RE: [Bug 62] inset in inset and undo/redo/ drawing problems
On 06-Jan-2002 bugzilla-daemon wrote: --- Additional Comments From [EMAIL PROTECTED] 2002-01-07 09:03 --- Ok, I don't see problems now (prolly from previous redo fixes) Well no I guess that I fixed the drawing problems with my commit yesterday! Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ I keep seeing spots in front of my eyes. Did you ever see a doctor? No, just spots.
Re: [Patch] fixed bug 59
On Sun, Jan 06, 2002 at 09:03:15PM +, John Levon wrote: On Sun, Jan 06, 2002 at 10:26:08PM +0200, Martin Vermeer wrote: On Fri, Jan 04, 2002 at 08:36:58AM +0100, Michael Schmitt wrote: - Insert a (math) inset into a table cell. Then, move the cursor along the current row. The lines at the bottom of other cells are erased (on screen) (#59) Fixed, and did some cleanup and documentation too, for my own consumption ;-) tested it, it's OK in all cases I know of except the problem with attachment 37 is still there. I think it should be applied anyway. If you like you can fix 37 later ;) The culprit of 37 is the call tabular-GetCellInset(cell)-draw(bv, font, baseline, dx, false); Perhaps I'll look at it later if I find time. good stuff ! + //lyxerr tabular-GetAscentOfRow(i) + // tabular-GetDescentOfRow(i) + // - the_locking_inset-ascent(bv, font) + // the_locking_inset-descent(bv, font) + // TEXT_TO_INSET_OFFSET endl; why is this bit here ? Wrote it before I found out that half of the parameters printed were irrelevant ;-) Actually I think it can go. regards john p.s. even better would be to see this section of code in a separate function !! g. Martin -- Martin Vermeer [EMAIL PROTECTED] Helsinki University of Technology Department of Surveying P.O. Box 1200, FIN-02015 HUT, Finland :wq msg31007/pgp0.pgp Description: PGP signature
Re: New insights on bugs
On 06-Jan-2002 Lars Gullik Bjønnes wrote: | - Some (biggest) memory leaks [snip] | createUndo(BufferView*,Undo::undo_kind,const Paragraph*,const Paragraph*) I really cannot see how and why this happens. Well I can and we ALWAYS had this problem. It's easy when undoing/redoing changes the substituted paragraphs are not freed but their pointer are lost. I put some code in the undo_funcs.C to handle this case but I crash with that code and I really cannot see why. So Michael if you have access next time to a good memory tool, have a look at undo_funcs.C and remove the comment on the #define at the beginning as this then activates the freeing of the paragraphs but this also gives a core dump on undo/redo operations, but maybe we can see what's wrong with that code then. | * This is occurring while in: | InsetTabular::insetInInsetY() [insettabular.C:703] | BufferView::fitLockedInsetCursor(int,int,int,int) | [BufferView2.C:499] Try this patch and see what happens: Well the below patch should not be neccessary! If we access that variable and it was not set before then we have a bug somewhere else in the code! I hope I fixed this with my last commits (yesterday). | InsetTabular::clone(const Buffer,bool)const [insettabular.C:175] | Paragraph::Paragraph(const Paragraph,bool) [paragraph.C:145] This look dangerous. Why this is just that an InsetTabular was cloned and then the undo operation locked the inset without setting the inset_y position, which then is accessed. The badest thing which can occur in such an operation is that the cursor is somewhere too much down and we will see an empty gray screen :), but no crush can occur in this situation, this may also explain the bugs we have open where we sometime see this situation, but as I told you I hope I fixed this. | #0 0x080cda79 in Paragraph::inInset (this=0x837c128) at | paragraph.C:2083 and how can inInset crash? [snip] Well I guess John explained this already! Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ wok, n.: Something to thwow at a wabbit.
Re: New insights on bugs
On 05-Jan-2002 John Levon wrote: hmm, this looks like the_locking_inset is referring to a deleted inset. I don't get this code too much ... Nope the_locking_inset is ok inset_y is not ok (it was not initialized), but I explained this already in my former mail! Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Life is like bein' on a mule team. Unless you're the lead mule, all the scenery looks about the same.
Re: KLyX -- am I missing something?
John == John Levon [EMAIL PROTECTED] writes: John you've missed the point. A checkbox is a checkbox is a checkbox. John A radio button is another thing altogether. Most toolkits have John visual clues for telling the difference. Xforms has this too. What about using roundbutton or round3dbutton for radio buttons? JMarc
Re: KLyX -- am I missing something?
On Mon, Jan 07, 2002 at 09:34:31AM +0100, Jean-Marc Lasgouttes wrote: John you've missed the point. A checkbox is a checkbox is a checkbox. John A radio button is another thing altogether. Most toolkits have John visual clues for telling the difference. Xforms has this too. What about using roundbutton or round3dbutton for radio buttons? oh, then it's a lyx bug. Didn't realise. regards john -- Unless everyone else on earth is attending meetings I haven't been told about. - /. paranoia at its finest
[PATCH] auto-apply for FormTabular
I think this is the simplest and best solution for now.
please apply
john
--
Unless everyone else on earth is attending meetings I haven't been told
about.
- /. paranoia at its finest
Index: src/frontends/xforms/ChangeLog
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/ChangeLog,v
retrieving revision 1.219
diff -u -p -r1.219 ChangeLog
--- src/frontends/xforms/ChangeLog 2002/01/04 15:14:22 1.219
+++ src/frontends/xforms/ChangeLog 2002/01/07 08:46:48
@@ -1,3 +1,12 @@
+2002-01-07 John Levon [EMAIL PROTECTED]
+
+ * FormTabular.h:
+ * FormTabular.C:
+ * form_tabular.h:
+ * form_tabular.C:
+ * forms/form_tabular.fd: ugly fix to get around
+ some of the UI problems (bug #110).
+
2002-01-04 Jürgen Spitzmüller [EMAIL PROTECTED]
* FormDocument.C: Recognize the default paper settings too
Index: src/frontends/xforms/FormTabular.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/FormTabular.C,v
retrieving revision 1.31
diff -u -p -r1.31 FormTabular.C
--- src/frontends/xforms/FormTabular.C 2001/12/20 14:52:15 1.31
+++ src/frontends/xforms/FormTabular.C 2002/01/07 08:46:52
@@ -153,6 +153,9 @@ void FormTabular::build()
void FormTabular::update()
{
+ if (closing_)
+ return;
+
if (!inset_ || !inset_-tabular.get())
return;
@@ -479,7 +482,20 @@ bool FormTabular::input(FL_OBJECT * ob,
string special;;
int cell = inset_-getActCell();
-
+
+// ugly hack to auto-apply the stuff that hasn't been
+// yet. don't let this continue to exist ...
+if (ob == dialog_-button_close) {
+closing_ = true;
+input(column_options_-input_column_width, 0);
+input(cell_options_-input_mcolumn_width, 0);
+input(column_options_-input_special_alignment, 0);
+input(cell_options_-input_special_multialign, 0);
+closing_ = false;
+ ok();
+return true;
+}
+
if (actCell_ != cell) {
update();
fl_set_object_label(dialog_-text_warning,
Index: src/frontends/xforms/FormTabular.h
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/FormTabular.h,v
retrieving revision 1.15
diff -u -p -r1.15 FormTabular.h
--- src/frontends/xforms/FormTabular.h 2001/12/19 16:13:21 1.15
+++ src/frontends/xforms/FormTabular.h 2002/01/07 08:46:52
@@ -90,6 +90,8 @@ private:
int actCell_;
/// The ButtonController
ButtonControllerOkCancelReadOnlyPolicy, xformsBC bc_;
+ /// if we are applying stuff during a close of the dialog
+ bool closing_;
};
Index: src/frontends/xforms/form_tabular.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/form_tabular.C,v
retrieving revision 1.21
diff -u -p -r1.21 form_tabular.C
--- src/frontends/xforms/form_tabular.C 2001/12/19 16:13:21 1.21
+++ src/frontends/xforms/form_tabular.C 2002/01/07 08:46:53
@@ -33,7 +33,7 @@ FD_form_tabular * FormTabular::build_tab
fl_set_button_shortcut(obj, scex(_(dummy)), 1);
}
fl_set_object_lsize(obj, FL_NORMAL_SIZE);
-fl_set_object_callback(obj, C_FormBaseDeprecatedCancelCB, 0);
+fl_set_object_callback(obj, C_FormBaseDeprecatedInputCB, 0);
fdui-input_tabular_column = obj = fl_add_input(FL_NORMAL_INPUT, 65, 260, 60, 30,
);
fl_set_object_lsize(obj, FL_NORMAL_SIZE);
fl_set_object_lalign(obj, FL_ALIGN_TOP_LEFT);
Index: src/frontends/xforms/form_tabular.h
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/form_tabular.h,v
retrieving revision 1.16
diff -u -p -r1.16 form_tabular.h
--- src/frontends/xforms/form_tabular.h 2001/12/19 16:13:21 1.16
+++ src/frontends/xforms/form_tabular.h 2002/01/07 08:46:53
@@ -5,7 +5,7 @@
#define FD_form_tabular_h_
/** Callbacks, globals and object handlers **/
-extern C void C_FormBaseDeprecatedCancelCB(FL_OBJECT *, long);
+extern C void C_FormBaseDeprecatedInputCB(FL_OBJECT *, long);
extern C void C_FormBaseDeprecatedInputCB(FL_OBJECT *, long);
Index: src/frontends/xforms/forms/form_tabular.fd
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/forms/form_tabular.fd,v
retrieving revision 1.14
diff -u -p -r1.14 form_tabular.fd
--- src/frontends/xforms/forms/form_tabular.fd 2001/12/19 16:13:21 1.14
+++ src/frontends/xforms/forms/form_tabular.fd 2002/01/07 08:46:55
@@ -64,7 +64,7 @@ shortcut:
resize: FL_RESIZE_ALL
gravity: FL_NoGravity FL_NoGravity
name: button_close
-callback: C_FormBaseDeprecatedCancelCB
+callback: C_FormBaseDeprecatedInputCB
argument: 0
Re: KLyX -- am I missing something?
John == John Levon [EMAIL PROTECTED] writes: John On Mon, Jan 07, 2002 at 09:34:31AM +0100, Jean-Marc Lasgouttes John wrote: you've missed the point. A checkbox is a checkbox is a John checkbox. A radio button is another thing altogether. Most John toolkits have visual clues for telling the difference. Xforms has this too. What about using roundbutton or round3dbutton for radio buttons? John oh, then it's a lyx bug. Didn't realise. Indeed. This probably dates back to the first versions... We should maybe change all the radio buttons (not checkbox) to round3dbutton for 1.2.0. Then people will be glad that the interface is much better dur tu GUI-I :) JMarc
Re: New insights on bugs
On 07-Jan-2002 Lars Gullik Bjønnes wrote: Class variables should be initialized anyway, especially pointers. Well if this is the rule just commit your change it surely won't do anything bad, the only thing we won't get debug notifies anymore and so we cannot be pointed to problems in the code by them. Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ No yak too dirty; no dumpster too hollow.
