Thesis submission time

[John Levon]

It's thesis submission time here, and all those who made the mistake of
not using LaTeX/LyX are desparately trying to fix the broken output
they've got.

In contrast my output was perfect first time, flawless, and I could focus
on fixing the actual content.

So once again thanks all for a great program ! (and I will be working on
it again soon when I have a working C++ compiler at home !)

thanks
john
 
-- 
"The words CON and NUL
Are reserved by the system.
Don't ask; they just are."

Re: Working on a Word to LyX importer

[Andre Poenitz]

 My work would go faster if I had a file describing the equivalent of LateX
 codes in  LyX codes [for example $\alpha$ is \begin_inset Formula \( \alpha  \)
 \end_inset ]. Does this document exists somewhere ?

None that I know of.

I think the LyX format is likely to change pretty fast, so it'd probably
better to improve any existing Word2LaTeX converter and work on a direct
LaTeX import to LyX or even on a direct Word importer for LyX. 

LyX's internals have improved significantly over the last months, so the
poor state of the external .lyx format does not reflect the current state.

Supporting the current .lyx format by something important as an importer
from Word would prabably hinder changes in this area.

That's my just personal opinion of course. 

Andre'

-- 
Andre' Poenitz  [EMAIL PROTECTED]

instant core dump

[Angus Leeming]

Running Current CVS results in an instant core dump on my machine (DEC 
Alpha). Does anybody else have this experience? If not, then I guess that 
this is something to do with these size_type changes (and all to remove 
warnings on the DEC!) Lars, do you have anymore changes to submit soon, or 
shall I investigate the code as is?

Angus

Re: src/ext_l10n.h

[Jean-Marc Lasgouttes]

 "John" == John Levon [EMAIL PROTECTED] writes:

John Er, whoops I added this to .cvsignore but I've just realised
John shouldn't this be removed from CVS altogether ?

John It is generated from autogen.sh so shouldn't it be removed from
John the CVS like configure etc.

You are right. I just did it.

JMarc

Re: compiling 1.1.5fix1 with DEC cxx and lyxstring

[Jean-Marc Lasgouttes]

 "Angus" == Angus Leeming [EMAIL PROTECTED] writes:

Angus Any ideas on the way forward, Jean-Marc? Everything works fine
Angus with std::string, but with lyxstring I get a whole pile of
Angus errors

All I can say is that this code has been changed in 6.2, which uses
std::string correctly.

JMarc

Re: NoGUI LyX

[Jean-Marc Lasgouttes]

 "Baruch" == Baruch Even [EMAIL PROTECTED] writes:

Baruch On 27 Sep 2000, Jean-Marc Lasgouttes wrote:
 and fix session (I have several bug reports in my mail folder which
 depend on this).

Baruch How about updating a known bug list and a todo list on the
Baruch developers website for all to see?

Baruch It will help development since we'll all be able to know what
Baruch bugs are pending and what urgent todo stuff is waiting. It
Baruch will also help new developers to find small tasks that they
Baruch can start with.

Even better would be a real bug-tracking database...

Concerning bugs I am aware of, you can find some of them in my last
posting about 1.1.5fix2.

JMarc

Re: Bug in CVS

[Jean-Marc Lasgouttes]

 "Dekel" == Dekel Tsur [EMAIL PROTECTED] writes:

Dekel But why sel is 0 ? The problem is in Intl::InitKeyMapper: Since
Dekel the "default" language is removed, n should be initialized to
Dekel 0. I've attached a patch that does both fixes.
   
Applied.

JMarc

Re: [PATCH] Complain about Qt2

[Jean-Marc Lasgouttes]

 "John" == John Levon [EMAIL PROTECTED] writes:

John This patch changes
John the kde m4 stuff so that if Qt2 is being picked up, configure
John will stop at the Qt setup rather than failing at the KDE
John compile, which is misleading.

Applied.

JMarc

Re: NoGUI LyX

[John Levon]

On 29 Sep 2000, Jean-Marc Lasgouttes wrote:

 Even better would be a real bug-tracking database...
 
 JMarc
 

as people pointed out last time I suggested such, like bugzilla, this
would require people with a knowledge of the codebase to adminstrate and
triage, prioritise etc.

BUT this would be real advantage if people actually use it, especially for
tracking developer's/user's comments on the bugs ...

I would offer to take such a responsibility myself but I really don't know
the code well enough to judge !

john

-- 
"The words CON and NUL
Are reserved by the system.
Don't ask; they just are."

Re: instant core dump

[Jean-Marc Lasgouttes]

 "Angus" == Angus Leeming [EMAIL PROTECTED] writes:

Angus Running Current CVS results in an instant core dump on my
Angus machine (DEC Alpha). Does anybody else have this experience? If
Angus not, then I guess that this is something to do with these
Angus size_type changes (and all to remove warnings on the DEC!)
Angus Lars, do you have anymore changes to submit soon, or shall I
Angus investigate the code as is?

At least you are able to compile... OK, I just commit the patch from
Dekel to cure a coredump releated to combox (please try it out, since
I can't compile). I also added the Qt2 patch.

My compile problems are related to Allan latest change, so you may be
able to help here:

mycxx -DHAVE_CONFIG_H -I. -I../../../../lyx-devel/src/frontends/xforms -I../../../src 
-I../../../../lyx-devel/src/ -I../../../../lyx-devel/src/frontends/ -I../../.. 
-I../../../../lyx-devel -I../../../../lyx-devel/src/cheaders 
-I/afs/rocq/home/meval/common/include -I/usr/local/include -g0 
-Wp,-MD,.deps/FormBase.pp -c ../../../../lyx-devel/src/frontends/xforms/FormBase.C -o 
FormBase.o
cxx: Error: ../../../../lyx-devel/src/frontends/xforms/FormBase.C, line 96: 
  "SigC::HandleObj, Scope_::operator Obj *() [with
  Obj=SigC::SlotData, Scope_=SigC::SlotExtend]" is inaccessible
if (u_) {
^

What does if(u_) mean when u_ is a Connection? I guess it should be
replaced with something else.

JMarc

Core dump on startup revistited

[Angus Leeming]

I get a core dump instantly when starting up the current CVS lyx. 

I have tracked down the problem to language.C's LangInit::initL() and the 
line:

if (languages.find(lyxrc.default_language) == languages.end()) {

The problem lies with lyxrc.default_language. Modifying the source slightly 
to:


std::cerr  "got to here 2"  std::endl;
std::cerr  "lyxrc.default_language "  lyxrc.default_language  
std::endl;
if (languages.find(lyxrc.default_language) == languages.end()) {

gdb is not much help:
(gdb) r
Starting program: /usr/users/aleem/OTHERS_CODE/lyx/devel/src/lyx
got to here 1
got to here 2
lyxrc.default_language Memory fault(coredump) 
  
But dbx does better:
(dbx) r
got to here 1
got to here 2
lyxrc.default_language signal Segmentation fault at   
[length__CQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__TvXv:1703 
+0x8,0x1200e6034]
(dbx) where
  0 
length__CQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__TvXv(this 
= 0x14006cf90) ["/usr/include/cxx/string~alt~deccxx_17E4996F":1703, 
0x1200e6034]
   1 
size__CQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__TvXv(this 
= 0x14006cf90) ["/usr/include/cxx/string~alt~deccxx_17E4996F":852, 
0x1200e5ffc]
   2 
__ls__3stdXRQ13std40basic_ostream__TcQ13std15char_traits__TcRCQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__Tv(os
 
= @1400704d8, s = @14006cf90) ["/usr/include/cxx/ostream.cc":1720, 
0x1200e6068]
   3 initL__8LangInitXv(this = 0x14006a760) ["language.C":131, 0x1201aeba0]
   4 __ct__8LangInitXv(this = 0x14006a760) ["language.C":34, 0x1201add2c]
   5 __init_sti__language_C_a0d15379() ["language.C":143, 0x1201aef08]
(dbx) quit
 

First, can someone explain the last few lines of language.C:
static
LangInit langinit;

bool LangInit::init = false;

Commenting them out allowed lyx to start, but loading the first document 
causes a similar problem:

