On 9/11/05 2:36 PM, "Helge Hafting" <[EMAIL PROTECTED]> wrote:
> I remember cvs went down a few weeks ago.
>
> There seems to be lot of development going on now, biut I still get
> this:
> $ cvs up
> /cvs/lyx: no such repository
>
> Should I change something? I couldn't find any new information
Could this patch be applied to development/win32/lyx.vcproj and
committed, please? This updates the MSVC build per Michael's recent
change.
Thanks
Rob
Index: ./development/win32/lyx.vcproj
===
RCS file: /cvs/lyx/lyx-devel/development/
Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes:
> to get sort?
>
> can you try to include instead.
>
> #include
>
Yes, that works.
Thanks
Rob
I get this:
ControlTexinfo.C
\lyx\lyx-devel\src\frontends\controllers\ControlTexinfo.C(61) : error
C2039: 'sort' : is not a member of 'std'
\lyx\lyx-devel\src\frontends\controllers\ControlTexinfo.C(61) : error
C3861: 'sort': identifier not found, even with argument-dependent lookup
I had to do th
The files src/insets/updatableinset.Ch were recently removed from the
build. Could this patch to development/win32/lyx.vcproj please be
applied?
Thanks
Rob
lyx.vcproj.diff
Description: lyx.vcproj.diff
Angus Leeming <[EMAIL PROTECTED]> writes:
>
> Michael Schmitt wrote:
> > Has anybody ever tried to compile LyX 1.4 on Windows?
>
> Asger ported LyX 1.4 to MSVC and Visual Studio back in January. Rob Bearman
> uses it to compile the whole tree in about 15 minutes. I u
> Asger Ottar Alstrup wrote:
> > Angus Leeming wrote:
> >
> >> My reading the WinMain docs at
> >>
> http://msdn.microsoft.com/library/en-us/winui/winui/windowsuse
rinterface/windowing/windows/windowreference/windowfunctions/winmain.asp
> >>
> >> tells me that it should...
> >>
> >> Any clues?
> >> "Bennett" == Bennett Helm <[EMAIL PROTECTED]> writes:
> >
> > Bennett> From what I can tell, it comes down to this. Rob is using
> > Bennett> Mac OS X 10.4 (=Darwin 8.0.0), whereas I use
> 10.3.9 (= Darwin
> > Bennett> 7.9.0).
> >
> > Do you use the same Qt version? Is it compiled in the
On 6/10/05 12:32 PM, "Bennett Helm" <[EMAIL PROTECTED]> wrote:
> I have a 512MB iMac G4 (1.25GHz); it's surely not my hardware that's
> the issue.
>
> Perhaps it's the configuration. Here's what I've got (compiled with
> gcc-3.3):
>
> ./configure --with-frontend=qt --without-x
> --prefix=/Applica
On 6/10/05 9:36 AM, "Bennett Helm" <[EMAIL PROTECTED]> wrote:
> But, frankly, the difference in speed introduced with the fontcache
> patch is trivial compared to the slowness of lyx-140 overall: it really
> is unusable on the Mac, and something else must account for the
> slowness.
>
> I'd love
t; Probably, but let's see if this code compiles with MSVC
> first. Rob Bearman
> has been pretty quick to tell us of problems in the past.
>
> --
> Angus
>
>
MSVC has no complaints.
Thanks
Rob
On 5/31/05 12:43 PM, "Angus Leeming" <[EMAIL PROTECTED]> wrote:
> Rob Bearman wrote:
>> Does anyone else see this? Fresh CVS check out/build. OS X 10.4. I've built
>> successfully on Tiger before.
>
> Nope. But different Qt versions may behave differentl
Does anyone else see this? Fresh CVS check out/build. OS X 10.4. I've built
successfully on Tiger before.
g++ -DHAVE_CONFIG_H -I. -I. -I../../../src -DQT_CLEAN_NAMESPACE
-DQT_GENUINE_STR -I../../../src -I../../../src/frontends -I../../../images
-I/Users/robbear/dev/qt-mac-free-3.3.4/include -I../
On 5/22/05 12:43 PM, "Angus Leeming" <[EMAIL PROTECTED]> wrote:
> Next problem. The lyx shortcut and lyx entry in the start menu are present
> only for the user who installed lyx. (In this case the administrator.) How
> do I get the installer to do so for all users? I guess that this is another
>
On 5/18/05 12:20 AM, "Georg Baum" <[EMAIL PROTECTED]> wrote:
> Rob Bearman wrote:
>
>> The problem seems to be tex2lyx's insertion of "\begin_layout Standard"
>> before the enumeration inside the box inset.
