On Fri, Apr 28, 2006 at 11:38:45AM +0200, Jean-Marc Lasgouttes wrote:
> > "Abdelrazak" == Abdelrazak Younes <[EMAIL PROTECTED]> writes:
>
> Abdelrazak> In my tree, there is no libqt-mt.la for Qt3/mingw. Maybe
> Abdelrazak> I've compiled as a dll. Qt4/mingw have no .la file either.
>
> So, nob
> "Enrico" == Enrico Forestieri <[EMAIL PROTECTED]> writes:
Enrico> I don't obtain an .la file when compiling Qt for mingw, but my
Enrico> "compiler" (ie, g++ -mno-cygwin) leaves in the lib directory a
Enrico> .prl file containing that info. This is the one tagged
Enrico> 'mingw-'.
So, can we
On Fri, Apr 28, 2006 at 10:16:56AM +0200, Jean-Marc Lasgouttes wrote:
> > "Enrico" == Enrico Forestieri <[EMAIL PROTECTED]> writes:
>
> Enrico> Strictly speaking, the LDFLAGS thing is a quick hack in order
> Enrico> to have configure test for Qt succeeding, otherwise libtool is
> Enrico> able
> "Abdelrazak" == Abdelrazak Younes <[EMAIL PROTECTED]> writes:
Abdelrazak> In my tree, there is no libqt-mt.la for Qt3/mingw. Maybe
Abdelrazak> I've compiled as a dll. Qt4/mingw have no .la file either.
So, nobody has these la files... Enrico, what is the version of qt
that gives you this?
> "Georg" == Georg Baum <[EMAIL PROTECTED]> writes:
Georg> Which is not very difficult since you wrote something like that
Georg> already.
I know :)
JMarc
On Friday 28 April 2006 09:57, Angus Leeming wrote:
> Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes:
> > Does Qt always have .la files for the cases that interest us? Does it
> > mean that we could do without the check for X11?
>
> .la files are generated by libtool. Qt don't use libtool themselv
Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes:
> Does Qt always have .la files for the cases that interest us? Does it
> mean that we could do without the check for X11?
.la files are generated by libtool. Qt don't use libtool themselves, so most
platforms won't have libqt.la files. My guess is
Jean-Marc Lasgouttes a écrit :
"Enrico" == Enrico Forestieri <[EMAIL PROTECTED]> writes:
Enrico> Strictly speaking, the LDFLAGS thing is a quick hack in order
Enrico> to have configure test for Qt succeeding, otherwise libtool is
Enrico> able to figure out what libraries are needed from libqt.l
Jean-Marc Lasgouttes wrote:
> You read my mind, Georg.
Which is not very difficult since you wrote something like that already.
Georg
Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes:
| Does Qt always have .la files for the cases that interest us?
No.
--
Lgb
> "Georg" == Georg Baum <[EMAIL PROTECTED]> writes:
Georg> 3. Remove all windows viewers and editors from
Georg> lib/configure.py, but keep the formats. On startup, use your
Georg> new function to query whether a format can be viewed/edited for
Georg> all formats with empty viewers/editors and
> "Enrico" == Enrico Forestieri <[EMAIL PROTECTED]> writes:
Enrico> Strictly speaking, the LDFLAGS thing is a quick hack in order
Enrico> to have configure test for Qt succeeding, otherwise libtool is
Enrico> able to figure out what libraries are needed from libqt.la.
Looks like we need a ver
> "Bo" == Bo Peng <[EMAIL PROTECTED]> writes:
>> - A way to avoid adding autoViewable and autoEditable properties
>> would be to set viewer and editor to "auto" as needed (after
>> reading lyxrc.defaults).
Bo> Define "when needed". I was doing so in configure.py.
I mean:
- read lyxrc.defaul
Bo Peng wrote:
> JMarc, before we continue this isAutoViewable business, you (we)
> really need to decide what to do with configure.py. We can
>
> 1. use my "auto" (nobody seems to like it) patch, and change
> isAutoViewable
>
> 2. disable \format for windows. We need basically a
>
> if os.nam
Angus Leeming wrote:
Hum, not sure... On my platform I don't have to set LDFLAGS manually and
I would like it to stay that way. I don't understand why you have to set
them Enrico but we should fix that.
Because he links against qt.lib rather than qt.dll?
(I seem to remember Michael Gerz r
> > Thank you. I (or you) can fix the enum{ View, Edit} thing after the
> > patch is confirmed to work.
>
> Urgh... I forgot that... But I have added autoconf support for
> -lshlwapi.
Not a problem. I now have a combined patch and will post it after I
test it later on windows.
JMarc, before we co
On Thu, Apr 27, 2006 at 08:33:08AM -0500, Bo Peng wrote:
> > It does make sense. Did you not read one my previous comments? I wrote that
> > you should use view and edit, not open and edit
>
> Because the windows term for view is open.
Unless it means 'Execute'...
Andre'
On Thu, Apr 27, 2006 at 09:45:02AM -0500, Bo Peng wrote:
> > > Bo> 1. test 2. check for cygwin
> > >
> > > I can't do any of these.
> >
> > I am doing both.
>
> Thank you. I (or you) can fix the enum{ View, Edit} thing after the
> patch is confirmed to work.
