscript 6.51 + > > The gs process starts but never exits. Why I don't know. One can go > into /proc and see the commandline arguments, but this doesn't explain > where it hangs. If you're curious I could try to strace gs. But for now > I'll just report the workaround. This is what happens when you are using a menu in lyx when rendering is finished by the ghostscript child. It is an xforms bug we can do nothing about. > If using ghostscript 5.50 it always works fine. > > On my system lyx 1.1.6-fix3 works with Ghostscript 6.51 when setting > \ps_command to "/usr/bin/gs " > in lyxrc.defaults or equivalent > > It's the first argument to execlp which must end with a space (works > also when setting the first string to " ") This is /thoroughly/ bizarre. There is no binary named "/usr/bin/gs " so I am amazed it works at all. Are you sure about this ? thanks john -- "This bulletin discusses three security vulnerabilities that are unrelated except in the sense that both affect ISA Server 2000" - Microsoft Product Security
Re: Updated bug list
Am Mittwoch, 7. November 2001 17:18 schrieb Jean-Marc Lasgouttes: > > "Juergen" == Juergen Spitzmueller <[EMAIL PROTECTED]> > > writes: > > Juergen> We could of course use geometry for the AMS-classes and > other Juergen> classes who don't support landscape natively. I > haven't Juergen> checked, but it seems to me that ejour belongs to > these ones, Juergen> too. Are we supposed to do this? What do the > LaTeX gurus say? > > I'm not a LaTeX guru, but we could have yet another tag in the layout > file to declare the options it know about (or a tag explicitely for > landscape). Something like: NoOption landscape which means that we have to chose geometry automatically if the user choses landscape? This would probably be the best, since then the authors of new layout files are responsible and we do not have to take care when a new stylefile comes in. If we can agree on this solution I can check which of the current classes do not support landscape and have a look if I can implement this. Hints on how to do this are very welcome, though. Jürgen. > JMarc
Re: Bug list update
Juergen Spitzmueller wrote: > Jean-Marc Lasgouttes wrote: > > I did the splitting of the xxx_apply stuff, but by mistake commited > > it before doing any testing :) Juergen, could you test whether the > > FormDocument still works and whether 'save defaults' works better > > than before? I'll do some testing of my own too. Note that any new > > bug should not be difficult to fix, I think. > > Thank you for doing this. I will have a closer look this evening! I had a closer look. I think I tried all possible settings, saved as class default and reopended. Seems that it works now as it should. Very good work. The only thing that became worse AFAICS is the "Use Class Defaults" button. It is now disabled most of the time (actually it obviously disables after any first change you do), which shouldn't happen. Is it possible that params.hasClassDefaults() doesn't give the correct values anymore? Many thanks anyway, Jürgen. > > JMarc
Re: [PATCH] minibuffer dropdown completion
On Wed, Nov 07, 2001 at 05:23:16PM +0100, Jean-Marc Lasgouttes wrote: > Yes I did. Do you use lyxstring? Here is a backtrace: ^ ah, d'oh, I forgot lyxstring was more picky than std::string. I'll have a look thanks john -- "This bulletin discusses three security vulnerabilities that are unrelated except in the sense that both affect ISA Server 2000" - Microsoft Product Security
Re: [PATCH] minibuffer dropdown completion