RE: [PATCH] auto-apply for FormTabular
On 07-Jan-2002 John Levon wrote: I think this is the simplest and best solution for now. Well I don't know John! What if I typed something there by error I now have no way to return to the original. If we do this then at least we have to ask the user if he would like us to apply the changes. But then we didn't ask before to (when we had the autoapply on entering data!), so maybe this is really the best choice we have now. I'll apply it in the meantime! Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ You can observe a lot just by watching. -- Yogi Berra
Re: New insights on bugs
On 07-Jan-2002 Lars Gullik Bjønnes wrote:
Pointer * pointer;
...
if (!pointer) {
pointer = new Pointer;
}
will always have problems.
I don't get what you mean here (especially in the contents we are talking
about), but as I told you I don't argue that all class variables should be
initialized I just say that if I would have done this with the variables in
cause I would not have discovered the hidden bug Michael pointed me too. So
my argument was that maybe it's sometimes better to not initialize them if
we know that the should be initialized before we use them.
Jug
--
-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._
Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED]
Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253
I-39100 Bozen Web: http://www.sad.it/~jug
-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._
Cole's Law:
Thinly sliced cabbage.
Re: [PATCH] auto-apply for FormTabular
On Mon, Jan 07, 2002 at 10:04:50AM +0100, Juergen Vigna wrote: Well I don't know John! What if I typed something there by error I now have no way to return to the original. If we do this then at least we there's no point worrying about dog poo on the streets when city hall is on fire ... or something. It's good enough I think... regards john -- Unless everyone else on earth is attending meetings I haven't been told about. - /. paranoia at its finest
Re: More bugs (=new list)
Kayvan == Kayvan A Sylvan [EMAIL PROTECTED] writes: Kayvan On Fri, Jan 04, 2002 at 04:30:51PM +0100, Jean-Marc Lasgouttes Kayvan wrote: Michael Only occurs when you click at menu item View/DVI; CTRL-D Michael works well (#122) I know how to fix that and will do so next week (once I can enjoy celibacy again...). Kayvan This sounds interesting... Do tell. ;-) What? The fix or the celibacy? For the latter, do like me: drive your wife and kid to Antwerpen train station tomorrow morning. Or maybe Antwerpen is a bit too far for you... JMarc
Re: [PATCH] FileInfo misuse
John == John Levon [EMAIL PROTECTED] writes: John this fixes most of the cases, and asserts on the others John please apply Applied locally. JMarc
Re: noweb2lyx updates for lyx-1.2.0cvs
Kayvan == Kayvan A Sylvan [EMAIL PROTECTED] writes: Kayvan Hi folks, Here is an update for the noweb2lyx.lyx example Kayvan file. This does not change noweb2lyx.in in any way. Applied locally JMarc
[Patch] fixed bug 59 attachment 37 too ;-)
As it says.
I must admit that I don't really quite understand this,
and thus don't like it. How can baseline 0 if not in
error?
Martin
--
Martin Vermeer [EMAIL PROTECTED]
Helsinki University of Technology
Department of Surveying
P.O. Box 1200, FIN-02015 HUT, Finland
:wq
Index: insettext.C
===
RCS file: /cvs/lyx/lyx-devel/src/insets/insettext.C,v
retrieving revision 1.245
diff -u -p -r1.245 insettext.C
--- insettext.C 2001/12/28 13:26:54 1.245
+++ insettext.C 2002/01/07 09:14:32
@@ -388,7 +388,8 @@ void InsetText::draw(BufferView * bv, Ly
if (!cleared (top_x == int(x))
((need_update(INIT|FULL)) || (top_baseline != baseline)
||(last_drawn_width != insetWidth))) {
- clearInset(bv, baseline, cleared);
+ // Condition necessary to eliminate bug 59 attachment 37
+ if (baseline 0) clearInset(bv, baseline, cleared);
}
top_x = int(x);
msg31021/pgp0.pgp
Description: PGP signature
Re: [Patch] fixed bug 59
Martin == Martin Vermeer [EMAIL PROTECTED] writes: Martin On Fri, Jan 04, 2002 at 08:36:58AM +0100, Michael Schmitt Martin wrote: - Insert a (math) inset into a table cell. Then, move the cursor along the current row. The lines at the bottom of other cells are erased (on screen) (#59) Martin Fixed, and did some cleanup and documentation too, for my own Martin consumption ;-) I applied it in my tree. What about a ChangeLog entry? I removed also the debugging statements said to be useless. What about the others? JMarc
Re: [PATCH] auto-apply for FormTabular
On 07-Jan-2002 John Levon wrote: It's good enough I think... Well yes but could you send me a ChangeLog? (I already applied the patch so please only the ChangeLog!) Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Mirrors should reflect a little before throwing back images. -- Jean Cocteau
Re: [PATCH] fix insets wider than screen
John == John Levon [EMAIL PROTECTED] writes: John I hadn't considered that inset_x could be 0. This patch fixes John bug #162 reported by Herbert. John please apply john Done. JMarc
Re: [PATCH] auto-apply for FormTabular
On Mon, Jan 07, 2002 at 10:20:21AM +0100, Juergen Vigna wrote: Well yes but could you send me a ChangeLog? (I already applied the patch so please only the ChangeLog!) the patch didn't include it ? 2002-01-07 John Levon [EMAIL PROTECTED] * FormTabular.h: * FormTabular.C: * form_tabular.h: * form_tabular.C: * forms/form_tabular.fd: ugly fix to get around some of the UI problems (bug #110). regards john -- Unless everyone else on earth is attending meetings I haven't been told about. - /. paranoia at its finest
Re: [Patch] fixed bug 59
On Mon, Jan 07, 2002 at 10:21:38AM +0100, Jean-Marc Lasgouttes wrote: Martin == Martin Vermeer [EMAIL PROTECTED] writes: Martin On Fri, Jan 04, 2002 at 08:36:58AM +0100, Michael Schmitt Martin wrote: - Insert a (math) inset into a table cell. Then, move the cursor along the current row. The lines at the bottom of other cells are erased (on screen) (#59) Martin Fixed, and did some cleanup and documentation too, for my own Martin consumption ;-) I applied it in my tree. What about a ChangeLog entry? Like: --- 2002-01-07 Martin Vermeer [EMAIL PROTECTED] * insettabular.C: * insettext.C: Fix for bug #59: Insert a (math) inset into a table cell. Then, move the cursor along the current row. The lines at the bottom of other cells are erased (on screen). The insettext.C fix eliminates bug illustrated by attachment 37. --- (Assumes the additional fix also applied) I removed also the debugging statements said to be useless. What about the others? IMO they can go if this is found to work flawlessly. Easily re-instated if needed. JMarc Martin -- Martin Vermeer [EMAIL PROTECTED] Helsinki University of Technology Department of Surveying P.O. Box 1200, FIN-02015 HUT, Finland :wq msg31026/pgp0.pgp Description: PGP signature
Re: KLyX -- am I missing something?
On Mon, Jan 07, 2002 at 09:58:46AM +0100, Jean-Marc Lasgouttes wrote:
We should maybe change all the radio buttons (not checkbox) to
round3dbutton for 1.2.0. Then people will be glad that the interface
do I win most boring patch ?
I've spared bandwidth so applier need to run make updatesrc in forms/ before
applying. Everything seems to looks OK I think ...
regards
john
--
Unless everyone else on earth is attending meetings I haven't been told
about.