(gdb) r
Starting program: /usr/users/aleem/OTHERS_CODE/lyx/devel/src/lyx
 
Program received signal SIGSEGV, Segmentation fault.
0x12031faa0 in 
__as__Q13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__TvXRCQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__Tv
 
()
(gdb) where
#0  0x12031faa0 in 
__as__Q13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__TvXRCQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__Tv
 
()
#1  0x12018ffb8 in __ct__12BufferParamsXv ()
#2  0x12017a36c in 
__ct__6BufferXRCQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__Tvjb 
()
#3  0x12018cd2c in 
newBuffer__13BufferStorageXRCQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__Tvjb
 
()
#4  0x12018e1dc in 
readFile__10BufferListXRCQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__Tvjb
 
()
#5  0x12018eff4 in 
loadLyXFile__10BufferListXRCQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__Tvjb
 
()
#6  0x1201d6a30 in 
Dispatch__7LyXFuncXiRCQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__Tv
 
()
#7  0x1202c7af0 in MenuCallback__Q17Menubar5PimplXP7flobjs_l ()
#8  0x1202c784c in C_Menubar_Pimpl_MenuCallback ()
#9  0x3ffbff3e87c in fl_object_qread () from /usr/local/lib/libforms.so.0.89
#10 0x3ffbff538e0 in fl_check_forms () from /usr/local/lib/libforms.so.0.89
#11 0x1202c16bc in runTime__10GUIRunTimeXv ()
#12 0x1201c088c in runTime__6LyXGUIXv ()
#13 0x1201c3394 in __ct__3LyXXPiPPc ()
#14 0x1201f7e0c in main ()
(gdb) quit
  
Any ideas?
Angus

Re: instant core dump

[Angus Leeming]

It means nothing. Remove it, leaving
u_.disconnect();
This is legal, fine, good, compiles, works, etc!

On Fri, 29 Sep 2000, Jean-Marc Lasgouttes wrote:
  "Angus" == Angus Leeming [EMAIL PROTECTED] writes:

 Angus Running Current CVS results in an instant core dump on my
 Angus machine (DEC Alpha). Does anybody else have this experience? If
 Angus not, then I guess that this is something to do with these
 Angus size_type changes (and all to remove warnings on the DEC!)
 Angus Lars, do you have anymore changes to submit soon, or shall I
 Angus investigate the code as is?

 At least you are able to compile... OK, I just commit the patch from
 Dekel to cure a coredump releated to combox (please try it out, since
 I can't compile). I also added the Qt2 patch.

 My compile problems are related to Allan latest change, so you may be
 able to help here:

 mycxx -DHAVE_CONFIG_H -I. -I../../../../lyx-devel/src/frontends/xforms
 -I../../../src -I../../../../lyx-devel/src/
 -I../../../../lyx-devel/src/frontends/ -I../../.. -I../../../../lyx-devel
 -I../../../../lyx-devel/src/cheaders -I/afs/rocq/home/meval/common/include
 -I/usr/local/include -g0 -Wp,-MD,.deps/FormBase.pp -c
 ../../../../lyx-devel/src/frontends/xforms/FormBase.C -o FormBase.o cxx:
 Error: ../../../../lyx-devel/src/frontends/xforms/FormBase.C, line 96:
 "SigC::HandleObj, Scope_::operator Obj *() [with
   Obj=SigC::SlotData, Scope_=SigC::SlotExtend]" is inaccessible
 if (u_) {
 ^

 What does if(u_) mean when u_ is a Connection? I guess it should be
 replaced with something else.

 JMarc

lyxformat/i18n

[Ralf Corsepius]

Hi,

When saving a lyx document in a "C" environment (LANG="C"),
lyx-1.1.5cvs saves its \lyxformat like this (2nd line of a lyx
document)
\lyxformat 2.16

When loading such a document in a German environment
(LANG="german"), lyx chokes on this directive and issues the
following error message:
Achtung: Benötige LyX-Format 2,16, die gewählte Datei enthält 2,00
[In English: Warning: need lyxformat 2,16 but found 2,00]

[BTW: Translating Warning with Achtung sucks - A Warning is a
"Warnung" in German - Achtung means "attention".]

When saving a lyx document in a German environment (LANG="german"),
this lyxformat is written to the file:
\lyxformat 2,16

Loading a lyx document containing \lyxformat 2,16 in any environment
gives the same warning as above.

AFAIS, lyx is trying to write the lyxformat's value as floats
(buffer.C and lyxlex.C), but doesn't consider that atof etc. are
subject to internationalization.

Running the tiny sample below in German, C and other environments
should demonstrate this issue.

#include locale.h
#include stdio.h
#include stdlib.h

int main()
{
  float a = 2.16 ;
  float b  ;
  char *lang = getenv("LANG");
  setlocale(LC_ALL,lang);
  b = atof( "2.17" );
  fprintf( stdout, "%f %f\n", a, b );
  return 0;
} 
 

Ralf

-- 
Ralf Corsepius 
Forschungsinstitut fuer Anwendungsorientierte Wissensverarbeitung
(FAW)
Helmholtzstr. 16, 89081 Ulm, Germany Tel: +49/731/501-8690
mailto:[EMAIL PROTECTED]   FAX: +49/731/501-999  
http://www.faw.uni-ulm.de

Re: Core dump on startup revistited

[Angus Leeming]

Compiled again now that Jean-Marc has submitted Dekel's combox patch. Can 
report that all is fine. No core dumps. Lyx is working again ;-)

The question still remains however: why are the last few lines of language.C 
needed?

static
LangInit langinit;

bool LangInit::init = false;

Angus

Small fixes to please picky C++ compiler

[Jean-Marc Lasgouttes]

Hello,

While compiling LyX in fussy mode of compaq cxx (std strict_ansi -w0),
the following warnings got printed:

| cxx: Info: ../sigc++/basic_signal.h, line 139: destructor for base
|   class "SigC::Signal_" is not virtual (D:basclsnondto)
|   detected during instantiation of class "SigC::Signal0void, Marsh
| [with Marsh=SigC::Marshalvoid]" at line 76 of
| "../../lyx-devel/src/insets/inseterror.h"
|   :public Signal_
| --^

| cxx: Info: ../sigc++/slot.h, line 410: destructor for base class
|   "SigC::HandleSigC::SlotData, SigC::SlotExtend" is not virtual
|   (D:basclsnondto)
|   detected during:
| instantiation of class "SigC::Slot1void, P1 [with P1=const
|   std::vectorstd::string, std::allocatorstd::string ]"
|   at line 284 of "../sigc++/basic_signal.h"
| instantiation of class "SigC::Signal1void, P1, Marsh [with
|   P1=const std::vectorstd::string,
|   std::allocatorstd::string ,
|   Marsh=SigC::Marshalvoid]" at line 79 of
|   "../../lyx-devel/src/frontends/Dialogs.h"
|:public HandleSlotData,SlotExtend
| ---^

They are fixed by adding virtual destructors to Signal_ and Handle
classes. The (trivial) patch is attached.

Also, when compiling with gcc 2.8.1, I get plenty of
../../sigc++/slot.h: In method `void Callback0void::call()':
../../sigc++/slot.h:260: warning: `return' with a value, in function returning void

This is harmless, but annoying nevertheless.

Hope this helps.

