Re: Whining about lyx1.2.1 ;-)))
On Fri, 4 Oct 2002, John Levon wrote: > On Fri, Oct 04, 2002 at 10:04:54AM -0400, Kuba Ober wrote: [...] > > 4. Sometime between win 3.1.0 and win98 certain windows apps were used to have > > "Keyboard" topic in their help system. I don't think that was such a bad > > idea. I personally wouldn't mind having a short lyx help document with all > > keybindings nicely formatted to fit a single sheet of paper (or two) (kind of > > a keybort cheatsheet). Now if anybody can point me to a complete list of > > keybindings for lyx in any form (whether in sources, config files, or > > elsewhere), I'd be happy to make such a cheatsheet. > > lyx-devel/lib/bind/ > > also some default bindings in src/lyx_main.C Also see the reference manual. Allan. (ARRae)
development process
If I had the same amount of time to spend to work on LyX I had in June I'd be pretty disappointed by the current speed of the development process. Fortunately I have not. Nevertheless I think we need to adjust the process in the future somehow. The 1.2 and 1.3 freezes somehow follow the same scheme: Something that was considered "necessary" for the release was not finished and nobody really helped out. This basically means we waste a potential of several "volunteer months" worth of work - and there is no "auto-correction" mechanism that breaks the dead-lock which occurs "if nobody does anything" I have no strong opinion or even an idea on how to solve that, I just want to say that there _is_ a problem. A serious one to be precise. Making people help out would probably the way to go if this were a company. But forcing something down the throat of volunteers may work once or twice, but that's not sustainable over years. I think we need some automatism to solve the dead-lock issue. Like "if pre1 is not out after a four week freeze, the freeze is revoked"... Hanging around indefinitely does not help anybody, neither the project nor the users. It's pretty irrelevant from a user's point of view whether he does not get an update because of an overly long freeze or because of on-going development without official releases. But in the latter case he gets at least the chance of an improvement... Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: Figures can't be updated, trashed document
Darren Freeman wrote: > Unpack the attached tgz anywhere. You will see the directory "Blah1 - > Blah2" and the symlink "blah-link" to this directory. > > LyX has problems when there are spaces in filenames etc. so the link is > a tempting fix. CD into the linked directory. LyX/LaTeX does not support spaces in the filename Herbert -- http://www.educat.hu-berlin.de/~voss/lyx/
Re: bidi_level
On Mon, Oct 07, 2002 at 10:20:30PM +0200, Dekel Tsur wrote: > On Mon, Oct 07, 2002 at 05:36:18PM +0200, Andre Poenitz wrote: > > > > Could anybody please explain in simple words what this bidi stuff is good > > for and how it works? > > It is related to RTL handling: > > For LTR text, bidi_level = 0 > For RTL text, bidi_level = 1 > For LTR text inside RTL text, bidi_level = 2 And RTL in LTR? And other nestings? Is this for display only? And what does bidi_start mean? And waht is bidi_start = -1? Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
autogen.sh: GNU m4 test is fake!
Hi I reported this before, but was ignored :(. I still believe this is wrong, though not that critical: In autogen.sh there is: if test x$ok = xno ; then echo "not found." fi which tests for the presence of any m4 compiler. However, "$ok" is either empty or "yes". So testing it against "no" is bullocks. Initialize "$ok" to "no", or replace the test by "if text x$ok != xyes". Cheers, Rob.
Re: Figures can't be updated, trashed document
Darren Freeman wrote: > > It is critical that the offending directory be like "Blah1 - Blah2" and > not just "Blah1 Blah2", the minus character has something to do with it. Could it be that the "-" is interpreted as a flag to the command, screwing up the whole meaning of the command's operation? On unix that's usually remedied by adding "--" inbetween the flags and the parameters. Just a thought. Any gurus out there that may confirm this? Regards, Rob.
Re: Figures can't be updated, trashed document
At 06:22 PM 07/10/2002, you wrote: >On Mon, Oct 07, 2002 at 05:00:04PM +0930, Darren Freeman wrote: > > > When I open the file with LyX, it freezes about 20% of the way down the > > document. As in, you scroll down and after hitting a certain part it > > freezes LyX. > >What do top or ps aux say about the status of the lyx process? I grabbed the output of ps aux for you but left it on another drive - woops. It was running external converters for the image anyway. You'll be more than compensated by the following =) > > And NO I can't post the document it's confidential. It's free software, > not > > free document =) > >Would you manage somehow to cut down the doc to the minimum producing the >lock-up? And is it then postable? My first look at the LyX file format =) I managed to come up with a minimal file which is 100% guaranteed to lock up LyX on my system. Hopefully yours too. It turns out that the fault was due to weird filenames and symlinks.. I couldn't believe what I had to preserve in the directory tree to keep it locking up. Do the following: Unpack the attached tgz anywhere. You will see the directory "Blah1 - Blah2" and the symlink "blah-link" to this directory. LyX has problems when there are spaces in filenames etc. so the link is a tempting fix. CD into the linked directory. You will see hang.lyx and fig.png Run LyX from this directory. Choose the open file dialogue. Something interesting happens: the path to the file you are looking at is the absolute path, not the symlink where you were. Hence there are spaces in this filename. If you open hang.lyx and unfold the figure, you will lock LyX while the external converters get confused by the filenames. If you instead open by changing to the path including the blah-link symlink, so that no spaces are present, everything is fine. Of course fig.png is empty, but if you stuck something in there it would now display it. It is critical that the offending directory be like "Blah1 - Blah2" and not just "Blah1 Blah2", the minus character has something to do with it. So you see how much of a fluke it was that I even caused this problem!! > > Have fun, > > Darren Freeman > > > >Martin Hope this helped.. Was a real bitch to figure out how to cause the fault in a directory-portable way! Have fun, Darren Freeman hang-lyx.tgz Description: application/compressed
Re: Save from Word as TXT, import, get extra spaces
On Mon, Oct 07, 2002 at 04:34:57PM +0930, Darren Freeman wrote: > I saved my Word DOC using Word 2000 (nooo!!!), as a TXT file. > > After importing into LyX, I found that there was an extra space at the end > of each paragraph. Possibly due to the extra control code DOS uses with a > newline. > > This just makes me want to go through the whole file and remove these > spaces.. I don't want no steenkin' DOS reminders in my shiny new report =) You can run dos2unix on the text file before importing it to LyX.
