Re: current cvs does not compile

[Helge Hafting]

On Thu, Apr 21, 2005 at 02:40:26PM +, Angus Leeming wrote:
 Helge Hafting wrote:
  filetools.C:1067: error: `SIGCHLD' undeclared (first use this function)
  filetools.C:1067: error: `sigaddset' undeclared (first use this function)
  filetools.C:1070: error: `SIG_BLOCK' undeclared (first use this function)
  filetools.C:1070: error: `sigprocmask' undeclared (first use this
  function) filetools.C:1091: error: `SIG_SETMASK' undeclared (first use
 
  Re-running autogen.sh and configure did not help, neither
  did fetching this file specifically from cvs.
 
 Hi, Helge. I've just committed this patch to cvs. There may be a little lag
 before it appears on the anoncvs mirror, so I'd apply my patch now.

It compiles now. :-)

Helge Hafting

Re: current cvs does not compile

[Helge Hafting]

On Thu, Apr 21, 2005 at 02:40:26PM +, Angus Leeming wrote:
> Helge Hafting wrote:
> > filetools.C:1067: error: `SIGCHLD' undeclared (first use this function)
> > filetools.C:1067: error: `sigaddset' undeclared (first use this function)
> > filetools.C:1070: error: `SIG_BLOCK' undeclared (first use this function)
> > filetools.C:1070: error: `sigprocmask' undeclared (first use this
> > function) filetools.C:1091: error: `SIG_SETMASK' undeclared (first use
> 
> > Re-running autogen.sh and configure did not help, neither
> > did fetching this file specifically from cvs.
> 
> Hi, Helge. I've just committed this patch to cvs. There may be a little lag
> before it appears on the anoncvs mirror, so I'd apply my patch now.

It compiles now. :-)

Helge Hafting

current cvs does not compile

[Helge Hafting]

 g++ -DHAVE_CONFIG_H -I. -I. -I../../src -I./.. -I../../boost -W -Wall 
-I/usr/X11R6/include -g -O -fno-exceptions -MT filetools.lo -MD -MP -MF 
.deps/filetools.Tpo -c filetools.C -o filetools.o
filetools.C: In function `const lyx::support::cmd_ret 
   lyx::support::RunCommand(const std::string)':
filetools.C:1065: error: `sigemptyset' undeclared (first use this function)
filetools.C:1065: error: (Each undeclared identifier is reported only once for 
   each function it appears in.)
filetools.C:1067: error: `SIGCHLD' undeclared (first use this function)
filetools.C:1067: error: `sigaddset' undeclared (first use this function)
filetools.C:1070: error: `SIG_BLOCK' undeclared (first use this function)
filetools.C:1070: error: `sigprocmask' undeclared (first use this function)
filetools.C:1091: error: `SIG_SETMASK' undeclared (first use this function)
filetools.C:1164:2: warning: #warning Why is this func deleting a file? (Lgb)
make[5]: *** [filetools.lo] Error 1
make[5]: Leaving directory `/usr/src/lyx-devel/src/support'
make[4]: *** [all-recursive] Error 1
make[4]: Leaving directory `/usr/src/lyx-devel/src/support'
make[3]: *** [all] Error 2
make[3]: Leaving directory `/usr/src/lyx-devel/src/support'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/usr/src/lyx-devel/src'
make[1]: *** [all] Error 2
make[1]: Leaving directory `/usr/src/lyx-devel/src'
make: *** [all-recursive] Error 1


Re-running autogen.sh and configure did not help, neither
did fetching this file specifically from cvs.

Helge Hafting

Re: current cvs does not compile

[Angus Leeming]

Helge Hafting wrote:
 filetools.C:1067: error: `SIGCHLD' undeclared (first use this function)
 filetools.C:1067: error: `sigaddset' undeclared (first use this function)
 filetools.C:1070: error: `SIG_BLOCK' undeclared (first use this function)
 filetools.C:1070: error: `sigprocmask' undeclared (first use this
 function) filetools.C:1091: error: `SIG_SETMASK' undeclared (first use

 Re-running autogen.sh and configure did not help, neither
 did fetching this file specifically from cvs.

Hi, Helge. I've just committed this patch to cvs. There may be a little lag
before it appears on the anoncvs mirror, so I'd apply my patch now.

Regards,
-- 
Angus

signal_14x.diff.gz
Description: GNU Zip compressed data

current cvs does not compile

[Helge Hafting]

 g++ -DHAVE_CONFIG_H -I. -I. -I../../src -I./.. -I../../boost -W -Wall 
-I/usr/X11R6/include -g -O -fno-exceptions -MT filetools.lo -MD -MP -MF 
.deps/filetools.Tpo -c filetools.C -o filetools.o
filetools.C: In function `const lyx::support::cmd_ret 
   lyx::support::RunCommand(const std::string&)':
filetools.C:1065: error: `sigemptyset' undeclared (first use this function)
filetools.C:1065: error: (Each undeclared identifier is reported only once for 
   each function it appears in.)
filetools.C:1067: error: `SIGCHLD' undeclared (first use this function)
filetools.C:1067: error: `sigaddset' undeclared (first use this function)
filetools.C:1070: error: `SIG_BLOCK' undeclared (first use this function)
filetools.C:1070: error: `sigprocmask' undeclared (first use this function)
filetools.C:1091: error: `SIG_SETMASK' undeclared (first use this function)
filetools.C:1164:2: warning: #warning Why is this func deleting a file? (Lgb)
make[5]: *** [filetools.lo] Error 1
make[5]: Leaving directory `/usr/src/lyx-devel/src/support'
make[4]: *** [all-recursive] Error 1
make[4]: Leaving directory `/usr/src/lyx-devel/src/support'
make[3]: *** [all] Error 2
make[3]: Leaving directory `/usr/src/lyx-devel/src/support'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/usr/src/lyx-devel/src'
make[1]: *** [all] Error 2
make[1]: Leaving directory `/usr/src/lyx-devel/src'
make: *** [all-recursive] Error 1


Re-running autogen.sh and configure did not help, neither
did fetching this file specifically from cvs.

Helge Hafting

Re: current cvs does not compile

[Angus Leeming]

Helge Hafting wrote:
> filetools.C:1067: error: `SIGCHLD' undeclared (first use this function)
> filetools.C:1067: error: `sigaddset' undeclared (first use this function)
> filetools.C:1070: error: `SIG_BLOCK' undeclared (first use this function)
> filetools.C:1070: error: `sigprocmask' undeclared (first use this
> function) filetools.C:1091: error: `SIG_SETMASK' undeclared (first use

> Re-running autogen.sh and configure did not help, neither
> did fetching this file specifically from cvs.

Hi, Helge. I've just committed this patch to cvs. There may be a little lag
before it appears on the anoncvs mirror, so I'd apply my patch now.

Regards,
-- 
Angus

signal_14x.diff.gz
Description: GNU Zip compressed data

Re: Something is rotten in current cvs

[Alfredo Braunstein]

Alfredo Braunstein wrote:

 Angus Leeming wrote:
 
 On Friday 26 March 2004 7:05 pm, Andre Poenitz wrote:
 On Thu, Mar 25, 2004 at 04:20:18PM +, Angus Leeming wrote:
  LyX is crashing when iterating over insets. Note the two added
  lines, below. That looks very, very wrong :-(

 This seems to be solvable. Attached a 'makepatch.sh' style patch
 that replaces Buffer::inset_iterator by some small wrapper around
 DocumentIterator.

  [Alfredo: Is this close to what you proposed for PosIterator?]

 Seem to work.

 Could you please verify this and apply if sound?

Ok, your testcase works fine with Andre's patch and this implementation of
forwardInset.

I've reimplemented forwardPar as well which was severely fsckd up.

I'll apply both shortly if there are no objections.

Alfredo
Index: dociterator.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/dociterator.C,v
retrieving revision 1.7
diff -u -p -u -r1.7 dociterator.C
--- dociterator.C	25 Mar 2004 09:16:15 -	1.7
+++ dociterator.C	27 Mar 2004 12:11:32 -
@@ -232,6 +232,37 @@ InsetBase * DocumentIterator::innerInset
 }
 
 
+void DocumentIterator::forwardPar()
+{
+	while (depth()) {
+		size_t old_depth = depth();
+		idx_type old_idx = idx();
+		par_type old_par = par();
+
+		forwardPos();
+
+		if (depth()  old_depth)
+			break;
+		if (depth() == old_depth 
+		(old_idx != idx() || old_par != par()))
+			break;
+	}
+}
+
+
+void DocumentIterator::forwardInset()
+{
+	while (depth()) {
+		size_t old_depth = depth();
+
+		forwardPos();
+
+		if (depth()  old_depth)
+			break;
+	}
+}
+
+
 void DocumentIterator::forwardPos()
 {
 	CursorSlice  top = back();
@@ -296,9 +327,9 @@ void DocumentIterator::forwardPos()
 
 void DocumentIterator::forwardChar()
 {
-	forwardPos(); 
-	while (size() != 0  pos() == lastpos())
-		forwardPos();
+	do {
+		forwardPos(); 
+	} while (size() != 0  pos() == lastpos());
 }
 
 
@@ -306,62 +337,6 @@ void DocumentIterator::backwardChar()
 {
 	lyxerr  not implemented  endl;
 	BOOST_ASSERT(false);
-}
-
-
-void DocumentIterator::forwardPar()
-{
-	CursorSlice  top = back();
-	lyxerr  XXX   *this  endl;
-
-	// move into an inset to the right if possible
-	InsetBase * n = 0;
-	if (top.pos() != lastpos()) {
-		// this is impossible for pos() == size()
-		if (inMathed()) {
-			n = (top.cell().begin() + top.pos())-nucleus();
-		} else {
-			if (paragraph().isInset(top.pos()))
-n = paragraph().getInset(top.pos());
-		}
-	}
-
-	if (n  n-isActive()) {
-		lyxerr  ... descend  endl;
-		push_back(CursorSlice(*n));
-		return;
-	}
-
-	// otherwise move on one cell back if possible
-	if (top.pos()  lastpos()) {
-		lyxerr  ... next pos  endl;
-		++top.pos();
-		return;
-	}
-
-	// otherwise move on one cell back if possible
-	if (top.par()  lastpar()) {
-		lyxerr  ... next par  endl;
-		++top.par();
-		top.pos() = 0;
-		return;
-	}
-
-	// otherwise try to move on one cell if possible
-	while (top.idx()  top.lastidx()) {
-		lyxerr  ... next idx 
-			  was:   top.idx()   max:   top.lastidx()  endl;
-		++top.idx();
-		top.par() = 0;
-		top.pos() = 0;
-		return;
-	}
-
-	// otherwise leave inset an jump over inset as a whole
-	pop_back();
-	// 'top' is invalid now...
-	if (size())
-		++back().pos();
 }
 
 

Re: Something is rotten in current cvs

[Andre Poenitz]

On Sat, Mar 27, 2004 at 12:59:19AM +0100, Alfredo Braunstein wrote:
 Andre Poenitz wrote:
 
  This seems to be solvable. Attached a 'makepatch.sh' style patch that
  replaces Buffer::inset_iterator by some small wrapper around
  DocumentIterator.
  
   [Alfredo: Is this close to what you proposed for PosIterator?]
 
 Yes... I wonder if private derivation would be preferable...

Fine with me if somebody writes all the 'using' stuff ;-)

 PS: you miss some goodies in dociterator, just shove in what you
 have and we fill the rest (In particular we need forwardInset for
 inset_iterator and backwardChar for PosIterator)

Thanks for the offer. I would if I could... Next scheduled 'full net
access' is on Thursday...

Andre'

Re: Something is rotten in current cvs

[Andre Poenitz]

On Sat, Mar 27, 2004 at 12:56:03AM +0100, Alfredo Braunstein wrote:
  Missing insetiterator.C?
 
 No, just DocIterator::forwardInset in dociterator.C.

  
void DocumentIterator::forwardInset()
{
forwardPos(); 
while (!empty()  (pos() == lastpos() || nextInset() == 0))
forwardPos();
}

This is inefficient for texted as it iterates over all possible cursor
positions and checks whether there is an inset there. It should be more
efficient to use the insetlist somehow.
 
 I suppose Andre' has it in his tree (seems to work)...

I really don't like the current situation where I can't even run a 'cvs
up' properly, let alone 'diff' or 'commit'.

Is there any other means to access LyX CVS apart from the usual?
[I could probably ssh to baywatch, cvs co a local copy and work from
there, but that's neither nice, nor fast, nor cheap...]

Andre'

Re: Something is rotten in current cvs

[Alfredo Braunstein]

Andre Poenitz wrote:

 On Sat, Mar 27, 2004 at 12:56:03AM +0100, Alfredo Braunstein wrote:
  Missing insetiterator.C?
 
 No, just DocIterator::forwardInset in dociterator.C.
 
   
 void DocumentIterator::forwardInset()
 {
 forwardPos();
 while (!empty()  (pos() == lastpos() || nextInset() == 0))
 forwardPos();
 }
 
 This is inefficient for texted as it iterates over all possible cursor
 positions and checks whether there is an inset there. It should be more
 efficient to use the insetlist somehow.

I think it is a complete no-issue for now, as iterating over all cursor
positions of the Userguide20 is still unnoticeable (sub second) here. If it
shows up in the profiles we can do it later. I'd really prefer to have the
simplest implementation now...

Alfredo

Re: Something is rotten in current cvs

[Andre Poenitz]

On Sat, Mar 27, 2004 at 01:16:33PM +0100, Alfredo Braunstein wrote:
 Ok, your testcase works fine with Andre's patch and this implementation of
 forwardInset.


+void DocumentIterator::forwardInset()
+{
+   while (depth()) {
+   size_t old_depth = depth();
+
+   forwardPos();
+
+   if (depth()  old_depth)
+   break;
+   }
+}

Aehm, forget the last mail. If I am understanding this correctly,
this touches only 'descendable insets', not all insets. This might be
sufficient for some usages but not for 'goto label' or similar.

-   forwardPos(); 
-   while (size() != 0  pos() == lastpos())
-   forwardPos();
+   do {
+   forwardPos(); 
+   } while (size() != 0  pos() == lastpos());

I don't like do...while loops. Certainly a matter of taste, though...

Andre'

Re: Something is rotten in current cvs

[Andre Poenitz]

On Sat, Mar 27, 2004 at 01:16:33PM +0100, Alfredo Braunstein wrote:
 I've reimplemented forwardPar as well which was severely fsckd up.

AKA not implemented at all?

 I'll apply both shortly if there are no objections.

None from my side...

Andre'

Re: Something is rotten in current cvs

[Andre Poenitz]

On Sat, Mar 27, 2004 at 02:11:16PM +0100, Alfredo Braunstein wrote:
 I think it is a complete no-issue for now, as iterating over all cursor
 positions of the Userguide20 is still unnoticeable (sub second) here. If it
 shows up in the profiles we can do it later. I'd really prefer to have the
 simplest implementation now...

Me too. 'twas just to sooth the naysayers...