JMarc

Index: handle.h
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/sigc++/handle.h,v
retrieving revision 1.2
diff -u -r1.2 handle.h
--- handle.h2000/06/12 11:55:08 1.2
+++ handle.h2000/09/29 12:49:09
@@ -84,6 +84,10 @@
 Obj *o=handle.obj();
 scope_.set(o,o,false);
}
+
+ // virtual destructor
+ virtual ~Handle() {}
+ 
   };
 
 #define HANDLE_CTORS(X,T,P)  \
Index: macros/basic_signal.h.m4
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/sigc++/macros/basic_signal.h.m4,v
retrieving revision 1.3
diff -u -r1.3 basic_signal.h.m4
--- macros/basic_signal.h.m42000/08/10 04:54:00 1.3
+++ macros/basic_signal.h.m42000/09/29 12:49:09
@@ -61,7 +61,7 @@
   SlotData* out_connect(SlotData *s);
 
   Signal_();
-  ~Signal_();
+  virtual ~Signal_();
 
 public:
   bool empty() const;

Re: more cxx warnings for buffer.C

[Jean-Marc Lasgouttes]

 "Lars" == Lars Gullik Bjønnes [EMAIL PROTECTED] writes:

Lars ../../lyx-devel/src/support/block.h, line 48: conversion to |
Lars integral type of smaller size could lose data (D:intconlosbit) |
Lars detected during: | instantiation of "void blockT,
Lars s::operator=(const blockT, s ) | [with T=Bullet, s=4UL]" at
Lars line 197 of | "../../lyx-devel/src/buffer.C" | implicit
Lars generation of "BufferParams | BufferParams::operator=(const
Lars BufferParams )" at line | 197 of "../../lyx-devel/src/buffer.C"
Lars | arr[i] == b[i]; | ^

Lars I don't quite understand tis one.

Well, I do not understand templates much, but in definitions taking a
block as argument like
bool operator==(block const  b) const {
shouldn't we have
bool operator==(blockfoo,bar const  b) const {

JMarc

Re: Core dump on startup revistited

[Jean-Marc Lasgouttes]

 "Angus" == Angus Leeming [EMAIL PROTECTED] writes:

Angus Compiled again now that Jean-Marc has submitted Dekel's combox
Angus patch. Can report that all is fine. No core dumps. Lyx is
Angus working again ;-)

Angus The question still remains however: why are the last few lines
Angus of language.C needed?

Angus static LangInit langinit;

Angus bool LangInit::init = false;

It is a bool telling whether the initialization of languages has been
done already, I guess.

JMarc

Re: lyxformat/i18n

[Jean-Marc Lasgouttes]

 "Ralf" == Ralf Corsepius [EMAIL PROTECTED] writes:

Ralf Hi, When saving a lyx document in a "C" environment (LANG="C"),
Ralf lyx-1.1.5cvs saves its \lyxformat like this (2nd line of a lyx
Ralf document) \lyxformat 2.16

Ralf When loading such a document in a German environment
Ralf (LANG="german"), lyx chokes on this directive and issues the
Ralf following error message: Achtung: Benötige LyX-Format 2,16, die
Ralf gewählte Datei enthält 2,00 [In English: Warning: need lyxformat
Ralf 2,16 but found 2,00]

Hello, Thanks for the report. We have a fix for that in the main tree,
and I'll try to port it to 1.1.5fix2 too.

Small question: does adding setlocale(LC_NUMERIC, "C") in you program
fix the problem? If it does, I guess LyX should just use that.

JMarc

Re: more cxx warnings for buffer.C

[Lars Gullik Bjønnes]

Jean-Marc Lasgouttes [EMAIL PROTECTED] writes:


| Well, I do not understand templates much, but in definitions taking a
| block as argument like
| bool operator==(block const  b) const {
| shouldn't we have
| bool operator==(blockfoo,bar const  b) const {

No, that should not be needed.
Does the warning go away if you do that?

One other thing is that operator== should be moved out of the class.

Lgb

Re: Core dump on startup revistited

[Angus Leeming]

 Angus The question still remains however: why are the last few lines
 Angus of language.C needed?

 Angus static LangInit langinit;
 Angus bool LangInit::init = false;

 It is a bool telling whether the initialization of languages has been
 done already, I guess.

Sorry, I didn't make myself clear. What I meant was, why does this static 
variable exist at all? It has scope only within the file and does nothing 
here other than initialise class static variable LangInit::init to true. 
Thereafter, this variable is reset to false. Why can't we get rid of 
"static LangInit langinit;" ?

Or am I misreading this code?

Angus

Re: more cxx warnings for buffer.C

[Jean-Marc Lasgouttes]

 "Lars" == Lars Gullik Bjønnes [EMAIL PROTECTED] writes:

Lars Jean-Marc Lasgouttes [EMAIL PROTECTED] writes: |
Lars Well, I do not understand templates much, but in definitions
Lars taking a | block as argument like | bool operator==(block const
Lars  b) const { | shouldn't we have | bool
Lars operator==(blockfoo,bar const  b) const {

Lars No, that should not be needed. Does the warning go away if you
Lars do that?

No, but I do not know what I am supposed to put there...

Lars One other thing is that operator== should be moved out of the
Lars class.

I meant to show the operator=, in fact.

JMarc

Re: Core dump on startup revistited

[Jean-Marc Lasgouttes]

 "Angus" == Angus Leeming [EMAIL PROTECTED] writes:

Angus The question still remains however: why are the last few lines
Angus of language.C needed?

Angus static LangInit langinit; bool LangInit::init = false;
  It is a bool telling whether the initialization of languages has
 been done already, I guess.

Angus Sorry, I didn't make myself clear. What I meant was, why does
Angus this static variable exist at all? It has scope only within the
Angus file and does nothing here other than initialise class static
Angus variable LangInit::init to true. Thereafter, this variable is
Angus reset to false. Why can't we get rid of "static LangInit
Angus langinit;" ?

I have to admit that I do not know, in fact. Lars?

JMarc

Re: Core dump on startup revistited

[Lars Gullik Bjønnes]

Angus Leeming [EMAIL PROTECTED] writes:

|  Angus The question still remains however: why are the last few lines
|  Angus of language.C needed?
| 
|  Angus static LangInit langinit;
|  Angus bool LangInit::init = false;
| 
|  It is a bool telling whether the initialization of languages has been
|  done already, I guess.
| 
| Sorry, I didn't make myself clear. What I meant was, why does this static 
| variable exist at all? It has scope only within the file and does nothing 
| here other than initialise class static variable LangInit::init to true. 
| Thereafter, this variable is reset to false. Why can't we get rid of 
| "static LangInit langinit;" ?

It make the LangInit::LangInit constructor be run before main is
started by the runtime environment.

The errors you are seeing your be a problem with ordering of
initializatoin of static global variables.

Feel free to rewrite this is a lazy singleton class.

Lgb

Re: lyxformat/i18n

[Ralf Corsepius]

Jean-Marc Lasgouttes wrote:
 
  "Ralf" == Ralf Corsepius [EMAIL PROTECTED] writes:
 
 Ralf Hi, When saving a lyx document in a "C" environment (LANG="C"),
 Ralf lyx-1.1.5cvs saves its \lyxformat like this (2nd line of a lyx
 Ralf document) \lyxformat 2.16
 
 Ralf When loading such a document in a German environment
 Ralf (LANG="german"), lyx chokes on this directive and issues the
 Ralf following error message: Achtung: Benötige LyX-Format 2,16, die
 Ralf gewählte Datei enthält 2,00 [In English: Warning: need lyxformat
 Ralf 2,16 but found 2,00]
 
 Hello, Thanks for the report. We have a fix for that in the main tree,
 and I'll try to port it to 1.1.5fix2 too.
 
 Small question: does adding setlocale(LC_NUMERIC, "C") in you program
 fix the problem?
It does.

 If it does, I guess LyX should just use that.
 
Yes, unless it has side effects somewhere else, this might be a
feasible workaround.

Ralf

-- 
Ralf Corsepius 
Forschungsinstitut fuer Anwendungsorientierte Wissensverarbeitung
(FAW)
Helmholtzstr. 16, 89081 Ulm, Germany Tel: +49/731/501-8690
mailto:[EMAIL PROTECTED]   FAX: +49/731/501-999  
http://www.faw.uni-ulm.de

Re: more cxx warnings for buffer.C

[Lars Gullik Bjønnes]

Jean-Marc Lasgouttes [EMAIL PROTECTED] writes:

|  "Lars" == Lars Gullik Bjønnes [EMAIL PROTECTED] writes:
| 
| Lars Jean-Marc Lasgouttes [EMAIL PROTECTED] writes: |
| Lars Well, I do not understand templates much, but in definitions
| Lars taking a | block as argument like | bool operator==(block const
| Lars  b) const { | shouldn't we have | bool
| Lars operator==(blockfoo,bar const  b) const {
| 
| Lars No, that should not be needed. Does the warning go away if you
| Lars do that?
| 
| No, but I do not know what I am supposed to put there...

blockT, s  operator=(blockT, s const  b) {
for (size_t i = 0; i  size(); ++i) {
arr[i] == b[i];
}
return *this;
}

perhaps

template class S, size_t l
blockT, s  operator=(blockS, l const  b) {
Assert(b.size() == size());
for (size_t i = 0; i  size(); ++i) {
arr[i] == b[i];
}
return *this;
}

| 
| Lars One other thing is that operator== should be moved out of the
| Lars class.
| 
| I meant to show the operator=, in fact.

ok.