Re: bidi_level
On Mon, Oct 07, 2002 at 05:36:18PM +0200, Andre Poenitz wrote: > > Could anybody please explain in simple words what this bidi stuff is good > for and how it works? It is related to RTL handling: For LTR text, bidi_level = 0 For RTL text, bidi_level = 1 For LTR text inside RTL text, bidi_level = 2
Re: [PATCH]: QtWrap
On Sat, Sep 28, 2002 at 06:04:35PM +0200, Juergen Spitzmueller wrote: > OK, here you are. Sorry for the confusion. Applied. thanks john -- "I will eat a rubber tire to the music of The Flight of the Bumblebee"
[1.2.1 RPM for Red Hat 8.0]
The following is available as ftp://ftp.sylvan.com/pub/lyx/lyx-1.2.1-1rh8-xforms089.i386.rpm Thanks, Milos! - Forwarded message from Milos Komarcevic <[EMAIL PROTECTED]> - Subject: 1.2.1 RPM for Red Hat 8.0 From: Milos Komarcevic <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] Date: 07 Oct 2002 15:39:06 +0100 Kayvan I have just uploaded the 1.2.1 RPM built on Red Hat 8.0 with xforms 0.89 to ftp://ftp.sylvan.com/incoming/forlyx Can you please make it available? Thanks, Milos - End forwarded message - -- 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) msg45595/pgp0.pgp Description: PGP signature
bidi_level
Could anybody please explain in simple words what this bidi stuff is good for and how it works? [I was just trying to fix the "cursor in front of displayed formula is broken" bug, but I do not even understand how it works in the "normal" case. This stuff is seriously broken...] Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
1.2.1 RPM for Red Hat 8.0
Kyvan I have just uploaded the 1.2.1 RPM built on Red Hat 8.0 with xforms 0.89 to ftp://ftp.sylvan.com/incoming/forlyx Can you please make it available? Thanks, Milos
Re: [PATCH] Handle unavailable viewers
Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes: | This patch handles gracefully the case where a viewer has the value of | "none" (program not found). This avoids problems reported by people | who do not have netscape installed (recent linux installation). Proper | support for mozilla (with file: URL) will come later. > | OK to apply? Yes. -- Lgb
Re: FormLog.C: Build log archaic?
Angus Leeming <[EMAIL PROTECTED]> writes:
| Well done, Rob.
>
| why not add a little helper function to xforms_helpers.[Ch]
>
| bool isActive(FL_OBJECT * ob)
| {
| return ob && ob->active > 0;
| }
yes.
--
Lgb
Re: [Patch] small bug in xforms_helpers.C ?
Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes: >> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes: > | Angus> I think that this is new to me (maybe I missed it following my | Angus> recent travels) but that it's definitely a bug. Could you apply | Angus> the fix please. > | Lars? Yes, it looks obvious. -- Lgb
Re: Whining about lyx1.2.1 ;-)))
On Mon, Oct 07, 2002 at 10:15:42AM -0400, Kuba Ober wrote: > Is this backported into 1.2.x branch? Don't think so and I don't think this will ever happen. There was a major change in the font handling code from 1.2 to 1.3 and I do not really want to look at the old stuff again... [Apart from that I'd like people to test the new code ;-}] Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: Whining about lyx1.2.1 ;-)))
> > 2. Using \mathrm doesn't seem to work: it only makes the first letter
> > roman, as in ${\mathrm r}oman$. It used to work fine in latest 1.6.x. To
> > say "k^th" with th in roman, you'd type C-m k^\mathrmth.
> This is fixed in 1.3.0.
Is this backported into 1.2.x branch?
Cheers, Kuba Ober
Re: New Xforms "Paragraph Layout" dialog. Comments?
Jean-Marc Lasgouttes wrote: >>"R" == R Lahaye <[EMAIL PROTECTED]> writes: > > > R> Thanks so much! That's exactly the information I needed. > > R> Please have a look at attached patch to ControlParagraph.[Ch]; > R> would you agree with this way of doing it? (I renamed "alignpos_" > R> to "alignpossible_" on the fly, cause it's a better name :) ). > > I do not agree with > > /// is alignment possible > - alignpos_ = layout->alignpossible; > + alignpossible_ = layout->alignpossible; > + > + /// set default alignment > + aligndefault_ = pp_->align(); > > I would use > + aligndefault_ = layout->align; > > What we want is the default of the textclass. Yes, thanks! I have fixed it here in my tree, and I'll keep it for later in my mega patch to Xforms. Rob.
Re: Menu system pauses for a long time
On Mon, Oct 07, 2002 at 04:02:06PM +0200, Jean-Marc Lasgouttes wrote: > Since it seems TC has had no time to look at this bug, I did a bit of > searching of my own (using an old sources snapshot from TC) and found > that the culprit is in handle_motion (xpopup.c) where the item number > cval is computed. > > The comment below says it all: > > /* if released on title bar, cval is zero. However, if there is no title, >change cval to -1 (invalid) if "too right" */ > > It happens that "too right" means "more than 1/3 of the menu width". > What's the reason for that? It is a real pain with LyX, when if the > File menu is open, clicking on Edit does nothing (not in the "too > right" area). iirc the problem was that that code doesn't have a good idea of where one title ends and the other begins. At least, that was the problem I hit when trying to fix the menus to behave in a bearable manner. regards john -- "I will eat a rubber tire to the music of The Flight of the Bumblebee"
Re: Menu system pauses for a long time
> "John" == John Levon <[EMAIL PROTECTED]> writes: John> On Mon, Oct 07, 2002 at 10:09:03AM +0200, Jean-Marc Lasgouttes John> wrote: >> This is a bug in xforms and I do not know what to do about it. The >> File menu will only close if you click at least in the second half >> of the popup. So if you click on Edit, it is too close and does not >> work. This is a deliberate action of xforms and no I do not know >> why this was done. John> I'm not so sure this is deliberate. I spent a while a bit ago John> trying to fix up the xforms menu code, but didn't really get John> anywhere. John> Have we ever actually complained to them ? Here is something I sent two years ago, along with the offending code. I never got an answer from TC as far as I know. It would be nice to investigate whether this clever code could be removed. JMarc - # To subscribers of the xforms list from Jean-Marc Lasgouttes <[EMAIL PROTECTED]> : > "Jean-Marc" == Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes: Jean-Marc> I have done a small example of a bug that have been Jean-Marc> plaguing LyX for a long time (file attached). Jean-Marc> In the following example (butchered from the demo popup.c), Jean-Marc> if you click on the button so that the menu appears, then Jean-Marc> clicking just to the right of the button fails to close the Jean-Marc> menu. This is a pain in LyX where the menu bar is a bunch Jean-Marc> of this buttons and opening 'Edit' when 'File' is open is a Jean-Marc> real problem. Since it seems TC has had no time to look at this bug, I did a bit of searching of my own (using an old sources snapshot from TC) and found that the culprit is in handle_motion (xpopup.c) where the item number cval is computed. The comment below says it all: /* if released on title bar, cval is zero. However, if there is no title, change cval to -1 (invalid) if "too right" */ It happens that "too right" means "more than 1/3 of the menu width". What's the reason for that? It is a real pain with LyX, when if the File menu is open, clicking on Edit does nothing (not in the "too right" area). Another things that confuses me is that the example I gave does use a title, so it should not have this problem. However, it certainly explains the problem seen with LyX. Hope this helps. Another question: having an existing popup, is it possible to add a separator to it? In our case, we do not know what the last entry is, since we use %x to convey information about the function linked to the entry. This means our current code is a bit more complicated than it should. JMarc _ To unsubscribe, send the message "unsubscribe" to [EMAIL PROTECTED] or see http://bob.usuhs.mil/mailserv/xforms.html XForms Home Page: http://world.std.com/~xforms List Archive: http://bob.usuhs.mil/mailserv/list-archives/
Re: [Patch] small bug in xforms_helpers.C ?