Andre'

Re: Something is rotten in current cvs

[Alfredo Braunstein]

Andre Poenitz wrote:

 +void DocumentIterator::forwardInset()
 +{
 + while (depth()) {
 + size_t old_depth = depth();
 +
 + forwardPos();
 +
 + if (depth()  old_depth)
 + break;
 + }
 +}
 
 Aehm, forget the last mail. If I am understanding this correctly,
 this touches only 'descendable insets', not all insets. This might be
 sufficient for some usages but not for 'goto label' or similar.

Ouch, you're right.

Alfredo

Re: Something is rotten in current cvs

[Alfredo Braunstein]

Alfredo Braunstein wrote:

> Angus Leeming wrote:
> 
>> On Friday 26 March 2004 7:05 pm, Andre Poenitz wrote:
>>> On Thu, Mar 25, 2004 at 04:20:18PM +, Angus Leeming wrote:
>>> > LyX is crashing when iterating over insets. Note the two added
>>> > lines, below. That looks very, very wrong :-(
>>>
>>> This seems to be solvable. Attached a 'makepatch.sh' style patch
>>> that replaces Buffer::inset_iterator by some small wrapper around
>>> DocumentIterator.
>>>
>>>  [Alfredo: Is this close to what you proposed for PosIterator?]
>>>
>>> Seem to work.
>>>
>>> Could you please verify this and apply if sound?

Ok, your testcase works fine with Andre's patch and this implementation of
forwardInset.

I've reimplemented forwardPar as well which was severely fsckd up.

I'll apply both shortly if there are no objections.

Alfredo
Index: dociterator.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/dociterator.C,v
retrieving revision 1.7
diff -u -p -u -r1.7 dociterator.C
--- dociterator.C	25 Mar 2004 09:16:15 -	1.7
+++ dociterator.C	27 Mar 2004 12:11:32 -
@@ -232,6 +232,37 @@ InsetBase * DocumentIterator::innerInset
 }
 
 
+void DocumentIterator::forwardPar()
+{
+	while (depth()) {
+		size_t old_depth = depth();
+		idx_type old_idx = idx();
+		par_type old_par = par();
+
+		forwardPos();
+
+		if (depth() > old_depth)
+			break;
+		if (depth() == old_depth &&
+		(old_idx != idx() || old_par != par()))
+			break;
+	}
+}
+
+
+void DocumentIterator::forwardInset()
+{
+	while (depth()) {
+		size_t old_depth = depth();
+
+		forwardPos();
+
+		if (depth() > old_depth)
+			break;
+	}
+}
+
+
 void DocumentIterator::forwardPos()
 {
 	CursorSlice & top = back();
@@ -296,9 +327,9 @@ void DocumentIterator::forwardPos()
 
 void DocumentIterator::forwardChar()
 {
-	forwardPos(); 
-	while (size() != 0 && pos() == lastpos())
-		forwardPos();
+	do {
+		forwardPos(); 
+	} while (size() != 0 && pos() == lastpos());
 }
 
 
@@ -306,62 +337,6 @@ void DocumentIterator::backwardChar()
 {
 	lyxerr << "not implemented" << endl;
 	BOOST_ASSERT(false);
-}
-
-
-void DocumentIterator::forwardPar()
-{
-	CursorSlice & top = back();
-	lyxerr << "XXX " << *this << endl;
-
-	// move into an inset to the right if possible
-	InsetBase * n = 0;
-	if (top.pos() != lastpos()) {
-		// this is impossible for pos() == size()
-		if (inMathed()) {
-			n = (top.cell().begin() + top.pos())->nucleus();
-		} else {
-			if (paragraph().isInset(top.pos()))
-n = paragraph().getInset(top.pos());
-		}
-	}
-
-	if (n && n->isActive()) {
-		lyxerr << "... descend" << endl;
-		push_back(CursorSlice(*n));
-		return;
-	}
-
-	// otherwise move on one cell back if possible
-	if (top.pos() < lastpos()) {
-		lyxerr << "... next pos" << endl;
-		++top.pos();
-		return;
-	}
-
-	// otherwise move on one cell back if possible
-	if (top.par() < lastpar()) {
-		lyxerr << "... next par" << endl;
-		++top.par();
-		top.pos() = 0;
-		return;
-	}
-
-	// otherwise try to move on one cell if possible
-	while (top.idx() < top.lastidx()) {
-		lyxerr << "... next idx" 
-			<< " was: " << top.idx() << " max: " << top.lastidx() << endl;
-		++top.idx();
-		top.par() = 0;
-		top.pos() = 0;
-		return;
-	}
-
-	// otherwise leave inset an jump over inset as a whole
-	pop_back();
-	// 'top' is invalid now...
-	if (size())
-		++back().pos();
 }
 
 

Re: Something is rotten in current cvs

[Andre Poenitz]

On Sat, Mar 27, 2004 at 12:59:19AM +0100, Alfredo Braunstein wrote:
> Andre Poenitz wrote:
> 
> > This seems to be solvable. Attached a 'makepatch.sh' style patch that
> > replaces Buffer::inset_iterator by some small wrapper around
> > DocumentIterator.
> > 
> >  [Alfredo: Is this close to what you proposed for PosIterator?]
> 
> Yes... I wonder if private derivation would be preferable...

Fine with me if somebody writes all the 'using' stuff ;-)

> PS: you miss some goodies in dociterator, just shove in what you
> have and we fill the rest (In particular we need forwardInset for
> inset_iterator and backwardChar for PosIterator)

Thanks for the offer. I would if I could... Next scheduled 'full net
access' is on Thursday...

Andre'

Re: Something is rotten in current cvs

[Andre Poenitz]

On Sat, Mar 27, 2004 at 12:56:03AM +0100, Alfredo Braunstein wrote:
> > Missing insetiterator.C?
> 
> No, just DocIterator::forwardInset in dociterator.C.

  
void DocumentIterator::forwardInset()
{
forwardPos(); 
while (!empty() && (pos() == lastpos() || nextInset() == 0))
forwardPos();
}

This is inefficient for texted as it iterates over all possible cursor
positions and checks whether there is an inset there. It should be more
efficient to use the insetlist somehow.
 
> I suppose Andre' has it in his tree ("seems to work")...

I really don't like the current situation where I can't even run a 'cvs
up' properly, let alone 'diff' or 'commit'.

Is there any other means to access LyX CVS apart from the usual?
[I could probably ssh to baywatch, cvs co a local copy and work from
there, but that's neither nice, nor fast, nor cheap...]

Andre'

Re: Something is rotten in current cvs

[Alfredo Braunstein]

Andre Poenitz wrote:

> On Sat, Mar 27, 2004 at 12:56:03AM +0100, Alfredo Braunstein wrote:
>> > Missing insetiterator.C?
>> 
>> No, just DocIterator::forwardInset in dociterator.C.
> 
>   
> void DocumentIterator::forwardInset()
> {
> forwardPos();
> while (!empty() && (pos() == lastpos() || nextInset() == 0))
> forwardPos();
> }
> 
> This is inefficient for texted as it iterates over all possible cursor
> positions and checks whether there is an inset there. It should be more
> efficient to use the insetlist somehow.

I think it is a complete no-issue for now, as iterating over all cursor
positions of the Userguide20 is still unnoticeable (sub second) here. If it
shows up in the profiles we can do it later. I'd really prefer to have the
simplest implementation now...

Alfredo

Re: Something is rotten in current cvs

[Andre Poenitz]

On Sat, Mar 27, 2004 at 01:16:33PM +0100, Alfredo Braunstein wrote:
> Ok, your testcase works fine with Andre's patch and this implementation of
> forwardInset.


+void DocumentIterator::forwardInset()
+{
+   while (depth()) {
+   size_t old_depth = depth();
+
+   forwardPos();
+
+   if (depth() > old_depth)
+   break;
+   }
+}

Aehm, forget the last mail. If I am understanding this correctly,
this touches only 'descendable insets', not all insets. This might be
sufficient for some usages but not for 'goto label' or similar.

-   forwardPos(); 
-   while (size() != 0 && pos() == lastpos())
-   forwardPos();
+   do {
+   forwardPos(); 
+   } while (size() != 0 && pos() == lastpos());

I don't like do...while loops. Certainly a matter of taste, though...

Andre'

Re: Something is rotten in current cvs

[Andre Poenitz]

On Sat, Mar 27, 2004 at 01:16:33PM +0100, Alfredo Braunstein wrote:
> I've reimplemented forwardPar as well which was severely fsckd up.

AKA "not implemented at all"?

> I'll apply both shortly if there are no objections.

None from my side...

Andre'

Re: Something is rotten in current cvs

[Andre Poenitz]

On Sat, Mar 27, 2004 at 02:11:16PM +0100, Alfredo Braunstein wrote:
> I think it is a complete no-issue for now, as iterating over all cursor
> positions of the Userguide20 is still unnoticeable (sub second) here. If it
> shows up in the profiles we can do it later. I'd really prefer to have the
> simplest implementation now...

Me too. 'twas just to sooth the naysayers...

Andre'

Re: Something is rotten in current cvs

[Alfredo Braunstein]

Andre Poenitz wrote:

> +void DocumentIterator::forwardInset()
> +{
> + while (depth()) {
> + size_t old_depth = depth();
> +
> + forwardPos();
> +
> + if (depth() > old_depth)
> + break;
> + }
> +}
> 
> Aehm, forget the last mail. If I am understanding this correctly,
> this touches only 'descendable insets', not all insets. This might be
> sufficient for some usages but not for 'goto label' or similar.

Ouch, you're right.

Alfredo

Re: Something is rotten in current cvs

[Andre Poenitz]

On Thu, Mar 25, 2004 at 05:02:50PM +, Angus Leeming wrote:
  // This is also a buffer property (ale)
  void Buffer::fillWithBibKeys(std::vectorstd::pairstring, string 
 
  for (inset_iterator it = inset_const_iterator_begin();
  it != inset_const_iterator_end(); ++it) {
  +   InsetBase * base = it.operator-();
  +   lyxerr  Base ==   base  std::endl;
 
 In fact, I have to add 
 if (!base)
 break;

There'd better be an ASSERT on that. So I suspect the initialization
going wrong somewhere already.

 for things to display. Thereafter, all sorts is messed up. Screenshot 
 attached. In no particular order:
 
 * Math macros are foo-barred. Have been for a while.

I know.

 * Table cells don't resize to accommodate the graphic once it has been 
 loaded. This one is new as of today.

Probably due to the removal of the 'manual' update calls. I guess you
have to call LyXText::redoParagraph on the respective outer paragraph
and update() now manually from whatever code runs when the graphics is
loaded.

Andre'

Re: Something is rotten in current cvs

[Andre Poenitz]

On Thu, Mar 25, 2004 at 04:20:18PM +, Angus Leeming wrote:
 LyX is crashing when iterating over insets. Note the two added lines, 
 below. That looks very, very wrong :-(

This seems to be solvable. Attached a 'makepatch.sh' style patch that
replaces Buffer::inset_iterator by some small wrapper around
DocumentIterator.

 [Alfredo: Is this close to what you proposed for PosIterator?]

Seem to work.

Could you please verify this and apply if sound?

Andre'


poenitz-2004-03-26_21:08:46.patch.gz
Description: application/gunzip

Re: Something is rotten in current cvs

[Angus Leeming]

On Friday 26 March 2004 7:05 pm, Andre Poenitz wrote:
 On Thu, Mar 25, 2004 at 04:20:18PM +, Angus Leeming wrote:
  LyX is crashing when iterating over insets. Note the two added
  lines, below. That looks very, very wrong :-(

 This seems to be solvable. Attached a 'makepatch.sh' style patch
 that replaces Buffer::inset_iterator by some small wrapper around
 DocumentIterator.

  [Alfredo: Is this close to what you proposed for PosIterator?]

 Seem to work.

 Could you please verify this and apply if sound?

Missing insetiterator.C?

Angus

Re: Something is rotten in current cvs

[Alfredo Braunstein]

Angus Leeming wrote:

 On Friday 26 March 2004 7:05 pm, Andre Poenitz wrote:
 On Thu, Mar 25, 2004 at 04:20:18PM +, Angus Leeming wrote:
  LyX is crashing when iterating over insets. Note the two added
  lines, below. That looks very, very wrong :-(

 This seems to be solvable. Attached a 'makepatch.sh' style patch
 that replaces Buffer::inset_iterator by some small wrapper around
 DocumentIterator.

  [Alfredo: Is this close to what you proposed for PosIterator?]

 Seem to work.

 Could you please verify this and apply if sound?
 
 Missing insetiterator.C?

No, just DocIterator::forwardInset in dociterator.C.

I suppose Andre' has it in his tree (seems to work)...

Alfredo

Re: Something is rotten in current cvs

[Alfredo Braunstein]

Andre Poenitz wrote:

 This seems to be solvable. Attached a 'makepatch.sh' style patch that
 replaces Buffer::inset_iterator by some small wrapper around
 DocumentIterator.
 
  [Alfredo: Is this close to what you proposed for PosIterator?]

Yes... I wonder if private derivation would be preferable...

Alfredo

PS: you miss some goodies in dociterator, just shove in what you have and we
fill the rest (In particular we need forwardInset for inset_iterator and
backwardChar for PosIterator)

Alfredo

Re: Something is rotten in current cvs

[Andre Poenitz]

On Thu, Mar 25, 2004 at 05:02:50PM +, Angus Leeming wrote:
> > // This is also a buffer property (ale)
> > void Buffer::fillWithBibKeys(std::vector >
> 
> > for (inset_iterator it = inset_const_iterator_begin();
> > it != inset_const_iterator_end(); ++it) {
> > +   InsetBase * base = it.operator->();
> > +   lyxerr << "Base == " << base << std::endl;
> 
> In fact, I have to add 
> if (!base)
> break;

There'd better be an ASSERT on that. So I suspect the initialization
going wrong somewhere already.

> for things to display. Thereafter, all sorts is messed up. Screenshot 
> attached. In no particular order:
> 
> * Math macros are foo-barred. Have been for a while.

I know.

> * Table cells don't resize to accommodate the graphic once it has been 
> loaded. This one is new as of today.

Probably due to the removal of the 'manual' update calls. I guess you
have to call LyXText::redoParagraph on the respective outer paragraph
and update() now manually from whatever code runs when the graphics is
loaded.

Andre'

Re: Something is rotten in current cvs

[Andre Poenitz]

On Thu, Mar 25, 2004 at 04:20:18PM +, Angus Leeming wrote:
> LyX is crashing when iterating over insets. Note the two added lines, 
> below. That looks very, very wrong :-(

This seems to be solvable. Attached a 'makepatch.sh' style patch that
replaces Buffer::inset_iterator by some small wrapper around
DocumentIterator.

 [Alfredo: Is this close to what you proposed for PosIterator?]

Seem to work.

Could you please verify this and apply if sound?