> "John" == John Levon <[EMAIL PROTECTED]> writes: John> On Mon, Nov 05, 2001 at 06:08:20PM +0100, Jean-Marc Lasgouttes John> wrote: >> Applied. Note that there still a crash with lyxstrings when typing >> "up" because up is a command in itself, and the code does an >> access after end of strings (on which lyxstring asserts). John> have you checked this recently ? It worked OK for me. Yes I did. Do you use lyxstring? Here is a backtrace: (gdb) bt #0 0x40128d21 in __kill () from /lib/libc.so.6 #1 0x40128996 in raise () at ../sysdeps/posix/raise.c:28 #2 0x4012a0b8 in abort () at ../sysdeps/generic/abort.c:139 #3 0x81d0ca7 in Letext () at ../../../lyx-devel/src/support/abort.C:9 #4 0x81d6b42 in lyxstring::operator[] (this=0xb0b4, pos=2) at ../../../lyx-devel/src/support/LAssert.h:24 #5 0x80b9312 in MiniBuffer::peek_event (this=0x82fd450, ob=0x82fd4d8, event=9, key=9) at ../../lyx-devel/src/minibuffer.C:163 #6 0x80b97a8 in C_MiniBuffer_peek_event (ob=0x82fd4d8, event=9, key=9) at ../../lyx-devel/src/minibuffer.C:263 #7 0x819ff7d in fl_handle_it () at ../../../../lyx-devel/src/frontends/xforms/form_maths_matrix.C:81 #8 0x81a0065 in fl_handle_object () at ../../../../lyx-devel/src/frontends/xforms/form_maths_matrix.C:81 #9 0x81a307a in fl_keyboard () at ../../../../lyx-devel/src/frontends/xforms/form_maths_matrix.C:81 #10 0x81a3475 in fl_handle_form () at ../../../../lyx-devel/src/frontends/xforms/form_maths_matrix.C:81 #11 0x81a3786 in do_keyboard () at ../../../../lyx-devel/src/frontends/xforms/form_maths_matrix.C:81 #12 0x81a3e6b in do_interaction_step () at ../../../../lyx-devel/src/frontends/xforms/form_maths_matrix.C:81 #13 0x81a44d9 in fl_treat_interaction_events () at ../../../../lyx-devel/src/frontends/xforms/form_maths_matrix.C:81 #14 0x81a4514 in fl_check_forms () at ../../../../lyx-devel/src/frontends/xforms/form_maths_matrix.C:81 #15 0x813f06d in GUIRunTime::runTime () at ../../../../lyx-devel/src/frontends/xforms/GUIRunTime.C:86 #16 0x809eee8 in LyXGUI::runTime (this=0x82a4040) at ../../lyx-devel/src/lyx_gui.C:315 #17 0x809fea2 in LyX::LyX (this=0xb4d0, argc=0xb4f0, argv=0xb534) at ../../lyx-devel/src/lyx_main.C:176 #18 0x80b8891 in main (argc=1, argv=0xb534) at ../../lyx-devel/src/main.C:38 (gdb) frame 5 #5 0x80b9312 in MiniBuffer::peek_event (this=0x82fd450, ob=0x82fd4d8, event=9, key=9) at ../../lyx-devel/src/minibuffer.C:163 163 test += tmp[test.length()]; This line 163 is a problem because "up" is a command in itself, and we try to read after end of string. Is that enough info? JMarc
Re: docbook
On Wed, Nov 07, 2001 at 05:22:39PM +0100, Andre Poenitz wrote: > InsetFormula::docbook() is seemingly never called. > > Any ideas why? Never mind linuxdoc is called... my fault. Andre' -- André Pönitz .. [EMAIL PROTECTED]
docbook
InsetFormula::docbook() is seemingly never called. Any ideas why? Andre' -- André Pönitz .. [EMAIL PROTECTED]
Re: Crash with mb drop-down
On Wed, Nov 07, 2001 at 10:09:10AM +0100, Lars Gullik Bjønnes wrote: > I can't see _why_ it does not make sense. because it's claiming that I call the string functions from the 1st line of completed() (for one). Anyway, I fixed it - I forgot xforms counts from 1 like some silly moo... wait for a patch ... john -- "This bulletin discusses three security vulnerabilities that are unrelated except in the sense that both affect ISA Server 2000" - Microsoft Product Security
Re: CVS Update: lyx-devel
On Wed, Nov 07, 2001 at 02:13:42PM +, Jose Abilio Oliveira Matos wrote:
> > The part does not look really MathMLish, does it?
>
> Nope. That is the _alt_ernative text version of the equation, like
> in html where for the img tag you have the alt attribute for those
> browsers who don't show graphics (lynx, w3m, ...)