>
> Yes. The attached fix is goi
1. Load test.lyx which has a box inset with an enumeration inside it
2. Export it to LaTex (test.tex)
3. Run tex2lyx on test.tex, creating tex2lyx_output.lyx
When tex2lyx_output.lyx is loaded in LyX, the box inset starts with a
blank line which disappears once the cursor is moved into the inset.
> QSettings gives us e.g. (transparent) access to the Windows registry,
> where all the Window native programs put there configuration stuff and
> "everybody" expects it. When not using Qt we've a choice of either
> programming this access ourselves or remain some kind of sevond class
> citizen by
On 5/16/05 1:36 AM, "Angus Leeming" <[EMAIL PROTECTED]> wrote:
> Rob Bearman wrote:
>>> p.s. Angus, I noticed, that you added some patches to compile LyX with
>>> MSVC. Could you post an instruction for the compilation of LyX 1.4CVS
>>> with MSVC.
On 5/15/05 11:59 AM, "Uwe Stöhr" <[EMAIL PROTECTED]> wrote:
> p.s. Angus, I noticed, that you added some patches to compile LyX with
> MSVC. Could you post an instruction for the compilation of LyX 1.4CVS
> with MSVC.
Instructions for compiling with Visual Studio.NET 2003 exist in
src/development
> Btw, I keep getting errors wrt some missing boost header
> (something with
> detaisl/msvc/while.hpp or such in the path) when I try to compile LyX
> with MSVC.
>
> Have you ever seen such problem?
>
> Andre'
No, I never have. I've been building successfully on three separate
machines.
Thanks
In case this is of concern...
\lyx\lyx-devel\src\mover.C(82) : warning C4172: returning address of
local variable or temporary
Mover const & Movers::operator()(string const & fmt) const
{
SpecialsMap::const_iterator const it = specials_.find(fmt);
return (it == specials_.end()) ?
Launch LyX, create new file, insert box inset (or perhaps any
collapsable inset), click on button to collapse, click on it again. This
crashes LyX. A portion of the Windows stack trace included. I see this
on both Windows and Mac.
lyx.exe!lyx::support::abort() Line 23 C++
lyx.exe
> Multiply defined symbols? Just remove lengthvalidator.*
>
> --
> Angus
>
Could the attached patch to development/win32/lyx.vcproj be applied,
please? This takes care of the lengthvalidator/validator changes for the
MSVC build.
Thanks
Rob
lyx.vcproj.diff
Description: lyx.vcproj.diff
> Rob Bearman wrote:
>
>> In other words, "noclobber" is the same as "bFailIfExists." When
>> noclobber==FALSE, you want the copy to overwrite. Instead, the current
>> code fails the copy.
>
> Good spot! I take it LyX chewed up something it s
In src/support/fs_extras.C, shouldn't it be thus:
void copy_file(path const & source, path const & target, bool noclobber)
{
#ifdef BOOST_POSIX
...
#endif
#ifdef BOOST_WINDOWS
- if (::CopyFile(source.string().c_str(), target.string().c_str(),
!noclobber) == 0) {
+ if (::CopyFile(sourc
> Does this work for you? (Everything is, of course, fine here.)
>
> If so, please post your local changes to config.h and I'll update
> development/Win32 accordingly. (I'll remove win32_kludge.diff also.)
>
> --
> Angus
>
Here's the patch to config.h. Everything builds fine, and it's nice to
> I propose to modify configure.ac as attached so that running
> autogen.sh produces the also attached changes to the generated
> src/config.h.in. You don't actually run autogen.sh at all of
> course, but I
> anticipate that you'd use these config.h.in changes as a
> template with
> which to
> And is this definition telling the truth? Ie, do you have the mktemp
> function as well as the _mktemp function?