Andre'


poenitz-2004-03-26_21:08:46.patch.gz
Description: application/gunzip

Re: Something is rotten in current cvs

[Angus Leeming]

On Friday 26 March 2004 7:05 pm, Andre Poenitz wrote:
> On Thu, Mar 25, 2004 at 04:20:18PM +, Angus Leeming wrote:
> > LyX is crashing when iterating over insets. Note the two added
> > lines, below. That looks very, very wrong :-(
>
> This seems to be solvable. Attached a 'makepatch.sh' style patch
> that replaces Buffer::inset_iterator by some small wrapper around
> DocumentIterator.
>
>  [Alfredo: Is this close to what you proposed for PosIterator?]
>
> Seem to work.
>
> Could you please verify this and apply if sound?

Missing insetiterator.C?

Angus

Re: Something is rotten in current cvs

[Alfredo Braunstein]

Angus Leeming wrote:

> On Friday 26 March 2004 7:05 pm, Andre Poenitz wrote:
>> On Thu, Mar 25, 2004 at 04:20:18PM +, Angus Leeming wrote:
>> > LyX is crashing when iterating over insets. Note the two added
>> > lines, below. That looks very, very wrong :-(
>>
>> This seems to be solvable. Attached a 'makepatch.sh' style patch
>> that replaces Buffer::inset_iterator by some small wrapper around
>> DocumentIterator.
>>
>>  [Alfredo: Is this close to what you proposed for PosIterator?]
>>
>> Seem to work.
>>
>> Could you please verify this and apply if sound?
> 
> Missing insetiterator.C?

No, just DocIterator::forwardInset in dociterator.C.

I suppose Andre' has it in his tree ("seems to work")...

Alfredo

Re: Something is rotten in current cvs

[Alfredo Braunstein]

Andre Poenitz wrote:

> This seems to be solvable. Attached a 'makepatch.sh' style patch that
> replaces Buffer::inset_iterator by some small wrapper around
> DocumentIterator.
> 
>  [Alfredo: Is this close to what you proposed for PosIterator?]

Yes... I wonder if private derivation would be preferable...

Alfredo

PS: you miss some goodies in dociterator, just shove in what you have and we
fill the rest (In particular we need forwardInset for inset_iterator and
backwardChar for PosIterator)

Alfredo

Something is rotten in current cvs

[Angus Leeming]

LyX is crashing when iterating over insets. Note the two added lines, 
below. That looks very, very wrong :-(

Angus

// This is also a buffer property (ale)
void Buffer::fillWithBibKeys(std::vectorstd::pairstring, string   
keys)
const
{
/// if this is a child document and the parent is already loaded
/// use the parent's list instead  [ale990412]
Buffer const * tmp = getMasterBuffer();
if (tmp != this) {
tmp-fillWithBibKeys(keys);
return;
}

for (inset_iterator it = inset_const_iterator_begin();
it != inset_const_iterator_end(); ++it) {
+   InsetBase * base = it.operator-();
+   lyxerr  Base ==   base  std::endl;
if (it-lyxCode() == InsetOld::BIBTEX_CODE) {

Base == 0x856e008
Base == 0x8547408
Base == 0x8568630
Base == 0

Program received signal SIGSEGV, Segmentation fault.
0x080a08a0 in 
Buffer::fillWithBibKeys(std::vectorstd::pairstd::string, 
std::string, std::allocatorstd::pairstd::string, std::string  
) const (
this=0x8528788, [EMAIL PROTECTED]) at ../../src/buffer.C:1214
1214if (it-lyxCode() == InsetOld::BIBTEX_CODE) {

(gdb) bt
#0  0x080a08a0 in 
Buffer::fillWithBibKeys(std::vectorstd::pairstd::string, 
std::string, std::allocatorstd::pairstd::string, std::string  
) const (
this=0x8528788, [EMAIL PROTECTED]) at ../../src/buffer.C:1214
#1  0x0819f51a in (anonymous namespace)::getNatbibLabel(Buffer const, 
std::string const, std::string const, std::string const, 
std::string const, bool, bool) ([EMAIL PROTECTED], 
[EMAIL PROTECTED], [EMAIL PROTECTED],
[EMAIL PROTECTED], [EMAIL PROTECTED], numerical=false, 
jura=false)
at ../../../src/insets/insetcite.C:57
#2  0x081a13a3 in InsetCitation::generateLabel(Buffer const) const (
this=0x8463310, [EMAIL PROTECTED]) at 
../../../src/insets/insetcommand.h:67
#3  0x081a159f in InsetCitation::getScreenLabel(Buffer const) const (
this=0x8463310, [EMAIL PROTECTED]) at 
../../../src/insets/insetcite.C:307
#4  0x081a374d in InsetCommand::metrics(MetricsInfo, Dimension) 
const (
this=0x8463310, [EMAIL PROTECTED], [EMAIL PROTECTED])
at ../../../src/insets/insetcommand.C:50

-- 
Angus

Something is rotten in current cvs

[Angus Leeming]

LyX is crashing when iterating over insets. Note the two added lines, 
below. That looks very, very wrong :-(

Angus

// This is also a buffer property (ale)
void Buffer::fillWithBibKeys(std::vector > & 
keys)
const
{
/// if this is a child document and the parent is already loaded
/// use the parent's list instead  [ale990412]
Buffer const * tmp = getMasterBuffer();
if (tmp != this) {
tmp->fillWithBibKeys(keys);
return;
}

for (inset_iterator it = inset_const_iterator_begin();
it != inset_const_iterator_end(); ++it) {
+   InsetBase * base = it.operator->();
+   lyxerr << "Base == " << base << std::endl;
if (it->lyxCode() == InsetOld::BIBTEX_CODE) {

Base == 0x856e008
Base == 0x8547408
Base == 0x8568630
Base == 0

Program received signal SIGSEGV, Segmentation fault.
0x080a08a0 in 
Buffer::fillWithBibKeys(std::vector, std::allocator > 
>&) const (
this=0x8528788, [EMAIL PROTECTED]) at ../../src/buffer.C:1214
1214if (it->lyxCode() == InsetOld::BIBTEX_CODE) {

(gdb) bt
#0  0x080a08a0 in 
Buffer::fillWithBibKeys(std::vector, std::allocator > 
>&) const (
this=0x8528788, [EMAIL PROTECTED]) at ../../src/buffer.C:1214
#1  0x0819f51a in (anonymous namespace)::getNatbibLabel(Buffer const&, 
std::string const&, std::string const&, std::string const&, 
std::string const&, bool, bool) ([EMAIL PROTECTED], 
[EMAIL PROTECTED], [EMAIL PROTECTED],
[EMAIL PROTECTED], [EMAIL PROTECTED], numerical=false, 
jura=false)
at ../../../src/insets/insetcite.C:57
#2  0x081a13a3 in InsetCitation::generateLabel(Buffer const&) const (
this=0x8463310, [EMAIL PROTECTED]) at 
../../../src/insets/insetcommand.h:67
#3  0x081a159f in InsetCitation::getScreenLabel(Buffer const&) const (
this=0x8463310, [EMAIL PROTECTED]) at 
../../../src/insets/insetcite.C:307
#4  0x081a374d in InsetCommand::metrics(MetricsInfo&, Dimension&) 
const (
this=0x8463310, [EMAIL PROTECTED], [EMAIL PROTECTED])
at ../../../src/insets/insetcommand.C:50

-- 
Angus

Re: Current CVS can not select with mouse

[Andre Poenitz]

On Mon, Mar 22, 2004 at 10:20:26AM +, Angus Leeming wrote:
 Kayvan A. Sylvan wrote:
 
  I don't know when this bug popped up, but I can no longer select
  with the mouse (click-hold-drag) although the SHIFT-move method of
  text selection seems to work.
 
 The problem is that the relevant code has been commented out. 
 Unfortunately, it doesn't compile, so the fix will require some 
 thought.
 
 bool BufferView::Pimpl::workAreaDispatch(FuncRequest const  cmd0)
 {
 ...
 #if 0

I am not sure this is the real reason. IIRC selection worked at some
point with this #if 0 'active'.

Andre'

Re: Current CVS can not select with mouse

[Andre Poenitz]

On Mon, Mar 22, 2004 at 10:20:26AM +, Angus Leeming wrote:
> Kayvan A. Sylvan wrote:
> 
> > I don't know when this bug popped up, but I can no longer select
> > with the mouse (click-hold-drag) although the SHIFT-move method of
> > text selection seems to work.
> 
> The problem is that the relevant code has been commented out. 
> Unfortunately, it doesn't compile, so the fix will require some 
> thought.
> 
> bool BufferView::Pimpl::workAreaDispatch(FuncRequest const & cmd0)
> {
> ...
> #if 0

I am not sure this is the real reason. IIRC selection worked at some
point with this #if 0 'active'.

Andre'

Re: Current CVS can not select with mouse

[Angus Leeming]

Kayvan A. Sylvan wrote:

 I don't know when this bug popped up, but I can no longer select
 with the mouse (click-hold-drag) although the SHIFT-move method of
 text selection seems to work.

The problem is that the relevant code has been commented out. 
Unfortunately, it doesn't compile, so the fix will require some 
thought.

bool BufferView::Pimpl::workAreaDispatch(FuncRequest const  cmd0)
{
...
#if 0
case LFUN_MOUSE_MOTION: {
if (!available())
return false;
FuncRequest cmd1 = cmd;
DispatchResult res = cur.inset().dispatch(cur, cmd);
if (fitCursor() || res.update()) {
update();
cur.updatePos();
}
return true;
}
#else
case LFUN_MOUSE_MOTION: 
#endif


-- 
Angus

[PATCH] Re: Current CVS: Insert-File- submenus grayed out

[Jean-Marc Lasgouttes]

 Kayvan == Kayvan A Sylvan [EMAIL PROTECTED] writes:

Kayvan I want to Insert-File-ASCII as Lines, but this option is
Kayvan grayed out along with LyX Document and Ascii as
Kayvan paragraphs.

This is cured by the following patch.

Andre', this patch also cures the handling of read-only buffers. I
moved the check for read-only to the end of getStatus, now that the
default for enable is false.

I'll commit shortly if there is no objection.

JMarc

Index: src/ChangeLog
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/ChangeLog,v
retrieving revision 1.1838
diff -u -p -r1.1838 ChangeLog
--- src/ChangeLog	22 Mar 2004 16:22:51 -	1.1838
+++ src/ChangeLog	22 Mar 2004 17:04:59 -
@@ -1,5 +1,8 @@
 2004-03-22  Jean-Marc Lasgouttes  [EMAIL PROTECTED]
 
+	* lyxfunc.C (getStatus): handle read-only buffers correctly;
+	handle LFUN_FILE_INSERT_*
+
 	* lyxrc.C (setDefaults, getDescription, output, read): 
 	* lyxrc.h: remove ps_command
 
Index: src/lyxfunc.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/lyxfunc.C,v
retrieving revision 1.585
diff -u -p -r1.585 lyxfunc.C
--- src/lyxfunc.C	18 Mar 2004 16:41:45 -	1.585
+++ src/lyxfunc.C	22 Mar 2004 17:05:00 -
@@ -287,25 +287,12 @@ FuncStatus LyXFunc::getStatus(FuncReques
 	setStatusMessage(N_(Command disabled));
 
 	// Check whether we need a buffer
-	if (!lyxaction.funcHasFlag(cmd.action, LyXAction::NoBuffer)) {
-		// Yes we need a buffer, do we have one?
-		if (buf) {
-			// yes
-			// Can we use a readonly buffer?
-			if (buf-isReadonly() 
-			!lyxaction.funcHasFlag(cmd.action,
-		   LyXAction::ReadOnly)) {
-// no
-setStatusMessage(N_(Document is read-only));
-flag.enabled(false);
-			}
-		} else {
-			// no
-			setStatusMessage(N_(Command not allowed with
-	out any document open));
-			flag.enabled(false);
-			return flag;
-		}
+	if (!lyxaction.funcHasFlag(cmd.action, LyXAction::NoBuffer)  !buf) {
+		// no, exit directly
+		setStatusMessage(N_(Command not allowed with
+out any document open));
+		flag.enabled(false);
+		return flag;
 	}
 
 	// I would really like to avoid having this switch and rather try to
@@ -503,15 +490,27 @@ FuncStatus LyXFunc::getStatus(FuncReques
 	case LFUN_SET_COLOR:
 	case LFUN_MESSAGE:
 	case LFUN_EXTERNAL_EDIT:
+	case LFUN_FILE_INSERT:
+	case LFUN_FILE_INSERT_ASCII:
+	case LFUN_FILE_INSERT_ASCII_PARA:
 		// these are handled in our dispatch()
 		break;
 
 	default:
 		cur.getStatus(cmd, flag);
-		return flag;
 	}
 
-	flag.enabled(enable);
+	if (!enable)
+		flag.enabled(false);
+
+	// Can we use a readonly buffer?
+	if (buf  buf-isReadonly() 
+	 !lyxaction.funcHasFlag(cmd.action, LyXAction::ReadOnly)
+	 !lyxaction.funcHasFlag(cmd.action, LyXAction::NoBuffer)) {
+		setStatusMessage(N_(Document is read-only));
+		flag.enabled(false);
+	}
+
 	return flag;
 }
 

Re: Current CVS can not select with mouse

[Angus Leeming]

Kayvan A. Sylvan wrote:

> I don't know when this bug popped up, but I can no longer select
> with the mouse (click-hold-drag) although the SHIFT-move method of
> text selection seems to work.

The problem is that the relevant code has been commented out. 
Unfortunately, it doesn't compile, so the fix will require some 
thought.

bool BufferView::Pimpl::workAreaDispatch(FuncRequest const & cmd0)
{
...
#if 0
case LFUN_MOUSE_MOTION: {
if (!available())
return false;
FuncRequest cmd1 = cmd;
DispatchResult res = cur.inset().dispatch(cur, cmd);
if (fitCursor() || res.update()) {
update();
cur.updatePos();
}
return true;
}
#else
case LFUN_MOUSE_MOTION: 
#endif


-- 
Angus

[PATCH] Re: Current CVS: Insert->File-> submenus grayed out

[Jean-Marc Lasgouttes]

> "Kayvan" == Kayvan A Sylvan <[EMAIL PROTECTED]> writes:

Kayvan> I want to Insert->File->ASCII as Lines, but this option is
Kayvan> grayed out along with "LyX Document" and "Ascii as
Kayvan> paragraphs".

This is cured by the following patch.

Andre', this patch also cures the handling of read-only buffers. I
moved the check for read-only to the end of getStatus, now that the
default for "enable" is false.

I'll commit shortly if there is no objection.