Lgb

Patch for 1.1.5fix

[Dekel Tsur]

Fix a problem with LTR table floats (the fix is not needed in 1.1.6 due to
the change in table code).

 patch.gz

Re: Patch for 1.1.5fix

[Jean-Marc Lasgouttes]

 "Dekel" == Dekel Tsur [EMAIL PROTECTED] writes:

Dekel Fix a problem with
Dekel LTR table floats (the fix is not needed in 1.1.6 due to the
Dekel change in table code).

Thanks. Applied.

JMarc

Re: NoGUI LyX

[Nicolas Thiery]

Hello,

 as people pointed out last time I suggested such, like bugzilla, this
 would require people with a knowledge of the codebase to adminstrate and
 triage, prioritise etc.
 
 BUT this would be real advantage if people actually use it, especially for
 tracking developer's/user's comments on the bugs ...
 
 I would offer to take such a responsibility myself but I really don't know
 the code well enough to judge !

SourceForge.net is providing support (CVS repository, anonymous ftp
site, Web page, Bug tracking system, mailing lists, forums ) for any
open-source project. I use it for my own projects. At least from the
purely technical point of view, setting up a bug-tracking system for
lyx out there would probably be a matter of minutes, + 1 day for them
to process the query. By the way, releasing the rpm's out there would
make them easily available since they are immediately added to the
database on www.rpmfind.net.

Have a good week-end,
Nicolas

PS: and since the thread is about GUIless export, oh yes this would be
a great feature!

-- 
Nicolas M. Thiéry "Isil", 412 Washington Avenue, 80403 Golden Colorado (USA)
Mél: [EMAIL PROTECTED], Tél: (303)273-5492, Fax: (303)273-3875
WWW: URL:http://www.mines.edu/~nthiery/

 PGP signature

Re: NoGUI LyX

[John Levon]

On Fri, 29 Sep 2000, Nicolas Thiery wrote:

 SourceForge.net is providing support (CVS repository, anonymous ftp
 site, Web page, Bug tracking system, mailing lists, forums ) for any
 open-source project. I use it for my own projects. At least from the
 purely technical point of view, setting up a bug-tracking system for
 lyx out there would probably be a matter of minutes, + 1 day for them
 to process the query. By the way, releasing the rpm's out there would
 make them easily available since they are immediately added to the
 database on www.rpmfind.net.

well indeed the set up isn't the problem, but the maintenance. If the bug
database becomes out of sync with the codebase, it's practically useless
... there are not enough volunteers to maintain such :(

john

-- 
"I'll tell you what justice is. Justice is a knee in the gut from the floor on the 
chin at
 night sneaky with a knife brought up down on the magazine of a battleship sand-bagged
 underhanded in the dark without a word of warning. Garroting."
- The bloated colonel with the big fat moustache

Various suggestions

[Nicolas Thiery]

Hello,

Here is a bunch of suggestions I collected other the last few months,
and be lazy to send earlier.

I will first describe my use of LyX at it is probably unusual, and
some of my suggestions are related to this. I have a very bad hand
writing, so I display my class notes during the class using my laptop
and a LCD projector. I also post the notes on the web in different
format (html, dvi, ps, text, PDF). The traditional way would be to
produce slides with latex/lyx compile them, and display them with gs
for example. But this is not enough for me, since I want to make minor
fixes to the notes during the class, or to add, say, results of
exercises we work on. Well, I simply display my notes directly with
LyX. And it works almost perfectly, LyX does an excellent job as an
informal presentation tool. It just needs to get used to scroll
through the notes, instead of switching from one slide to the other as
in a usual presentation.

Thanks so much for this great versatile tool!

For reference: http://www.mines.edu/Academic/courses/math_cs/macs358/

 - In math equations, \circ is displayed on the screen as an unknown
   macro, instead of being drawn. So far I am using \bullet instead
   but it's ugly. I need it in three weeks for my class on composition
   of functions :-)

 - I am missing the following starred environments with amsmath:
   notation*, problem*, ... Are they non-standard ?

 - When entering emphasized text, striking right arrow at the end of the
   line should allow to exit the emphasized text.

 - Bug in the users guide: it says LyX uses \epsfig, whereas it really
   uses includegraphics to import figures (which is The Right Thing).

 - Could the file browser for including graphics display thumbnails of
   the figure as, say, in the xfig file browser ?

 - LyX support for "\DeclareGraphicsExtensions" ?

   To be able to generate both postscript and PDF documents, I have my
   graphics files in both .eps and .pdf format. My problem is that if
   I specify a graphics file without an extension "bla", lyx does not
   find it. On the other hand, if I specify the full file name
   "bla.eps", the compilation with pdflatex does not work.

   What I currently do is to specify the full name, so that the figure
   appears correctly during the presentation, and then I have use
   little perl hack to remove the extension from the exported latex
   file before running pdflatex on it. It works, but yuck

   One way to get around this would be to have LyX support the
   standard latex graphics extensions mechanism. Specifically when the
   name of the figure is specified without extensions, LyX would try
   the standard extensions in order.

 - LyX support for "\graphicspath"

   It would be nice to have LyX support the graphicspath latex
   feature. Right now, LyX does not find the figure do display it in
   the LyX buffer. More important, the compilation fails. On the other
   hand, exporting as latex, and compiling by hand works. It seems to
   be related to the compilation in the temporary directory.

   I attached to this email an tar.gz with an example.

 - With the Format-Character menu, clicking on apply switch back and
   forth between the original and the modified versions. Why not.
   However, it's counter intuitive that clicking on OK switches back
   to the original !

 - Support for the hyperref package would be cool.

 - Could  LyX-Code be defined in amsmath/amsbook classes ?

 - My class notes are split up, with one file per chapter. I want to
   be able to compile each chapter separately, as well as to compile
   the full document. The problem is that I have some math-macros that
   are used in several chapters at the same time. To be able to
   compile each chapter separately, I need to define them at the
   beginning of each chapter. But then, when I compile the full
   document, LaTeX complains about multiply defined macros.

   What would be the best way to get around this ?

 - For my use of LyX as presentation tool, I need to be able to
   quickly scroll the text to some precise position, so as to display
   the precise part of the document I want, and hide the solutions of
   the exercises. Moving the cursor around does not work well, since
   it makes jumps if there are figures or big equations. Right now I
   use the scrollbar. I would prefer to have keyboard shortcuts (Say
   Ctrl-Up arrow and Ctrl-down arrow, for scrolling up or down by a
   fixed amount (e.g. 1cm). That would save me from looking once again
   foolish trying to figure out where the mouse cursor disappeared.

 - I use many different paragraph styles, and had to define many
   shortcuts of them. I would find more practical to have a unique
   shortcut, with name completion.

   Example: to get to the "Problem" paragraph style, hit M-p PrTAB.
   It could be on some other shortcut than M-p, to avoid conflicts.

   Also, when using the popout paragraph style menu, PgUp and PgDown
   should probably also 

Re: Working on a Word to LyX importer

[Duncan Simpson]

word2x is designed to be retargetable, so adding another output format is not 
a major problem and neither is chaning or replacing it later. If you want to 
do the same in wv it is probably a lot harder (last time there is no clean 
separation between the front and back end in wv).

If some C++ master could turn my OLE acrhive support into OLE streams as C++ 
ifstreams then it would be a huge boost for word2x 2,0 (and help word2x 0.00x, 
and no doubt tools for other OLE writing applications---most prominently the 
rest of M$ office).

-- 
Duncan (-:
"software industry, the: unique industry where selling substandard goods is
legal and you can charge extra for fixing the problems."