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes: Angus> I think that this is new to me (maybe I missed it following my Angus> recent travels) but that it's definitely a bug. Could you apply Angus> the fix please. Lars? JMarc
Re: An Addition for "emacs.bind"
> "John" == John Weiss <[EMAIL PROTECTED]> writes: John> Well, after upgrading to v1.2.x, I began the task of merging John> changes in with my local customizations. Imagine my glee at John> seeing the "command-sequence" binding! At last --- we can add a John> proper char-transpose to the Emacs bindings! What's wrong with \bind "C-t""chars-transpose" that we already have? JMarc
Re: filesize limit on the docs list
> "Mate" == Mate Wierdl <[EMAIL PROTECTED]> writes: Mate> There is a 60K filesize limit on the docs list; should I raise Mate> it? If yes, to what? I think 150k would be a good limit, since it allows to send even our larger files. JMarc
Re: [Patch] small bug in xforms_helpers.C ?
On Monday 07 October 2002 2:37 pm, Jean-Marc Lasgouttes wrote:
> > "R" == R Lahaye <[EMAIL PROTECTED]> writes:
>
> R> Hi,
>
> R> In xforms_helpers.C, there is the following in function
> R> formatted():
>
> R>string::size_type const nxtpos1 = sin.find(' ', curpos);
> R> string::size_type const nxtpos2 = sin.find('\n', curpos);
> R> string::size_type const nxtpos = std::min(nxtpos1,
> nxtpos1);
>
> R> The last line looks suspicious to me. The minimum of the
> same R> integers is most probably a typo. I suppose it should
> be R> "std::min(nxtpos1, nxtpos2)".
>
> R> If so, attached patch fixes that.
>
> Angus, what is the status of this? The code definitely seems
> buggy...
>
> JMarc
I think that this is new to me (maybe I missed it following my
recent travels) but that it's definitely a bug. Could you apply
the fix please.
Angus
Re: New Xforms "Paragraph Layout" dialog. Comments?
> "R" == R Lahaye <[EMAIL PROTECTED]> writes: R> Thanks so much! That's exactly the information I needed. R> Please have a look at attached patch to ControlParagraph.[Ch]; R> would you agree with this way of doing it? (I renamed "alignpos_" R> to "alignpossible_" on the fly, cause it's a better name :) ). I do not agree with /// is alignment possible - alignpos_ = layout->alignpossible; + alignpossible_ = layout->alignpossible; + + /// set default alignment + aligndefault_ = pp_->align(); I would use + aligndefault_ = layout->align; What we want is the default of the textclass. JMarc
Re: My favourite showstopper (#575 and #438)
> "Michael" == Michael Schmitt <[EMAIL PROTECTED]> writes: Michael> BTW: Are there any plans to release a 1.2.2 version before Michael> 1.3.0 is out? (I always work with the latest patches in the Michael> cvs, of course; I am just curious) Yes, I'd like to do that. JMarc
Re: Menu system pauses for a long time
On Mon, Oct 07, 2002 at 10:09:03AM +0200, Jean-Marc Lasgouttes wrote: > This is a bug in xforms and I do not know what to do about it. The > File menu will only close if you click at least in the second half of > the popup. So if you click on Edit, it is too close and does not work. > This is a deliberate action of xforms and no I do not know why this > was done. I'm not so sure this is deliberate. I spent a while a bit ago trying to fix up the xforms menu code, but didn't really get anywhere. Have we ever actually complained to them ? john -- "I will eat a rubber tire to the music of The Flight of the Bumblebee"
Re: [Patch] small bug in xforms_helpers.C ?
> "R" == R Lahaye <[EMAIL PROTECTED]> writes:
R> Hi,
R> In xforms_helpers.C, there is the following in function
R> formatted():
R> string::size_type const nxtpos1 = sin.find(' ', curpos);
R> string::size_type const nxtpos2 = sin.find('\n', curpos);
R> string::size_type const nxtpos = std::min(nxtpos1, nxtpos1);
R> The last line looks suspicious to me. The minimum of the same
R> integers is most probably a typo. I suppose it should be
R> "std::min(nxtpos1, nxtpos2)".
R> If so, attached patch fixes that.
Angus, what is the status of this? The code definitely seems buggy...
JMarc
Re: [PATCH] automatic reconfiguration
> "Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes: Lars> I thought we agreed to apply it? OK, I applied it. Note that more cleanup is possible/needed. JMarc
[PATCH] Handle unavailable viewers
This patch handles gracefully the case where a viewer has the value of
"none" (program not found). This avoids problems reported by people
who do not have netscape installed (recent linux installation). Proper
support for mozilla (with file: URL) will come later.
OK to apply?
JMarc
Index: src/ChangeLog
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/ChangeLog,v
retrieving revision 1.948
diff -u -p -r1.948 ChangeLog
--- src/ChangeLog 7 Oct 2002 09:41:59 - 1.948
+++ src/ChangeLog 7 Oct 2002 13:24:45 -
@@ -1,3 +1,10 @@
+2002-10-07 Jean-Marc Lasgouttes <[EMAIL PROTECTED]>
+
+ * lyxrc.C (read): treat a viewer or converter command of "none" as
+ if it were empty.
+
+ * MenuBackend.C (expandFormats): for an update, also allow the
+ formats that are not viewable
2002-10-07 André Pönitz <[EMAIL PROTECTED]>
Index: src/MenuBackend.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/MenuBackend.C,v
retrieving revision 1.55
diff -u -p -r1.55 MenuBackend.C
--- src/MenuBackend.C 10 Sep 2002 11:50:13 - 1.55
+++ src/MenuBackend.C 7 Oct 2002 13:24:45 -
@@ -352,7 +352,7 @@ void expandFormats(MenuItem::Kind kind,
action = LFUN_PREVIEW;
break;
case MenuItem::UpdateFormats:
- formats = Exporter::GetExportableFormats(buf, true);
+ formats = Exporter::GetExportableFormats(buf, false);
action = LFUN_UPDATE;
break;
default:
Index: src/lyxrc.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/lyxrc.C,v
retrieving revision 1.131
diff -u -p -r1.131 lyxrc.C
--- src/lyxrc.C 28 Aug 2002 08:30:26 - 1.131
+++ src/lyxrc.C 7 Oct 2002 13:24:45 -
@@ -977,7 +977,8 @@ int LyXRC::read(string const & filename)
if (lexrc.next()) {
flags = lexrc.getString();
}
- if (command.empty() || command == "none") {
+ if (command.empty()
+ || token(command, ' ', 0) == "none") {
converters.erase(from, to);
} else {
converters.add(from, to, command, flags);
@@ -991,6 +992,8 @@ int LyXRC::read(string const & filename)
}
if (lexrc.next()) {
command = lexrc.getString();
+if (token(command, ' ', 0) == "none")
+ command.erase();
}
formats.setViewer(format, command);
break;
Re: word `Greek` in LyX
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes: >> As follows from a discussion we had earlier with David Karstrup, it >> would be better to replace this "Greek" by "abc" and avoid to >> translate it at all. Angus> I vaguely remember the discussion, but didn't see a patch, then Angus> or now. Indeed :) JMarc
Re: FormLog.C: Build log archaic?