JMarc

Index: src/ChangeLog
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/ChangeLog,v
retrieving revision 1.1838
diff -u -p -r1.1838 ChangeLog
--- src/ChangeLog	22 Mar 2004 16:22:51 -	1.1838
+++ src/ChangeLog	22 Mar 2004 17:04:59 -
@@ -1,5 +1,8 @@
 2004-03-22  Jean-Marc Lasgouttes  <[EMAIL PROTECTED]>
 
+	* lyxfunc.C (getStatus): handle read-only buffers correctly;
+	handle LFUN_FILE_INSERT_*
+
 	* lyxrc.C (setDefaults, getDescription, output, read): 
 	* lyxrc.h: remove ps_command
 
Index: src/lyxfunc.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/lyxfunc.C,v
retrieving revision 1.585
diff -u -p -r1.585 lyxfunc.C
--- src/lyxfunc.C	18 Mar 2004 16:41:45 -	1.585
+++ src/lyxfunc.C	22 Mar 2004 17:05:00 -
@@ -287,25 +287,12 @@ FuncStatus LyXFunc::getStatus(FuncReques
 	setStatusMessage(N_("Command disabled"));
 
 	// Check whether we need a buffer
-	if (!lyxaction.funcHasFlag(cmd.action, LyXAction::NoBuffer)) {
-		// Yes we need a buffer, do we have one?
-		if (buf) {
-			// yes
-			// Can we use a readonly buffer?
-			if (buf->isReadonly() &&
-			!lyxaction.funcHasFlag(cmd.action,
-		   LyXAction::ReadOnly)) {
-// no
-setStatusMessage(N_("Document is read-only"));
-flag.enabled(false);
-			}
-		} else {
-			// no
-			setStatusMessage(N_("Command not allowed with"
-	"out any document open"));
-			flag.enabled(false);
-			return flag;
-		}
+	if (!lyxaction.funcHasFlag(cmd.action, LyXAction::NoBuffer) && !buf) {
+		// no, exit directly
+		setStatusMessage(N_("Command not allowed with"
+"out any document open"));
+		flag.enabled(false);
+		return flag;
 	}
 
 	// I would really like to avoid having this switch and rather try to
@@ -503,15 +490,27 @@ FuncStatus LyXFunc::getStatus(FuncReques
 	case LFUN_SET_COLOR:
 	case LFUN_MESSAGE:
 	case LFUN_EXTERNAL_EDIT:
+	case LFUN_FILE_INSERT:
+	case LFUN_FILE_INSERT_ASCII:
+	case LFUN_FILE_INSERT_ASCII_PARA:
 		// these are handled in our dispatch()
 		break;
 
 	default:
 		cur.getStatus(cmd, flag);
-		return flag;
 	}
 
-	flag.enabled(enable);
+	if (!enable)
+		flag.enabled(false);
+
+	// Can we use a readonly buffer?
+	if (buf && buf->isReadonly() 
+	&& !lyxaction.funcHasFlag(cmd.action, LyXAction::ReadOnly)
+	&& !lyxaction.funcHasFlag(cmd.action, LyXAction::NoBuffer)) {
+		setStatusMessage(N_("Document is read-only"));
+		flag.enabled(false);
+	}
+
 	return flag;
 }
 

Current CVS can not select with mouse

[Kayvan A. Sylvan]

I don't know when this bug popped up, but I can no longer select with
the mouse (click-hold-drag) although the SHIFT-move method of text
selection seems to work.

-- 
Kayvan A. Sylvan  | Proud husband of   | Father to my kids:
Sylvan Associates, Inc.   | Laura Isabella Sylvan  | Katherine Yelena (8/8/89)
http://sylvan.com/~kayvan | crown of her husband | Robin Gregory (2/28/92)

Current CVS: Can not work in math inset

[Kayvan A. Sylvan]

I can create a math inset (C-m) but I can not navigate into it.

Trying to move the cursor in causes the cursor to bounce to the top of
the document.

-- 
Kayvan A. Sylvan  | Proud husband of   | Father to my kids:
Sylvan Associates, Inc.   | Laura Isabella Sylvan  | Katherine Yelena (8/8/89)
http://sylvan.com/~kayvan | crown of her husband | Robin Gregory (2/28/92)

Current CVS: Insert-File- submenus grayed out

[Kayvan A. Sylvan]

I want to Insert-File-ASCII as Lines, but this option is grayed out
along with LyX Document and Ascii as paragraphs.

-- 
Kayvan A. Sylvan  | Proud husband of   | Father to my kids:
Sylvan Associates, Inc.   | Laura Isabella Sylvan  | Katherine Yelena (8/8/89)
http://sylvan.com/~kayvan | crown of her husband | Robin Gregory (2/28/92)

Current CVS can not select with mouse

[Kayvan A. Sylvan]

I don't know when this bug popped up, but I can no longer select with
the mouse (click-hold-drag) although the SHIFT-move method of text
selection seems to work.

-- 
Kayvan A. Sylvan  | Proud husband of   | Father to my kids:
Sylvan Associates, Inc.   | Laura Isabella Sylvan  | Katherine Yelena (8/8/89)
http://sylvan.com/~kayvan | "crown of her husband" | Robin Gregory (2/28/92)

Current CVS: Can not work in math inset

[Kayvan A. Sylvan]

I can create a math inset (C-m) but I can not navigate into it.

Trying to move the cursor in causes the cursor to bounce to the top of
the document.

-- 
Kayvan A. Sylvan  | Proud husband of   | Father to my kids:
Sylvan Associates, Inc.   | Laura Isabella Sylvan  | Katherine Yelena (8/8/89)
http://sylvan.com/~kayvan | "crown of her husband" | Robin Gregory (2/28/92)

Current CVS: Insert->File-> submenus grayed out

[Kayvan A. Sylvan]

I want to Insert->File->ASCII as Lines, but this option is grayed out
along with "LyX Document" and "Ascii as paragraphs".

-- 
Kayvan A. Sylvan  | Proud husband of   | Father to my kids:
Sylvan Associates, Inc.   | Laura Isabella Sylvan  | Katherine Yelena (8/8/89)
http://sylvan.com/~kayvan | "crown of her husband" | Robin Gregory (2/28/92)

current cvs

[Alfredo Braunstein]

just cvsed up and loaded the userguide:

1) cursor starts in some strange position. going back gives a crash

2) The Edit menu gives a crash

Alfredo

current cvs

[Alfredo Braunstein]

just cvsed up and loaded the userguide:

1) cursor starts in some strange position. going back gives a crash

2) The Edit menu gives a crash

Alfredo

Re: current cvs enters loop...

[Alfredo Braunstein]

Juergen Spitzmueller wrote:

 ...when right-clicking on a collapsable.
 The console loops on HANDLE LFUN_FINISHED_RIGHT.

Also when trying to exit any inset with the arrow keys (loops on the
corresponding LFUN_FINISHED_{UP,DOWN,LEFT,RIGHT}).

Alfredo

Re: current cvs enters loop...

[Alfredo Braunstein]

Juergen Spitzmueller wrote:

> ...when right-clicking on a collapsable.
> The console loops on HANDLE LFUN_FINISHED_RIGHT.

Also when trying to exit any inset with the arrow keys (loops on the
corresponding LFUN_FINISHED_{UP,DOWN,LEFT,RIGHT}).

Alfredo

current cvs enters loop...

[Juergen Spitzmueller]

...when right-clicking on a collapsable.
The console loops on HANDLE LFUN_FINISHED_RIGHT.

Jürgen.

current cvs enters loop...

[Juergen Spitzmueller]

...when right-clicking on a collapsable.
The console loops on HANDLE LFUN_FINISHED_RIGHT.

Jürgen.

Re: Current CVS: Problem with text-wrap float

[Juergen Spitzmueller]

Kayvan A. Sylvan wrote:
 This is with the current CVS:

 When I create a text-wrap float and try to modify its parameters (right
 mouse click and click on Right and set up 25 col%, for example),
 hitting OK or Apply does not work. Nothing is saved and the settings
 are lost.

AFAICS there's something fishy with passing the params to the dialogs (not 
only in floatflt, also in the other floats). If you export to LaTeX, then 
you'll see that the changes _will_ get applied, but they won't be read in if 
you are right-clicking on the inset (string2params problem?).

I'd be very interested on any hint how this could be fixed. I have a patch for 
sideways[figure|table] support lying around...

Jürgen.

Re: Current CVS: Problem with text-wrap float

[Angus Leeming]

Juergen Spitzmueller wrote:

 Kayvan A. Sylvan wrote:
 This is with the current CVS:

 When I create a text-wrap float and try to modify its parameters
 (right mouse click and click on Right and set up 25 col%, for
 example), hitting OK or Apply does not work. Nothing is saved
 and the settings are lost.
 
 AFAICS there's something fishy with passing the params to the
 dialogs (not only in floatflt, also in the other floats). If you
 export to LaTeX, then you'll see that the changes _will_ get
 applied, but they won't be read in if you are right-clicking on the
 inset (string2params problem?).
 
 I'd be very interested on any hint how this could be fixed. I have a
 patch for sideways[figure|table] support lying around...

Well, you receive a string, so print it out:
lyxerr  str  endl;
Now convert it to a param:
float_param param;
string2param(str, param);
Now convert it back and print:
string str2 = param2string(param);
lyxerr  str2  endl;

Are these two print statements the same?

-- 
Angus

Re: Current CVS: Problem with text-wrap float

[Juergen Spitzmueller]

Angus Leeming wrote:
 Well, you receive a string, so print it out:
         lyxerr  str  endl;
 Now convert it to a param:
         float_param param;
         string2param(str, param);
 Now convert it back and print:
         string str2 = param2string(param);
         lyxerr  str2  endl;

 Are these two print statements the same?

Yes.
I added some more debug statements. In InsetFloatParams::read, I added those:

if (lex.isOK()) {
lex.next();
string token = lex.getString();
lyxerr  Should be placement:   token  '\n'  endl;  //-- here
if (token == placement) {
lex.next();
placement = lex.getString();
} else {
// take countermeasures
lex.pushToken(token);
}
lex.next();
token = lex.getString();
lyxerr  Should be wide:   token  '\n'  endl; // -- and here
if (token == wide) {
lex.next();
string const tmptoken = lex.getString();
wide = (tmptoken == true);
} else {
lyxerr  InsetFloat::Read:: Missing wide!
endl;
// take countermeasures
lex.pushToken(token);
}
}

The output is:

Should be placement: table
Should be wide: table
InsetFloat::Read:: Missing wide!

but only when I right-click the inset (open the dialog). Lyx seems to have no 
problems with reading the inset when I open a file.

Ideas?
Jürgen.

Re: Current CVS: Problem with text-wrap float

[Angus Leeming]

Juergen Spitzmueller wrote:

 The output is:
 
 Should be placement: table
 Should be wide: table
 InsetFloat::Read:: Missing wide!
 
 but only when I right-click the inset (open the dialog). Lyx seems
 to have no problems with reading the inset when I open a file.

I'll try and have a look this evening.

-- 
Angus

Re: Current CVS: Problem with text-wrap float

[Juergen Spitzmueller]

Angus Leeming wrote:
 I'll try and have a look this evening.

Thanks.
Jürgen.

Re: Current CVS: Problem with text-wrap float

[Juergen Spitzmueller]

Kayvan A. Sylvan wrote:
> This is with the current CVS:
>
> When I create a text-wrap float and try to modify its parameters (right
> mouse click and click on "Right" and set up "25" "col%", for example),
> hitting "OK" or "Apply" does not work. Nothing is saved and the settings
> are lost.

AFAICS there's something fishy with passing the params to the dialogs (not 
only in floatflt, also in the other floats). If you export to LaTeX, then 
you'll see that the changes _will_ get applied, but they won't be read in if 
you are right-clicking on the inset (string2params problem?).

I'd be very interested on any hint how this could be fixed. I have a patch for 
sideways[figure|table] support lying around...

Jürgen.

Re: Current CVS: Problem with text-wrap float

[Angus Leeming]

Juergen Spitzmueller wrote:

> Kayvan A. Sylvan wrote:
>> This is with the current CVS:
>>
>> When I create a text-wrap float and try to modify its parameters
>> (right mouse click and click on "Right" and set up "25" "col%", for
>> example), hitting "OK" or "Apply" does not work. Nothing is saved
>> and the settings are lost.
> 
> AFAICS there's something fishy with passing the params to the
> dialogs (not only in floatflt, also in the other floats). If you
> export to LaTeX, then you'll see that the changes _will_ get
> applied, but they won't be read in if you are right-clicking on the
> inset (string2params problem?).
> 
> I'd be very interested on any hint how this could be fixed. I have a
> patch for sideways[figure|table] support lying around...

Well, you receive a string, so print it out:
lyxerr << str << endl;
Now convert it to a param:
float_param param;
string2param(str, param);
Now convert it back and print:
string str2 = param2string(param);
lyxerr << str2 << endl;

Are these two print statements the same?

-- 
Angus

Re: Current CVS: Problem with text-wrap float

[Juergen Spitzmueller]

Angus Leeming wrote:
> Well, you receive a string, so print it out:
>         lyxerr << str << endl;
> Now convert it to a param:
>         float_param param;
>         string2param(str, param);
> Now convert it back and print:
>         string str2 = param2string(param);
>         lyxerr << str2 << endl;
>
> Are these two print statements the same?

Yes.
I added some more debug statements. In InsetFloatParams::read, I added those:

if (lex.isOK()) {
lex.next();
string token = lex.getString();
lyxerr << "Should be placement: " << token << '\n' << endl;  //<-- here
if (token == "placement") {
lex.next();
placement = lex.getString();
} else {
// take countermeasures
lex.pushToken(token);
}
lex.next();
token = lex.getString();
lyxerr << "Should be wide: " << token << '\n' << endl; // <-- and here
if (token == "wide") {
lex.next();
string const tmptoken = lex.getString();
wide = (tmptoken == "true");
} else {
lyxerr << "InsetFloat::Read:: Missing wide!"
   << endl;
// take countermeasures
lex.pushToken(token);
}
}

The output is:

Should be placement: table
Should be wide: table
InsetFloat::Read:: Missing wide!

but only when I right-click the inset (open the dialog). Lyx seems to have no 
problems with reading the inset when I open a file.

Ideas?
Jürgen.

Re: Current CVS: Problem with text-wrap float

[Angus Leeming]

Juergen Spitzmueller wrote:

> The output is:
> 
> Should be placement: table
> Should be wide: table
> InsetFloat::Read:: Missing wide!
> 
> but only when I right-click the inset (open the dialog). Lyx seems
> to have no problems with reading the inset when I open a file.

I'll try and have a look this evening.

-- 
Angus

Re: Current CVS: Problem with text-wrap float

[Juergen Spitzmueller]

Angus Leeming wrote:
> I'll try and have a look this evening.

Thanks.
Jürgen.

Current CVS: Problem with text-wrap float

[Kayvan A. Sylvan]

This is with the current CVS:

When I create a text-wrap float and try to modify its parameters (right mouse
click and click on Right and set up 25 col%, for example), hitting OK
or Apply does not work. Nothing is saved and the settings are lost.

---Kayvan
-- 
Kayvan A. Sylvan  | Proud husband of   | Father to my kids:
Sylvan Associates, Inc.   | Laura Isabella Sylvan  | Katherine Yelena (8/8/89)
http://sylvan.com/~kayvan | crown of her husband | Robin Gregory (2/28/92)

Current CVS: Problem with text-wrap float

[Kayvan A. Sylvan]

This is with the current CVS:

When I create a text-wrap float and try to modify its parameters (right mouse
click and click on "Right" and set up "25" "col%", for example), hitting "OK"
or "Apply" does not work. Nothing is saved and the settings are lost.