>
> > Maybe I should re-formulate question:
> > In an ideal world, how should LyX's docbook export of math stuff look like?
>
> Use the MathML, of course. ;-)
Something like this:
\frac{1}{2}
12
?
> No I don't, but I will look for it, now I'm leaving for my classes.
>
> A quick search in google show this site that has in the end a list
> of elements for MathML 1 (representation):
>
> http://tech.irt.org/articles/js081/
That's a start...
Andre'
--
André Pönitz .. [EMAIL PROTECTED]
Re: 1.1.6fix3 pictures no longer display ( eps ghostscript preview )
On Mon, 24 Sep 2001 07:25:08 -0700 john wrote: > well as someone said that doesn't work for everyone. > > If I could reproduce the problem I could debug it :( > > john Sorry if this is a bit outdated, but this bug still seems to exist. In figinset.C there seems to be a problem with execlp if one uses ghostscript 6.51 + The gs process starts but never exits. Why I don't know. One can go into /proc and see the commandline arguments, but this doesn't explain where it hangs. If you're curious I could try to strace gs. But for now I'll just report the workaround. If using ghostscript 5.50 it always works fine. On my system lyx 1.1.6-fix3 works with Ghostscript 6.51 when setting \ps_command to "/usr/bin/gs " in lyxrc.defaults or equivalent It's the first argument to execlp which must end with a space (works also when setting the first string to " ") Morten Ps: the xhost trick didn't do it for me and I'm not on the lyx list. -- - The Penguin's 1. We are better together than alone - Postulate 2. If you push something hard enough it will fall over - Morten S. Nielsen Dept. of Manuf. Engineering and Management - mailto:[EMAIL PROTECTED] Building 425, 2. floor, DK-2800 Lyngby
Re: CVS Update: lyx-devel
On Wed, Nov 07, 2001 at 02:51:53PM +0100, Andre Poenitz wrote: > > The part does not look really MathMLish, does it? Nope. That is the _alt_ernative text version of the equation, like in html where for the img tag you have the alt attribute for those browsers who don't show graphics (lynx, w3m, ...) > Maybe I should re-formulate question: > In an ideal world, how should LyX's docbook export of math stuff look like? Use the MathML, of course. ;-) > Andre' > > PS: Do you know some concise list of common MathML elements? I know the > "specification", but that's at least 90% incomprehensible for me... No I don't, but I will look for it, now I'm leaving for my classes. A quick search in google show this site that has in the end a list of elements for MathML 1 (representation): http://tech.irt.org/articles/js081/ > -- > André Pönitz .. [EMAIL PROTECTED] -- José
Re: CVS Update: lyx-devel
On Wed, Nov 07, 2001 at 01:49:21PM +, Jose Abilio Oliveira Matos wrote: > An Equation is a formal mathematical equation (with an optional rather > than required title). > > If the MathML Module is used, Equation can also contain the mml:math > element. > > Example > > >"http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd";> > Fermat's Last Theorum > x^n + y^n ≠ z^n ∀ n ≠ 2 > > /equation> > > As you can see from this example before MathML it was used text or > graphics for the mathematics. The part does not look really MathMLish, does it? Maybe I should re-formulate question: In an ideal world, how should LyX's docbook export of math stuff look like? Andre' PS: Do you know some concise list of common MathML elements? I know the "specification", but that's at least 90% incomprehensible for me... -- André Pönitz .. [EMAIL PROTECTED]
Re: CVS Update: lyx-devel
On Wed, Nov 07, 2001 at 02:24:05PM +0100, Andre Poenitz wrote: > > What format does docbook use for math in general? Until the advent of MathML none in particular. >From DocBook "The Definitive Guide" http://www.docbook.org/tdg/en/html/equation.html * Element equation Description An Equation is a formal mathematical equation (with an optional rather than required title). If the MathML Module is used, Equation can also contain the mml:math element. Example http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd";> Fermat's Last Theorum x^n + y^n ≠ z^n ∀ n ≠ 2 /equation> * As you can see from this example before MathML it was used text or graphics for the mathematics. > Andre' > > -- > André Pönitz .. [EMAIL PROTECTED] -- José