I should note that for filetools.C, I must prepend the underscore on
popen and pclose. That might not have been clear since I stated that the
existing kludge isn't necessary.
Thanks
> And is this definition telling the truth? Ie, do you have the mktemp
> function as well as the _mktemp function?
Yes.
> Rob,
>
> same idea but this time for tempname.C. It appears from the
> docs that we
> just need a few underscores. Does the attached patch enable
> you to compile?
>
> mktemp:
> http://msdn.microsoft.com/library/default.asp?url=/library/en-
> us/vclib/html/_crt__mktemp.2c_._wmktemp.asp
> Equ
> Rob,
>
> it seems that popen et al. are all documented members of the
> Windows API,
> so I fail to see why you need to disable (win32_kludge.diff)
> RunCommand in
> src/support/filetools.C. The one possible problem seems to be
> that popen
> and pclose are documented as having a _ prefix.
Based on yesterday's changes, could this patch be committed please? It
contains the updates to config.h and lyx.vcproj in the development\win32
directory.
Thanks
Rob
msvc.diff
Description: msvc.diff
> Given that we have:
> src/frontends/qt2/QLyXKeySym.C:
> typedef map EncodingMap;
> EncodingMap encoding_map;
>
> The change below is obviously an improvement, even though
> there's no reason
> why a sane compiler shouldn't compile the original too.
>
> Nonetheless, can I ask you to con
> Rob, what happens if you define CXX_GLOBAL_CSTD in your config.h ?
> Does it mean you no longer need these two kludges, or does it
> break things
> elsewhere?
>
> Index: src/DepTable.C
> ===
> RCS file: /usr/local/lyx/cvsroot/lyx-d
> Anyway, I read it that win32_kludge.diff should now be something like
> the attached. Correct?
>
> Angus
>
>
Yes, that's just what I have.
Rob
> Please add these (with appropriate define/undef) to your config.h:
>
> /* Define to 1 if you have the header file. */
> /* #undef HAVE_SYS_UTIME_H */
> /* Define to 1 if you have the header file. */
> #define HAVE_UTIME_H 1
>
> and check that the attached rob.diff does the job. (It may be
>
> > > Looking at the other bits of win32_kludge.diff, can you
> > > confirm that 'time'
> > > and 'isalpha' are not in namespace std?
> > >
> >
> > Correct.
>
> I goofed. 'isalpha' *is* in namespace std.
>
Argh. The compiler doesn't agree with the docs and says isalpha is not a
member of std.
> > Looking at the other bits of win32_kludge.diff, can you
> > confirm that 'time'
> > and 'isalpha' are not in namespace std?
> >
>
> Correct.
I goofed. 'isalpha' *is* in namespace std.
> I think that you should modify your config.h so that
> HAVE_SYS_TIME_H is not
> defined...
Ok...
>
> --- src/frontends/LyXView.C 2005/04/26 10:30:22 1.52
> +++ src/frontends/LyXView.C 2005/04/26 15:39:37
> @@ -39,8 +39,10 @@
>
> #include
>
> +#ifndef _WIN32
> #ifdef HAVE_SY
> How interesting! HAVE_SELECT and HAVE_SOCKET are definitely
> not defined
> using mingw/minsys.
>
> HAVE_DUP2 is defined on mingw/minsys (my very bad), so we really need
> another test to prevent USE_ISPELL from being defined. I'll
> think on it.
>
> Looking at your error messages below, I'm
> Looking at your error messages below, I'm frankly surprised
> by things going
> wrong in lyxserver.C Why are you compiling code invoking
> mkfifo and fcntl.
> Is HAVE_MKFIFO defined?
I'm only using what was defined in the original commit of config.h (I've
always depended on the kindness of str
In case you're interested in my config.h and if it helps resolve any
confusion from my previous message, I'm including that here for your
reference.
Thanks
Rob
config.h
Description: config.h
> the changes I've just committed will probably break your
> build. To help you
> resurrect things, I attach the current config.h.in from which
> config.h is
> generated. I also attach the changes from the previous
> config.h.in. The
> new macros HAVE_DUP2, HAVE_SELECT, HAVE_SOCKET and USE_ISPEL
> > Yes, that's what I'm using
> > (http://kde-cygwin.sourceforge.net/qt3-win32/compile-net.php from
> > Asger's notes in the development\win32\readme.txt). As far as I can
> > tell, I've got the latest of that branch, anyway - a cvs
> diff shows no
> > changes. Should I be using something else?