---Kayvan
-- 
Kayvan A. Sylvan  | Proud husband of   | Father to my kids:
Sylvan Associates, Inc.   | Laura Isabella Sylvan  | Katherine Yelena (8/8/89)
http://sylvan.com/~kayvan | "crown of her husband" | Robin Gregory (2/28/92)

[friendly rant] current cvs status

[Alfredo Braunstein]

Current cvs is crashing everywhere.

One of the problems is that we have lyxtext using the global cursor, when
the relevant slice of this cursor can belong elsewhere.
This can be a nightmare to debug (and maintain). IMVHO, we don't have enough
willing manpower to afford that.

For instance, in text*.C there are calls to 

LyXText::cursor() (who accesses the back of the LCursor, i.e.
LCursor::cursor())
BufferView::cursor().{pos(),par()} (who accesses LCursor::current)

In some cases, these two calls are mixed toghether for getting the cursor
par and pos (see for instance setCurrentFont):

void LyXText::setCurrentFont()
{
LCursor  cur = bv()-cursor();
pos_type pos = cur.pos();
ParagraphList::iterator pit = cursorPar();
...

and cursorPar uses LyXText::cursor.

I think that currently this /may/ be possible to debug (maybe with some
documentation effort from Andre) but it can be a very hard. (in this
particular case, some interpolation makes me believe that
cursorPar/LyXText::cursor should be fixed... is that true?)

IMHO we may have done a step which is a bit too long: we need to settle for
a setup which is not a nightmare to maintain (like *not* using the global
cursor in LyXText members but passing some explicit pit/pos to those that
need it, or something equivalent) /before/ making a switch that breaks
everything.

Of course, I may be messing up things, since I don't understand the cursor
code completely. For instance, should the LyXText ever try to use the
cursor tip, or should it always use the current slice. Does the dispatch
code ensures that the current slice belongs to the LyXText before
dispatching to it? Etc etc. If I am, forget me.

Thoughts?

Alfredo

[friendly rant] current cvs status

[Alfredo Braunstein]

Current cvs is crashing everywhere.

One of the problems is that we have lyxtext using the global cursor, when
the relevant slice of this cursor can belong elsewhere.
This can be a nightmare to debug (and maintain). IMVHO, we don't have enough
willing manpower to afford that.

For instance, in text*.C there are calls to 

LyXText::cursor() (who accesses the back of the LCursor, i.e.
LCursor::cursor())
BufferView::cursor().{pos(),par()} (who accesses LCursor::current)

In some cases, these two calls are mixed toghether for getting the cursor
par and pos (see for instance setCurrentFont):

void LyXText::setCurrentFont()
{
LCursor & cur = bv()->cursor();
pos_type pos = cur.pos();
ParagraphList::iterator pit = cursorPar();
...

and cursorPar uses LyXText::cursor.

I think that currently this /may/ be possible to debug (maybe with some
documentation effort from Andre) but it can be a very hard. (in this
particular case, some interpolation makes me believe that
cursorPar/LyXText::cursor should be fixed... is that true?)

IMHO we may have done a step which is a bit too long: we need to settle for
a setup which is not a nightmare to maintain (like "*not* using the global
cursor in LyXText members but passing some explicit pit/pos to those that
need it", or something equivalent) /before/ making a switch that breaks
everything.

Of course, I may be messing up things, since I don't understand the cursor
code completely. For instance, should the LyXText ever try to use the
cursor tip, or should it always use the "current" slice. Does the dispatch
code ensures that the "current" slice belongs to the LyXText before
dispatching to it? Etc etc. If I am, forget me.

Thoughts?

Alfredo

Re: Current CVS lyx: selection does not work

[Andre Poenitz]

On Tue, Jan 20, 2004 at 02:06:39PM -0800, Kayvan A. Sylvan wrote:
 You can not select text in the current CVS lyx. Neither hold-left-mouse and
 drag nor they keyboard method (shift+arrow keys) work.

I am happy enough that it doesn't crash...

Andre'

-- 
Those who desire to give up Freedom in order to gain Security, will not have,
nor do they deserve, either one. (T. Jefferson or B. Franklin or both...)

Current CVS does not compile

[Kayvan A. Sylvan]

g++ -DHAVE_CONFIG_H -I. -I../../../lyx/src/mathed -I../../src 
-I../../../lyx/src/mathed/../ -I../../../lyx/boost -I/usr/X11R6/include -O2 
-fno-exceptions -W -Wall -mms-bitfields -MT formulabase.lo -MD -MP -MF 
.deps/formulabase.Tpo -c ../../../lyx/src/mathed/formulabase.C
../../../lyx/src/mathed/formulabase.C:303:2: warning: #warning FIXME
../../../lyx/src/mathed/formulabase.C:789:2: warning: #warning pretty ugly
../../../lyx/src/mathed/formulabase.C: In function `void 
   mathDispatchCreation(LCursor, const FuncRequest, bool)':
../../../lyx/src/mathed/formulabase.C:862: error: `InsetFormula' undeclared 
   (first use this function)
../../../lyx/src/mathed/formulabase.C:862: error: (Each undeclared identifier 
   is reported only once for each function it appears in.)
../../../lyx/src/mathed/formulabase.C:862: error: `f' undeclared (first use 
   this function)
../../../lyx/src/mathed/formulabase.C:862: error: parse error before `;' token
../../../lyx/src/mathed/formulabase.C:878: error: parse error before `(' token
../../../lyx/src/mathed/formulabase.C: In function `void mathDispatch(LCursor, 
   const FuncRequest)':
../../../lyx/src/mathed/formulabase.C:916: error: cannot convert `
   InsetFormulaMacro*' to `UpdatableInset*' for argument `2' to `bool 
   unnamed::openNewInset(LCursor, UpdatableInset*)'
../../../lyx/src/mathed/formulabase.C:923: error: parse error before `;' token
make[3]: *** [formulabase.lo] Error 1


-- 
Kayvan A. Sylvan  | Proud husband of   | Father to my kids:
Sylvan Associates, Inc.   | Laura Isabella Sylvan  | Katherine Yelena (8/8/89)
http://sylvan.com/~kayvan | crown of her husband | Robin Gregory (2/28/92)

Re: Current CVS does not compile

[Lars Gullik Bjønnes]

Kayvan A. Sylvan [EMAIL PROTECTED] writes:

| g++ -DHAVE_CONFIG_H -I. -I../../../lyx/src/mathed -I../../src 
-I../../../lyx/src/mathed/../ -I../../../lyx/boost -I/usr/X11R6/include -O2 
-fno-exceptions -W -Wall -mms-bitfields -MT formulabase.lo -MD -MP -MF 
.deps/formulabase.Tpo -c ../../../lyx/src/mathed/formulabase.C
| ../../../lyx/src/mathed/formulabase.C:303:2: warning: #warning FIXME
| ../../../lyx/src/mathed/formulabase.C:789:2: warning: #warning pretty ugly
| ../../../lyx/src/mathed/formulabase.C: In function `void 
|mathDispatchCreation(LCursor, const FuncRequest, bool)':
| ../../../lyx/src/mathed/formulabase.C:862: error: `InsetFormula' undeclared 
|(first use this function)
| ../../../lyx/src/mathed/formulabase.C:862: error: (Each undeclared identifier 

autogen, configure.

-- 
Lgb

Re: Current CVS does not compile

[Kayvan A. Sylvan]

On Mon, Jan 26, 2004 at 05:58:54PM +0100, Lars Gullik Bjønnes wrote:
 Kayvan A. Sylvan [EMAIL PROTECTED] writes:
 
 | g++ -DHAVE_CONFIG_H -I. -I../../../lyx/src/mathed -I../../src 
 -I../../../lyx/src/mathed/../ -I../../../lyx/boost -I/usr/X11R6/include -O2 
 -fno-exceptions -W -Wall -mms-bitfields -MT formulabase.lo -MD -MP -MF 
 .deps/formulabase.Tpo -c ../../../lyx/src/mathed/formulabase.C
 | ../../../lyx/src/mathed/formulabase.C:303:2: warning: #warning FIXME
 | ../../../lyx/src/mathed/formulabase.C:789:2: warning: #warning pretty ugly
 | ../../../lyx/src/mathed/formulabase.C: In function `void 
 |mathDispatchCreation(LCursor, const FuncRequest, bool)':
 | ../../../lyx/src/mathed/formulabase.C:862: error: `InsetFormula' undeclared 
 |(first use this function)
 | ../../../lyx/src/mathed/formulabase.C:862: error: (Each undeclared identifier 
 
 autogen, configure.
 

Thanks. Fixed.
-- 
Kayvan A. Sylvan  | Proud husband of   | Father to my kids:
Sylvan Associates, Inc.   | Laura Isabella Sylvan  | Katherine Yelena (8/8/89)
http://sylvan.com/~kayvan | crown of her husband | Robin Gregory (2/28/92)

Re: Current CVS lyx: selection does not work

[Andre Poenitz]

On Tue, Jan 20, 2004 at 02:06:39PM -0800, Kayvan A. Sylvan wrote:
> You can not select text in the current CVS lyx. Neither hold-left-mouse and
> drag nor they keyboard method (shift+arrow keys) work.

I am happy enough that it doesn't crash...

Andre'

-- 
Those who desire to give up Freedom in order to gain Security, will not have,
nor do they deserve, either one. (T. Jefferson or B. Franklin or both...)

Current CVS does not compile

[Kayvan A. Sylvan]

g++ -DHAVE_CONFIG_H -I. -I../../../lyx/src/mathed -I../../src 
-I../../../lyx/src/mathed/../ -I../../../lyx/boost -I/usr/X11R6/include -O2 
-fno-exceptions -W -Wall -mms-bitfields -MT formulabase.lo -MD -MP -MF 
.deps/formulabase.Tpo -c ../../../lyx/src/mathed/formulabase.C
../../../lyx/src/mathed/formulabase.C:303:2: warning: #warning FIXME
../../../lyx/src/mathed/formulabase.C:789:2: warning: #warning pretty ugly
../../../lyx/src/mathed/formulabase.C: In function `void 
   mathDispatchCreation(LCursor&, const FuncRequest&, bool)':
../../../lyx/src/mathed/formulabase.C:862: error: `InsetFormula' undeclared 
   (first use this function)
../../../lyx/src/mathed/formulabase.C:862: error: (Each undeclared identifier 
   is reported only once for each function it appears in.)
../../../lyx/src/mathed/formulabase.C:862: error: `f' undeclared (first use 
   this function)
../../../lyx/src/mathed/formulabase.C:862: error: parse error before `;' token
../../../lyx/src/mathed/formulabase.C:878: error: parse error before `(' token
../../../lyx/src/mathed/formulabase.C: In function `void mathDispatch(LCursor&, 
   const FuncRequest&)':
../../../lyx/src/mathed/formulabase.C:916: error: cannot convert `
   InsetFormulaMacro*' to `UpdatableInset*' for argument `2' to `bool 
   ::openNewInset(LCursor&, UpdatableInset*)'
../../../lyx/src/mathed/formulabase.C:923: error: parse error before `;' token
make[3]: *** [formulabase.lo] Error 1


-- 
Kayvan A. Sylvan  | Proud husband of   | Father to my kids:
Sylvan Associates, Inc.   | Laura Isabella Sylvan  | Katherine Yelena (8/8/89)
http://sylvan.com/~kayvan | "crown of her husband" | Robin Gregory (2/28/92)

Re: Current CVS does not compile

[Lars Gullik Bjønnes]

"Kayvan A. Sylvan" <[EMAIL PROTECTED]> writes:

| g++ -DHAVE_CONFIG_H -I. -I../../../lyx/src/mathed -I../../src 
-I../../../lyx/src/mathed/../ -I../../../lyx/boost -I/usr/X11R6/include -O2 
-fno-exceptions -W -Wall -mms-bitfields -MT formulabase.lo -MD -MP -MF 
.deps/formulabase.Tpo -c ../../../lyx/src/mathed/formulabase.C
| ../../../lyx/src/mathed/formulabase.C:303:2: warning: #warning FIXME
| ../../../lyx/src/mathed/formulabase.C:789:2: warning: #warning pretty ugly
| ../../../lyx/src/mathed/formulabase.C: In function `void 
|mathDispatchCreation(LCursor&, const FuncRequest&, bool)':
| ../../../lyx/src/mathed/formulabase.C:862: error: `InsetFormula' undeclared 
|(first use this function)
| ../../../lyx/src/mathed/formulabase.C:862: error: (Each undeclared identifier 

autogen, configure.

-- 
Lgb

Re: Current CVS does not compile

[Kayvan A. Sylvan]

On Mon, Jan 26, 2004 at 05:58:54PM +0100, Lars Gullik Bjønnes wrote:
> "Kayvan A. Sylvan" <[EMAIL PROTECTED]> writes:
> 
> | g++ -DHAVE_CONFIG_H -I. -I../../../lyx/src/mathed -I../../src 
> -I../../../lyx/src/mathed/../ -I../../../lyx/boost -I/usr/X11R6/include -O2 
> -fno-exceptions -W -Wall -mms-bitfields -MT formulabase.lo -MD -MP -MF 
> .deps/formulabase.Tpo -c ../../../lyx/src/mathed/formulabase.C
> | ../../../lyx/src/mathed/formulabase.C:303:2: warning: #warning FIXME
> | ../../../lyx/src/mathed/formulabase.C:789:2: warning: #warning pretty ugly
> | ../../../lyx/src/mathed/formulabase.C: In function `void 
> |mathDispatchCreation(LCursor&, const FuncRequest&, bool)':
> | ../../../lyx/src/mathed/formulabase.C:862: error: `InsetFormula' undeclared 
> |(first use this function)
> | ../../../lyx/src/mathed/formulabase.C:862: error: (Each undeclared identifier 
> 
> autogen, configure.
> 

Thanks. Fixed.
-- 
Kayvan A. Sylvan  | Proud husband of   | Father to my kids:
Sylvan Associates, Inc.   | Laura Isabella Sylvan  | Katherine Yelena (8/8/89)
http://sylvan.com/~kayvan | "crown of her husband" | Robin Gregory (2/28/92)

Current CVS lyx: selection does not work

[Kayvan A. Sylvan]

You can not select text in the current CVS lyx. Neither hold-left-mouse and
drag nor they keyboard method (shift+arrow keys) work.

Best regards,
---Kayvan
-- 
Kayvan A. Sylvan  | Proud husband of   | Father to my kids:
Sylvan Associates, Inc.   | Laura Isabella Sylvan  | Katherine Yelena (8/8/89)
http://sylvan.com/~kayvan | crown of her husband | Robin Gregory (2/28/92)

Current CVS lyx: selection does not work

[Kayvan A. Sylvan]

You can not select text in the current CVS lyx. Neither hold-left-mouse and
drag nor they keyboard method (shift+arrow keys) work.