On Monday 07 October 2002 1:35 pm, Rob Lahaye wrote:
> Angus Leeming wrote:
> > Maybe we're looking at the wrong object. We should be
> > looking at the input object to check its activation state,
> > not at the label.
> >
> > Look in the xforms source (object.c). fl_activate_object and
> > fl_deactivate_object most definitely DO set the active flag
> > appropriately.
>
> No, I'm not looking at the wrong object.
> I just found out:
>
> FL_OBJECT *object;
>
> object is active: object->active = 1
> object is inactive: object->active = -1
>
> So using "if (object->active)" always executes the true
> case!!! We should not treat it implicitly as a bool.
>
> Instead we should use:
>object->active > 0 for active
> and
>object->active < 0 for inactive
>
> Ha, silly, isn't it!
>
> Cheers,
> Rob.
Well done, Rob.
why not add a little helper function to xforms_helpers.[Ch]
bool isActive(FL_OBJECT * ob)
{
return ob && ob->active > 0;
}
Angus
aleem@thorax:src$ grep ACTIVATED *.h
flinternal.h:#define DEACTIVATED -1 /* must be none zero and
negative */
Re: FormLog.C: Build log archaic?
Angus Leeming wrote: > > Maybe we're looking at the wrong object. We should be looking at > the input object to check its activation state, not at the label. > > Look in the xforms source (object.c). fl_activate_object and > fl_deactivate_object most definitely DO set the active flag > appropriately. No, I'm not looking at the wrong object. I just found out: FL_OBJECT *object; object is active: object->active = 1 object is inactive: object->active = -1 So using "if (object->active)" always executes the true case!!! We should not treat it implicitly as a bool. Instead we should use: object->active > 0 for active and object->active < 0 for inactive Ha, silly, isn't it! Cheers, Rob.
Re: word `Greek` in LyX
On Monday 07 October 2002 1:03 pm, Jean-Marc Lasgouttes wrote: > > "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes: > > Angus> On Sunday 06 October 2002 7:34 pm, Vitaly Lipatov wrote: > >> Please pay attention for 'Greek' in the Math Panel. It word > >> have not to translate in this context. > > Angus> form_maths_panel.C is autogenerated from > form_maths_panel.fd, Angus> so I can't apply your patch. > > Angus> Having said that, I don't see what is wrong anyway. > Passing Angus> "Greek" to the _() function is the right thing > to do and IS Angus> done, here at least. > > As follows from a discussion we had earlier with David > Karstrup, it would be better to replace this "Greek" by "abc" > and avoid to translate it at all. I vaguely remember the discussion, but didn't see a patch, then or now. Angus
Re: word `Greek` in LyX
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes: Angus> On Sunday 06 October 2002 7:34 pm, Vitaly Lipatov wrote: >> Please pay attention for 'Greek' in the Math Panel. It word have >> not to translate in this context. Angus> form_maths_panel.C is autogenerated from form_maths_panel.fd, Angus> so I can't apply your patch. Angus> Having said that, I don't see what is wrong anyway. Passing Angus> "Greek" to the _() function is the right thing to do and IS Angus> done, here at least. As follows from a discussion we had earlier with David Karstrup, it would be better to replace this "Greek" by "abc" and avoid to translate it at all. JMarc
Re: word `Greek` in LyX
On Sunday 06 October 2002 7:34 pm, Vitaly Lipatov wrote: > Please pay attention for 'Greek' in the Math Panel. > It word have not to translate in this context. form_maths_panel.C is autogenerated from form_maths_panel.fd, so I can't apply your patch. Having said that, I don't see what is wrong anyway. Passing "Greek" to the _() function is the right thing to do and IS done, here at least. Regards, Angus
Re: crash moving with arrow in a fraction
On Fri, Oct 04, 2002 at 04:18:39PM -0400, Dr. Richard E. Hawkins wrote:
> I had a fraction inside a comment. I put n in the numerator, asked for
> parenthesis in the denominator, entered n-r, exited the parentheses,
> added an exclamation point, and hit up-arrow to get back to the
> numerator. Both numerator and denominatro became highlighted, and
Hm...
> ived signal SIGSEGV, Segmentation fault.
> 0x822695d in MathAtom::MathAtom (this=0x87fbb40, at=@0x87fbb3c)
> at math_atom.C:38
> 38 : nucleus_(at.nucleus_ ? at.nucleus_->clone() : 0)
> (gdb) bt
> #0 0x822695d in MathAtom::MathAtom (this=0x87fbb40, at=@0x87fbb3c)
> at math_atom.C:38
> #1 0x8234d0d in vector >::_M_insert_aux (
> this=0x87faaf8, __position=0x87fbb44, __x=@0xbfbfd03c)
> at /usr/include/g++/stl_construct.h:48
> #2 0x8233f85 in MathArray::insert (this=0x87faaf8, pos=1,
> t=@0xbfbfd03c)
> at /usr/include/g++/stl_vector.h:348
> #3 0x82279f0 in MathCursor::plainInsert (this=0x880dfc0, t=@0xbfbfd03c)
> at math_cursor.C:349
> #4 0x8227c19 in MathCursor::insert (this=0x880dfc0, c=33 '!')
> at math_cursor.C:367
> #5 0x822e464 in MathCursor::interpret (this=0x880dfc0, c=33 '!')
> at math_cursor.C:1337
[This ^^^ looks like we are still busy inserting '!' and there was no
Cursor-Up yet... anyway.]
> #6 0x821cdd9 in InsetFormulaBase::localDispatch (this=0x8802000,
> cmd=@0xbfbfdfac) at formulabase.C:695
> #7 0x8164421 in LyXFunc::dispatch (this=0x857ba00, ev=@0xbfbfe044,
> verbose=false) at lyxfunc.C:754
> #8 0x816229c in LyXFunc::processKeySym (this=0xbfbfe0fc,
> keysym=0xbfbfe0fc,
> state=shift) at lyxfunc.C:239
> #9 0x805de9a in BufferView::Pimpl::workAreaKeyPress (this=0x860e480,
> key={
> px = 0xbfbfe178, pn = {pi_ = 0x4}}, state=shift)
I don't know anything about the boost stuff, but pi_ is defined as some
pointer. A value of 0x4 simply looks wrong here.
And then I loose track. Maybe one of the boost/signal/frontend gurus can
explain what the next two lines mean. [Btw why are we passing the
LyXKeySymPtr as value? Wouldn't a const ref do as well, given that copying
is more expensive than passing the ref even if it is a shared_ptr?