Re: Various suggestions

[Nicolas Thiery]

Oops, I forgot the archive. Here it is.

Nicolas
-- 
Nicolas M. Thiéry "Isil", 412 Washington Avenue, 80403 Golden Colorado (USA)
Mél: [EMAIL PROTECTED], Tél: (303)273-5492, Fax: (303)273-3875
WWW: URL:http://www.mines.edu/~nthiery/

 lyx-graphicspath-bug.tgz
 PGP signature

Thesis submission time

[John Levon]

It's thesis submission time here, and all those who made the mistake of
not using LaTeX/LyX are desparately trying to fix the broken output
they've got.

In contrast my output was perfect first time, flawless, and I could focus
on fixing the actual content.

So once again thanks all for a great program ! (and I will be working on
it again soon when I have a working C++ compiler at home !)

thanks
john
 
-- 
"The words CON and NUL
Are reserved by the system.
Don't ask; they just are."

Re: Working on a Word to LyX importer

[Andre Poenitz]

> My work would go faster if I had a file describing the equivalent of LateX
> codes in  LyX codes [for example $\alpha$ is \begin_inset Formula \( \alpha  \)
> \end_inset ]. Does this document exists somewhere ?

None that I know of.

I think the LyX format is likely to change pretty fast, so it'd probably
better to improve any existing Word2LaTeX converter and work on a direct
LaTeX import to LyX or even on a direct Word importer for LyX. 

LyX's internals have improved significantly over the last months, so the
poor state of the external .lyx format does not reflect the current state.

Supporting the current .lyx format by something important as an importer
from Word would prabably hinder changes in this area.

That's my just personal opinion of course. 

Andre'

-- 
Andre' Poenitz  [EMAIL PROTECTED]

instant core dump

[Angus Leeming]

Running Current CVS results in an instant core dump on my machine (DEC 
Alpha). Does anybody else have this experience? If not, then I guess that 
this is something to do with these size_type changes (and all to remove 
warnings on the DEC!) Lars, do you have anymore changes to submit soon, or 
shall I investigate the code as is?

Angus

Re: src/ext_l10n.h

[Jean-Marc Lasgouttes]

> "John" == John Levon <[EMAIL PROTECTED]> writes:

John> Er, whoops I added this to .cvsignore but I've just realised
John> shouldn't this be removed from CVS altogether ?

John> It is generated from autogen.sh so shouldn't it be removed from
John> the CVS like configure etc.

You are right. I just did it.

JMarc

Re: compiling 1.1.5fix1 with DEC cxx and lyxstring

[Jean-Marc Lasgouttes]

> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:

Angus> Any ideas on the way forward, Jean-Marc? Everything works fine
Angus> with std::string, but with lyxstring I get a whole pile of
Angus> errors

All I can say is that this code has been changed in 6.2, which uses
std::string correctly.

JMarc

Re: NoGUI LyX

[Jean-Marc Lasgouttes]

> "Baruch" == Baruch Even <[EMAIL PROTECTED]> writes:

Baruch> On 27 Sep 2000, Jean-Marc Lasgouttes wrote:
>> and fix session (I have several bug reports in my mail folder which
>> depend on this).

Baruch> How about updating a known bug list and a todo list on the
Baruch> developers website for all to see?

Baruch> It will help development since we'll all be able to know what
Baruch> bugs are pending and what urgent todo stuff is waiting. It
Baruch> will also help new developers to find small tasks that they
Baruch> can start with.

Even better would be a real bug-tracking database...

Concerning bugs I am aware of, you can find some of them in my last
posting about 1.1.5fix2.

JMarc

Re: Bug in CVS

[Jean-Marc Lasgouttes]

> "Dekel" == Dekel Tsur <[EMAIL PROTECTED]> writes:

Dekel> But why sel is 0 ? The problem is in Intl::InitKeyMapper: Since
Dekel> the "default" language is removed, n should be initialized to
Dekel> 0. I've attached a patch that does both fixes.
   
Applied.

JMarc

Re: [PATCH] Complain about Qt2

[Jean-Marc Lasgouttes]

> "John" == John Levon <[EMAIL PROTECTED]> writes:

John> This patch changes
John> the kde m4 stuff so that if Qt2 is being picked up, configure
John> will stop at the Qt setup rather than failing at the KDE
John> compile, which is misleading.

Applied.

JMarc

Re: NoGUI LyX

[John Levon]

On 29 Sep 2000, Jean-Marc Lasgouttes wrote:

> Even better would be a real bug-tracking database...
> 
> JMarc
> 

as people pointed out last time I suggested such, like bugzilla, this
would require people with a knowledge of the codebase to adminstrate and
triage, prioritise etc.

BUT this would be real advantage if people actually use it, especially for
tracking developer's/user's comments on the bugs ...

I would offer to take such a responsibility myself but I really don't know
the code well enough to judge !

john

-- 
"The words CON and NUL
Are reserved by the system.
Don't ask; they just are."

Re: instant core dump

[Jean-Marc Lasgouttes]

> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:

Angus> Running Current CVS results in an instant core dump on my
Angus> machine (DEC Alpha). Does anybody else have this experience? If
Angus> not, then I guess that this is something to do with these
Angus> size_type changes (and all to remove warnings on the DEC!)
Angus> Lars, do you have anymore changes to submit soon, or shall I
Angus> investigate the code as is?

At least you are able to compile... OK, I just commit the patch from
Dekel to cure a coredump releated to combox (please try it out, since
I can't compile). I also added the Qt2 patch.

My compile problems are related to Allan latest change, so you may be
able to help here:

mycxx -DHAVE_CONFIG_H -I. -I../../../../lyx-devel/src/frontends/xforms -I../../../src 
-I../../../../lyx-devel/src/ -I../../../../lyx-devel/src/frontends/ -I../../.. 
-I../../../../lyx-devel -I../../../../lyx-devel/src/cheaders 
-I/afs/rocq/home/meval/common/include -I/usr/local/include -g0 
-Wp,-MD,.deps/FormBase.pp -c ../../../../lyx-devel/src/frontends/xforms/FormBase.C -o 
FormBase.o
cxx: Error: ../../../../lyx-devel/src/frontends/xforms/FormBase.C, line 96: 
  "SigC::Handle::operator Obj *() [with
  Obj=SigC::SlotData, Scope_=SigC::SlotExtend]" is inaccessible
if (u_) {
^

What does if(u_) mean when u_ is a Connection? I guess it should be
replaced with something else.

JMarc

Core dump on startup revistited

[Angus Leeming]

I get a core dump instantly when starting up the current CVS lyx. 

I have tracked down the problem to language.C's LangInit::initL() and the 
line:

if (languages.find(lyxrc.default_language) == languages.end()) {

The problem lies with lyxrc.default_language. Modifying the source slightly 
to:


std::cerr << "got to here 2" << std::endl;
std::cerr << "lyxrc.default_language " << lyxrc.default_language << 
std::endl;
if (languages.find(lyxrc.default_language) == languages.end()) {

gdb is not much help:
(gdb) r
Starting program: /usr/users/aleem/OTHERS_CODE/lyx/devel/src/lyx
got to here 1
got to here 2
lyxrc.default_language Memory fault(coredump) 
  
But dbx does better:
(dbx) r
got to here 1
got to here 2
lyxrc.default_language signal Segmentation fault at   
[length__CQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__TvXv:1703 
+0x8,0x1200e6034]
(dbx) where
>  0 
length__CQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__TvXv(this 
= 0x14006cf90) ["/usr/include/cxx/string~alt~deccxx_17E4996F":1703, 
0x1200e6034]
   1 
size__CQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__TvXv(this 
= 0x14006cf90) ["/usr/include/cxx/string~alt~deccxx_17E4996F":852, 
0x1200e5ffc]
   2 
__ls__3stdXRQ13std40basic_ostream__TcQ13std15char_traits__TcRCQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__Tv(os
 
= @1400704d8, s = @14006cf90) ["/usr/include/cxx/ostream.cc":1720, 
0x1200e6068]
   3 initL__8LangInitXv(this = 0x14006a760) ["language.C":131, 0x1201aeba0]
   4 __ct__8LangInitXv(this = 0x14006a760) ["language.C":34, 0x1201add2c]
   5 __init_sti__language_C_a0d15379() ["language.C":143, 0x1201aef08]
(dbx) quit
 