Best regards,
---Kayvan
-- 
Kayvan A. Sylvan  | Proud husband of   | Father to my kids:
Sylvan Associates, Inc.   | Laura Isabella Sylvan  | Katherine Yelena (8/8/89)
http://sylvan.com/~kayvan | "crown of her husband" | Robin Gregory (2/28/92)

Re: crash with current cvs

[Andre Poenitz]

On Wed, Jan 07, 2004 at 12:24:43AM +0100, Alfredo Braunstein wrote:
 open UG, click on first footnote - 
 
 #0  0x080ff99c in Row::pos() const ()
 #1  0x0811a765 in Paragraph::getRow(int) (this=0x83c7a88, pos=0)
 at ../../src/paragraph.C:1831
 #2  0x0813ef5d in LyXText::cursorX(LyXCursor const) const (this=0x8586eb4,
 [EMAIL PROTECTED]) at ../../src/text.C:1873
 #3  0x0813eef0 in LyXText::cursorX() const (this=0x8586eb4)
 at ../../src/text.C:1860
 #4  0x08145e98 in LyXText::setCursorIntern(int, int, bool, bool) (
 this=0x8586eb4, par=0, pos=0, setfont=true, boundary=false)
 at ../../src/text2.C:1215
 #5  0x08145a3f in LyXText::setCursor(int, int, bool, bool) (this=0x8586eb4,
 par=0, pos=0, setfont=true, boundary=false) at ../../src/text2.C:1155
 
 A quick inspection shows that uncollapsing the cursor is about to be set on
 the first par of the inset that for some reason seems to have 0 rows.

The reason is probably that the row cache is empty as the row was never
drawn.

I am not sure how a proper solution would look like.

 I think that this was somewhat recently introduced (maybe a couple of
 weeks?).

Maybe.

Andre'

-- 
Those who desire to give up Freedom in order to gain Security, will not have,
nor do they deserve, either one. (T. Jefferson or B. Franklin or both...)

Re: crash with current cvs

[Andre Poenitz]

On Wed, Jan 07, 2004 at 09:07:24AM +0100, Andre' Poenitz wrote:
 The reason is probably that the row cache is empty as the row was never
 drawn.
 
 I am not sure how a proper solution would look like.

The attached patch paints over the problem...

Andre'
Index: paragraph.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/paragraph.C,v
retrieving revision 1.348
diff -u -p -r1.348 paragraph.C
--- paragraph.C 1 Dec 2003 13:35:40 -   1.348
+++ paragraph.C 7 Jan 2004 08:09:00 -
@@ -64,7 +64,7 @@ using std::ostringstream;
 
 
 Paragraph::Paragraph()
-   : y(0), height(0), begin_of_body_(0),
+   : rows(1), y(0), height(0), begin_of_body_(0),
  pimpl_(new Paragraph::Pimpl(this))
 {
itemdepth = 0;
@@ -73,7 +73,8 @@ Paragraph::Paragraph()
 
 
 Paragraph::Paragraph(Paragraph const  par)
-   : y(0), height(0), text_(par.text_), begin_of_body_(par.begin_of_body_),
+   : rows(par.rows), y(0), height(0),
+ text_(par.text_), begin_of_body_(par.begin_of_body_),
  pimpl_(new Paragraph::Pimpl(*par.pimpl_, this))
 {

Re: crash with current cvs

[Alfredo Braunstein]

Andre Poenitz wrote:

 On Wed, Jan 07, 2004 at 09:07:24AM +0100, Andre' Poenitz wrote:
 The reason is probably that the row cache is empty as the row was never
 drawn.
 
 I am not sure how a proper solution would look like.
 
 The attached patch paints over the problem...

;-)

Note that rows are needed on setCursor only for setting the correct
bv-x_target (I think). I wonder if we can simply avoid that somehow.

Alfredo

Re: crash with current cvs

[Andre Poenitz]

On Wed, Jan 07, 2004 at 09:22:32AM +0100, Alfredo Braunstein wrote:
 Andre Poenitz wrote:
 
  On Wed, Jan 07, 2004 at 09:07:24AM +0100, Andre' Poenitz wrote:
  The reason is probably that the row cache is empty as the row was never
  drawn.
  
  I am not sure how a proper solution would look like.
  
  The attached patch paints over the problem...
 
 ;-)
 
 Note that rows are needed on setCursor only for setting the correct
 bv-x_target (I think). I wonder if we can simply avoid that somehow.

Well, an 'if (rows.empty())' in this place might work then as well.

Andre'

-- 
Those who desire to give up Freedom in order to gain Security, will not have,
nor do they deserve, either one. (T. Jefferson or B. Franklin or both...)

Re: crash with current cvs

[Andre Poenitz]

On Wed, Jan 07, 2004 at 09:22:32AM +0100, Alfredo Braunstein wrote:
 Andre Poenitz wrote:
 
  On Wed, Jan 07, 2004 at 09:07:24AM +0100, Andre' Poenitz wrote:
  The reason is probably that the row cache is empty as the row was never
  drawn.
  
  I am not sure how a proper solution would look like.
  
  The attached patch paints over the problem...
 
 ;-)
 
 Note that rows are needed on setCursor only for setting the correct
 bv-x_target (I think). I wonder if we can simply avoid that somehow.

See attached.

Andre'

-- 
Those who desire to give up Freedom in order to gain Security, will not have,
nor do they deserve, either one. (T. Jefferson or B. Franklin or both...)
Index: text.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/text.C,v
retrieving revision 1.512
diff -u -p -r1.512 text.C
--- text.C  18 Dec 2003 04:43:06 -  1.512
+++ text.C  7 Jan 2004 08:34:04 -
@@ -1870,6 +1870,8 @@ int LyXText::cursorY() const
 int LyXText::cursorX(LyXCursor const  cur) const
 {
ParagraphList::iterator pit = getPar(cur);
+   if (pit-rows.empty())
+   return 0;
Row const  row = *pit-getRow(cur.pos());
pos_type pos= cur.pos();
pos_type cursor_vpos= 0;

Re: crash with current cvs

[Alfredo Braunstein]

Andre Poenitz wrote:

 Note that rows are needed on setCursor only for setting the correct
 bv-x_target (I think). I wonder if we can simply avoid that somehow.
 
 See attached.

Solved thanks.

Btw, the first rebreak after uncollapsing is wrong, but there are other
width problems so hopefully they are all related ;-).

Alfredo

Re: crash with current cvs

[Andre Poenitz]

On Wed, Jan 07, 2004 at 12:24:43AM +0100, Alfredo Braunstein wrote:
> open UG, click on first footnote -> 
> 
> #0  0x080ff99c in Row::pos() const ()
> #1  0x0811a765 in Paragraph::getRow(int) (this=0x83c7a88, pos=0)
> at ../../src/paragraph.C:1831
> #2  0x0813ef5d in LyXText::cursorX(LyXCursor const&) const (this=0x8586eb4,
> [EMAIL PROTECTED]) at ../../src/text.C:1873
> #3  0x0813eef0 in LyXText::cursorX() const (this=0x8586eb4)
> at ../../src/text.C:1860
> #4  0x08145e98 in LyXText::setCursorIntern(int, int, bool, bool) (
> this=0x8586eb4, par=0, pos=0, setfont=true, boundary=false)
> at ../../src/text2.C:1215
> #5  0x08145a3f in LyXText::setCursor(int, int, bool, bool) (this=0x8586eb4,
> par=0, pos=0, setfont=true, boundary=false) at ../../src/text2.C:1155
> 
> A quick inspection shows that uncollapsing the cursor is about to be set on
> the first par of the inset that for some reason seems to have 0 rows.

The reason is probably that the row cache is empty as the row was never
drawn.

I am not sure how a proper solution would look like.

> I think that this was somewhat recently introduced (maybe a couple of
> weeks?).

Maybe.

Andre'

-- 
Those who desire to give up Freedom in order to gain Security, will not have,
nor do they deserve, either one. (T. Jefferson or B. Franklin or both...)

Re: crash with current cvs

[Andre Poenitz]

On Wed, Jan 07, 2004 at 09:07:24AM +0100, Andre' Poenitz wrote:
> The reason is probably that the row cache is empty as the row was never
> drawn.
> 
> I am not sure how a proper solution would look like.

The attached patch paints over the problem...

Andre'
Index: paragraph.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/paragraph.C,v
retrieving revision 1.348
diff -u -p -r1.348 paragraph.C
--- paragraph.C 1 Dec 2003 13:35:40 -   1.348
+++ paragraph.C 7 Jan 2004 08:09:00 -
@@ -64,7 +64,7 @@ using std::ostringstream;
 
 
 Paragraph::Paragraph()
-   : y(0), height(0), begin_of_body_(0),
+   : rows(1), y(0), height(0), begin_of_body_(0),
  pimpl_(new Paragraph::Pimpl(this))
 {
itemdepth = 0;
@@ -73,7 +73,8 @@ Paragraph::Paragraph()
 
 
 Paragraph::Paragraph(Paragraph const & par)
-   : y(0), height(0), text_(par.text_), begin_of_body_(par.begin_of_body_),
+   : rows(par.rows), y(0), height(0),
+ text_(par.text_), begin_of_body_(par.begin_of_body_),
  pimpl_(new Paragraph::Pimpl(*par.pimpl_, this))
 {

Re: crash with current cvs

[Alfredo Braunstein]

Andre Poenitz wrote:

> On Wed, Jan 07, 2004 at 09:07:24AM +0100, Andre' Poenitz wrote:
>> The reason is probably that the row cache is empty as the row was never
>> drawn.
>> 
>> I am not sure how a proper solution would look like.
> 
> The attached patch paints over the problem...

;-)

Note that rows are needed on setCursor only for setting the correct
bv->x_target (I think). I wonder if we can simply avoid that somehow.

Alfredo

Re: crash with current cvs

[Andre Poenitz]

On Wed, Jan 07, 2004 at 09:22:32AM +0100, Alfredo Braunstein wrote:
> Andre Poenitz wrote:
> 
> > On Wed, Jan 07, 2004 at 09:07:24AM +0100, Andre' Poenitz wrote:
> >> The reason is probably that the row cache is empty as the row was never
> >> drawn.
> >> 
> >> I am not sure how a proper solution would look like.
> > 
> > The attached patch paints over the problem...
> 
> ;-)
> 
> Note that rows are needed on setCursor only for setting the correct
> bv->x_target (I think). I wonder if we can simply avoid that somehow.

Well, an 'if (rows.empty())' in this place might work then as well.

Andre'

-- 
Those who desire to give up Freedom in order to gain Security, will not have,
nor do they deserve, either one. (T. Jefferson or B. Franklin or both...)

Re: crash with current cvs

[Andre Poenitz]

On Wed, Jan 07, 2004 at 09:22:32AM +0100, Alfredo Braunstein wrote:
> Andre Poenitz wrote:
> 
> > On Wed, Jan 07, 2004 at 09:07:24AM +0100, Andre' Poenitz wrote:
> >> The reason is probably that the row cache is empty as the row was never
> >> drawn.
> >> 
> >> I am not sure how a proper solution would look like.
> > 
> > The attached patch paints over the problem...
> 
> ;-)
> 
> Note that rows are needed on setCursor only for setting the correct
> bv->x_target (I think). I wonder if we can simply avoid that somehow.

See attached.

Andre'

-- 
Those who desire to give up Freedom in order to gain Security, will not have,
nor do they deserve, either one. (T. Jefferson or B. Franklin or both...)
Index: text.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/text.C,v
retrieving revision 1.512
diff -u -p -r1.512 text.C
--- text.C  18 Dec 2003 04:43:06 -  1.512
+++ text.C  7 Jan 2004 08:34:04 -
@@ -1870,6 +1870,8 @@ int LyXText::cursorY() const
 int LyXText::cursorX(LyXCursor const & cur) const
 {
ParagraphList::iterator pit = getPar(cur);
+   if (pit->rows.empty())
+   return 0;
Row const & row = *pit->getRow(cur.pos());
pos_type pos= cur.pos();
pos_type cursor_vpos= 0;

Re: crash with current cvs

[Alfredo Braunstein]

Andre Poenitz wrote:

>> Note that rows are needed on setCursor only for setting the correct
>> bv->x_target (I think). I wonder if we can simply avoid that somehow.
> 
> See attached.

Solved thanks.

Btw, the first rebreak after uncollapsing is wrong, but there are other
width problems so hopefully they are all related ;-).

Alfredo

crash with current cvs

[Alfredo Braunstein]

open UG, click on first footnote - 

#0  0x080ff99c in Row::pos() const ()
#1  0x0811a765 in Paragraph::getRow(int) (this=0x83c7a88, pos=0)
at ../../src/paragraph.C:1831
#2  0x0813ef5d in LyXText::cursorX(LyXCursor const) const (this=0x8586eb4,
[EMAIL PROTECTED]) at ../../src/text.C:1873
#3  0x0813eef0 in LyXText::cursorX() const (this=0x8586eb4)
at ../../src/text.C:1860
#4  0x08145e98 in LyXText::setCursorIntern(int, int, bool, bool) (
this=0x8586eb4, par=0, pos=0, setfont=true, boundary=false)
at ../../src/text2.C:1215
#5  0x08145a3f in LyXText::setCursor(int, int, bool, bool) (this=0x8586eb4,
par=0, pos=0, setfont=true, boundary=false) at ../../src/text2.C:1155

A quick inspection shows that uncollapsing the cursor is about to be set on
the first par of the inset that for some reason seems to have 0 rows.

I think that this was somewhat recently introduced (maybe a couple of
weeks?).

Alfredo

crash with current cvs

[Alfredo Braunstein]

open UG, click on first footnote -> 

#0  0x080ff99c in Row::pos() const ()
#1  0x0811a765 in Paragraph::getRow(int) (this=0x83c7a88, pos=0)
at ../../src/paragraph.C:1831
#2  0x0813ef5d in LyXText::cursorX(LyXCursor const&) const (this=0x8586eb4,
[EMAIL PROTECTED]) at ../../src/text.C:1873
#3  0x0813eef0 in LyXText::cursorX() const (this=0x8586eb4)
at ../../src/text.C:1860
#4  0x08145e98 in LyXText::setCursorIntern(int, int, bool, bool) (
this=0x8586eb4, par=0, pos=0, setfont=true, boundary=false)
at ../../src/text2.C:1215
#5  0x08145a3f in LyXText::setCursor(int, int, bool, bool) (this=0x8586eb4,
par=0, pos=0, setfont=true, boundary=false) at ../../src/text2.C:1155

A quick inspection shows that uncollapsing the cursor is about to be set on
the first par of the inset that for some reason seems to have 0 rows.

I think that this was somewhat recently introduced (maybe a couple of
weeks?).

Alfredo

Re: Auto-build failure of current CVS

[Kayvan A. Sylvan]

On Tue, Jul 29, 2003 at 06:10:36AM -0700, Kayvan A. Sylvan wrote:
 Problem with make dist:
 

This patch fixes the automated build problems.