- /. paranoia at its finest
Index: src/frontends/xforms/ChangeLog
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/ChangeLog,v
retrieving revision 1.219
diff -u -p -r1.219 ChangeLog
--- src/frontends/xforms/ChangeLog 2002/01/04 15:14:22 1.219
+++ src/frontends/xforms/ChangeLog 2002/01/07 09:35:26
@@ -1,3 +1,8 @@
+2002-01-07 John Levon [EMAIL PROTECTED]
+
+ * form_*:
+ * forms/form_*: change radio buttons to roundbutton
+
2002-01-04 Jürgen Spitzmüller [EMAIL PROTECTED]
* FormDocument.C: Recognize the default paper settings too
Index: src/frontends/xforms/forms/form_document.C.patch
===
RCS file:
/usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/forms/form_document.C.patch,v
retrieving revision 1.6
diff -u -p -r1.6 form_document.C.patch
--- src/frontends/xforms/forms/form_document.C.patch2001/03/15 13:37:03 1.6
+++ src/frontends/xforms/forms/form_document.C.patch2002/01/07 09:35:36
@@ -1,6 +1,6 @@
form_document.CWed Oct 4 18:07:46 2000
-+++ form_document.C.orig Wed Oct 4 18:07:23 2000
-@@ -9,6 +9,8 @@
+--- form_document.CMon Jan 7 09:19:31 2002
form_document.C.newMon Jan 7 09:19:24 2002
+@@ -10,6 +10,8 @@
#include stdlib.h
#include form_document.h
#include FormDocument.h
@@ -9,25 +9,25 @@
FD_form_tabbed_document::~FD_form_tabbed_document()
{
-@@ -390,10 +392,18 @@
+@@ -543,10 +545,18 @@
fdui-form = fl_bgn_form(FL_NO_BOX, 440, 345);
fdui-form-u_vdata = this;
+ fl_set_border_width(-1);
obj = fl_add_box(FL_FLAT_BOX, 0, 0, 440, 345, );
-- fdui-bmtable_bullet_panel = obj = fl_add_pixmapbutton(FL_NORMAL_BUTTON, 90, 105,
260, 180, );
+- fdui-bmtable_bullet_panel = obj = fl_add_pixmapbutton(FL_NORMAL_BUTTON, 135, 105,
+260, 180, );
-fl_set_object_boxtype(obj, FL_NO_BOX);
-fl_set_pixmapbutton_file(obj,
_(/nfs/sinco/source/lyx/lyx-devel/lib/images/psnfss2.xpm));
+ fl_set_border_width(-3);
-+ fdui-bmtable_bullet_panel = obj = fl_add_bmtable(1, 90, 105, 265, 180, );
++ fdui-bmtable_bullet_panel = obj = fl_add_bmtable(1, 135, 105, 260, 180, );
+ fl_set_object_callback(obj, C_FormBaseDeprecatedInputCB, BULLETBMTABLE);
+ fl_set_object_lcol(obj, FL_BLUE);
+ fl_set_object_boxtype(obj, FL_UP_BOX);
+ fl_set_bmtable_pixmap_file(obj, 6, 6,
+ LibFileSearch(images,
-+ standard,
++ standard,
+ xpm).c_str());
+ fl_set_border_width(-1);
- obj = fl_add_frame(FL_ENGRAVED_FRAME, 95, 20, 255, 70, );
- fdui-choice_bullet_size = obj = fl_add_choice(FL_NORMAL_CHOICE, 15, 45, 65, 30,
idex(_(Size|#z)));
- fl_set_button_shortcut(obj, scex(_(Size|#z)), 1);
+ obj = fl_add_frame(FL_ENGRAVED_FRAME, 135, 20, 260, 55, );
+ {
+ char const * const dummy = N_(Size|#z);
Index: src/frontends/xforms/forms/form_document.fd
===
RCS file:
/usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/forms/form_document.fd,v
retrieving revision 1.18
diff -u -p -r1.18 form_document.fd
--- src/frontends/xforms/forms/form_document.fd 2001/11/12 14:28:12 1.18
+++ src/frontends/xforms/forms/form_document.fd 2002/01/07 09:35:37
@@ -328,7 +328,7 @@ argument:
class: FL_BEGIN_GROUP
type: 0
-box: 0 10 10 0
+box: 0 0 0 0
boxtype: FL_NO_BOX
colors: FL_COL1 FL_MCOL
alignment: FL_ALIGN_CENTER
@@ -344,7 +344,7 @@ callback:
argument:
-class: FL_CHECKBUTTON
+class: FL_ROUNDBUTTON
type: RADIO_BUTTON
box: 255 40 120 30
boxtype: FL_NO_BOX
@@ -362,7 +362,7 @@ callback: C_FormBaseDeprecatedInputCB
argument: INPUT
-class: FL_CHECKBUTTON
+class: FL_ROUNDBUTTON
type: RADIO_BUTTON
box: 255 75 120 30
boxtype: FL_NO_BOX
@@ -928,7 +928,7 @@ argument: INPUT
class: FL_BEGIN_GROUP
type: 0
-box: 0 10 10 0
+box: 0 0 0 0
boxtype: FL_NO_BOX
colors: FL_COL1 FL_MCOL
alignment: FL_ALIGN_CENTER
@@ -944,7 +944,7 @@ callback:
argument:
-class: FL_CHECKBUTTON
+class: FL_ROUNDBUTTON
type: RADIO_BUTTON
box: 300 30 120 30
boxtype: FL_NO_BOX
@@ -962,7 +962,7 @@ callback: C_FormBaseDeprecatedInputCB
argument: INPUT
-class: FL_CHECKBUTTON
+class: FL_ROUNDBUTTON
type:
Re: [Patch] fixed bug 59
Martin == Martin Vermeer [EMAIL PROTECTED] writes: Martin Like: [snip] Martin (Assumes the additional fix also applied) Thanks. I applied the additional fix too. JMarc
Re: [Patch] fixed bug 59
On 07-Jan-2002 Jean-Marc Lasgouttes wrote: I removed also the debugging statements said to be useless. What about the others? Jean-Marc I applied the InsetTabular patch about clearing the cell and also added a ChangeLog entry. Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Schapiro's Explanation: The grass is always greener on the other side -- but that's because they use more manure.
Re: [PATCH] auto-apply for FormTabular
On 07-Jan-2002 John Levon wrote: On Mon, Jan 07, 2002 at 10:20:21AM +0100, Juergen Vigna wrote: Well yes but could you send me a ChangeLog? (I already applied the patch so please only the ChangeLog!) the patch didn't include it ? Sorry I didn't see it! Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Yea from the table of my memory I'll wipe away all trivial fond records. -- Hamlet
Re: [Patch] fixed bug 59
Juergen == Juergen Vigna [EMAIL PROTECTED] writes: Juergen On 07-Jan-2002 Jean-Marc Lasgouttes wrote: I removed also the debugging statements said to be useless. What about the others? Juergen Jean-Marc I applied the InsetTabular patch about clearing the Juergen cell and also added a ChangeLog entry. You mean martin's patch for #59? Are you going to commit soon? Otherwise I'll have to revert it before commiting. JMarc
Re: [Patch] fixed bug 59
On 07-Jan-2002 Jean-Marc Lasgouttes wrote: Thanks. I applied the additional fix too. Jean-Marc I just applied the one in InsetTabular (sorry but I had all ready and changed some stuff so I commited already). If you'd like I apply the one in InsetText too, otherwise I let it to you if you're already ready for commit. Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Mate, this parrot wouldn't VOOM if you put four million volts through it! -- Monty Python
Re: [Patch] fixed bug 59
Juergen == Juergen Vigna [EMAIL PROTECTED] writes: Juergen On 07-Jan-2002 Jean-Marc Lasgouttes wrote: Thanks. I applied the additional fix too. Juergen Jean-Marc I just applied the one in InsetTabular (sorry but I Juergen had all ready and changed some stuff so I commited already). Juergen If you'd like I apply the one in InsetText too, otherwise I Juergen let it to you if you're already ready for commit. I'll do it right now. JMarc
Re: KLyX -- am I missing something?
John == John Levon [EMAIL PROTECTED] writes: John On Mon, Jan 07, 2002 at 09:58:46AM +0100, Jean-Marc Lasgouttes John wrote: We should maybe change all the radio buttons (not checkbox) to round3dbutton for 1.2.0. Then people will be glad that the interface John do I win most boring patch ? I'm sure people will love it, as dull as it may sound from the technical point of view. What do the changes to non-fd files mean? Did you double check them? And did you try round3dbuttons too? I'm not sure which one looks best (==less ugly). JMarc
Re: KLyX -- am I missing something?
On Mon, Jan 07, 2002 at 11:17:59AM +0100, Jean-Marc Lasgouttes wrote: I'm sure people will love it, as dull as it may sound from the technical point of view. What do the changes to non-fd files mean? Did you double check them? the only change should be to the form_document patch ? And that's just a rediff essentially as things moved around a little... And did you try round3dbuttons too? I'm not sure which one looks best (==less ugly). whoops, didn't see that one. hmm, do you want to sedall ROUNDBUTTON ROUND3DBUTTON or should I and resubmit ? thanks john -- Unless everyone else on earth is attending meetings I haven't been told about. - /. paranoia at its finest
Re: KLyX -- am I missing something?
John == John Levon [EMAIL PROTECTED] writes: And did you try round3dbuttons too? I'm not sure which one looks best (==less ugly). John whoops, didn't see that one. hmm, do you want to sedall John ROUNDBUTTON ROUND3DBUTTON or should I and resubmit ? I'll do it. JMarc
Re: Solaris complilation bug fix: sinf/cosf missing (addendum)
John == John Levon [EMAIL PROTECTED] writes: John On Thu, Jan 03, 2002 at 10:54:17AM -0800, Gary Hall wrote: I forgot to mention that an include directive for sunmath.h is required for src/insets/figinset.C. John What lyx version ? I think your system is a little broken, we John don't mention sinf or cosf in figinset.C John I don't even know what they are ... Presumably version of sin/cos which operates on floats instead of doubles... I remember reports on solaris mentionning problems with cmath or math.h, but I do not really remember the details. JMarc
Re: Solaris complilation bug fix: sinf/cosf missing (addendum)
Gary == Gary Hall [EMAIL PROTECTED] writes: Gary I forgot to mention that an include directive for sunmath.h is Gary required for src/insets/figinset.C. What version of solaris? What compiler? If it is gcc, what does 'gcc -v' say? JMarc
Re: [PATCH] Let the gnome frontend compile
On Thursday 03 January 2002 4:35 am, Michael Koziarski wrote: On Thu, Jan 03, 2002 at 01:44:40PM +1000 or thereabouts, Allan Rae wrote: On Thu, 3 Jan 2002, Michael A. Koziarski wrote: Thanks to John Levon for his assistance. Please apply Applied. Allan. (ARRae) Thanks, as an aside the problem was that src/frontends/controllers/ControlDialog.C doesn't #include bufferview.h. It should, shouldn't it? Cheers, There is no ControlDialog.C. Do you mean the .h file? Perhaps you'd like to send a patch and an explanation about why we should #include a file in a .h file if not absolutely necessary? Happy New Year to you all, by the way! Angus
RE: Question for our STL experts!
On 07-Jan-2002 Juergen Vigna wrote: I ask this because it seems to me that it does only insert a new item but does not substitute the other one and then when searching for the LyXText of this BufferView we have multiple choices. Well it really seems it does not! This could be a reason for a whole lot of bugs we don't know where they come from! I debuged this and while in the cached_text variable there is the NEW version of the LyXText when doing a: Cache::iterator it = cache.find(bv); I always get the old one! Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Usage: fortune -P [] -a [xsz] [Q: [file]] [rKe9] -v6[+] dataspec ... inputdir
Re: euro as special character
Herbert == Herbert Voss [EMAIL PROTECTED] writes: Herbert I only wanted to know why I can get the Euro Symbol in all X Herbert applications with AltGr-e but not inside LyX. I have no Herbert knowledge about this keyboard stuff. What font do you use in these applications? ISO-8859-15? Did you tell LyX to use this encoding? JMarc
Compile problem --- where to #include a missing header file?
Compiliation here dies in BufferView.C:
cxx: Error: box.h, line 45: #349 no operator matches these operands
operand types are: std::ostream char [8]
return o x1,y1: b.x1 , b.y1
-^
cxx: Error: box.h, line 46: #135 namespace std has no member endl
x2,y2: b.x2 , b.y2 std::endl;
-^
Where is the correct place to #include ostream ?
Note that box.h contians the following inlined function:
inline std::ostream operator(std::ostream o, Box b)
{
return o x1,y1: b.x1 , b.y1
x2,y2: b.x2 , b.y2 std::endl;
}
Does that mean that I should #include support/LOstream.h in box.h or in
BufferView.C?
Angus
Re: Question for our STL experts!
On 07-Jan-2002 Lars Gullik Bjønnes wrote: The BufferView in actual use should _never_ be in the cache. So when you search in the cach for a BufferView and find the one you want to use, it should also be removed from the cache. Care to explain? I don't think I did that part of the code someone else did ;), so I don't understand what you mean above! (and and the old BufferView should be inserted in the cache _if_ it is still in use.) Well we surely don't know in the InsetText if the BufferView is still in use or not! So we only can insert the new BufferView. But this is irrelevant as for now we only have 1 bufferview, so the case that the BufferView changes can not be true for the moment (and therefore hardly be tested!) Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Wagner's music is better than it sounds. -- Mark Twain
Re: Question for our STL experts!