First, can someone explain the last few lines of language.C:
static
LangInit langinit;

bool LangInit::init = false;

Commenting them out allowed lyx to start, but loading the first document 
causes a similar problem:

(gdb) r
Starting program: /usr/users/aleem/OTHERS_CODE/lyx/devel/src/lyx
 
Program received signal SIGSEGV, Segmentation fault.
0x12031faa0 in 
__as__Q13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__TvXRCQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__Tv
 
()
(gdb) where
#0  0x12031faa0 in 
__as__Q13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__TvXRCQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__Tv
 
()
#1  0x12018ffb8 in __ct__12BufferParamsXv ()
#2  0x12017a36c in 
__ct__6BufferXRCQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__Tvjb 
()
#3  0x12018cd2c in 
newBuffer__13BufferStorageXRCQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__Tvjb
 
()
#4  0x12018e1dc in 
readFile__10BufferListXRCQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__Tvjb
 
()
#5  0x12018eff4 in 
loadLyXFile__10BufferListXRCQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__Tvjb
 
()
#6  0x1201d6a30 in 
Dispatch__7LyXFuncXiRCQ13std60basic_string__TcQ13std15char_traits__TcQ13std13allocator__Tv
 
()
#7  0x1202c7af0 in MenuCallback__Q17Menubar5PimplXP7flobjs_l ()
#8  0x1202c784c in C_Menubar_Pimpl_MenuCallback ()
#9  0x3ffbff3e87c in fl_object_qread () from /usr/local/lib/libforms.so.0.89
#10 0x3ffbff538e0 in fl_check_forms () from /usr/local/lib/libforms.so.0.89
#11 0x1202c16bc in runTime__10GUIRunTimeXv ()
#12 0x1201c088c in runTime__6LyXGUIXv ()
#13 0x1201c3394 in __ct__3LyXXPiPPc ()
#14 0x1201f7e0c in main ()
(gdb) quit
  
Any ideas?
Angus

Re: instant core dump

[Angus Leeming]

It means nothing. Remove it, leaving
u_.disconnect();
This is legal, fine, good, compiles, works, etc!

On Fri, 29 Sep 2000, Jean-Marc Lasgouttes wrote:
> > "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
>
> Angus> Running Current CVS results in an instant core dump on my
> Angus> machine (DEC Alpha). Does anybody else have this experience? If
> Angus> not, then I guess that this is something to do with these
> Angus> size_type changes (and all to remove warnings on the DEC!)
> Angus> Lars, do you have anymore changes to submit soon, or shall I
> Angus> investigate the code as is?
>
> At least you are able to compile... OK, I just commit the patch from
> Dekel to cure a coredump releated to combox (please try it out, since
> I can't compile). I also added the Qt2 patch.
>
> My compile problems are related to Allan latest change, so you may be
> able to help here:
>
> mycxx -DHAVE_CONFIG_H -I. -I../../../../lyx-devel/src/frontends/xforms
> -I../../../src -I../../../../lyx-devel/src/
> -I../../../../lyx-devel/src/frontends/ -I../../.. -I../../../../lyx-devel
> -I../../../../lyx-devel/src/cheaders -I/afs/rocq/home/meval/common/include
> -I/usr/local/include -g0 -Wp,-MD,.deps/FormBase.pp -c
> ../../../../lyx-devel/src/frontends/xforms/FormBase.C -o FormBase.o cxx:
> Error: ../../../../lyx-devel/src/frontends/xforms/FormBase.C, line 96:
> "SigC::Handle::operator Obj *() [with
>   Obj=SigC::SlotData, Scope_=SigC::SlotExtend]" is inaccessible
> if (u_) {
> ^
>
> What does if(u_) mean when u_ is a Connection? I guess it should be
> replaced with something else.
>
> JMarc

lyxformat/i18n

[Ralf Corsepius]

Hi,

When saving a lyx document in a "C" environment (LANG="C"),
lyx-1.1.5cvs saves its \lyxformat like this (2nd line of a lyx
document)
\lyxformat 2.16

When loading such a document in a German environment
(LANG="german"), lyx chokes on this directive and issues the
following error message:
Achtung: Benötige LyX-Format 2,16, die gewählte Datei enthält 2,00
[In English: Warning: need lyxformat 2,16 but found 2,00]

[BTW: Translating Warning with Achtung sucks - A Warning is a
"Warnung" in German - Achtung means "attention".]

When saving a lyx document in a German environment (LANG="german"),
this lyxformat is written to the file:
\lyxformat 2,16

Loading a lyx document containing \lyxformat 2,16 in any environment
gives the same warning as above.

AFAIS, lyx is trying to write the lyxformat's value as floats
(buffer.C and lyxlex.C), but doesn't consider that atof etc. are
subject to internationalization.

Running the tiny sample below in German, C and other environments
should demonstrate this issue.

#include 
#include 
#include 

int main()
{
  float a = 2.16 ;
  float b  ;
  char *lang = getenv("LANG");
  setlocale(LC_ALL,lang);
  b = atof( "2.17" );
  fprintf( stdout, "%f %f\n", a, b );
  return 0;
} 
 

Ralf

-- 
Ralf Corsepius 
Forschungsinstitut fuer Anwendungsorientierte Wissensverarbeitung
(FAW)
Helmholtzstr. 16, 89081 Ulm, Germany Tel: +49/731/501-8690
mailto:[EMAIL PROTECTED]   FAX: +49/731/501-999  
http://www.faw.uni-ulm.de

Re: Core dump on startup revistited

[Angus Leeming]

Compiled again now that Jean-Marc has submitted Dekel's combox patch. Can 
report that all is fine. No core dumps. Lyx is working again ;-)

The question still remains however: why are the last few lines of language.C 
needed?

static
LangInit langinit;

bool LangInit::init = false;

Angus

Small fixes to please picky C++ compiler

[Jean-Marc Lasgouttes]

Hello,

While compiling LyX in fussy mode of compaq cxx (std strict_ansi -w0),
the following warnings got printed:

| cxx: Info: ../sigc++/basic_signal.h, line 139: destructor for base
|   class "SigC::Signal_" is not virtual (D:basclsnondto)
|   detected during instantiation of class "SigC::Signal0
| [with Marsh=SigC::Marshal]" at line 76 of
| "../../lyx-devel/src/insets/inseterror.h"
|   :public Signal_
| --^

| cxx: Info: ../sigc++/slot.h, line 410: destructor for base class
|   "SigC::Handle" is not virtual
|   (D:basclsnondto)
|   detected during:
| instantiation of class "SigC::Slot1 [with P1=const
|   std::vector &]"
|   at line 284 of "../sigc++/basic_signal.h"
| instantiation of class "SigC::Signal1 [with
|   P1=const std::vector &,
|   Marsh=SigC::Marshal]" at line 79 of
|   "../../lyx-devel/src/frontends/Dialogs.h"
|:public Handle
| ---^

They are fixed by adding virtual destructors to Signal_ and Handle
classes. The (trivial) patch is attached.

Also, when compiling with gcc 2.8.1, I get plenty of
../../sigc++/slot.h: In method `void Callback0::call()':
../../sigc++/slot.h:260: warning: `return' with a value, in function returning void

This is harmless, but annoying nevertheless.

Hope this helps.

