On Aug 3, 2007, at 5:21 PM, Jean-Marc Lasgouttes wrote:
Bennett Helm <[EMAIL PROTECTED]> writes:
Won't compile for me on Mac:
Did you fix it?
Yes. (./autogen.sh)
Bennett
Bennett Helm <[EMAIL PROTECTED]> writes:
> Won't compile for me on Mac:
Did you fix it?
JMarc
Darren Freeman <[EMAIL PROTECTED]> writes:
> I'm on the 1.5 branch and it built fine under Linux without
> reconfiguring or anything.
Thanks.
JMarc
On Thu, 2007-08-02 at 16:05 +0200, Jean-Marc Lasgouttes wrote:
> Bennett Helm <[EMAIL PROTECTED]> writes:
>
> >> Looks like gzstream.* did not get compiled and/or linked. Does running
> >> 'make clean && make ' in src/support improve the situation?
> >
> > Nope -- same problem.
>
> Did the makefi
Bennett Helm <[EMAIL PROTECTED]> writes:
>> Looks like gzstream.* did not get compiled and/or linked. Does running
>> 'make clean && make ' in src/support improve the situation?
>
> Nope -- same problem.
Did the makefile get recreated automatically? (it did for me)
Does running ./autogen.sh help
On Aug 2, 2007, at 6:07 AM, Jean-Marc Lasgouttes wrote:
Bennett Helm <[EMAIL PROTECTED]> writes:
On Aug 1, 2007, at 11:26 AM, Jean-Marc Lasgouttes wrote:
"Bo Peng" <[EMAIL PROTECTED]> writes:
OK. Please apply to trunk and branch. Frankly, I am too tired to
debug
boost::iostreams.
Agreed. I
Bennett Helm <[EMAIL PROTECTED]> writes:
> On Aug 1, 2007, at 11:26 AM, Jean-Marc Lasgouttes wrote:
>
>> "Bo Peng" <[EMAIL PROTECTED]> writes:
>>> OK. Please apply to trunk and branch. Frankly, I am too tired to
>>> debug
>>> boost::iostreams.
>>
>> Agreed. I would have liked to upgrade to boost 1.
On Aug 1, 2007, at 11:26 AM, Jean-Marc Lasgouttes wrote:
"Bo Peng" <[EMAIL PROTECTED]> writes:
OK. Please apply to trunk and branch. Frankly, I am too tired to
debug
boost::iostreams.
Agreed. I would have liked to upgrade to boost 1.34.1 too, but this
will be for 1.5.2. The patches are in n
"Bo Peng" <[EMAIL PROTECTED]> writes:
> OK. Please apply to trunk and branch. Frankly, I am too tired to debug
> boost::iostreams.
Agreed. I would have liked to upgrade to boost 1.34.1 too, but this
will be for 1.5.2. The patches are in now, but I did not take much
time to test them. People, pleas
> To be more precise, here is the patch that I propose to apply
> (slightly cleaned up version of Bo's). I added automake stuff, what
> needs to be done for cmake?
OK. Please apply to trunk and branch. Frankly, I am too tired to debug
boost::iostreams.
Is there any chance to fix middle-button pas
[EMAIL PROTECTED] (Jürgen Spitzmüller) writes:
> Huh?
> http://marc.info/?l=lyx-devel&m=118595157304160&w=2
Just when I thought you were gone already...
Have a nice vacation,
JMarc
Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes:
> I propose to apply the patch to 1.5.1svn now, wait a bit to see what
> happens and then release 1.5.1 as it stands. Would it make sense?
To be more precise, here is the patch that I propose to apply
(slightly cleaned up version of Bo's). I added
Jean-Marc Lasgouttes wrote:
> I propose to apply the patch to 1.5.1svn now, wait a bit to see what
> happens and then release 1.5.1 as it stands. Would it make sense?
>
> I agree this is a delay, but we cannot afford to be wrong another time.
FWIW, I agree.
> I see that Jose and Juergen carefully
"Bo Peng" <[EMAIL PROTECTED]> writes:
>> I tested the compression feature on all kinds of documents including 1.4
>> documents and the Linux document from Bo. I did not find any problems.
>> The same code has also been used in many 1.4 releases and works fine there.
>
> Did you upload this installe
Andre Poenitz <[EMAIL PROTECTED]> writes:
>> With this patch things seem to work fine. I can open your file and save
>> & open files with compression enabled. I'm uploading 1.5.0-2 installers
>> right now.
>
> Please do not distribute new stuff unless this is _thoroughly_ tested.