>
> If you can report these problems on the mailing list
> (mirrored on gmane at
> gmane.comp.kde.devel.cygwin), that'd be great! Especially if you can
> create a small standalone program demonstrating the problem.
>
> --
> Angus
It's quite possible that I'm mistaken about having the latest CVS o
> Rob, you are using Qt/Windows Free Edition, aren't you? I am
> quite sure
> that the problem in within the Qt library, not within LyX.
> However, the
> Qt/Win Free guys have done a good job in fixing memory leaks
> and other
> bugs (e.g., the library used to cause 100% CPU usage after a few
Bring up LyX, create a new document, then do nothing but watch the Mem
Usage column under the Windows task manager. LyX gobbles up 4k every
five seconds or so. I've left LyX running unattended and wondered why
the thing always seems to crash when I come back to it. This probably
explains it.
Sorry
On Windows, I noticed that menu invocation results in the appearance of
a new blank taskbar button which goes away as soon as the menu is
released. Hierarchical menus result in even more blank taskbar buttons,
again going away once the submenu is released. This does not happen with
Ruurd's version.
> Many thanks for bearing with me through this, Rob. I hope
> that the next
> commit will remove ispell.C, lyxserver.C, lyxsocket.C,
> client/client.C,
> support/os_win32.C and support/socktools.C from the kludge.diff file
> entirely.
If we could make the kludge file go away altogether, I'd b
> CVSROOT: /usr/local/lyx/cvsroot
> Module name: lyx-devel
> Repository: lyx-devel/src/support/
> Changes by: [EMAIL PROTECTED] 05/04/26 12:30:24
>
> Modified files:
> lyx-devel/src/: Bidi.C Bidi.h ChangeLog coordcache.C
> coordcache.h ispell.C lyxserve
> Modified files:
> lyx-devel/src/support/: ChangeLog userinfo.C
> lyx-devel/development/Win32/: win32_kludge.diff
>
> Log message:
> Add Windows-specific code to userinfo.C.
Under MSVC, the userinfo.C compile gives a PASCAL macro redefinition
warning, the conflict being betwe
> Ok, then could you try the attached patch please.
>
> --
> Angus
>
Yes, this works great. Thank you.
Rob
> > As for the second question, I'm able to define an explicit
> > converter as you specified and the graphic loads correctly. I
> > removed the convertDefault.sh file to make sure I wasn't being
> > fooled that that was being run. Reinstating it, though, and then
> > removing the explicit converte
> You mentioned earlier that you couldn't run LyX from the console? I
> wonder why not. Here Start->Run "command", followed by
> /lyx -dbg graphics,file > lyx.log
> works perfectly.
What I meant to say is that I can't get debug output displayed to the
console, or anywhere else for that matter
[PATCH 13x, 14x] Windows packaging fix
>
> This is just a head's up of a fix verified by (prompted by too :) Rob
> Bearman.
>
> Windows can pass the name of the lyx binary on the command
> line as either
> "lyx" or "lyx.exe". The former way can bre
> Ahhh. Thanks, Jean-Marc. I see that we already have the test
> and that Rob
> merely needs me to modify development/Win32/config.h.
>
> Looks Ok to commit?
>
> --
> Angus
This built fine.
Thanks
Rob
> Incidentally, do graphics now load asynchronously onto the
> LyX screen for
> you? Ie, given an eps file can you continue doing stuff
> whilst the graphic
> inset cyles throgh "loading", "scaling", etc?
>
> Also, can you define a graphics converter explicitly? Ie,
> rather than let
> lyx fall
>
> Modified files:
> lyx-devel/src/support/: ChangeLog forkedcall.C forkedcall.h
>
> Log message:
> MSVC build fixes.