Re: Bug list update
Jean-Marc Lasgouttes wrote: > I did the splitting of the xxx_apply stuff, but by mistake commited > it before doing any testing :) Juergen, could you test whether the > FormDocument still works and whether 'save defaults' works better > than before? I'll do some testing of my own too. Note that any new > bug should not be difficult to fix, I think. Thank you for doing this. I will have a closer look this evening! Jürgen. > JMarc
Re: CVS Update: lyx-devel
On Wed, Nov 07, 2001 at 02:22:43PM +, [EMAIL PROTECTED] wrote: > Log message: > Improved support for docbook export in inset text. What format does docbook use for math in general? Andre' -- André Pönitz .. [EMAIL PROTECTED]
Re: Bug list update
> "Juergen" == Juergen Spitzmueller <[EMAIL PROTECTED]> writes: Juergen> Well, IMHO it is much better. It is cleaner from a devel's Juergen> and more comfortable from a user's point of view. I would be Juergen> very grateful if you could do it, since then it will be done Juergen> *right* ;-) To elaborate on what I have been writing earlier: the 'save default' feature is not yet using the new stuff, so it is not supposed to work better as before (I'm working on it). However, I'd like to know whether FormDocument works as before. Sorry for the inconvenience. JMarc
Re: Bug list update
> "Juergen" == Juergen Spitzmueller <[EMAIL PROTECTED]> writes: Juergen> - Start LyX; open new file; open the document dialog; change Juergen> separation from "indent" to "skip"; "save as class defaults"; Juergen> close document; open new file; open document dialog -> Juergen> separation is reset to "indent" again! Juergen> Strange. We use the same LFUN_LAYOUT_SAVE_DEFAULT than in Juergen> 1.1.6, where this works. Unfortunately, I do not have the Juergen> slightest idea where to look at. JMarc, can you help here, Juergen> please? I did the splitting of the xxx_apply stuff, but by mistake commited it before doing any testing :) Juergen, could you test whether the FormDocument still works and whether 'save defaults' works better than before? I'll do some testing of my own too. Note that any new bug should not be difficult to fix, I think. Things to try out are related to language change, textclass change and bullets (those for which I had to be particularly careful). I'll add the missing changelogs too. JMarc
Re: pipestream, 5th attempt
On Wed, Nov 07, 2001 at 10:30:31AM +0100, Lars Gullik Bjønnes wrote: > How hard would it be to have > > pipestream(char const * const cmd); > > as the signature, and split it into arguments inside the create > method? Not really hard, but we can't use arguments with spaces then or we would have to do our own commandline parsing. I think I don't like any of that. > pipestream(std::vector const &); > > is too heavy I guess. Certainly. > (or perhaps not... I kindo like it, except for the char* part. > > Perhaps multiple constructors, for one to 10 args? If you wish. Actually, I'd provide only the one and two arg version and would add others only if they are needed. And I'd like to keep the char*[] version anyway as safe fall back Andre' -- André Pönitz .. [EMAIL PROTECTED]
math-extern
I think I have found a fairly reliable way to figure out where to put extra "invisible multiplications" for Maple. But first you have to promise not to beat me too hard, ok? Every pair of adjacent chars with code LM_TC_VAR (i.e. the code a "normal" 'abc..' gets initially) is split by extra '*'. Then I feed the expression to Maple's syntax checker, parse the output and put in an extra '*' if it complains. This is repeated until the syntax checker remains silent or a fixed number of iterations is reached. So zx^2 + 4zxy + 4zy^2 plus "M-x math-extern maple factor" indeed yields z(x + 2y)^2 Would be nice if somebody played a bit with it and provided me with a stream of small examples that don't work as they should. Finding some is definitely not hard since almost nothing works right now... Andre' -- André Pönitz .. [EMAIL PROTECTED]