> at BufferView_pimpl.C:406
> #10 0x806c26e in
> boost::detail::function::void_function_obj_invoker2 :bind_t boost::shared_ptr, key_modifier::state>,
> boost::_bi::list3, boost::arg<1>,
> boost::arg<2> > >, void, boost::shared_ptr,
> key_modifier::state>::invoke (function_obj_ptr={obj_ptr = 0x8609d90,
> const_obj_ptr = 0x8609d90, func_ptr = 0x8609d90}, a0=0xbfbfe3fc,
> a1=shift) at ../boost/boost/bind/mem_fn_template.hpp:213
> #11 0x83ebdcb in boost::signal2,
> key_modifier::state, boost::last_value, int, less,
> boost::function2,
> key_modifier::state, boost::empty_function_policy,
> boost::empty_function_mixin, allocator >
> >::operator() (
> this=0xbfbfeb08, a1=0xbfbfeb08, a2=shift)
> at ../../../boost/boost/function/function_template.hpp:328
It looks ok below this:
> #12 0x83e61d4 in XWorkArea::work_area_handler (ob=0x860fe00, event=9,
> key=33,
> xev=0x285d6280) at XWorkArea.C:503
> #13 0x83e4ad2 in C_XWorkArea_work_area_handler (ob=0x860fe00, event=9,
> key=33,
> xev=0x285d6280) at XWorkArea.C:143
> #14 0x2859fb66 in fl_handle_it (obj=0x860fe00, event=9, mx=286, my=732,
> key=33, xev=0x285d6280) at objects.c:1542
> #15 0x2859fc28 in fl_handle_object (obj=0x860fe00, event=9, mx=286,
> my=732,
> key=33, xev=0x285d6280) at objects.c:1572
> #16 0x2858fb1b in fl_keyboard (form=0x85fc700, key=33, x=286, y=732,
> xev=0x285d6280) at forms.c:1286
> #17 0x2858ff68 in fl_handle_form (form=0x85fc700, event=9, key=33,
Andre'
--
Those who desire to give up Freedom in order to gain Security,
will not have, nor do they deserve, either one. (T. Jefferson)
Re: Figures can't be updated, trashed document
Darren Freeman wrote: > Dear list, > > Now here's one very *serious* bug. > > After inserting a bunch of EPS figures, I found that one had a small > mistake. So I fixed the FIG source with xfig, ran transfig etc. and > created a new EPS file with the same name as the one I had already > inserted. Hi, Darren. The resulting LaTeX run uses the correct version of the figure, so your output should be fine. The bug lies in LyX's rendering of the image on the screen. LyX stores the pixmap using the file name as the identifier. LyX 1.2.X doesn't bother to check if the image changes. The bug has been squashed in CVS and so all should be fine in LyX 1.3. Regards, Angus
Re: [PATCH]: fix compilation for xforms 0.88
Angus Leeming <[EMAIL PROTECTED]> writes: | Angus Leeming wrote: > >> Trivial, tedious patch. Ok to apply? >> >> Angus > | Oops. Just goes to show that it may be trivial but you've still got to get | it right. Attached does the right thing. > | A Yes, ok. -- Lgb
Re: Can't tell if a cross-reference is emphasised
On Mon, Oct 07, 2002 at 04:48:01PM +0930, Darren Freeman wrote:
> I like to put my figure references in bold, as in:
>
> Figure 3
>
> But you can't tell if the cross reference itself is bold. I can imagine
> situations where you only notice days after publication that the number
> wasn't bold.
>
> Is this a missing feature?
You shouldn't be using LyX like this.
The proper way of achieving the above, is to use the PrettyRef reference type
(without writing Figure before the reference, and without marking anything in
bold) and add to the preamble the line
\newrefformat{fig}{\textbf{Figure~\ref{#1}}}
Re: [PATCH]: fix compilation for xforms 0.88
Angus Leeming wrote:
> Trivial, tedious patch. Ok to apply?
>
> Angus
Oops. Just goes to show that it may be trivial but you've still got to get
it right. Attached does the right thing.
A
Index: ChangeLog
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/ChangeLog,v
retrieving revision 1.554
diff -u -p -r1.554 ChangeLog
--- ChangeLog 4 Oct 2002 10:39:01 - 1.554
+++ ChangeLog 7 Oct 2002 10:17:27 -
@@ -1,3 +1,9 @@
+2002-10-07 Angus Leeming <[EMAIL PROTECTED]>
+
+ * FormBase.C (hide):
+ * FormBaseDeprecated.C (hide): wrap that call to fl_hide_tooltip inside
+ #ifdef required xforms version.
+
2002-10-04 Angus Leeming <[EMAIL PROTECTED]>
* form_graphics.fd: un-cripple dialog by reverting yesterday's change.
Index: FormBase.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/FormBase.C,v
retrieving revision 1.61
diff -u -p -r1.61 FormBase.C
--- FormBase.C 3 Oct 2002 11:25:00 - 1.61
+++ FormBase.C 7 Oct 2002 10:17:27 -
@@ -28,8 +28,10 @@
extern "C" {
+#if FL_VERSION < 1 && FL_REVISION < 89
// This should be in forms.h but isn't
void fl_hide_tooltip();
+#endif
// Callback function invoked by xforms when the dialog is closed by the
// window manager
@@ -164,9 +166,11 @@ void FormBase::show()
void FormBase::hide()
{
+#if FL_VERSION < 1 && FL_REVISION < 89
// Does no harm if none is visible and ensures that the tooltip form
// is hidden should the dialog be closed from the keyboard.
fl_hide_tooltip();
+#endif
// xforms sometimes tries to process a hint-type MotionNotify, and
// use XQueryPointer, without verifying if the window still exists.
Index: FormBaseDeprecated.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/FormBaseDeprecated.C,v
retrieving revision 1.35
diff -u -p -r1.35 FormBaseDeprecated.C
--- FormBaseDeprecated.C 3 Oct 2002 11:25:00 - 1.35
+++ FormBaseDeprecated.C 7 Oct 2002 10:17:27 -
@@ -32,8 +32,10 @@
extern "C" {
+#if FL_VERSION > 0 || FL_REVISION >= 89
// This should be in forms.h but isn't
void fl_hide_tooltip();
+#endif
// Callback function invoked by xforms when the dialog is closed by the
// window manager
@@ -173,9 +175,11 @@ void FormBaseDeprecated::show()
void FormBaseDeprecated::hide()
{
+#if FL_VERSION > 0 || FL_REVISION >= 89
// Does no harm if none is visible and ensures that the tooltip form
// is hidden should the dialog be closed from the keyboard.
fl_hide_tooltip();
+#endif
// xforms sometimes tries to process a hint-type MotionNotify, and
// use XQueryPointer, without verifying if the window still exists.
[PATCH]: fix compilation for xforms 0.88
Trivial, tedious patch. Ok to apply?