JMarc

Index: handle.h
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/sigc++/handle.h,v
retrieving revision 1.2
diff -u -r1.2 handle.h
--- handle.h2000/06/12 11:55:08 1.2
+++ handle.h2000/09/29 12:49:09
@@ -84,6 +84,10 @@
 Obj *o=handle.obj();
 scope_.set(o,o,false);
}
+
+ // virtual destructor
+ virtual ~Handle() {}
+ 
   };
 
 #define HANDLE_CTORS(X,T,P)  \
Index: macros/basic_signal.h.m4
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/sigc++/macros/basic_signal.h.m4,v
retrieving revision 1.3
diff -u -r1.3 basic_signal.h.m4
--- macros/basic_signal.h.m42000/08/10 04:54:00 1.3
+++ macros/basic_signal.h.m42000/09/29 12:49:09
@@ -61,7 +61,7 @@
   SlotData* out_connect(SlotData *s);
 
   Signal_();
-  ~Signal_();
+  virtual ~Signal_();
 
 public:
   bool empty() const;

Re: more cxx warnings for buffer.C

[Jean-Marc Lasgouttes]

> "Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes:

Lars> ../../lyx-devel/src/support/block.h, line 48: conversion to |
Lars> integral type of smaller size could lose data (D:intconlosbit) |
Lars> detected during: | instantiation of "void block s>::operator=(const block &) | [with T=Bullet, s=4UL]" at
Lars> line 197 of | "../../lyx-devel/src/buffer.C" | implicit
Lars> generation of "BufferParams | ::operator=(const
Lars> BufferParams &)" at line | 197 of "../../lyx-devel/src/buffer.C"
Lars> | arr[i] == b[i]; | ^

Lars> I don't quite understand tis one.

Well, I do not understand templates much, but in definitions taking a
block as argument like
bool operator==(block const & b) const {
shouldn't we have
bool operator==(block const & b) const {

JMarc

Re: Core dump on startup revistited

[Jean-Marc Lasgouttes]

> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:

Angus> Compiled again now that Jean-Marc has submitted Dekel's combox
Angus> patch. Can report that all is fine. No core dumps. Lyx is
Angus> working again ;-)

Angus> The question still remains however: why are the last few lines
Angus> of language.C needed?

Angus> static LangInit langinit;

Angus> bool LangInit::init = false;

It is a bool telling whether the initialization of languages has been
done already, I guess.

JMarc

Re: lyxformat/i18n

[Jean-Marc Lasgouttes]

> "Ralf" == Ralf Corsepius <[EMAIL PROTECTED]> writes:

Ralf> Hi, When saving a lyx document in a "C" environment (LANG="C"),
Ralf> lyx-1.1.5cvs saves its \lyxformat like this (2nd line of a lyx
Ralf> document) \lyxformat 2.16

Ralf> When loading such a document in a German environment
Ralf> (LANG="german"), lyx chokes on this directive and issues the
Ralf> following error message: Achtung: Benötige LyX-Format 2,16, die
Ralf> gewählte Datei enthält 2,00 [In English: Warning: need lyxformat
Ralf> 2,16 but found 2,00]

Hello, Thanks for the report. We have a fix for that in the main tree,
and I'll try to port it to 1.1.5fix2 too.

Small question: does adding setlocale(LC_NUMERIC, "C") in you program
fix the problem? If it does, I guess LyX should just use that.

JMarc

Re: more cxx warnings for buffer.C

[Lars Gullik Bjønnes]

Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes:


| Well, I do not understand templates much, but in definitions taking a
| block as argument like
| bool operator==(block const & b) const {
| shouldn't we have
| bool operator==(block const & b) const {

No, that should not be needed.
Does the warning go away if you do that?

One other thing is that operator== should be moved out of the class.

Lgb

Re: Core dump on startup revistited

[Angus Leeming]

> Angus> The question still remains however: why are the last few lines
> Angus> of language.C needed?
>
> Angus> static LangInit langinit;
> Angus> bool LangInit::init = false;
>
> It is a bool telling whether the initialization of languages has been
> done already, I guess.

Sorry, I didn't make myself clear. What I meant was, why does this static 
variable exist at all? It has scope only within the file and does nothing 
here other than initialise class static variable LangInit::init to true. 
Thereafter, this variable is reset to false. Why can't we get rid of 
"static LangInit langinit;" ?

Or am I misreading this code?

Angus

Re: more cxx warnings for buffer.C

[Jean-Marc Lasgouttes]

> "Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes:

Lars> Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes: |
Lars> Well, I do not understand templates much, but in definitions
Lars> taking a | block as argument like | bool operator==(block const
Lars> & b) const { | shouldn't we have | bool
Lars> operator==(block const & b) const {

Lars> No, that should not be needed. Does the warning go away if you
Lars> do that?

No, but I do not know what I am supposed to put there...

Lars> One other thing is that operator== should be moved out of the
Lars> class.

I meant to show the operator=, in fact.

JMarc

Re: Core dump on startup revistited

[Jean-Marc Lasgouttes]

> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:

Angus> The question still remains however: why are the last few lines
Angus> of language.C needed?
>>
Angus> static LangInit langinit; bool LangInit::init = false;
>>  It is a bool telling whether the initialization of languages has
>> been done already, I guess.

Angus> Sorry, I didn't make myself clear. What I meant was, why does
Angus> this static variable exist at all? It has scope only within the
Angus> file and does nothing here other than initialise class static
Angus> variable LangInit::init to true. Thereafter, this variable is
Angus> reset to false. Why can't we get rid of "static LangInit
Angus> langinit;" ?

I have to admit that I do not know, in fact. Lars?

JMarc

Re: Core dump on startup revistited

[Lars Gullik Bjønnes]

Angus Leeming <[EMAIL PROTECTED]> writes:

| > Angus> The question still remains however: why are the last few lines
| > Angus> of language.C needed?
| >
| > Angus> static LangInit langinit;
| > Angus> bool LangInit::init = false;
| >
| > It is a bool telling whether the initialization of languages has been
| > done already, I guess.
| 
| Sorry, I didn't make myself clear. What I meant was, why does this static 
| variable exist at all? It has scope only within the file and does nothing 
| here other than initialise class static variable LangInit::init to true. 
| Thereafter, this variable is reset to false. Why can't we get rid of 
| "static LangInit langinit;" ?

It make the LangInit::LangInit constructor be run before main is
started by the runtime environment.

The errors you are seeing your be a problem with ordering of
initializatoin of static global variables.

Feel free to rewrite this is a lazy singleton class.

Lgb

Re: lyxformat/i18n

[Ralf Corsepius]

Jean-Marc Lasgouttes wrote:
> 
> > "Ralf" == Ralf Corsepius <[EMAIL PROTECTED]> writes:
> 
> Ralf> Hi, When saving a lyx document in a "C" environment (LANG="C"),
> Ralf> lyx-1.1.5cvs saves its \lyxformat like this (2nd line of a lyx
> Ralf> document) \lyxformat 2.16
> 
> Ralf> When loading such a document in a German environment
> Ralf> (LANG="german"), lyx chokes on this directive and issues the
> Ralf> following error message: Achtung: Benötige LyX-Format 2,16, die
> Ralf> gewählte Datei enthält 2,00 [In English: Warning: need lyxformat
> Ralf> 2,16 but found 2,00]
> 
> Hello, Thanks for the report. We have a fix for that in the main tree,
> and I'll try to port it to 1.1.5fix2 too.
> 
> Small question: does adding setlocale(LC_NUMERIC, "C") in you program
> fix the problem?
It does.

> If it does, I guess LyX should just use that.
 
Yes, unless it has side effects somewhere else, this might be a
feasible workaround.

Ralf

-- 
Ralf Corsepius 
Forschungsinstitut fuer Anwendungsorientierte Wissensverarbeitung
(FAW)
Helmholtzstr. 16, 89081 Ulm, Germany Tel: +49/731/501-8690
mailto:[EMAIL PROTECTED]   FAX: +49/731/501-999  
http://www.faw.uni-ulm.de

Re: more cxx warnings for buffer.C

[Lars Gullik Bjønnes]

Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes:

| > "Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes:
| 
| Lars> Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes: |
| Lars> Well, I do not understand templates much, but in definitions
| Lars> taking a | block as argument like | bool operator==(block const
| Lars> & b) const { | shouldn't we have | bool
| Lars> operator==(block const & b) const {
| 
| Lars> No, that should not be needed. Does the warning go away if you
| Lars> do that?
| 
| No, but I do not know what I am supposed to put there...

block & operator=(block const & b) {
for (size_t i = 0; i < size(); ++i) {
arr[i] == b[i];
}
return *this;
}

perhaps

template 
block & operator=(block const & b) {
Assert(b.size() == size());
for (size_t i = 0; i < size(); ++i) {
arr[i] == b[i];
}
return *this;
}

| 
| Lars> One other thing is that operator== should be moved out of the
| Lars> class.
| 
| I meant to show the operator=, in fact.

ok.