Urgh... I forgot that... But I have ad
On Thu, Apr 27, 2006 at 05:02:20PM +0200, Jean-Marc Lasgouttes wrote:
> > "Enrico" == Enrico Forestieri <[EMAIL PROTECTED]> writes:
>
> Enrico> Currently I have to add libraries using LDFLAGS at configure
> Enrico> time:
>
> Enrico> LDFLAGS="-L$QTDIR/lib -lqt -lkernel32 -luser32 -lgdi32
> Enr
On Thu, Apr 27, 2006 at 05:34:37PM +0200, Abdelrazak Younes wrote:
> Angus Leeming a écrit :
> >Abdelrazak Younes <[EMAIL PROTECTED]> writes:
> >>Hum, not sure... On my platform I don't have to set LDFLAGS manually and
> >>I would like it to stay that way. I don't understand why you have to set
>
> I guess 'empty format' should mean 'empty viewer'. Besides that I like
> the principle of the patch.
Yes.
> - please do not pass the mode as a string to autoOpenFile but an enum
> or a bool.
Will do.
> - if "open" and "edit" are really separate operations, it probably
> means that canAuto
Angus Leeming a écrit :
Abdelrazak Younes <[EMAIL PROTECTED]> writes:
Hum, not sure... On my platform I don't have to set LDFLAGS manually and
I would like it to stay that way. I don't understand why you have to set
them Enrico but we should fix that.
Because he links against qt.lib rather th
Abdelrazak Younes <[EMAIL PROTECTED]> writes:
> Hum, not sure... On my platform I don't have to set LDFLAGS manually and
> I would like it to stay that way. I don't understand why you have to set
> them Enrico but we should fix that.
Because he links against qt.lib rather than qt.dll?
(I seem to
Bo Peng a écrit :
Bo> 1. test 2. check for cygwin
I can't do any of these.
I am doing both.
Thank you. I (or you) can fix the enum{ View, Edit} thing after the
patch is confirmed to work.
Bo> 3. add autoconf things for -lshlwapi
Isn't there a place already where one has to add a bunch of -
> "Enrico" == Enrico Forestieri <[EMAIL PROTECTED]> writes:
Enrico> Currently I have to add libraries using LDFLAGS at configure
Enrico> time:
Enrico> LDFLAGS="-L$QTDIR/lib -lqt -lkernel32 -luser32 -lgdi32
Enrico> -lcomdlg32 \ -ladvapi32 -lshell32 -lole32 -loleaut32 -luuid
Enrico> -limm32 -lw
> > Bo> 1. test 2. check for cygwin
> >
> > I can't do any of these.
>
> I am doing both.
Thank you. I (or you) can fix the enum{ View, Edit} thing after the
patch is confirmed to work.
> > Bo> 3. add autoconf things for -lshlwapi
> >
> > Isn't there a place already where one has to add a bunch o
On Thu, Apr 27, 2006 at 02:28:46PM +0200, Jean-Marc Lasgouttes wrote:
> > "Bo" == Bo Peng <[EMAIL PROTECTED]> writes:
> Bo> Please:
>
> Bo> 1. test 2. check for cygwin
>
> I can't do any of these.
I am doing both.
> Bo> 3. add autoconf things for -lshlwapi
>
> Isn't there a place already
Abdelrazak Younes <[EMAIL PROTECTED]> writes:
> As there are only two options, a bool might be cleaner:
> bool autoOpenFile(string const & filename, bool const edit = false)
> {
> command = edit? "edit": "open";
> }
> But that's not very important anyway...
Scott Meyers says that an enum is
> It does make sense. Did you not read one my previous comments? I wrote that
> you should use view and edit, not open and edit
Because the windows term for view is open.
> Besides that, think about other platforms: They will probably not use a
> string, so think of the above as the special windo
Bo Peng a écrit :
(maybe after taking into account JMarc
comments about string and enum)?
That string (open or edit) will actually be used as command. I do not
think it makes much sense to
if (mode == OPEN)
command = "open"
else if (mode == EDIT)
command = "edit")
As there are only two o
Bo Peng wrote:
> That string (open or edit) will actually be used as command. I do not
> think it makes much sense to
>
> if (mode == OPEN)
> command = "open"
> else if (mode == EDIT)
> command = "edit")
>
> and blah...
It does make sense. Did you not read one my previous comments? I wrote
> (maybe after taking into account JMarc
> comments about string and enum)?
That string (open or edit) will actually be used as command. I do not
think it makes much sense to
if (mode == OPEN)
command = "open"
else if (mode == EDIT)
command = "edit")
and blah...
Bo
Jean-Marc Lasgouttes a écrit :
"Bo" == Bo Peng <[EMAIL PROTECTED]> writes:
Bo> 1. test 2. check for cygwin
I can't do any of these.
As you two seem to have reached an agreement about what was the way to
go, why don't you commit Bo (maybe after taking into account JMarc
comments about strin
> "Bo" == Bo Peng <[EMAIL PROTECTED]> writes:
Bo> Dear list, Attached is a patch that
Bo> 1. add autoViewable to Format. This will return true, under
Bo> windows, if a file extension is associated with a file.
Bo> 2. If the format of a file is empty, but is autoViewable, because
Bo> of a cha
Dear list,
Attached is a patch that
1. add autoViewable to Format. This will return true, under windows,
if a file extension is associated with a file.
2. If the format of a file is empty, but is autoViewable, because of a
change in graph.C, it will appear in the menu.
3. File with empty format
35 matches
Mail list logo