Angus
Index: ChangeLog
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/ChangeLog,v
retrieving revision 1.554
diff -u -p -r1.554 ChangeLog
--- ChangeLog 4 Oct 2002 10:39:01 - 1.554
+++ ChangeLog 7 Oct 2002 10:13:05 -
@@ -1,3 +1,9 @@
+2002-10-07 Angus Leeming <[EMAIL PROTECTED]>
+
+ * FormBase.C (hide):
+ * FormBaseDeprecated.C (hide): wrap that call to fl_hide_tooltip inside
+ #ifdef required xforms version.
+
2002-10-04 Angus Leeming <[EMAIL PROTECTED]>
* form_graphics.fd: un-cripple dialog by reverting yesterday's change.
Index: FormBase.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/FormBase.C,v
retrieving revision 1.61
diff -u -p -r1.61 FormBase.C
--- FormBase.C 3 Oct 2002 11:25:00 - 1.61
+++ FormBase.C 7 Oct 2002 10:13:05 -
@@ -28,8 +28,10 @@
extern "C" {
+#if FL_VERSION < 1 && FL_REVISION < 89
// This should be in forms.h but isn't
void fl_hide_tooltip();
+#endif
// Callback function invoked by xforms when the dialog is closed by the
// window manager
@@ -164,9 +166,11 @@ void FormBase::show()
void FormBase::hide()
{
+#if FL_VERSION < 1 && FL_REVISION < 89
// Does no harm if none is visible and ensures that the tooltip form
// is hidden should the dialog be closed from the keyboard.
fl_hide_tooltip();
+#endif
// xforms sometimes tries to process a hint-type MotionNotify, and
// use XQueryPointer, without verifying if the window still exists.
Index: FormBaseDeprecated.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/frontends/xforms/FormBaseDeprecated.C,v
retrieving revision 1.35
diff -u -p -r1.35 FormBaseDeprecated.C
--- FormBaseDeprecated.C 3 Oct 2002 11:25:00 - 1.35
+++ FormBaseDeprecated.C 7 Oct 2002 10:13:05 -
@@ -32,8 +32,10 @@
extern "C" {
+#if FL_VERSION < 1 && FL_REVISION < 89
// This should be in forms.h but isn't
void fl_hide_tooltip();
+#endif
// Callback function invoked by xforms when the dialog is closed by the
// window manager
@@ -173,9 +175,11 @@ void FormBaseDeprecated::show()
void FormBaseDeprecated::hide()
{
+#if FL_VERSION < 1 && FL_REVISION < 89
// Does no harm if none is visible and ensures that the tooltip form
// is hidden should the dialog be closed from the keyboard.
fl_hide_tooltip();
+#endif
// xforms sometimes tries to process a hint-type MotionNotify, and
// use XQueryPointer, without verifying if the window still exists.
Re: Help needed: autogen.sh problem - LyX 1.2.2cvs fails with SuSE8.1
[EMAIL PROTECTED] (Lars Gullik Bjønnes) writes: | [EMAIL PROTECTED] writes: > | | "Garst R. Reese" <[EMAIL PROTECTED]> schrieb am 06.10.2002, 18:27:34: >> >>> This comes from all of the unused, unsupported, unnecessary, and buggy, >>> qt, gtk, gnome crud in autogen.sh etc. --- talk to Jean Marc. >> | | I found out that it is sufficient to do the following: >> | | In file aclocal.m4: | | line 8: remove "gtk--.m4", "gnome--.m4", "gnome.m4" >> | | In file configure.in: | | comment out lines 68-73, | | comment out lines 207-228 >> | | With these modifications, I was able to configure and compile - except | | for one thing: Option "--with-included-string" does not work with | | gcc 3.2 due to a compile problem in the math code. Which brings me | | to the following question: Is option "--with-included-string" still | | relevant for gcc 3.2 or just intended for old systems? And how about | | "--with-included-gettext"? > | IMO old systems. But of course --with-included-string won't currently work on old systems either. -- Lgb
Re: configure fails on redhat 8.0
Mate Wierdl <[EMAIL PROTECTED]> writes:
| On Thu, Oct 03, 2002 at 07:46:22PM +0200, Lars Gullik Bj?nnes wrote:
>> Mate Wierdl <[EMAIL PROTECTED]> writes:
>>
>> | On Thu, Oct 03, 2002 at 11:39:02AM +0200, Lars Gullik Bj?nnes wrote:
>> >> Jos? Ab?lio Oliveira Matos <[EMAIL PROTECTED]> writes:
>> >>
>> >> | Hi,
>> >> | configure fails with this message:
>> >>
>> >> and this is not rh-8.0 specific.
>> >
>> | How about this one, after running a plain
>> >
>> | ./configure
>> >
>> | I get
>> >
>> | ...
>> | checking for a good enough C++ compiler... g++
>> | checking for c++... g++
>> | checking whether the C++ compiler (g++ ) works... no
>> | configure: error: installation or configuration problem: C++ compiler cannot
>> | create executables.
>>
>> And you checked what error in config.log that it actally failed on?
>
| configure:3094: checking for c++
| configure:3126: checking whether the C++ compiler (g++ ) works
| configure:3142: g++ -o conftestconftest.C 1>&5
| /usr/bin/ld: cannot find -lgcc_s
| collect2: ld returned 1 exit status
| configure: failed program was:
>
| #line 3137 "configure"
| #include "confdefs.h"
>
| int main(){return(0);}
So it seems that somehting is wrong with your installation.
--
Lgb
Re: Help needed: autogen.sh problem - LyX 1.2.2cvs fails with SuSE8.1
[EMAIL PROTECTED] writes: | "Garst R. Reese" <[EMAIL PROTECTED]> schrieb am 06.10.2002, 18:27:34: > >> This comes from all of the unused, unsupported, unnecessary, and buggy, >> qt, gtk, gnome crud in autogen.sh etc. --- talk to Jean Marc. > | I found out that it is sufficient to do the following: > | In file aclocal.m4: | line 8: remove "gtk--.m4", "gnome--.m4", "gnome.m4" > | In file configure.in: | comment out lines 68-73, | comment out lines 207-228 > | With these modifications, I was able to configure and compile - except | for one thing: Option "--with-included-string" does not work with | gcc 3.2 due to a compile problem in the math code. Which brings me | to the following question: Is option "--with-included-string" still | relevant for gcc 3.2 or just intended for old systems? And how about | "--with-included-gettext"? IMO old systems. -- Lgb
Re: some patch for i18n
Andre Poenitz <[EMAIL PROTECTED]> writes:
| On Sun, Oct 06, 2002 at 11:35:23PM +0400, Vitaly Lipatov wrote:
>> Please add _() for vertical fill. It is not translated now.
>>
>> --- /src/text.C.old 2002-10-05 00:55:04 +040
>> +++ src/text.C 2002-10-06 23:07:09 +0400
>> @@ -3281,7 +3281,7 @@
>> break;
>> }
>> case VSpace:: VFILL:
>> - str = prefix + " (vertical fill)";
>> + str = prefix + _(" (vertical fill)");
>> ty1 = ty2 = start;
>> by1 = by2 = end;
>> break;
>
| Looks sensible.