On 07-Jan-2002 Lars Gullik Bjønnes wrote: mapstring, string translate; translate[Hei] = Hello; translate.insert(make_pair(Hei, Hi)); then the insert will not work since Hei is already set. So in this case we should never use insert but always the assignment operator? Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ One good turn asketh another. -- John Heywood
Re: Question for our STL experts!
On 07-Jan-2002 Lars Gullik Bjønnes wrote: I am pretty sure you did it... (at least I am quite sure that I didn't, I only did the BufferView TextCache.) Well anyway it has to be fixed so I would need an experts council what I should do, I would do this: -cache.insert(make_pair(bv, cached_text)); +cache[bv] = cached_text; (and I had this consturct also before but we would have to look at the cvs history to see who changed this to insert, but do we really care?). True... but we must have it on the radar screen. ... But why then is a cache really needed... remind me please. (why can't the insettext just have a text pointer that stays the same all the time as long as the texts owner is the same as the bufferview used? Hmm... ) Because of the radar thing, should we change back to non multiple BufferView? I don't think this is so much slower is it? And anyway we will need it some time so it's better we have this there right now so we can see how slow we get with it. Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ In the long run, every program becomes rococco, and then rubble. -- Alan Perlis
Re: Compile problem --- where to #include a missing header file?
On Monday 07 January 2002 1:26 pm, you wrote:
| Note that box.h contians the following inlined function:
| inline std::ostream operator(std::ostream o, Box b)
| {
| return o x1,y1: b.x1 , b.y1
| x2,y2: b.x2 , b.y2 std::endl;
| }
| Does that mean that I should #include support/LOstream.h in box.h or in
| BufferView.C?
Do you really need this to be inlined?
and .h files should be selfcontaining (imho), so if you cannot move
this code out of line (you probably should) then LOstream should be
included in the .h file. Of course when not inlining you only need
forward references to ostream and that is done with iosfwd.
Good. I have a nice rule to add to my list of coding rules. Thank you!
As for moving this function out of line, this isn't my code. I was just
trying to get cvs to compile after two weeks away from a computer! I'm happy
to create a new file box.C if you want me to, but it seems a little like
overkill...
Angus
Re: More rfc stuff
On Sat, Jan 05, 2002 at 12:44:23AM +0100, Lars Gullik Bjønnes wrote:
There is still one huge obstacle (I have removed the next/prev from
Paragraph and changed Row to store an iterator): make the cursor store
a iterator and not a pointer to a paragraph.
I think the outside world could adapt mathed's notion of a 'cursor' at
some point of time. The 'mathcursor' is basically a triple of two
'mathiterators' and a 'bool', the iterators represent the position of the
selection anchor and the actual cursor, the bool says whether the selection
is active.
Now the 'complicated' part is a 'mathiterator' (see
src/mathed/math_iterator.[Ch] for complete description ;-)). This is
basically a vector of tripels (inset *, cell in inset, position in cell).
Maybe an example: Suppose we have $a+\frac{b}{\sqrt{cd}}$ and the cursor
is between 'c' and 'd'. Than the cursor iterator would look like
(hull,
0 /* hull has only one 'cell', we count from */
2 /* position left of frac: 0 is 'a', 1 is '+', 2 is the frac */ )
(frac,
0 /* frac has two cells, 0 is denominator, 1 is numerator */
1 /* we are 'left of the d')
Navigation is pretty easy using this representation (and there is no need
for inset locking etc...)
Andre'
--
André Pönitz .. [EMAIL PROTECTED]
Re: Question for our STL experts!
On 07-Jan-2002 Lars Gullik Bjønnes wrote: but I'd also say that it is an error trying to store a bufferview in the cache when it is already there. F.ex. the TextCache requires that the BufferView stored therein is _not_ viewed in any LyXView. I think that the cache in insettext very well could use the same semantics. Well I still don't understand what you mean, but I fixed it up (hopefully in the sense I interpreted out of your message). I'll commit soon then you can have a look at the new code. Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ I saw what you did and I know who you are.
File-Open aborts!
File-Open now aboards in this code!
if (LyXReadLink(File, Link)) {
Buffer += - ;
Buffer += Link;
// This gives the FileType of the file that
// is really pointed too after resolving all
// symlinks. This is not necessarily the same
// as the type of Link (which could again be a
// link). Is that intended?
// JV 199902
fileInfo.newFile(File);
- Buffer += fileInfo.typeIndicator();
}
}
Backtrace following,
Jug
#1 0x4027a8ba in raise (sig=6) at ../sysdeps/posix/raise.c:27
in ../sysdeps/posix/raise.c
(gdb) up
#2 0x4027c062 in abort () at ../sysdeps/generic/abort.c:88
in ../sysdeps/generic/abort.c
(gdb) up
#3 0x081b73df in lyx::abort () at abort.C:9
/sinco/source/lyx/lyx-devel/src/support/abort.C:9:91:beg:0x81b73df
Current language: auto; currently c++
(gdb) up
#4 0x081c78db in void lyx::Assertbool (assertion=false) at LAssert.h:24
/sinco/source/lyx/lyx-devel/src/support/LAssert.h:24:484:beg:0x81c78db
(gdb) up
#5 0x081b642c in FileInfo::typeIndicator (this=0xbfffede8) at FileInfo.C:178
/sinco/source/lyx/lyx-devel/src/support/FileInfo.C:178:3205:beg:0x81b642c
(gdb) up
#6 0x0815526d in FileDialog::Private::Reread (this=0x8328418) at FormFiledialog.C:325
/sinco/source/lyx/lyx-devel/src/frontends/xforms/FormFiledialog.C:325:7001:beg:0x815526d
(gdb) bt
#0 0x4027aae1 in __kill () from /lib/i686/libc.so.6
#1 0x4027a8ba in raise (sig=6) at ../sysdeps/posix/raise.c:27
#2 0x4027c062 in abort () at ../sysdeps/generic/abort.c:88
#3 0x081b73df in lyx::abort () at abort.C:9
#4 0x081c78db in void lyx::Assertbool (assertion=false) at LAssert.h:24
#5 0x081b642c in FileInfo::typeIndicator (this=0xbfffede8) at FileInfo.C:178
#6 0x0815526d in FileDialog::Private::Reread (this=0x8328418) at FormFiledialog.C:325
#7 0x08156ab6 in FileDialog::Private::Select (this=0x8328418, title=@0xb270,
path=@0xb288, mask=@0xbfffefd8, suggested=@0xb1f8) at FormFiledialog.C:733
#8 0x081545ac in FileDialog::Select (this=0xb268, path=@0xb288,
mask=@0xb208, suggested=@0xb1f8) at FileDialog.C:75
#9 0x080ab98d in LyXFunc::open (this=0x82dad28, fname=@0xb418) at lyxfunc.C:1813
#10 0x080a8c9b in LyXFunc::dispatch (this=0x82dad28, ac=165,
do_not_use_this_arg=@0xb458) at lyxfunc.C:1272
#11 0x0819d81a in Menubar::Pimpl::MenuCallback (ob=0x8305300, button=1) at
Menubar_pimpl.C:586
#12 0x0819b398 in C_Menubar_Pimpl_MenuCallback (ob=0x8305300, button=1) at
Menubar_pimpl.C:79
#13 0x4004e87f in fl_object_qread () from /usr/X11R6/lib/libforms.so.0.88
#14 0x4005cb39 in fl_check_forms () from /usr/X11R6/lib/libforms.so.0.88
#15 0x081577f1 in GUIRunTime::runTime () at GUIRunTime.C:86
#16 0x0809cb4d in LyXGUI::runTime (this=0x82bca18) at lyx_gui.C:318
#17 0x0809d310 in LyX::LyX (this=0xb610, argc=0xb670, argv=0xb6d4) at
../src/lyx_main.C:176
#18 0x080bf0df in main (argc=1, argv=0xb6d4) at ../src/main.C:38
#19 0x40268627 in __libc_start_main (main=0x80bf010 main, argc=1, ubp_av=0xb6d4,
init=0x804ef78 _init, fini=0x8255364 _fini, rtld_fini=0x4000dcd4 _dl_fini,
stack_end=0xb6cc) at ../sysdeps/generic/libc-start.c:129
(gdb) _in ../sysdeps/posix/raise.c
(gdb)
--
-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._
Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED]
Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253
I-39100 Bozen Web: http://www.sad.it/~jug
-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._
Sic transit gloria Monday!
RE: File-Open aborts!
On 07-Jan-2002 Juergen Vigna wrote: File-Open now aboards in this code! Same for save! Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Humor in the Court: Q: ...any suggestions as to what prevented this from being a murder trial instead of an attempted murder trial? A: The victim lived.
Re: [PATCH] trivial mathed patch
On Mon, Jan 07, 2002 at 12:40:09PM +, Angus Leeming wrote: André, here is a trivial patch for mathed. Applied. I also get lots and lots of warning messages so: cxx: Warning: math_scriptinset.h, line 37: #610-D int MathScriptInset::ascent(const MathInset *) const does not match MathInset::ascent -- virtual function override intended? int ascent(MathInset const * nuc) const; Please send me a complete list, otherwise I'll probably get a similar one from Jean-Marc at some poitn of time ;-} I also get a number of conversion to integral type of smaller size warnings when using size() et al, but not so many. Dito. Andre' -- André Pönitz .. [EMAIL PROTECTED]
Re: [PATCH] FileInfo misuse
On Saturday 05 January 2002 10:15 pm, you wrote:
this fixes most of the cases, and asserts on the others
please apply
john
John,
I've just been perusing the FileInfo patch that you posted and that Jean-Marc
applied. The patched code results in:
Index: src/frontends/xforms/xforms_helpers.C, line 461:
FileInfo d(name);
// FIXME: what is this supposed to do ?
// .newFile doesn't do what you think it does ...
if (!d.isOK() || !d.isDir()) {
d.newFile(dir);
}
if (!d.isDir()) {
...
}
I'd guess that I wrote this code (a long, long time ago because I don't
remember newFile at all!) and so should probably answer you.
Can you tell me what newFile does? I've just looked quickly at the code in
FileInfo.C and it seems to me that newFile resets the FileInfo to examine
dir not name. This, it seems, is exactly what I meant to do, all that
time ago, so I don't follow your FIXME comment. In fairness, I think that the
final line of code above should also have an isOK() check too but you seem
pretty knowledgable on all this.
Insight is, as always, appreciated.
Angus
Re: File-Open aborts!
On Monday 07 January 2002 2:55 pm, Juergen Vigna wrote:
File-Open now aboards in this code!
if (LyXReadLink(File, Link)) {
Buffer += - ;
Buffer += Link;
// This gives the FileType of the file that
// is really pointed too after resolving all
// symlinks. This is not necessarily the same
// as the type of Link (which could again be a
// link). Is that intended?