>
> Patches:
> http://www.lyx.org/cgi-bin/viewcvs.cgi/lyx-devel/src/support/C
> hangeLog?r1=1.336&r2=1.337
> http://www.lyx.org/cgi-bin/viewcvs.cgi/lyx-devel/src/su
Pardon my obvious ignorance. I'm reading the wiki now on loading images
and I'll give better feedback after I've installed the appropriate
tools.
Thanks
Rob
> Incidentally, do graphics now load asynchronously onto the
> LyX screen for
> you? Ie, given an eps file can you continue doing stuff
> whilst the graphic
> inset cyles throgh "loading", "scaling", etc?
>
> Also, can you define a graphics converter explicitly? Ie,
> rather than let
> lyx fall
> I think that this one is fine, although it should become
>
> #if defined (_WIN32) && defined(_MSC_VER)
Yes, this continues to work for me.
> Well, given that the idea is eventually to remove this file entirely,
> perhaps you'll first see whether this cures the problem:
>
> - int const t
With the recent changes to forkedcontr.h and forkedcall.*, I found that
I needed to reinstate the typedef shown below to forkedcall.h and
forkedcontr.h:
#ifdef _WIN32
typedef int pid_t;
#endif
I also needed to reinstate the #include "os_win32.h" to forkedcall.C,
which I think is necessary to hand
I've modified the prebuilt development/win32/package.C file based on the
recent changes to package.C.in. I'd appeciate it if this could be
committed.
Thanks
Rob
package.C.diff
Description: package.C.diff
> This will probably mean that Windows users (is that just Michael
> Schmitt?) will suffer from some temporary breakage of the
> Windows port
I am also building 1.4 for Windows regularly, but only with the Visual
Studio build process, fyi. By the way, will these patches eliminate the
need to ap
Does it make sense to add the --disable-concept-checks flag to the notes
in the README.MacOSX file? As far as I can tell from earlier threads
this was the solution to get the OS X build working and that information
might be helpful for new builders to come.
Thanks
Rob
> I can't comment on 1.4's behaviour as I'm reading this from a Windows
> box that only has 1.3 installed. Nonethless, why wouldn't you type:
>
> Insert->Math->Inline Formula
> 1\underline2>right arrow>3
>
> ?
>
> Angus
Most likely out of LyX newbie ignorance, alas. Nonetheless, your
suggested
In 1.3.5 I can do the following to get a 3-digit number with the middle
digit underlined in math mode (eg, 1_2_3)
1. Insert->Math->Inline Formula
2. Type a number
3. Insert->TeX
4. Type "underline" (without the quotes)
5. Type a number
6. Press right-arrow
7. Type a number
With the current 1.4 CV
> > It should look like C:\Desktop\UserGuide.lyx, I believe.
> >
> > Probably the ~/ part of MakeDisplayPath should only be done on Unix.
> > On windows, I guess there is some API to turn get those shorter file
> > names.
>
> Mmmm, but this is not too important either way. Perhaps Asger
> might
Submitted in case this is useful for anyone else. I've attached two
.vcproj files which can be added as existing projects to lyx.sln in
order to build tex2lyx.exe after lyx.exe is built under Visual Studio.
Use the following steps:
1. Copy tex2lyx.vcproj and tex2lyx_build.vcproj to development/win
> Ok, I did it the other way round. Try again.
>
> --
> Angus
>
Yes, it's fine with the change made in context.h instead.
Thanks
Rob
> I'm guessing this class/struct issue with Context might be
> the problem?
>
To answer my own question, yes, that's exactly the problem. Changing the
"class Context;" line in tex2lyx.h to "struct Context;" eliminates the
unresolved externals, the compiler warnings, and allows tex2lyx.exe to
bui
> >> What happens if you explicitly convert the FLAG_ITEM terms in
> >> table.C's
> >> invocation of parse_text_in_inset to unsigned?
> >>
> >> parse_text_in_inset(p, os, unsigned(FLAG_ITEM), false,
> context);
> >
> > The change didn't have any effect on the link error.
>
> Just for fun, w
> What happens if you explicitly convert the FLAG_ITEM terms in
> table.C's
> invocation of parse_text_in_inset to unsigned?
>
> parse_text_in_inset(p, os, unsigned(FLAG_ITEM), false, context);
The change didn't have any effect on the link error.