>
| Ok to apply?
Yes.
--
Lgb
Re: LyX allows space followed by guarded space
On Mon, Oct 07, 2002 at 04:49:55PM +0930, Darren Freeman wrote: > It is possible, when converting a regular space to a guarded space, that > you forget to remove the space. Result is space followed by guarded space. > Easy to miss. > > Does this conflict with the WYSIWYM philosophy? The existence of "guarded spaces". Sort of... yes. But philosophy is one side, the other is practicability. Sometimes they are "just needed". > Or is it legal to put all > sorts of extra spaces in if they are guarded? Legal, yes. But not nice. You should avoid them whenever possible. Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: LyX allows space followed by guarded space
> "Darren" == Darren Freeman <[EMAIL PROTECTED]> writes: Darren> Dear list, I don't know if this is a bug or not, I'm new. Darren> It is possible, when converting a regular space to a guarded Darren> space, that you forget to remove the space. Result is space Darren> followed by guarded space. Easy to miss. Darren> Does this conflict with the WYSIWYM philosophy? Or is it legal Darren> to put all sorts of extra spaces in if they are guarded? I think that trying to enforce that would create more trouble than it is worth. An alternative would be to make spaces optionally visible. JMarc
Re: Graphics: LyX and LaTeX scales have resolution of 1 %.
Andre Poenitz wrote: > On Mon, Oct 07, 2002 at 05:21:26PM +0900, Rob Lahaye wrote: > >>But an easy one (I believe). >>I'll try to add it in my Xforms patch here, which is going to move into >>CVS after 1.3 release. > > > How intrusive are the changes in your "big patch" btw? See below, for a rough impression. Do you want me to send you the patch file itself? >>I'm not so sure to make LyX View scaling also as a float. The 1 percent >>resolution >>for LyX View scaling at present is probably more than enough, isn't it? > > > Well... on a 1024 screen 1% is 10 pixels... ok. Personally I don't care > for the display on screen. Print output is more important Agreed. So I'll go for the LaTeX/output float-scaling only. Cheers, Rob. New file: src/frontends/xforms/checkedwidgets.C New file: src/frontends/xforms/checkedwidgets.h controllers/ButtonController.tmpl | 13 controllers/ButtonControllerBase.C | 35 ++ controllers/ButtonControllerBase.h | 26 + controllers/ControlDialog.tmpl |6 controllers/ControlInset.tmpl |6 controllers/ControlParagraph.C | 13 controllers/ControlParagraph.h |6 controllers/ControlSpellchecker.h |4 xforms/FeedbackController.C|6 xforms/FormAboutlyx.C | 52 +-- xforms/FormAboutlyx.h |4 xforms/FormBibtex.C| 108 +++ xforms/FormBrowser.C |5 xforms/FormCitation.C | 21 - xforms/FormDocument.C | 564 +++-- xforms/FormFloat.C | 240 --- xforms/FormFloat.h |4 xforms/FormGraphics.C | 194 ++-- xforms/FormGraphics.h |1 xforms/FormLog.C | 26 - xforms/FormMathsPanel.C|2 xforms/FormParagraph.C | 489 xforms/FormParagraph.h |4 xforms/FormPrint.C | 141 - xforms/FormPrint.h |5 xforms/FormSearch.C| 36 +- xforms/FormShowFile.C |7 xforms/FormSpellchecker.C | 192 +++- xforms/FormSpellchecker.h |8 xforms/FormTabular.C | 46 +-- xforms/FormWrap.C | 99 ++ xforms/FormWrap.h |4 xforms/Makefile.am |2 xforms/RadioButtonGroup.C | 37 +- xforms/RadioButtonGroup.h | 21 - xforms/xforms_helpers.C| 10 xforms/forms/form_aboutlyx.fd | 149 +++-- xforms/forms/form_bibtex.fd| 139 + xforms/forms/form_browser.fd | 21 - xforms/forms/form_citation.fd | 30 - xforms/forms/form_document.fd | 456 ++--- xforms/forms/form_float.fd | 190 xforms/forms/form_graphics.fd | 71 +--- xforms/forms/form_maths_panel.fd | 67 ++-- xforms/forms/form_paragraph.fd | 485 +-- xforms/forms/form_print.fd | 16 - xforms/forms/form_ref.fd | 32 +- xforms/forms/form_search.fd| 117 +++ xforms/forms/form_spellchecker.fd | 79 +++-- xforms/forms/form_texinfo.fd |4 xforms/forms/form_wrap.fd | 113 ++- 51 files changed, 2268 insertions(+), 2138 deletions(-)
Re: Figures can't be updated, trashed document
On Mon, Oct 07, 2002 at 05:00:04PM +0930, Darren Freeman wrote: > When I open the file with LyX, it freezes about 20% of the way down the > document. As in, you scroll down and after hitting a certain part it > freezes LyX. What do top or ps aux say about the status of the lyx process? > And NO I can't post the document it's confidential. It's free software, not > free document =) Would you manage somehow to cut down the doc to the minimum producing the lock-up? And is it then postable? > Have fun, > Darren Freeman > Martin msg4/pgp0.pgp Description: PGP signature
Re: Graphics: LyX and LaTeX scales have resolution of 1 %.
On Mon, Oct 07, 2002 at 05:21:26PM +0900, Rob Lahaye wrote: > But an easy one (I believe). > I'll try to add it in my Xforms patch here, which is going to move into > CVS after 1.3 release. How intrusive are the changes in your "big patch" btw? > Your 'visibility' criterion applies to the final LaTeX output, right? Yes. > I think it's easy to add a float-scaling here. > > I'm not so sure to make LyX View scaling also as a float. The 1 percent > resolution > for LyX View scaling at present is probably more than enough, isn't it? Well... on a 1024 screen 1% is 10 pixels... ok. Personally I don't care for the display on screen. Print output is more important Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: Graphics: LyX and LaTeX scales have resolution of 1 %.
Andre Poenitz wrote: > On Mon, Oct 07, 2002 at 08:39:23AM +0900, Rob Lahaye wrote: > >>The way the LyX and LaTeX scales are dealt with in the graphics code, >>implies that the scale variable is stored as an integer from 1 to 100. >>Fractional scales are not possible. Is that OK, or is it too restrictive? > > > 1% of page width is about 0.2 mm which _is_ visible. > > Even if it were not, I'd like to be able to leave things untouched when > imported/exported so yes, "82.47%" should be possible. Not an urgent item, > though... But an easy one (I believe). I'll try to add it in my Xforms patch here, which is going to move into CVS after 1.3 release. Your 'visibility' criterion applies to the final LaTeX output, right? I think it's easy to add a float-scaling here. I'm not so sure to make LyX View scaling also as a float. The 1 percent resolution for LyX View scaling at present is probably more than enough, isn't it? Regards, Rob.
Re: Failed to import MS Word document
On Mon, Oct 07, 2002 at 01:14:45AM +0930, Darren Freeman wrote:
> Dear list,
>
> I am a newbie to LyX, I have only just finished the tutorial.