// JV 199902
fileInfo.newFile(File);
- Buffer += fileInfo.typeIndicator();
}
}
It appears that you need a fileInfo.isOK() check once you've (re-)created
fileInfo and before you start to use it.
Angus (who's posted similar queries...)
Re: [PATCH] trivial mathed patch
Andre == Andre Poenitz [EMAIL PROTECTED] writes: Andre Please send me a complete list, otherwise I'll probably get a Andre similar one from Jean-Marc at some poitn of time ;-} Not soon. I can will only compile on cxx if explicitely requested to, until september. So Angus is probably our new cxx annoyance dept chief :) JMarc
Re: File-Open aborts!
Juergen == Juergen Vigna [EMAIL PROTECTED] writes:
File- Open now aboards in this code!
Juergen if (LyXReadLink(File, Link)) { Buffer += -
Juergen ; Buffer += Link;
Probably a consequence of John's recent FileInfo misuse cleanup (and
associated asserts :)
JMarc
Re: [PATCH] trivial mathed patch
On Monday 07 January 2002 3:10 pm, Andre Poenitz wrote: [snip] Please send me a complete list, otherwise I'll probably get a similar one from Jean-Marc at some poitn of time ;-} Ok. Recompiling now. I'll post you the (edited) make log privately to save others' bandwidth. Never fear, it won't be tooo big! Angus
Re: More rfc stuff
On Mon, Jan 07, 2002 at 03:34:46PM +0100, Lars Gullik Bjønnes wrote:
| Navigation is pretty easy using this representation (and there is no need
| for inset locking etc...)
If we are lucky we will be able to ditch the locking inset all over.
(and perhaps to use the same cursor all over as well, even inside
mathed?)
That would be _my_ plan (the big re-unification of mathed and The Rest of
the World) (not that I am too fond of re-unifications...)
We have to figure out, how the (inset*, cell, pos) translates to the
outside world, though.
From a first glance it should be possible without to much effort but
would need two insets for the text: A ParagraphInset, whose cells are the
lines on screen, and a TextInset containing the paragraphs.
I already wrote a few lines for proper \mbox{} support for mathed which
essentially does exactly this - but it's far from finished. I could
try to finish that and once this is done we have a starting point for the
TextInset conversion.
Andre'
--
André Pönitz .. [EMAIL PROTECTED]
Re: euro as special character
Herbert Voss [EMAIL PROTECTED] writes: an euro, but I'm not able to insert it with the keyboard. in all insets, like label, figure a.s.o., I get the currency symbol with AltGr-e (the sputnik) which becomes after closing the inset window the eurosign. but in lyx main-gui - nothing What does # xmodmap -pke | grep 26 tell you? Mine is: keycode 26 = e E EuroSign currency This means AltGR-e gives me the Euro sign, which does not work in any application I have tried. AltGR-E gives results in the Euro sign if ISO-8859-15 is selected, but it is technically wrong. Infact, it points to the currency sign, which is the sputnik ¤ you mentioned. I don't have lyx-cvs, so I can't comment on its behaviour. I can only get the sputnik within LyX 1.1.6. Don't use a broken fix to work around a broken system. It will still be broken. Thomas [EMAIL PROTECTED]
Re: Question for our STL experts!
On 07-Jan-2002 Andre Poenitz wrote: There is no way to have multiple entries for the same key in a std::map. One would have to use a std::multimap if that feature was needed. 'insert' creates a new entry or overwrite an ol;d entry with the same key, exactly what operator[]() would do. Well it seems it does not at least not in the implementation that I use! It seems that the insert is ignored! Anyway I fixed it in another way. Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Laugh, and the world ignores you. Crying doesn't help either.
Re: [PATCH] FileInfo misuse
On Monday 07 January 2002 3:59 pm, you wrote:
When I wrote the newFile it did excatly what I wanted it do. - run
the stat on the new file.
nice for loops etc.
I think it still does, which just leaves John's comment to understand!
Angus
FileInfo FileInfo::newFile(string const path, bool link)
{
fname = path;
status = 0;
err = NoErr;
dostat(link);
return *this;
}
FileInfo FileInfo::newFile(int fildes)
{
status = 0;
err = NoErr;
status = fstat(fildes, buf);
if (status) err = errno;
return *this;
}
Bug 52: Spellchecker reports success if not started and closed
This trivial patch fixes this bug I think. Could people confirm before I close the bug report... Angus Index: src/frontends/controllers/ControlSpellchecker.C === RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/controllers/ControlSpellchecker.C,v retrieving revision 1.12 diff -u -p -r1.12 ControlSpellchecker.C --- src/frontends/controllers/ControlSpellchecker.C 2001/10/23 16:32:16 1.12 +++ src/frontends/controllers/ControlSpellchecker.C 2002/01/07 16:07:12 @@ -233,17 +233,18 @@ void ControlSpellchecker::clearParams() lv_.view()-endOfSpellCheck(); - // show closing message - view().partialUpdate(2); + // show closing message if any words were checked. + if (count_ 0) + view().partialUpdate(2); // reset values to initial rtl_ = false; - word_ = ; + word_.erase(); newval_ = 0.0; oldval_ = 0; newvalue_ = 0; count_ = 0; - message_ = ; + message_.erase(); stop_ = false; result_ = SpellBase::ISP_OK; speller_ = 0;
Re: euro as special character
Thomas Steffen wrote: Herbert Voss [EMAIL PROTECTED] writes: an euro, but I'm not able to insert it with the keyboard. in all insets, like label, figure a.s.o., I get the currency symbol with AltGr-e (the sputnik) which becomes after closing the inset window the eurosign. but in lyx main-gui - nothing What does # xmodmap -pke | grep 26 tell you? Mine is: keycode 26 = e E EuroSign currency voss@maria:~ xmodmap -pke | grep 26 keycode 26 = e E EuroSign This means AltGR-e gives me the Euro sign, which does not work in any application I have tried. AltGR-E gives results in the Euro sign if ISO-8859-15 is selected, but it is technically wrong. Infact, it points to the currency sign, which is the sputnik ¤ you mentioned. I don't have lyx-cvs, so I can't this is not a sputnik for me, but an eurosymbol, because I have the ...-15 charset. comment on its behaviour. I can only get the sputnik within LyX 1.1.6. Again: I have only a problem with the Euro Symbol in the LyX main-gui (in 1.1.6fix3 and cvs)! In ALL other x-applications it works well. Herbert -- http://www.lyx.org/help/
Re: euro as special character
Herbert == Herbert Voss [EMAIL PROTECTED] writes: Herbert Again: I have only a problem with the Euro Symbol in the LyX Herbert main-gui (in 1.1.6fix3 and cvs)! In ALL other x-applications Herbert it works well. I'll try to experiment with that at home tomorrow. JMarc
Re: Question for our STL experts!
On Mon, Jan 07, 2002 at 04:44:23PM +0100, Juergen Vigna wrote: 'insert' creates a new entry or overwrite an ol;d entry with the same key, exactly what operator[]() would do. Well it seems it does not at least not in the implementation that I use! I don't believe you. Try again... ;-) reference operator[](const key_type x) is per Standard identical to (*((insert(make_pair(x, T(.first)).second so it _has_ to be the same. Andre' -- André Pönitz .. [EMAIL PROTECTED]
Re: Question for our STL experts!
On Mon, Jan 07, 2002 at 05:41:11PM +0100, Lars Gullik Bjønnes wrote: AFAIK insert will not insert a new key if it already exists. (only for map and set of course) From CD2: 23.1.2 Associative containers[lib.associative.reqmts] - a_eq.insert(t) iterator inserts t and returns the iterator pointing to the newly inserted ele- ment. - Andre' -- André Pönitz .. [EMAIL PROTECTED]
Re: [PATCH] FileInfo misuse
On Mon, Jan 07, 2002 at 03:28:09PM +, Angus Leeming wrote:
FileInfo d(name);
// FIXME: what is this supposed to do ?
// .newFile doesn't do what you think it does ...
if (!d.isOK() || !d.isDir()) {
d.newFile(dir);
}
if (!d.isDir()) {
yes I'm wrong sorry, lax of me, the code is OK other than the needed isOK
check. Is it ok for you to fix my error ?
thanks
john
--
Nobody uses formal methods for anything other than generating papers
with more authors than readers.
- Victor Yodaiken
Re: [PATCH] FileInfo misuse
On Monday 07 January 2002 5:36 pm, John Levon wrote:
On Mon, Jan 07, 2002 at 03:28:09PM +, Angus Leeming wrote:
FileInfo d(name);
// FIXME: what is this supposed to do ?
// .newFile doesn't do what you think it does ...
if (!d.isOK() || !d.isDir()) {
d.newFile(dir);
}
if (!d.isDir()) {
yes I'm wrong sorry, lax of me, the code is OK other than the needed isOK
check. Is it ok for you to fix my error ?
thanks
john
Sure. Glad that I had things right!
A
Re: Bug 52: Spellchecker reports success if not started and closed
On Mon, Jan 07, 2002 at 04:18:24PM +, Angus Leeming wrote: This trivial patch fixes this bug I think. Could people confirm before I close the bug report... works for me, great ! john -- Nobody uses formal methods for anything other than generating papers with more authors than readers. - Victor Yodaiken
box
see attachment
--
André Pönitz .. [EMAIL PROTECTED]
Index: box.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/box.C,v
retrieving revision 1.1
diff -u -p -r1.1 box.C
--- box.C 2002/01/07 14:36:10 1.1
+++ box.C 2002/01/07 17:58:53
@@ -25,7 +25,7 @@ bool Box::contained(int x, int y)
}
-ostream operator(ostream o, Box b)
+ostream operator(ostream o, Box const b)
{
return o x1,y1: b.x1 , b.y1
x2,y2: b.x2 , b.y2 std::endl;
Index: box.h
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/box.h,v
retrieving revision 1.5
diff -u -p -r1.5 box.h
--- box.h 2002/01/07 14:36:10 1.5
+++ box.h 2002/01/07 17:58:53
@@ -38,6 +38,6 @@ struct Box {
};
-std::ostream operator(std::ostream , Box );
+std::ostream operator(std::ostream , Box const );
#endif // BOX_H
Help tracking down bug #164
Hi guys, Jürgen can't reproduce bug #164 but it's still affecting me with the latest CVS snapshot. The bug is described in the following attachment. http://cvs.koziarski.com/showattachment.cgi?attach_id=41 -- Cheers Koz
Re: spellchecker crash
On Mon, Jan 07, 2002 at 06:34:33PM -0500, Richard E. Hawkins wrote:
Don't think I've seen *this* one before.
Program received signal SIGSEGV, Segmentation fault.
there is a gaping hole here where you forget to tell us what you actually
did ..