Thanks
Rob
I've been working on building a .vcproj file for tex2lyx and am down to
two unresolved externals before the thing works. The problem has to do
with the handle_tabular function in table.C, and the
parse_text_in_insert function in text.C. I was wondering if anyone had
any insight into this.
Note tha
> If you send the patch to me again, I'll apply it. Sorry for the delay.
>
> JMarc
>
Thank you! Here they are again, including my comments:
win32_kludge.diff - This is the fixed
development\win32\win32_kludge.diff file (a replacement for the existing
file, not a diff of the diff file).
msvc.di
> > Asking me nicely should suffice.
> >
> > --
> > Lgb
>
> That's easy enough. Please grant me access to the
> lyx-devel\development\win32 directory and refer me to any
> commit guideline docs that I should read.
>
> Thanks
> Rob
Apologies if I'm unclear on protocol as I haven't heard b
> Asking me nicely should suffice.
>
> --
> Lgb
That's easy enough. Please grant me access to the lyx-devel\development\win32
directory and refer me to any commit guideline docs that I should read.
Thanks
Rob
> Maybe you could get karma to commit yourself to the
> lyx-devel\development\win32 directory?
>
> Regards,
> Asger
>
If that's helpful, sure. Is there a karma application I should fill out?
Thanks
Rob
> Here's a summary of what needs to be done to fix the vcproj build:
>
> development\win32\package.C:
> Change GetEnv to getEnv and/or update from package.C.in
>
> development\win32\lyx.vcproj
> Add src\support\environment.C
> Remove src\support\putenv.C
>
> src\lyxserver.C (win32_kludge.diff)
>
Here's a summary of what needs to be done to fix the vcproj build:
development\win32\package.C:
Change GetEnv to getEnv and/or update from package.C.in
development\win32\lyx.vcproj
Add src\support\environment.C
Remove src\support\putenv.C
src\lyxserver.C (win32_kludge.diff)
Fix by updating broke
> A fixed VS lyxserver.C is attached. When I find the time,
> I'll update the
> patch.
>
> Regards,
> Asger
>
Thank you. Also, I think the vcproj file will need to be updated to
remove putenv.C and add environment.C. Unfortunately, environment.C
doesn't build under MSVC:
Compiling...
environm
> CVSROOT: /usr/local/lyx/cvsroot
> Module name: lyx-devel
> Repository: lyx-devel/src/
> Changes by: [EMAIL PROTECTED] 05/02/15 12:04:40
>
> Modified files:
> lyx-devel/src/: ChangeLog lyxserver.C
>
> Log message:
> Squash MSVC warning "local variable 'fd' used witho
> I am interested to a certain degree. However, it would help
> me a lot to
> have a quick look if the attachments were not declared as octet-stream
> but as, say, application/x-bzip or similar.
>
> Andre'
>
Apologies for the delay. I guess my mailer defaults to octet-stream for
types it doesn'
> Thank *you*.
>
> --
> Angus
>
>
I'll return the thanks again if someone can give me a pointer on how to
set up the LyX installation tree based on a vcproj build. The
win32\readme.txt file hints that this information was given previously
somewhere in the January list, but I can't find it beyo
I was able to complete a fresh vcproj build for the first time after the
latest commits. Thank you for the fixes. I've attached the build log in
case anyone's interested in the 30 warnings.
Thanks
Rob
buildlog.htm.bz2
Description: buildlog.htm.bz2
> Try to build a release build, and see if that helps. If not, you can
> manually inspect the files in the moc_files folders to see if their
> custom build step is correct.
No difference with the release build. Looking at the "mocinput" folder
of
the vcproj file, it's clear there's a bunch of m
> and is there any heads up short of my updating my CVS tree and
watching for it?
Never mind this particular question. I found the lyx-cvs list which was
documented in the web pages if only I had bothered to read it.
Thanks
Rob
I've applied Asger's MSVC patch
(http://thread.gmane.org/gmane.editors.lyx.devel/41102) but I'm having
problems with the Windows build under Visual Studio.NET 2003. I get a
bunch of failures trying to find certain *_moc.c files that didn't seem
to have been generated:
/echo on
-- Build started
88 matches
Mail list logo