Lgb

Patch for 1.1.5fix

[Dekel Tsur]

Fix a problem with LTR table floats (the fix is not needed in 1.1.6 due to
the change in table code).

 patch.gz

Re: Patch for 1.1.5fix

[Jean-Marc Lasgouttes]

> "Dekel" == Dekel Tsur <[EMAIL PROTECTED]> writes:

Dekel> Fix a problem with
Dekel> LTR table floats (the fix is not needed in 1.1.6 due to the
Dekel> change in table code).

Thanks. Applied.

JMarc

Re: NoGUI LyX

[Nicolas Thiery]

Hello,

> as people pointed out last time I suggested such, like bugzilla, this
> would require people with a knowledge of the codebase to adminstrate and
> triage, prioritise etc.
> 
> BUT this would be real advantage if people actually use it, especially for
> tracking developer's/user's comments on the bugs ...
> 
> I would offer to take such a responsibility myself but I really don't know
> the code well enough to judge !

SourceForge.net is providing support (CVS repository, anonymous ftp
site, Web page, Bug tracking system, mailing lists, forums ) for any
open-source project. I use it for my own projects. At least from the
purely technical point of view, setting up a bug-tracking system for
lyx out there would probably be a matter of minutes, + 1 day for them
to process the query. By the way, releasing the rpm's out there would
make them easily available since they are immediately added to the
database on www.rpmfind.net.

Have a good week-end,
Nicolas

PS: and since the thread is about GUIless export, oh yes this would be
a great feature!

-- 
Nicolas M. Thiéry "Isil", 412 Washington Avenue, 80403 Golden Colorado (USA)
Mél: [EMAIL PROTECTED], Tél: (303)273-5492, Fax: (303)273-3875
WWW: 

 PGP signature

Re: NoGUI LyX

[John Levon]

On Fri, 29 Sep 2000, Nicolas Thiery wrote:

> SourceForge.net is providing support (CVS repository, anonymous ftp
> site, Web page, Bug tracking system, mailing lists, forums ) for any
> open-source project. I use it for my own projects. At least from the
> purely technical point of view, setting up a bug-tracking system for
> lyx out there would probably be a matter of minutes, + 1 day for them
> to process the query. By the way, releasing the rpm's out there would
> make them easily available since they are immediately added to the
> database on www.rpmfind.net.

well indeed the set up isn't the problem, but the maintenance. If the bug
database becomes out of sync with the codebase, it's practically useless
... there are not enough volunteers to maintain such :(

john

-- 
"I'll tell you what justice is. Justice is a knee in the gut from the floor on the 
chin at
 night sneaky with a knife brought up down on the magazine of a battleship sand-bagged
 underhanded in the dark without a word of warning. Garroting."
- The bloated colonel with the big fat moustache

Various suggestions

[Nicolas Thiery]

Hello,

Here is a bunch of suggestions I collected other the last few months,
and be lazy to send earlier.

I will first describe my use of LyX at it is probably unusual, and
some of my suggestions are related to this. I have a very bad hand
writing, so I display my class notes during the class using my laptop
and a LCD projector. I also post the notes on the web in different
format (html, dvi, ps, text, PDF). The traditional way would be to
produce slides with latex/lyx compile them, and display them with gs
for example. But this is not enough for me, since I want to make minor
fixes to the notes during the class, or to add, say, results of
exercises we work on. Well, I simply display my notes directly with
LyX. And it works almost perfectly, LyX does an excellent job as an
informal presentation tool. It just needs to get used to scroll
through the notes, instead of switching from one slide to the other as
in a usual presentation.

Thanks so much for this great versatile tool!

For reference: http://www.mines.edu/Academic/courses/math_cs/macs358/

 - In math equations, \circ is displayed on the screen as an unknown
   macro, instead of being drawn. So far I am using \bullet instead
   but it's ugly. I need it in three weeks for my class on composition
   of functions :-)

 - I am missing the following starred environments with amsmath:
   notation*, problem*, ... Are they non-standard ?

 - When entering emphasized text, striking right arrow at the end of the
   line should allow to exit the emphasized text.

 - Bug in the users guide: it says LyX uses \epsfig, whereas it really
   uses includegraphics to import figures (which is The Right Thing).

 - Could the file browser for including graphics display thumbnails of
   the figure as, say, in the xfig file browser ?

 - LyX support for "\DeclareGraphicsExtensions" ?

   To be able to generate both postscript and PDF documents, I have my
   graphics files in both .eps and .pdf format. My problem is that if
   I specify a graphics file without an extension "bla", lyx does not
   find it. On the other hand, if I specify the full file name
   "bla.eps", the compilation with pdflatex does not work.

   What I currently do is to specify the full name, so that the figure
   appears correctly during the presentation, and then I have use
   little perl hack to remove the extension from the exported latex
   file before running pdflatex on it. It works, but yuck

   One way to get around this would be to have LyX support the
   standard latex graphics extensions mechanism. Specifically when the
   name of the figure is specified without extensions, LyX would try
   the standard extensions in order.

 - LyX support for "\graphicspath"

   It would be nice to have LyX support the graphicspath latex
   feature. Right now, LyX does not find the figure do display it in
   the LyX buffer. More important, the compilation fails. On the other
   hand, exporting as latex, and compiling by hand works. It seems to
   be related to the compilation in the temporary directory.

   I attached to this email an tar.gz with an example.

 - With the Format->Character menu, clicking on apply switch back and
   forth between the original and the modified versions. Why not.
   However, it's counter intuitive that clicking on OK switches back
   to the original !

 - Support for the hyperref package would be cool.

 - Could  LyX-Code be defined in amsmath/amsbook classes ?

 - My class notes are split up, with one file per chapter. I want to
   be able to compile each chapter separately, as well as to compile
   the full document. The problem is that I have some math-macros that
   are used in several chapters at the same time. To be able to
   compile each chapter separately, I need to define them at the
   beginning of each chapter. But then, when I compile the full
   document, LaTeX complains about multiply defined macros.

   What would be the best way to get around this ?

 - For my use of LyX as presentation tool, I need to be able to
   quickly scroll the text to some precise position, so as to display
   the precise part of the document I want, and hide the solutions of
   the exercises. Moving the cursor around does not work well, since
   it makes jumps if there are figures or big equations. Right now I
   use the scrollbar. I would prefer to have keyboard shortcuts (Say
   Ctrl-Up arrow and Ctrl-down arrow, for scrolling up or down by a
   fixed amount (e.g. 1cm). That would save me from looking once again
   foolish trying to figure out where the mouse cursor disappeared.

 - I use many different paragraph styles, and had to define many
   shortcuts of them. I would find more practical to have a unique
   shortcut, with name completion.

   Example: to get to the "Problem" paragraph style, hit M-p Pr.
   It could be on some other shortcut than M-p, to avoid conflicts.

   Also, when using the popout paragraph style menu, PgUp and PgDown
   should probably also move 

Re: Working on a Word to LyX importer

[Duncan Simpson]

word2x is designed to be retargetable, so adding another output format is not 
a major problem and neither is chaning or replacing it later. If you want to 
do the same in wv it is probably a lot harder (last time there is no clean 
separation between the front and back end in wv).

If some C++ master could turn my OLE acrhive support into OLE streams as C++ 
ifstreams then it would be a huge boost for word2x 2,0 (and help word2x 0.00x, 
and no doubt tools for other OLE writing applications---most prominently the 
rest of M$ office).

-- 
Duncan (-:
"software industry, the: unique industry where selling substandard goods is
legal and you can charge extra for fixing the problems."

Re: Various suggestions

[Nicolas Thiery]

Oops, I forgot the archive. Here it is.

Nicolas
-- 
Nicolas M. Thiéry "Isil", 412 Washington Avenue, 80403 Golden Colorado (USA)
Mél: [EMAIL PROTECTED], Tél: (303)273-5492, Fax: (303)273-3875
WWW: 

 lyx-graphicspath-bug.tgz
 PGP signature