0x820854c in FormSpellchecker::input (this=0x83653c8, obj=0x8452000)
at /usr/include/g++/std/bastring.h:220
220 { return assign (s, traits::length (s)); }
and this doesn't tell us which bit of code is problematic. But it is probably
one of the string blah = fl_get_browser_line()
can you tell us what you did just prior to this crash ?
regards
john
--
I went to set up a Yahoo ID for my dog. (Don't ask, but the DOG'S email was
cluttering my inbox).
- Ruthless Advisorette
Re: spellchecker crash
On Mon, Jan 07, 2002 at 06:55:42PM -0500, [EMAIL PROTECTED] wrote:
I double-clicked on a choice, to replace bMy with My or some such.
I believe the crash was immediate (the change certainly didn't make it
into the emergency file, but the prior change did.)
well I don't really see how it could happen, but certainly ther attached
patch is more robusy.
please apply
john
--
I went to set up a Yahoo ID for my dog. (Don't ask, but the DOG'S email was
cluttering my inbox).
- Ruthless Advisorette
Index: src/frontends/xforms/ChangeLog
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/ChangeLog,v
retrieving revision 1.224
diff -u -p -r1.224 ChangeLog
--- src/frontends/xforms/ChangeLog 2002/01/07 18:42:29 1.224
+++ src/frontends/xforms/ChangeLog 2002/01/08 00:06:47
@@ -1,3 +1,7 @@
+2002-01-08 John Levon [EMAIL PROTECTED]
+
+ * FormSpellchecker.C: fix possible crash on clicking a suggestion
+
2002-01-07 Angus Leeming [EMAIL PROTECTED]
* FormDocument.C:
Index: src/frontends/xforms/FormSpellchecker.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/FormSpellchecker.C,v
retrieving revision 1.3
diff -u -p -r1.3 FormSpellchecker.C
--- src/frontends/xforms/FormSpellchecker.C 2001/09/18 12:22:08 1.3
+++ src/frontends/xforms/FormSpellchecker.C 2002/01/08 00:06:47
@@ -81,12 +81,17 @@ ButtonPolicy::SMInput FormSpellchecker::
controller().options();
} else if (obj == dialog_-browser) {
int sel = fl_get_browser(dialog_-browser);
- if (clickline_==sel) {
+ if (sel 1)
+ return ButtonPolicy::SMI_VALID;
+
+ if (clickline_ == sel) {
string tmp = fl_get_input(dialog_-input);
controller().replace(tmp);
}
+
clickline_ = sel;
- string tmp = fl_get_browser_line(dialog_-browser, clickline_);
+ char const * cptmp = fl_get_browser_line(dialog_-browser, clickline_);
+ string tmp = (cptmp) ? cptmp : ;
fl_set_input(dialog_-input, tmp.c_str());
}
Bugzilla Upgraded to 2.14.1
Following the security advisory of 10 minutes ago, I've upgraded bugzilla to 2.14.1, fixing a few possible exploits. I'm looking forward to 2.16 to allow bugs to affect multiple versions, etc. The logs for cvs.koziarski.com imply that nothing untoward happened during our period of vulnerability. If only all web tool updates were as easy as cvs update :) Once again though, I'd like someone to volunteer a place for me to scp / ftp the bugs backups to. JIC something happens here. Backups are produced daily and are ~3m. -- Cheers Koz
Re: Help tracking down bug #164
On Tue, Jan 08, 2002 at 09:41:52AM +1300, Michael A. Koziarski wrote: Jürgen can't reproduce bug #164 but it's still affecting me with the latest CVS snapshot. I can't reproduce it either. CVS from yesterday 6pm or so... Andre' -- André Pönitz .. [EMAIL PROTECTED]
Re: multiple paragraphs in footnote causes error.
Richard E. Hawkins wrote: This works in a footnote as a single paragraph. When broken, it causes Too many }'s errors. The original 0 was simply the coldest liquid Fahrenheit could produce with ice and salt. It was later adjusted about 2 degrees to accomodate for sea level. Note that on the Kelvin and Renkin scales, 0 does have meaning; the absolute absence of heat energy. I don't understand why this should cause an error? give an example file Herbert -- http://www.lyx.org/help/
Re: Help tracking down bug #164
On Tue, 8 Jan 2002, Andre Poenitz wrote: On Tue, Jan 08, 2002 at 09:41:52AM +1300, Michael A. Koziarski wrote: Jürgen can't reproduce bug #164 but it's still affecting me with the latest CVS snapshot. I can't reproduce it either. CVS from yesterday 6pm or so... Better make that three against 'cos I can't reproduce it either using mouse or keyboard -- tried selecting cells in various ways and nothing bad happened. Current CVS. Allan. (ARRae)
Re: Ascii import
On Tue, Jan 08, 2002 at 02:58:56AM -0400, Garst R. Reese wrote: I was sent an ascii file that used Ctrl-M's as paragraph terminators. It looks good in vi, but LyX is ignoring the Ctrl-M's completely. I finally opened the file in Netscape and pasted it to LyX. What does LyX recognize as a text paragraph terminator? Is this a bug? I got the same effect importing as lines or paragraphs. The standard line seperator under *nix is Ctrl-J, so I guess Ctrl-M gets interpreted as 'simple space' You could try to replace the Ctrl-M's by Ctrl-J's by running e.g. tr '\015' '\012' in.txt out.txt Andre' -- André Pönitz .. [EMAIL PROTECTED]
Re: Question for our STL experts!
On Tue, Jan 08, 2002 at 12:16:56AM +0100, Lars Gullik Bjønnes wrote: Now read about pairiterator, bool a_eq.insert(t). And then read about [] again in 23.3.1.2. Indeed, I missed that. I stand corrected. Andre' -- André Pönitz .. [EMAIL PROTECTED]
RE: [Bug 62] inset in inset and undo/redo/ drawing problems
On 06-Jan-2002 bugzilla-daemon wrote: > --- Additional Comments From [EMAIL PROTECTED] 2002-01-07 09:03 --- > Ok, I don't see problems now (prolly from previous redo fixes) Well no I guess that I fixed the drawing problems with my commit yesterday! Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ "I keep seeing spots in front of my eyes." "Did you ever see a doctor?" "No, just spots."
Re: [Patch] fixed bug 59
On Sun, Jan 06, 2002 at 09:03:15PM +, John Levon wrote: > On Sun, Jan 06, 2002 at 10:26:08PM +0200, Martin Vermeer wrote: > > > On Fri, Jan 04, 2002 at 08:36:58AM +0100, Michael Schmitt wrote: > > > - Insert a (math) inset into a table cell. Then, move the cursor along the >current > > > row. The lines at the bottom of other cells are erased (on screen) (#59) > > > > Fixed, and did some cleanup and documentation too, for my own > > consumption ;-) > > tested it, it's OK in all cases I know of except the problem with attachment 37 is >still there. > I think it should be applied anyway. If you like you can fix 37 later ;) The culprit of 37 is the call tabular->GetCellInset(cell)->draw(bv, font, baseline, dx, false); Perhaps I'll look at it later if I find time. > good stuff ! > > > + //lyxerr << " " << tabular->GetAscentOfRow(i) << > > + // " " << tabular->GetDescentOfRow(i) << > > + // " - " << the_locking_inset->ascent(bv, font) ><< > > + // " " << the_locking_inset->descent(bv, font) << > > + // " " << TEXT_TO_INSET_OFFSET << endl; > > why is this bit here ? Wrote it before I found out that half of the parameters printed were irrelevant ;-) Actually I think it can go. > regards > john > > p.s. even better would be to see this section of code in a separate function !! g. Martin -- Martin Vermeer [EMAIL PROTECTED] Helsinki University of Technology Department of Surveying P.O. Box 1200, FIN-02015 HUT, Finland :wq msg31007/pgp0.pgp Description: PGP signature
Re: New insights on bugs
On 06-Jan-2002 Lars Gullik Bjønnes wrote: >| - Some (biggest) memory leaks [snip] >| createUndo(BufferView*,Undo::undo_kind,const Paragraph*,const Paragraph*) > > I really cannot see how and why this happens. Well I can and we ALWAYS had this problem. It's easy when undoing/redoing changes the substituted paragraphs are not freed but their pointer are lost. I put some code in the undo_funcs.C to handle this case but I crash with that code and I really cannot see why. So Michael if you have access next time to a good memory tool, have a look at undo_funcs.C and remove the comment on the #define at the beginning as this then activates the freeing of the paragraphs but this also gives a core dump on undo/redo operations, but maybe we can see what's wrong with that code then. >| * This is occurring while in: >| InsetTabular::insetInInsetY() [insettabular.C:703] >| BufferView::fitLockedInsetCursor(int,int,int,int) >| [BufferView2.C:499] > > Try this patch and see what happens: Well the below patch should not be neccessary! If we access that variable and it was not set before then we have a bug somewhere else in the code! I hope I fixed this with my last commits (yesterday). >| InsetTabular::clone(const Buffer&,bool)const [insettabular.C:175] >| Paragraph::Paragraph(const Paragraph&,bool) [paragraph.C:145] > > This look dangerous. Why this is just that an InsetTabular was cloned and then the undo operation locked the inset without setting the inset_y position, which then is accessed. The badest thing which can occur in such an operation is that the cursor is somewhere too much down and we will see an empty gray screen :), but no crush can occur in this situation, this may also explain the bugs we have open where we sometime see this situation, but as I told you I hope I fixed this. >| #0 0x080cda79 in Paragraph::inInset (this=0x837c128) at >| paragraph.C:2083 > > and how can inInset crash? [snip] Well I guess John explained this already! Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ wok, n.: Something to thwow at a wabbit.
Re: New insights on bugs
On 05-Jan-2002 John Levon wrote: > hmm, this looks like the_locking_inset is referring to a deleted inset. I don't > get this code too much ... Nope the_locking_inset is ok inset_y is not ok (it was not initialized), but I explained this already in my former mail! Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Life is like bein' on a mule team. Unless you're the lead mule, all the scenery looks about the same.
Re: KLyX -- am I missing something?
> "John" == John Levon <[EMAIL PROTECTED]> writes: John> you've missed the point. A checkbox is a checkbox is a checkbox. John> A radio button is another thing altogether. Most toolkits have John> visual clues for telling the difference. Xforms has this too. What about using roundbutton or round3dbutton for radio buttons? JMarc
Re: KLyX -- am I missing something?
On Mon, Jan 07, 2002 at 09:34:31AM +0100, Jean-Marc Lasgouttes wrote: > John> you've missed the point. A checkbox is a checkbox is a checkbox. > John> A radio button is another thing altogether. Most toolkits have > John> visual clues for telling the difference. > > Xforms has this too. What about using roundbutton or round3dbutton for > radio buttons? oh, then it's a lyx bug. Didn't realise. regards john -- "Unless everyone else on earth is attending meetings I haven't been told about." - /. paranoia at its finest
[PATCH] auto-apply for FormTabular
I think this is the simplest and best solution for now.
please apply
john
--
"Unless everyone else on earth is attending meetings I haven't been told
about."