-- 
Kayvan A. Sylvan  | Proud husband of   | Father to my kids:
Sylvan Associates, Inc.   | Laura Isabella Sylvan  | Katherine Yelena (8/8/89)
http://sylvan.com/~kayvan | crown of her husband | Robin Gregory (2/28/92)
Index: ./config/Makefile.am
===
RCS file: /cvs/lyx/lyx-devel/config/Makefile.am,v
retrieving revision 1.9
diff -u -u -r1.9 Makefile.am
--- ./config/Makefile.am2003/03/26 01:20:24 1.9
+++ ./config/Makefile.am2003/07/31 07:57:10
@@ -1,7 +1,7 @@
 include $(top_srcdir)/config/common.am
 
 EXTRA_DIST = common.am codeset.m4 libtool.m4 gettext.m4 \
-   lyxinclude.m4 lyxinclude25x.m4 lyxinclude213.m4 \
+   lyxinclude.m4 lyxinclude25x.m4 \
glibc21.m4 iconv.m4 isc-posix.m4 lcmessage.m4 progtest.m4 \
qt.m4 gtk--.m4 gnome--.m4 gnome.m4 pkg.m4 xforms.m4 pspell.m4 \
-   aspell.m4 relyx_configure.ac relyx_configure.in cygwin.m4
+   aspell.m4 cygwin.m4
Index: ./src/Makefile.am
===
RCS file: /cvs/lyx/lyx-devel/src/Makefile.am,v
retrieving revision 1.182
diff -u -u -r1.182 Makefile.am
--- ./src/Makefile.am   2003/07/28 14:40:26 1.182
+++ ./src/Makefile.am   2003/07/31 07:57:10
@@ -12,7 +12,8 @@
Variables.C \
Variables.h \
tracer.C \
-   tracer.h
+   tracer.h \
+   paper.h
 
 bin_PROGRAMS = lyx
 
Index: ./src/support/Makefile.am
===
RCS file: /cvs/lyx/lyx-devel/src/support/Makefile.am,v
retrieving revision 1.62
diff -u -u -r1.62 Makefile.am
--- ./src/support/Makefile.am   2003/07/28 23:05:58 1.62
+++ ./src/support/Makefile.am   2003/07/31 07:57:10
@@ -8,7 +8,7 @@
 INCLUDES = -I$(srcdir)/../ $(BOOST_INCLUDES)
 
 EXTRA_DIST = path_defines.C.in lyxstring.C lyxstring.h \
-os_unix.C os_win32.C os_os2.C
+os_unix.C os_win32.C os_os2.C gzstream.h
 
 if USE_LYXSTRING
 LYXSTRING = lyxstring.C lyxstring.h

Re: Auto-build failure of current CVS

[Lars Gullik Bjønnes]

Kayvan A. Sylvan [EMAIL PROTECTED] writes:

| On Tue, Jul 29, 2003 at 06:10:36AM -0700, Kayvan A. Sylvan wrote:
|  Problem with make dist:
|  
| 
| This patch fixes the automated build problems.

Yeah. this looks ok.

Unless somebody beats me to it, I will commit this later tonight.

-- 
Lgb

Re: Auto-build failure of current CVS

[Kayvan A. Sylvan]

On Tue, Jul 29, 2003 at 06:10:36AM -0700, Kayvan A. Sylvan wrote:
> Problem with "make dist":
> 

This patch fixes the automated build problems.

-- 
Kayvan A. Sylvan  | Proud husband of   | Father to my kids:
Sylvan Associates, Inc.   | Laura Isabella Sylvan  | Katherine Yelena (8/8/89)
http://sylvan.com/~kayvan | "crown of her husband" | Robin Gregory (2/28/92)
Index: ./config/Makefile.am
===
RCS file: /cvs/lyx/lyx-devel/config/Makefile.am,v
retrieving revision 1.9
diff -u -u -r1.9 Makefile.am
--- ./config/Makefile.am2003/03/26 01:20:24 1.9
+++ ./config/Makefile.am2003/07/31 07:57:10
@@ -1,7 +1,7 @@
 include $(top_srcdir)/config/common.am
 
 EXTRA_DIST = common.am codeset.m4 libtool.m4 gettext.m4 \
-   lyxinclude.m4 lyxinclude25x.m4 lyxinclude213.m4 \
+   lyxinclude.m4 lyxinclude25x.m4 \
glibc21.m4 iconv.m4 isc-posix.m4 lcmessage.m4 progtest.m4 \
qt.m4 gtk--.m4 gnome--.m4 gnome.m4 pkg.m4 xforms.m4 pspell.m4 \
-   aspell.m4 relyx_configure.ac relyx_configure.in cygwin.m4
+   aspell.m4 cygwin.m4
Index: ./src/Makefile.am
===
RCS file: /cvs/lyx/lyx-devel/src/Makefile.am,v
retrieving revision 1.182
diff -u -u -r1.182 Makefile.am
--- ./src/Makefile.am   2003/07/28 14:40:26 1.182
+++ ./src/Makefile.am   2003/07/31 07:57:10
@@ -12,7 +12,8 @@
Variables.C \
Variables.h \
tracer.C \
-   tracer.h
+   tracer.h \
+   paper.h
 
 bin_PROGRAMS = lyx
 
Index: ./src/support/Makefile.am
===
RCS file: /cvs/lyx/lyx-devel/src/support/Makefile.am,v
retrieving revision 1.62
diff -u -u -r1.62 Makefile.am
--- ./src/support/Makefile.am   2003/07/28 23:05:58 1.62
+++ ./src/support/Makefile.am   2003/07/31 07:57:10
@@ -8,7 +8,7 @@
 INCLUDES = -I$(srcdir)/../ $(BOOST_INCLUDES)
 
 EXTRA_DIST = path_defines.C.in lyxstring.C lyxstring.h \
-os_unix.C os_win32.C os_os2.C
+os_unix.C os_win32.C os_os2.C gzstream.h
 
 if USE_LYXSTRING
 LYXSTRING = lyxstring.C lyxstring.h

Re: Auto-build failure of current CVS

[Lars Gullik Bjønnes]

"Kayvan A. Sylvan" <[EMAIL PROTECTED]> writes:

| On Tue, Jul 29, 2003 at 06:10:36AM -0700, Kayvan A. Sylvan wrote:
| > Problem with "make dist":
| > 
| 
| This patch fixes the automated build problems.

Yeah. this looks ok.

Unless somebody beats me to it, I will commit this later tonight.

-- 
Lgb

Re: Auto-build failure of current CVS

[Kayvan A. Sylvan]

Problem with make dist:

 ? po/ui_l10n.pot
 ? src/frontends/xforms/lyx_forms.h
 ? src/frontends/xforms/lyx_xpm.h
 cvs server: Updating .
 P ChangeLog
 P configure.ac
[...]
 P src/tex2lyx/text.C
 cvs server: Updating src/xtl
 Using autoconf (GNU Autoconf) 2.57
 Locating GNU m4... /usr/bin/m4
 Generate acinclude.m4... done.
 Building macros...
 .
 lib/reLyX
 done.
 Building config header template...
 .
 done.
 Building Makefile templates...
 .
 boost/libs/regex/src/Makefile.am: trailing backslash on last line
 src/insets/Makefile.am: trailing backslash on last line
 lib/reLyX
 done.
 Building configure...
.
lib/reLyX
 done.
 Building lib/configure ... done.
 
 run ./configure ; make
 
 Running ./configure --with-frontend=xforms
 configuring LyX version 1.4.0cvs
 WARNING: This is a development version. Expect bugs.
 checking whether to enable maintainer-specific portions of Makefiles... no
[...]
 config.status: creating noweb2lyx
 
 Configuration
   Host type:  i686-pc-linux-gnu
   Special build flags:warnings assertions xforms-image-loader compression
   C   Compiler:   gcc
   C   Compiler flags: -g -O2
   C++ Compiler:   g++ (2.96)
   C++ Compiler flags: -g -O -fno-exceptions -ftemplate-depth-30 
 -Wno-non-template-friend -W -Wall
   Linker flags:   
   Frontend:   xforms
 libXpm version:   4.11
 libforms version: 1.0.0
   LyX binary dir: /usr/local/bin
   LyX files dir:  /usr/local/share/lyx
 
 Configuration of LyX was successful.
 Type 'make' to compile the program,
 and then 'make install' to install it.
 { test ! -d lyx-1.4.0cvs || { find lyx-1.4.0cvs -type d ! -perm -200 -exec chmod u+w 
 {} ';'  rm -fr lyx-1.4.0cvs; }; }
 mkdir lyx-1.4.0cvs
 /bin/sh ./config/mkinstalldirs lyx-1.4.0cvs/./config lyx-1.4.0cvs/development 
 lyx-1.4.0cvs/intl lyx-1.4.0cvs/po lyx-1.4.0cvs/sourcedoc lyx-1.4.0cvs/src 
 lyx-1.4.0cvs/src/frontends/xforms
 mkdir lyx-1.4.0cvs/./config
 mkdir lyx-1.4.0cvs/development
 mkdir lyx-1.4.0cvs/intl
 mkdir lyx-1.4.0cvs/po
 mkdir lyx-1.4.0cvs/sourcedoc
 mkdir lyx-1.4.0cvs/src
 mkdir lyx-1.4.0cvs/src/frontends
 mkdir lyx-1.4.0cvs/src/frontends/xforms
 list='config development intl po boost src sourcedoc lib'; for subdir in $list; do \
   if test $subdir = .; then :; else \
 test -d lyx-1.4.0cvs/$subdir \
 || mkdir lyx-1.4.0cvs/$subdir \
 || exit 1; \
 (cd $subdir  \
   make  \
 top_distdir=. \
 distdir=../lyx-1.4.0cvs/$subdir \
 distdir) \
   || exit 1; \
   fi; \
 done
 make[1]: Entering directory `/home/kayvan/src/lyx/config'
 make[1]: *** No rule to make target `lyxinclude213.m4', needed by `distdir'.  Stop.
 make[1]: Leaving directory `/home/kayvan/src/lyx/config'

-- 
Kayvan A. Sylvan  | Proud husband of   | Father to my kids:
Sylvan Associates, Inc.   | Laura Isabella Sylvan  | Katherine Yelena (8/8/89)
http://sylvan.com/~kayvan | crown of her husband | Robin Gregory (2/28/92)

Re: Auto-build failure of current CVS

[Kayvan A. Sylvan]

Problem with "make dist":

> ? po/ui_l10n.pot
> ? src/frontends/xforms/lyx_forms.h
> ? src/frontends/xforms/lyx_xpm.h
> cvs server: Updating .
> P ChangeLog
> P configure.ac
[...]
> P src/tex2lyx/text.C
> cvs server: Updating src/xtl
> Using autoconf (GNU Autoconf) 2.57
> Locating GNU m4... /usr/bin/m4
> Generate acinclude.m4... done.
> Building macros...
> .
> lib/reLyX
> done.
> Building config header template...
> .
> done.
> Building Makefile templates...
> .
> boost/libs/regex/src/Makefile.am: trailing backslash on last line
> src/insets/Makefile.am: trailing backslash on last line
> lib/reLyX
> done.
> Building configure...
>.
>lib/reLyX
> done.
> Building lib/configure ... done.
> 
> run "./configure ; make"
> 
> Running ./configure --with-frontend=xforms
> configuring LyX version 1.4.0cvs
> WARNING: This is a development version. Expect bugs.
> checking whether to enable maintainer-specific portions of Makefiles... no
[...]
> config.status: creating noweb2lyx
> 
> Configuration
>   Host type:  i686-pc-linux-gnu
>   Special build flags:warnings assertions xforms-image-loader compression
>   C   Compiler:   gcc
>   C   Compiler flags: -g -O2
>   C++ Compiler:   g++ (2.96)
>   C++ Compiler flags: -g -O -fno-exceptions -ftemplate-depth-30 
> -Wno-non-template-friend -W -Wall
>   Linker flags:   
>   Frontend:   xforms
> libXpm version:   4.11
> libforms version: 1.0.0
>   LyX binary dir: /usr/local/bin
>   LyX files dir:  /usr/local/share/lyx
> 
> Configuration of LyX was successful.
> Type 'make' to compile the program,
> and then 'make install' to install it.
> { test ! -d lyx-1.4.0cvs || { find lyx-1.4.0cvs -type d ! -perm -200 -exec chmod u+w 
> {} ';' && rm -fr lyx-1.4.0cvs; }; }
> mkdir lyx-1.4.0cvs
> /bin/sh ./config/mkinstalldirs lyx-1.4.0cvs/./config lyx-1.4.0cvs/development 
> lyx-1.4.0cvs/intl lyx-1.4.0cvs/po lyx-1.4.0cvs/sourcedoc lyx-1.4.0cvs/src 
> lyx-1.4.0cvs/src/frontends/xforms
> mkdir lyx-1.4.0cvs/./config
> mkdir lyx-1.4.0cvs/development
> mkdir lyx-1.4.0cvs/intl
> mkdir lyx-1.4.0cvs/po
> mkdir lyx-1.4.0cvs/sourcedoc
> mkdir lyx-1.4.0cvs/src
> mkdir lyx-1.4.0cvs/src/frontends
> mkdir lyx-1.4.0cvs/src/frontends/xforms
> list='config development intl po boost src sourcedoc lib'; for subdir in $list; do \
>   if test "$subdir" = .; then :; else \
> test -d lyx-1.4.0cvs/$subdir \
> || mkdir lyx-1.4.0cvs/$subdir \
> || exit 1; \
> (cd $subdir && \
>   make  \
> top_distdir="." \
> distdir=../lyx-1.4.0cvs/$subdir \
> distdir) \
>   || exit 1; \
>   fi; \
> done
> make[1]: Entering directory `/home/kayvan/src/lyx/config'
> make[1]: *** No rule to make target `lyxinclude213.m4', needed by `distdir'.  Stop.
> make[1]: Leaving directory `/home/kayvan/src/lyx/config'

-- 
Kayvan A. Sylvan  | Proud husband of   | Father to my kids:
Sylvan Associates, Inc.   | Laura Isabella Sylvan  | Katherine Yelena (8/8/89)
http://sylvan.com/~kayvan | "crown of her husband" | Robin Gregory (2/28/92)

Re: current cvs random crashes

[Alfredo Braunstein]

Lars Gullik Bjnnes wrote:

 Alfredo Braunstein [EMAIL PROTECTED] writes:
 
 | It wasn't, but it doesn't make a difference. I got the same bt. Then
 | retrying under gdb, I got also this one:
 
 Actaully it makes a huge difference, no the line numbers and files
 match up.

Of course. I meant the same crash.

 And no signal raised...

This difference is for running under gdb. I get the same as before if I
don't:

#0  0xe002 in ?? ()
(gdb) bt
#0  0xe002 in ?? ()
#1  0x42028a73 in abort () from /lib/tls/libc.so.6
#2  0x08340783 in lyx::abort() () at /home/ab/lyx-cvs/src/support/abort.C:19
#3  0x080f460a in error_handler (err_sig=11)
at /home/ab/lyx-cvs/src/lyx_main.C:208
#4  signal handler called
#5  0x0812c1a2 in Paragraph::getPositionOfInset(Inset const*) const (
this=0x869dcc8, inset=0x86252f0) at /home/ab/lyx-cvs/src/paragraph.C:742
#6  0x0814d821 in LyXText::workWidth(Inset const*) const (this=0x869dcc8,
inset=0x86252f0) at /home/ab/lyx-cvs/src/text.C:155

 
 what is it here?