>
> I tried importing a Word file and failed with the following cut-and-paste
> from the console:
>
> --8<-8<--8<--
>
> reLyX directory is: /usr/share/lyx/reLyX
> reLyX, the LaTeX to LyX translator. Revision date 2001/08/31
> Reading LaTeX command syntax
> (124-228199735.tex: Splitting Preamble
> Creating LyX preamble
> Reading layout file
> Cleaning... Expecting `\end{itemize}', got \end{enumerate} in `
> ' at /usr/share/lyx/reLyX/Text/TeX.pm line 402, line 62.
> Translating... Expecting `\end{itemize}', got \end{enumerate} in `
> ' at /usr/share/lyx/reLyX/Text/TeX.pm line 402, line 62.
> Problem with Layout Stack!
> Compilation failed in require at /usr//bin/reLyX line 76.
> Exited due to fatal Error!
>
> --8<-8<--8<--
>
> Yes, the file really was called 124-228199735.doc as it is an academic
> paper submission I wish to polish off using LyX (my supervisor LOVES LyX!).
>
> System info:
> Fresh install of Mandrake 9.0 - barely changed anything
> LyX version 1.2.1
> Athlon, KDE 3
>
> Word file:
> Saved with MS Word 2000, containing PNG images and around 10,000 words of text.
>
> I hope this hasn't been reported before as I am a bit of a n00b =)
>
> Have fun,
> Darren Freeman
This is due to the use of wvWare before reLyX -- I think it produces
a wrong LaTeX file. You could try to use wvCleanLatex (in the lib
directory) manually, and look at the produced LaTeX file. And fix it
manually if needed. (Alternatively, the intermediate LaTeX file should
be somewhere in your /tmp: 124-228199735.tex I suppose.)
Martin
msg45552/pgp0.pgp
Description: PGP signature
Re: Feature request: default units
> "Darren" == Darren Freeman <[EMAIL PROTECTED]> writes: Darren> Dear list, I don't know if this feature is really missing or Darren> if I'm just not looking hard enough. Darren> I want to be able to select my default units to be cm rather Darren> than in. Currently I have to click on "in" every time I change Darren> a figure width. This is indeed a problem... I think there is some clever code somewhere that says that if your paper size is 'metric' like a4, the unit will be cm. This is of course not good enough. Note however that you can directly type "10cm" and it will ignore the unit box. Feel free to file an enhancement request on bugzilla. JMarc
Re: Graphics: LyX and LaTeX scales have resolution of 1 %.
On Mon, Oct 07, 2002 at 08:39:23AM +0900, Rob Lahaye wrote: > The way the LyX and LaTeX scales are dealt with in the graphics code, > implies that the scale variable is stored as an integer from 1 to 100. > Fractional scales are not possible. Is that OK, or is it too restrictive? 1% of page width is about 0.2 mm which _is_ visible. Even if it were not, I'd like to be able to leave things untouched when imported/exported so yes, "82.47%" should be possible. Not an urgent item, though... Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: Menu system pauses for a long time
> "Darren" == Darren Freeman <[EMAIL PROTECTED]> writes: Darren> Dear list, I started seriously using Lyx last night, I'll give Darren> one post per issue that I'm raising. Sorry if any are dupes Darren> but there's more than I care to look for on Bugzilla =) Darren> When you start browsing a menu, say File, then decide to Darren> change to a new menu, say Insert, there is an incredible delay Darren> before updating to the new menu. Usually it just ignores you Darren> and you have to click back on the document and try again. This is a bug in xforms and I do not know what to do about it. The File menu will only close if you click at least in the second half of the popup. So if you click on Edit, it is too close and does not work. This is a deliberate action of xforms and no I do not know why this was done. JMarc
Re: some patch for i18n
On Sun, Oct 06, 2002 at 11:35:23PM +0400, Vitaly Lipatov wrote:
> Please add _() for vertical fill. It is not translated now.
>
> --- /src/text.C.old 2002-10-05 00:55:04 +040
> +++ src/text.C 2002-10-06 23:07:09 +0400
> @@ -3281,7 +3281,7 @@
> break;
> }
> case VSpace:: VFILL:
> - str = prefix + " (vertical fill)";
> + str = prefix + _(" (vertical fill)");
> ty1 = ty2 = start;
> by1 = by2 = end;
> break;
Looks sensible.
Ok to apply?
Andre'
--
Those who desire to give up Freedom in order to gain Security,
will not have, nor do they deserve, either one. (T. Jefferson)
Re: Help needed: autogen.sh problem - LyX 1.2.2cvs fails with SuSE 8.1
> "Garst" == Garst R Reese <[EMAIL PROTECTED]> writes: Garst> This comes from all of the unused, unsupported, unnecessary, Garst> and buggy, qt, gtk, gnome crud in autogen.sh etc. --- talk to Garst> Jean Marc. This comes mostly from the fact that the configure scripts of 1.2.x have not been ported to autoconf 2.5x and that they will proobably not be. I do not care a lot about it since lyx is distributed with already working configure scripts. I will nevertheless take a look at the changes that have been proposed and if they look clean enough, I'll try to do it. However, this will be after 1.2.2, methinks. Garst> "Trying to remove it is bound to introduce new bugs. We have Garst> no time for that... See what happened with 1.2.1. There have been so many 'fixes' there' that we have at least 3 bad regressions from 1.2.0. This is not acceptable. JMarc
Figures can't be updated, trashed document
Dear list, Now here's one very *serious* bug. After inserting a bunch of EPS figures, I found that one had a small mistake. So I fixed the FIG source with xfig, ran transfig etc. and created a new EPS file with the same name as the one I had already inserted. Only there was seemingly no way to make LyX import the changes. I couldn't delete the figure and reinsert it, no no no, I got the same mistake even though it was seemingly banished from the disc. I know it was fixed because I checked the EPS. I could import the FIG file and it was definitely fixed, but the EPS was a worry. It's as if the figure is stored in the LyX file even after deleting the figure. It gets worse. After really screwing around with adding figures in different ways to get this figure in, I gave up and just left it with the mistake in the figure. Then saved the file and closed LyX. When I open the file with LyX, it freezes about 20% of the way down the document. As in, you scroll down and after hitting a certain part it freezes LyX. I'm so seriously screwed if I don't figure that out. I didn't do much to the text, but I had managed to import the text from Word and fix it up. All I had accomplished in 5 hours was to import a Word file as TXT into LyX and then reformat it. Now I may not even have that. I don't know that the crash is caused by the figure drama but it's suspicious isn't it? Maybe two internal figures with the same name, and LyX is locking? Anyway I'm going to give LyX another go, but there's no longer any hope if finishing this document by the due date. I really wanted to use LyX since my supervisor and coworker both love it. Stats: LyX 1.2.1 Fresh install of Mandrake 9.0 Athlon And NO I can't post the document it's confidential. It's free software, not free document =) Have fun, Darren Freeman