- /. paranoia at its finest
Index: src/frontends/xforms/ChangeLog
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/ChangeLog,v
retrieving revision 1.219
diff -u -p -r1.219 ChangeLog
--- src/frontends/xforms/ChangeLog 2002/01/04 15:14:22 1.219
+++ src/frontends/xforms/ChangeLog 2002/01/07 08:46:48
@@ -1,3 +1,12 @@
+2002-01-07 John Levon <[EMAIL PROTECTED]>
+
+ * FormTabular.h:
+ * FormTabular.C:
+ * form_tabular.h:
+ * form_tabular.C:
+ * forms/form_tabular.fd: ugly fix to get around
+ some of the UI problems (bug #110).
+
2002-01-04 Jürgen Spitzmüller <[EMAIL PROTECTED]>
* FormDocument.C: Recognize the default paper settings too
Index: src/frontends/xforms/FormTabular.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/FormTabular.C,v
retrieving revision 1.31
diff -u -p -r1.31 FormTabular.C
--- src/frontends/xforms/FormTabular.C 2001/12/20 14:52:15 1.31
+++ src/frontends/xforms/FormTabular.C 2002/01/07 08:46:52
@@ -153,6 +153,9 @@ void FormTabular::build()
void FormTabular::update()
{
+ if (closing_)
+ return;
+
if (!inset_ || !inset_->tabular.get())
return;
@@ -479,7 +482,20 @@ bool FormTabular::input(FL_OBJECT * ob,
string special;;
int cell = inset_->getActCell();
-
+
+// ugly hack to auto-apply the stuff that hasn't been
+// yet. don't let this continue to exist ...
+if (ob == dialog_->button_close) {
+closing_ = true;
+input(column_options_->input_column_width, 0);
+input(cell_options_->input_mcolumn_width, 0);
+input(column_options_->input_special_alignment, 0);
+input(cell_options_->input_special_multialign, 0);
+closing_ = false;
+ ok();
+return true;
+}
+
if (actCell_ != cell) {
update();
fl_set_object_label(dialog_->text_warning,
Index: src/frontends/xforms/FormTabular.h
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/FormTabular.h,v
retrieving revision 1.15
diff -u -p -r1.15 FormTabular.h
--- src/frontends/xforms/FormTabular.h 2001/12/19 16:13:21 1.15
+++ src/frontends/xforms/FormTabular.h 2002/01/07 08:46:52
@@ -90,6 +90,8 @@ private:
int actCell_;
/// The ButtonController
ButtonController bc_;
+ /// if we are applying stuff during a close of the dialog
+ bool closing_;
};
Index: src/frontends/xforms/form_tabular.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/form_tabular.C,v
retrieving revision 1.21
diff -u -p -r1.21 form_tabular.C
--- src/frontends/xforms/form_tabular.C 2001/12/19 16:13:21 1.21
+++ src/frontends/xforms/form_tabular.C 2002/01/07 08:46:53
@@ -33,7 +33,7 @@ FD_form_tabular * FormTabular::build_tab
fl_set_button_shortcut(obj, scex(_(dummy)), 1);
}
fl_set_object_lsize(obj, FL_NORMAL_SIZE);
-fl_set_object_callback(obj, C_FormBaseDeprecatedCancelCB, 0);
+fl_set_object_callback(obj, C_FormBaseDeprecatedInputCB, 0);
fdui->input_tabular_column = obj = fl_add_input(FL_NORMAL_INPUT, 65, 260, 60, 30,
"");
fl_set_object_lsize(obj, FL_NORMAL_SIZE);
fl_set_object_lalign(obj, FL_ALIGN_TOP_LEFT);
Index: src/frontends/xforms/form_tabular.h
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/form_tabular.h,v
retrieving revision 1.16
diff -u -p -r1.16 form_tabular.h
--- src/frontends/xforms/form_tabular.h 2001/12/19 16:13:21 1.16
+++ src/frontends/xforms/form_tabular.h 2002/01/07 08:46:53
@@ -5,7 +5,7 @@
#define FD_form_tabular_h_
/** Callbacks, globals and object handlers **/
-extern "C" void C_FormBaseDeprecatedCancelCB(FL_OBJECT *, long);
+extern "C" void C_FormBaseDeprecatedInputCB(FL_OBJECT *, long);
extern "C" void C_FormBaseDeprecatedInputCB(FL_OBJECT *, long);
Index: src/frontends/xforms/forms/form_tabular.fd
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/forms/form_tabular.fd,v
retrieving revision 1.14
diff -u -p -r1.14 form_tabular.fd
--- src/frontends/xforms/forms/form_tabular.fd 2001/12/19 16:13:21 1.14
+++ src/frontends/xforms/forms/form_tabular.fd 2002/01/07 08:46:55
@@ -64,7 +64,7 @@ shortcut:
resize: FL_RESIZE_ALL
gravity: FL_NoGravity FL_NoGravity
name: button_close
-callback: C_FormBaseDeprecatedCancelCB
+callback:
Re: KLyX -- am I missing something?
> "John" == John Levon <[EMAIL PROTECTED]> writes: John> On Mon, Jan 07, 2002 at 09:34:31AM +0100, Jean-Marc Lasgouttes John> wrote: you've missed the point. A checkbox is a checkbox is a John> checkbox. A radio button is another thing altogether. Most John> toolkits have visual clues for telling the difference. >> Xforms has this too. What about using roundbutton or round3dbutton >> for radio buttons? John> oh, then it's a lyx bug. Didn't realise. Indeed. This probably dates back to the first versions... We should maybe change all the radio buttons (not checkbox) to round3dbutton for 1.2.0. Then people will be glad that the interface is much better dur tu GUI-I :) JMarc
Re: New insights on bugs
On 07-Jan-2002 Lars Gullik Bjønnes wrote: > Class variables should be initialized anyway, especially pointers. Well if this is the rule just commit your change it surely won't do anything bad, the only thing we won't get debug notifies anymore and so we cannot be pointed to problems in the code by them. Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ No yak too dirty; no dumpster too hollow.
RE: [PATCH] auto-apply for FormTabular
On 07-Jan-2002 John Levon wrote: > > I think this is the simplest and best solution for now. Well I don't know John! What if I typed something there by error I now have no way to return to the original. If we do this then at least we have to ask the user if he would like us to apply the changes. But then we didn't ask before to (when we had the autoapply on entering data!), so maybe this is really the best choice we have now. I'll apply it in the meantime! Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ You can observe a lot just by watching. -- Yogi Berra
Re: New insights on bugs
On 07-Jan-2002 Lars Gullik Bjønnes wrote:
> Pointer * pointer;
> ...
> if (!pointer) {
>pointer = new Pointer;
> }
>
> will always have problems.
I don't get what you mean here (especially in the contents we are talking
about), but as I told you I don't argue that all class variables should be
initialized I just say that if I would have done this with the variables in
cause I would not have discovered the hidden bug Michael pointed me too. So
my argument was that maybe it's sometimes better to not initialize them if
we know that the should be initialized before we use them.
Jug
--
-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._
Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED]
Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253
I-39100 Bozen Web: http://www.sad.it/~jug
-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._
Cole's Law:
Thinly sliced cabbage.
Re: [PATCH] auto-apply for FormTabular
On Mon, Jan 07, 2002 at 10:04:50AM +0100, Juergen Vigna wrote: > Well I don't know John! What if I typed something there by error I now > have no way to return to the original. If we do this then at least we there's no point worrying about dog poo on the streets when city hall is on fire ... or something. It's good enough I think... regards john -- "Unless everyone else on earth is attending meetings I haven't been told about." - /. paranoia at its finest
Re: More bugs (=new list)
> "Kayvan" == Kayvan A Sylvan <[EMAIL PROTECTED]> writes: Kayvan> On Fri, Jan 04, 2002 at 04:30:51PM +0100, Jean-Marc Lasgouttes Kayvan> wrote: Michael> Only occurs when you click at menu item "View/DVI"; CTRL-D Michael> works well (#122) >> I know how to fix that and will do so next week (once I can enjoy >> celibacy again...). Kayvan> This sounds interesting... Do tell. ;-) What? The fix or the celibacy? For the latter, do like me: drive your wife and kid to Antwerpen train station tomorrow morning. Or maybe Antwerpen is a bit too far for you... JMarc
Re: [PATCH] FileInfo misuse
> "John" == John Levon <[EMAIL PROTECTED]> writes: John> this fixes most of the cases, and asserts on the others John> please apply Applied locally. JMarc
Re: noweb2lyx updates for lyx-1.2.0cvs
> "Kayvan" == Kayvan A Sylvan <[EMAIL PROTECTED]> writes: Kayvan> Hi folks, Here is an update for the noweb2lyx.lyx example Kayvan> file. This does not change noweb2lyx.in in any way. Applied locally JMarc
[Patch] fixed bug 59 attachment 37 too ;-)
As it says.
I must admit that I don't really quite understand this,
and thus don't like it. How can baseline < 0 if not in
error?