// returns -1 if inset not found
int Paragraph::getPositionOfInset(Inset const * inset) const
{
// Find the entry.
InsetList::const_iterator it = insetlist.begin();
InsetList::const_iterator end = insetlist.end();
for (; it != end; ++it)
if (it-inset == inset)
return it-pos;
return -1;
}

(gdb) p it
$1 =
{iteratorstd::random_access_iterator_tag,InsetList::InsetTable,int,const
InsetList::InsetTable*,const InsetList::InsetTable = {No data fields},
  _M_current = 0x6e65704f}
(gdb) p *it
$3 = (const InsetTable ) @0x6e65704f: Cannot access memory at address
0x6e65704f


Regards, Alfredo

Re: current cvs random crashes

[Alfredo Braunstein]

Lars Gullik Bjønnes wrote:

> Alfredo Braunstein <[EMAIL PROTECTED]> writes:
> 
> | It wasn't, but it doesn't make a difference. I got the same bt. Then
> | retrying under gdb, I got also this one:
> 
> Actaully it makes a huge difference, no the line numbers and files
> match up.

Of course. I meant the same crash.

> And no signal raised...

This difference is for running under gdb. I get the same as before if I
don't:

#0  0xe002 in ?? ()
(gdb) bt
#0  0xe002 in ?? ()
#1  0x42028a73 in abort () from /lib/tls/libc.so.6
#2  0x08340783 in lyx::abort() () at /home/ab/lyx-cvs/src/support/abort.C:19
#3  0x080f460a in error_handler (err_sig=11)
at /home/ab/lyx-cvs/src/lyx_main.C:208
#4  
#5  0x0812c1a2 in Paragraph::getPositionOfInset(Inset const*) const (
this=0x869dcc8, inset=0x86252f0) at /home/ab/lyx-cvs/src/paragraph.C:742
#6  0x0814d821 in LyXText::workWidth(Inset const*) const (this=0x869dcc8,
inset=0x86252f0) at /home/ab/lyx-cvs/src/text.C:155

> 
> what is "it" here?


// returns -1 if inset not found
int Paragraph::getPositionOfInset(Inset const * inset) const
{
// Find the entry.
InsetList::const_iterator it = insetlist.begin();
InsetList::const_iterator end = insetlist.end();
for (; it != end; ++it)
if (it->inset == inset)
return it->pos;
return -1;
}

(gdb) p it
$1 =
{> = {},
  _M_current = 0x6e65704f}
(gdb) p *it
$3 = (const InsetTable &) @0x6e65704f: Cannot access memory at address
0x6e65704f


Regards, Alfredo

current cvs random crashes

[Alfredo Braunstein]

With current cvs, only by opening and closing some documents randomly, at
some point I get this abort.

I doesn't happend very often (and doesn't depend on the particular
documents).

Regards, Alfredo


#0  0xe002 in ?? ()
(gdb) bt
#0  0xe002 in ?? ()
#1  0x42028a73 in abort () from /lib/tls/libc.so.6
#2  0x082c56d3 in lyx::abort() () at /home/ab/lyx-cvs/src/support/abort.C:19
#3  0x080d5eab in error_handler (err_sig=11)
at /home/ab/lyx-cvs/src/lyx_main.C:208
#4  signal handler called
#5  Paragraph::getPositionOfInset(Inset const*) const (this=0x8647868,
inset=0x865e0d8) at stl_iterator.h:596
#6  0x081204da in LyXText::workWidth(Inset const*) const (this=0x8647868,
inset=0x865e0d8) at stl_list.h:142
#7  0x081bdc27 in UpdatableInset::getMaxWidth(BufferView*, UpdatableInset
const*) const (this=0x865e0d8, bv=0x6)
at /home/ab/lyx-cvs/src/insets/updatableinset.C:145
#8  0x081bdc1a in UpdatableInset::getMaxWidth(BufferView*, UpdatableInset
const*) const (this=0x865e12c, bv=0x6) at inset.h:214
#9  0x081b875a in InsetText::getMaxWidth(BufferView*, UpdatableInset const*)
const (this=0x865e12c, bv=0x8499c38, inset=0x865e12c)
at /home/ab/lyx-cvs/src/insets/insettext.C:1932
#10 0x081b370c in InsetText::textWidth(BufferView*, bool) const (
this=0x865e12c, bv=0x8499c38, fordraw=false)
at /home/ab/lyx-cvs/src/insets/insettext.C:304
#11 0x0812044c in LyXText::workWidth() const (this=0x858c528)
at /home/ab/lyx-cvs/src/text.C:142
---Type return to continue, or q return to quit---
#12 0x081224c2 in LyXText::rowBreakPoint(Row const) const (this=0x858c528,
[EMAIL PROTECTED]) at /home/ab/lyx-cvs/src/text.C:778
#13 0x08124996 in LyXText::breakAgain(std::_List_iteratorRow, Row, Row*)
(
this=0x858c528, rit=
{_List_iterator_base = {_M_node = 0x85faf50}, No data fields})
at stl_list.h:138
#14 0x081308d4 in LyXText::checkParagraph(std::_List_iteratorParagraph,
Paragraph, Paragraph*, int) (this=0x858c528, pit=
{_List_iterator_base = {_M_node = 0x865a230}, No data fields},
pos=12) at stl_list.h:91
#15 0x08130c4d in LyXText::updateInset(Inset*) (this=0x858c528,
inset=0x865ec68) at stl_list.h:91
#16 0x081b49d4 in InsetText::updateInsetInInset(BufferView*, Inset*) (
this=0x865e12c, bv=0x8499c38, inset=0x865ec68)
at /home/ab/lyx-cvs/src/insets/insettext.C:759
#17 0x081c0f8f in InsetCollapsable::updateInsetInInset(BufferView*, Inset*)
(
this=0x6, bv=0x8499c38, in=0x42130a14)
at /home/ab/lyx-cvs/src/insets/insetcollapsable.C:411
#18 0x0806afe0 in BufferView::Pimpl::updateInset(Inset*) (this=0x8499c88,
inset=0x865ec68) at /home/ab/lyx-cvs/src/BufferView_pimpl.C:1346
#19 0x0806358c in BufferView::updateInset(Inset*) (this=0x0,
inset=0x865ec68)
at /home/ab/lyx-cvs/src/BufferView.C:614
#20 0x082c2cee in grfx::PreviewedInset::imageReady(grfx::PreviewImage
const) co---Type return to continue, or q return to quit---
nst (this=0x865de18, [EMAIL PROTECTED])
at /home/ab/lyx-cvs/src/graphics/PreviewedInset.C:125
#21 0x082c2d92 in
boost::detail::function::void_function_obj_invoker1boost::_bi::bind_tvoid,
boost::_mfi::cmf1void, grfx::PreviewedInset, grfx::PreviewImage const,
boost::_bi::list2boost::_bi::valuegrfx::PreviewedInset*, boost::arg1 
, void, grfx::PreviewImage
const::invoke(boost::detail::function::any_pointer, grfx::PreviewImage
const) (function_obj_ptr=
  {obj_ptr = 0x0, const_obj_ptr = 0x0, func_ptr = 0, data = },
[EMAIL PROTECTED]) at mem_fn_template.hpp:166
#22 0x082bfc04 in boost::signal1void, grfx::PreviewImage const,
boost::last_valuevoid, int, std::lessint, boost::function1void,
grfx::PreviewImage const, std::allocatorboost::function_base 
::operator()(grfx::PreviewImage const) (this=0x866bcb0, [EMAIL PROTECTED]) at
function_template.hpp:321
#23 0x082beb44 in grfx::PreviewLoader::Impl::finishedGenerating(int, int) (
this=0x866bca8, pid=30629, retval=-1073750881) at shared_ptr.hpp:243
#24 0x082c07dd in
boost::detail::function::void_function_obj_invoker2boost::_bi::bind_tvoid,
boost::_mfi::mf2void, grfx::PreviewLoader::Impl, int, int,
boost::_bi::list3boost::_bi::valuegrfx::PreviewLoader::Impl*,
boost::arg1, boost::arg2  , void, int,
int::invoke(boost::detail::function::any_pointer, int, int)
(function_obj_ptr=
  {obj_ptr = 0x6, const_obj_ptr = 0x6, func_ptr = 0x6, data = \006},
a0=30629, a1=0) at mem_fn_template.hpp:213
#25 0x082d14f8 in boost::signal2void, int, int, boost::last_valuevoid,
int, s---Type return to continue, or q return to quit---
td::lessint, boost::function2void, int, int,
std::allocatorboost::function_base  ::operator()(int, int)
(this=0x8645f88, a1=0, a2=0)
at function_template.hpp:321
#26 0x082d0856 in ForkedProcess::emitSignal() (this=0x0) at
shared_ptr.hpp:237
#27 0x082d3e82 in ForkedcallsController::timer() (this=0x8371b00)
at /home/ab/lyx-cvs/src/support/forkedcontr.C:135
#28 0x082d42b1 in
boost::detail::function::void_function_obj_invoker0boost::_bi::bind_tvoid,
boost::_mfi

current cvs random crashes

[Alfredo Braunstein]

With current cvs, only by opening and closing some documents randomly, at
some point I get this abort.

I doesn't happend very often (and doesn't depend on the particular
documents).

Regards, Alfredo


#0  0xe002 in ?? ()
(gdb) bt
#0  0xe002 in ?? ()
#1  0x42028a73 in abort () from /lib/tls/libc.so.6
#2  0x082c56d3 in lyx::abort() () at /home/ab/lyx-cvs/src/support/abort.C:19
#3  0x080d5eab in error_handler (err_sig=11)
at /home/ab/lyx-cvs/src/lyx_main.C:208
#4  
#5  Paragraph::getPositionOfInset(Inset const*) const (this=0x8647868,
inset=0x865e0d8) at stl_iterator.h:596
#6  0x081204da in LyXText::workWidth(Inset const*) const (this=0x8647868,
inset=0x865e0d8) at stl_list.h:142
#7  0x081bdc27 in UpdatableInset::getMaxWidth(BufferView*, UpdatableInset
const*) const (this=0x865e0d8, bv=0x6)
at /home/ab/lyx-cvs/src/insets/updatableinset.C:145
#8  0x081bdc1a in UpdatableInset::getMaxWidth(BufferView*, UpdatableInset
const*) const (this=0x865e12c, bv=0x6) at inset.h:214
#9  0x081b875a in InsetText::getMaxWidth(BufferView*, UpdatableInset const*)
const (this=0x865e12c, bv=0x8499c38, inset=0x865e12c)
at /home/ab/lyx-cvs/src/insets/insettext.C:1932
#10 0x081b370c in InsetText::textWidth(BufferView*, bool) const (
this=0x865e12c, bv=0x8499c38, fordraw=false)
at /home/ab/lyx-cvs/src/insets/insettext.C:304
#11 0x0812044c in LyXText::workWidth() const (this=0x858c528)
at /home/ab/lyx-cvs/src/text.C:142
---Type  to continue, or q  to quit---
#12 0x081224c2 in LyXText::rowBreakPoint(Row const&) const (this=0x858c528,
[EMAIL PROTECTED]) at /home/ab/lyx-cvs/src/text.C:778
#13 0x08124996 in LyXText::breakAgain(std::_List_iterator<Row, Row&, Row*>)
(
this=0x858c528, rit=
{<_List_iterator_base> = {_M_node = 0x85faf50}, })
at stl_list.h:138
#14 0x081308d4 in LyXText::checkParagraph(std::_List_iterator<Paragraph,
Paragraph&, Paragraph*>, int) (this=0x858c528, pit=
{<_List_iterator_base> = {_M_node = 0x865a230}, },
pos=12) at stl_list.h:91
#15 0x08130c4d in LyXText::updateInset(Inset*) (this=0x858c528,
inset=0x865ec68) at stl_list.h:91
#16 0x081b49d4 in InsetText::updateInsetInInset(BufferView*, Inset*) (
this=0x865e12c, bv=0x8499c38, inset=0x865ec68)
at /home/ab/lyx-cvs/src/insets/insettext.C:759
#17 0x081c0f8f in InsetCollapsable::updateInsetInInset(BufferView*, Inset*)
(
this=0x6, bv=0x8499c38, in=0x42130a14)
at /home/ab/lyx-cvs/src/insets/insetcollapsable.C:411
#18 0x0806afe0 in BufferView::Pimpl::updateInset(Inset*) (this=0x8499c88,
inset=0x865ec68) at /home/ab/lyx-cvs/src/BufferView_pimpl.C:1346
#19 0x0806358c in BufferView::updateInset(Inset*) (this=0x0,
inset=0x865ec68)
at /home/ab/lyx-cvs/src/BufferView.C:614
#20 0x082c2cee in grfx::PreviewedInset::imageReady(grfx::PreviewImage
const&) co---Type  to continue, or q  to quit---
nst (this=0x865de18, [EMAIL PROTECTED])
at /home/ab/lyx-cvs/src/graphics/PreviewedInset.C:125
#21 0x082c2d92 in
boost::detail::function::void_function_obj_invoker1<boost::_bi::bind_t<void,
boost::_mfi::cmf1<void, grfx::PreviewedInset, grfx::PreviewImage const&>,
boost::_bi::list2<boost::_bi::value<grfx::PreviewedInset*>, boost::arg<1> >
>, void, grfx::PreviewImage
const&>::invoke(boost::detail::function::any_pointer, grfx::PreviewImage
const&) (function_obj_ptr=
  {obj_ptr = 0x0, const_obj_ptr = 0x0, func_ptr = 0, data = ""},
[EMAIL PROTECTED]) at mem_fn_template.hpp:166
#22 0x082bfc04 in boost::signal1<void, grfx::PreviewImage const&,
boost::last_value, int, std::less, boost::function1<void,
grfx::PreviewImage const&, std::allocator >
>::operator()(grfx::PreviewImage const&) (this=0x866bcb0, [EMAIL PROTECTED]) at
function_template.hpp:321
#23 0x082beb44 in grfx::PreviewLoader::Impl::finishedGenerating(int, int) (
this=0x866bca8, pid=30629, retval=-1073750881) at shared_ptr.hpp:243
#24 0x082c07dd in
boost::detail::function::void_function_obj_invoker2<boost::_bi::bind_t<void,
boost::_mfi::mf2<void, grfx::PreviewLoader::Impl, int, int>,
boost::_bi::list3<boost::_bi::value<grfx::PreviewLoader::Impl*>,
boost::arg<1>, boost::arg<2> > >, void, int,
int>::invoke(boost::detail::function::any_pointer, int, int)
(function_obj_ptr=
  {obj_ptr = 0x6, const_obj_ptr = 0x6, func_ptr = 0x6, data = "\006"},
a0=30629, a1=0) at mem_fn_template.hpp:213
#25 0x082d14f8 in boost::signal2<void, int, int, boost::last_value,
int, s---Type  to continue, or q  to quit---
td::less, boost::function2<void, int, int,
std::allocator > >::operator()(int, int)
(this=0x8645f88, a1=0, a2=0)
at function_template.hpp:321
#26 0x082d0856 in ForkedProcess::emitSignal() (this=0x0) at
shared_ptr.hpp:237
#27 0x082d3e82 in ForkedcallsController::timer() (this=0x8371b00)
at /home/ab/lyx-cvs/src/support/forkedcon