Martin
--
Martin Vermeer [EMAIL PROTECTED]
Helsinki University of Technology
Department of Surveying
P.O. Box 1200, FIN-02015 HUT, Finland
:wq
Index: insettext.C
===
RCS file: /cvs/lyx/lyx-devel/src/insets/insettext.C,v
retrieving revision 1.245
diff -u -p -r1.245 insettext.C
--- insettext.C 2001/12/28 13:26:54 1.245
+++ insettext.C 2002/01/07 09:14:32
@@ -388,7 +388,8 @@ void InsetText::draw(BufferView * bv, Ly
if (!cleared && (top_x == int(x))
&& ((need_update&(INIT|FULL)) || (top_baseline != baseline)
||(last_drawn_width != insetWidth))) {
- clearInset(bv, baseline, cleared);
+ // Condition necessary to eliminate bug 59 attachment 37
+ if (baseline > 0) clearInset(bv, baseline, cleared);
}
top_x = int(x);
msg31021/pgp0.pgp
Description: PGP signature
Re: [Patch] fixed bug 59
> "Martin" == Martin Vermeer <[EMAIL PROTECTED]> writes: Martin> On Fri, Jan 04, 2002 at 08:36:58AM +0100, Michael Schmitt Martin> wrote: >> - Insert a (math) inset into a table cell. Then, move the cursor >> along the current row. The lines at the bottom of other cells are >> erased (on screen) (#59) Martin> Fixed, and did some cleanup and documentation too, for my own Martin> consumption ;-) I applied it in my tree. What about a ChangeLog entry? I removed also the debugging statements said to be useless. What about the others? JMarc
Re: [PATCH] auto-apply for FormTabular
On 07-Jan-2002 John Levon wrote: > It's good enough I think... Well yes but could you send me a ChangeLog? (I already applied the patch so please only the ChangeLog!) Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Mirrors should reflect a little before throwing back images. -- Jean Cocteau
Re: [PATCH] fix insets wider than screen
> "John" == John Levon <[EMAIL PROTECTED]> writes: John> I hadn't considered that inset_x could be < 0. This patch fixes John> bug #162 reported by Herbert. John> please apply john Done. JMarc
Re: [PATCH] auto-apply for FormTabular
On Mon, Jan 07, 2002 at 10:20:21AM +0100, Juergen Vigna wrote: > Well yes but could you send me a ChangeLog? (I already applied the patch > so please only the ChangeLog!) the patch didn't include it ? 2002-01-07 John Levon <[EMAIL PROTECTED]> * FormTabular.h: * FormTabular.C: * form_tabular.h: * form_tabular.C: * forms/form_tabular.fd: ugly fix to get around some of the UI problems (bug #110). regards john -- "Unless everyone else on earth is attending meetings I haven't been told about." - /. paranoia at its finest
Re: [Patch] fixed bug 59
On Mon, Jan 07, 2002 at 10:21:38AM +0100, Jean-Marc Lasgouttes wrote: > > > "Martin" == Martin Vermeer <[EMAIL PROTECTED]> writes: > > Martin> On Fri, Jan 04, 2002 at 08:36:58AM +0100, Michael Schmitt > Martin> wrote: > >> - Insert a (math) inset into a table cell. Then, move the cursor > >> along the current row. The lines at the bottom of other cells are > >> erased (on screen) (#59) > > Martin> Fixed, and did some cleanup and documentation too, for my own > Martin> consumption ;-) > > I applied it in my tree. What about a ChangeLog entry? Like: --- 2002-01-07 Martin Vermeer <[EMAIL PROTECTED]> * insettabular.C: * insettext.C: Fix for bug #59: Insert a (math) inset into a table cell. Then, move the cursor along the current row. The lines at the bottom of other cells are erased (on screen). The insettext.C fix eliminates bug illustrated by attachment 37. --- (Assumes the additional fix also applied) > I removed also the debugging statements said to be useless. What about > the others? IMO they can go if this is found to work flawlessly. Easily re-instated if needed. > JMarc > Martin -- Martin Vermeer [EMAIL PROTECTED] Helsinki University of Technology Department of Surveying P.O. Box 1200, FIN-02015 HUT, Finland :wq msg31026/pgp0.pgp Description: PGP signature
Re: KLyX -- am I missing something?
On Mon, Jan 07, 2002 at 09:58:46AM +0100, Jean-Marc Lasgouttes wrote:
> We should maybe change all the radio buttons (not checkbox) to
> round3dbutton for 1.2.0. Then people will be glad that the interface
do I win most boring patch ?
I've spared bandwidth so applier need to run "make updatesrc" in forms/ before
applying. Everything seems to looks OK I think ...
regards
john
--
"Unless everyone else on earth is attending meetings I haven't been told
about."
- /. paranoia at its finest
Index: src/frontends/xforms/ChangeLog
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/ChangeLog,v
retrieving revision 1.219
diff -u -p -r1.219 ChangeLog
--- src/frontends/xforms/ChangeLog 2002/01/04 15:14:22 1.219
+++ src/frontends/xforms/ChangeLog 2002/01/07 09:35:26
@@ -1,3 +1,8 @@
+2002-01-07 John Levon <[EMAIL PROTECTED]>
+
+ * form_*:
+ * forms/form_*: change radio buttons to roundbutton
+
2002-01-04 Jürgen Spitzmüller <[EMAIL PROTECTED]>
* FormDocument.C: Recognize the default paper settings too
Index: src/frontends/xforms/forms/form_document.C.patch
===
RCS file:
/usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/forms/form_document.C.patch,v
retrieving revision 1.6
diff -u -p -r1.6 form_document.C.patch
--- src/frontends/xforms/forms/form_document.C.patch2001/03/15 13:37:03 1.6
+++ src/frontends/xforms/forms/form_document.C.patch2002/01/07 09:35:36
@@ -1,6 +1,6 @@
form_document.CWed Oct 4 18:07:46 2000
-+++ form_document.C.orig Wed Oct 4 18:07:23 2000
-@@ -9,6 +9,8 @@
+--- form_document.CMon Jan 7 09:19:31 2002
form_document.C.newMon Jan 7 09:19:24 2002
+@@ -10,6 +10,8 @@
#include
#include "form_document.h"
#include "FormDocument.h"
@@ -9,25 +9,25 @@
FD_form_tabbed_document::~FD_form_tabbed_document()
{
-@@ -390,10 +392,18 @@
+@@ -543,10 +545,18 @@
fdui->form = fl_bgn_form(FL_NO_BOX, 440, 345);
fdui->form->u_vdata = this;
+ fl_set_border_width(-1);
obj = fl_add_box(FL_FLAT_BOX, 0, 0, 440, 345, "");
-- fdui->bmtable_bullet_panel = obj = fl_add_pixmapbutton(FL_NORMAL_BUTTON, 90, 105,
260, 180, "");
+- fdui->bmtable_bullet_panel = obj = fl_add_pixmapbutton(FL_NORMAL_BUTTON, 135, 105,
+260, 180, "");
-fl_set_object_boxtype(obj, FL_NO_BOX);
-fl_set_pixmapbutton_file(obj,
_("/nfs/sinco/source/lyx/lyx-devel/lib/images/psnfss2.xpm"));
+ fl_set_border_width(-3);
-+ fdui->bmtable_bullet_panel = obj = fl_add_bmtable(1, 90, 105, 265, 180, "");
++ fdui->bmtable_bullet_panel = obj = fl_add_bmtable(1, 135, 105, 260, 180, "");
+ fl_set_object_callback(obj, C_FormBaseDeprecatedInputCB, BULLETBMTABLE);
+ fl_set_object_lcol(obj, FL_BLUE);
+ fl_set_object_boxtype(obj, FL_UP_BOX);
+ fl_set_bmtable_pixmap_file(obj, 6, 6,
+ LibFileSearch("images",
-+ "standard",
++ "standard",
+ "xpm").c_str());
+ fl_set_border_width(-1);
- obj = fl_add_frame(FL_ENGRAVED_FRAME, 95, 20, 255, 70, "");
- fdui->choice_bullet_size = obj = fl_add_choice(FL_NORMAL_CHOICE, 15, 45, 65, 30,
idex(_("Size|#z")));
- fl_set_button_shortcut(obj, scex(_("Size|#z")), 1);
+ obj = fl_add_frame(FL_ENGRAVED_FRAME, 135, 20, 260, 55, "");
+ {
+ char const * const dummy = N_("Size|#z");
Index: src/frontends/xforms/forms/form_document.fd
===
RCS file:
/usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/forms/form_document.fd,v
retrieving revision 1.18
diff -u -p -r1.18 form_document.fd
--- src/frontends/xforms/forms/form_document.fd 2001/11/12 14:28:12 1.18
+++ src/frontends/xforms/forms/form_document.fd 2002/01/07 09:35:37
@@ -328,7 +328,7 @@ argument:
class: FL_BEGIN_GROUP
type: 0
-box: 0 10 10 0
+box: 0 0 0 0
boxtype: FL_NO_BOX
colors: FL_COL1 FL_MCOL
alignment: FL_ALIGN_CENTER
@@ -344,7 +344,7 @@ callback:
argument:
-class: FL_CHECKBUTTON
+class: FL_ROUNDBUTTON
type: RADIO_BUTTON
box: 255 40 120 30
boxtype: FL_NO_BOX
@@ -362,7 +362,7 @@ callback: C_FormBaseDeprecatedInputCB
argument: INPUT
-class: FL_CHECKBUTTON
+class: FL_ROUNDBUTTON
type: RADIO_BUTTON
box: 255 75 120 30
boxtype: FL_NO_BOX
@@ -928,7 +928,7 @@ argument: INPUT
class: FL_BEGIN_GROUP
type: 0
-box: 0 10 10 0
+box: 0 0 0 0
boxtype: FL_NO_BOX
colors: FL_COL1 FL_MCOL
alignment: FL_ALIGN_CENTER
@@ -944,7 +944,7 @@ callback:
argument:
-class: FL_CHECKBUTTON
+class: FL_ROUNDBUTTON
type: RADIO_BUTTON
box: 300 30 120 30
boxtype: FL_NO_BOX
@@ -962,7 +962,7 @@ callback: C_FormBaseDeprecatedInputCB
argument: INPUT
-class:
Re: [Patch] fixed bug 59
> "Martin" == Martin Vermeer <[EMAIL PROTECTED]> writes: Martin> Like: [snip] Martin> (Assumes the additional fix also applied) Thanks. I applied the additional fix too. JMarc
Re: [Patch] fixed bug 59
On 07-Jan-2002 Jean-Marc Lasgouttes wrote: > I removed also the debugging statements said to be useless. What about > the others? Jean-Marc I applied the InsetTabular patch about clearing the cell and also added a ChangeLog entry. Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Schapiro's Explanation: The grass is always greener on the other side -- but that's because they use more manure.
Re: [PATCH] auto-apply for FormTabular
On 07-Jan-2002 John Levon wrote: > On Mon, Jan 07, 2002 at 10:20:21AM +0100, Juergen Vigna wrote: > >> Well yes but could you send me a ChangeLog? (I already applied the patch >> so please only the ChangeLog!) > > the patch didn't include it ? Sorry I didn't see it! Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Yea from the table of my memory I'll wipe away all trivial fond records. -- Hamlet
Re: [Patch] fixed bug 59
> "Juergen" == Juergen Vigna <[EMAIL PROTECTED]> writes: Juergen> On 07-Jan-2002 Jean-Marc Lasgouttes wrote: >> I removed also the debugging statements said to be useless. What >> about the others? Juergen> Jean-Marc I applied the InsetTabular patch about clearing the Juergen> cell and also added a ChangeLog entry. You mean martin's patch for #59? Are you going to commit soon? Otherwise I'll have to revert it before commiting. JMarc
Re: [Patch] fixed bug 59
On 07-Jan-2002 Jean-Marc Lasgouttes wrote: > Thanks. I applied the additional fix too. Jean-Marc I just applied the one in InsetTabular (sorry but I had all ready and changed some stuff so I commited already). If you'd like I apply the one in InsetText too, otherwise I let it to you if you're already ready for commit. Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Italienallee 13/N Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39100 Bozen Web: http://www.sad.it/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Mate, this parrot wouldn't VOOM if you put four million volts through it! -- Monty Python