I propose to ap
> I tested the compression feature on all kinds of documents including 1.4
> documents and the Linux document from Bo. I did not find any problems.
> The same code has also been used in many 1.4 releases and works fine there.
Did you upload this installer? I think this is the best we can do for
no
Andre Poenitz wrote:
With this patch things seem to work fine. I can open your file and save
& open files with compression enabled. I'm uploading 1.5.0-2 installers
right now.
Please do not distribute new stuff unless this is _thoroughly_ tested.
I tested the compression feature on all kinds
Bo Peng wrote:
> My suggestion is to remove this menu item for now (and only under
> windows), using the following patch. How to properly fix this issue is
> something for Jose/Jurgen to decide.
I think if the boost issue can not be sorted out quickly and reliabaly, we
should revert to the 1.4 me
Bo Peng wrote:
> I have to say that, if this is indeed boost::iostreams' fault, I would
> be really disappointed at the quality of boost libraries.
There is no single 'quality of boost libraries'. There are excellent
libraries, but there are also useless or complicated ones. And there are
unmain
On Wed, Aug 01, 2007 at 01:13:41AM -0500, Bo Peng wrote:
> One last attempt: according to
> http://www.boost.org/libs/iostreams/doc/installation.html, NO_ZLIB is
> by default 1 under windows. I define -DNO_ZLIB=0 when I build boost,
> and the problem persists.
Maybe the code checks with #ifndef N
On Wed, Aug 01, 2007 at 12:09:17AM +0200, Joost Verburg wrote:
> Bo Peng wrote:
> >If your windows build, with this patch, can open the attached
> >FAQ_comp.lyx (compressed with lyx 1.5.0 linux, official), and maybe a
> >1.4.x compressed file, you can release 1.5.0-1 with compression
> >enabled. Ot
One last attempt: according to
http://www.boost.org/libs/iostreams/doc/installation.html, NO_ZLIB is
by default 1 under windows. I define -DNO_ZLIB=0 when I build boost,
and the problem persists. What irritates me is that there should be
some warning if zlib is not used, and if it is not used, the
I just compiled lyx against boost 1.34.1. Unfortunately, the problem
persists. So, unless someone can write to boost-devel, and be advised
with the correct way to write/build boost::iostreams under windows, we
have to revert to gzstreams for 1.5.1.
Bo
On 7/31/07, Pavel Sanda <[EMAIL PROTECTED]> wrote:
> > > You want me to remove all traces of 1.5.0? Or only the windows
> > > version?
>
> at least for linux this is not possible, 1.5.0 tarballs already flying
> mirrored over the net.
And this is not needed because only windows builds have this pr
> > You want me to remove all traces of 1.5.0? Or only the windows
> > version?
at least for linux this is not possible, 1.5.0 tarballs already flying
mirrored over the net.
pavel
Bo Peng wrote:
If your windows build, with this patch, can open the attached
FAQ_comp.lyx (compressed with lyx 1.5.0 linux, official), and maybe a
1.4.x compressed file, you can release 1.5.0-1 with compression
enabled. Otherwise, please disable compression.
With this patch things seem to work
On Tue, Jul 31, 2007 at 11:39:08PM +0200, Jean-Marc Lasgouttes wrote:
> "Bo Peng" <[EMAIL PROTECTED]> writes:
> >> You want me to remove all traces of 1.5.0? Or only the windows
> >> version?
> >
> > Just remove the windows installers. Joost is trying out a patch I sent
> > to him (revert to gzstre
On Tue, Jul 31, 2007 at 11:14:23PM +0200, Jean-Marc Lasgouttes wrote:
> Andre Poenitz <[EMAIL PROTECTED]> writes:
> > On Tue, Jul 31, 2007 at 02:25:59PM -0500, Bo Peng wrote:
> >> > Maybe it could be better to write own iostream with zlib compression? Or
> >> > you
> >> > could find one in the net
Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes:
> I made all the files unreadable. And added a short announcement to the
> web site.
And I think we can wait until tomorrow for further action.
"Bo Peng" <[EMAIL PROTECTED]> writes:
>> You want me to remove all traces of 1.5.0? Or only the windows
>> version?
>
> Just remove the windows installers. Joost is trying out a patch I sent
> to him (revert to gzstream used in 1.4.x). If it works, he can release
> an updated installer with the pat
> You want me to remove all traces of 1.5.0? Or only the windows
> version?
Just remove the windows installers. Joost is trying out a patch I sent
to him (revert to gzstream used in 1.4.x). If it works, he can release
an updated installer with the patch.
Then, we can see if updating boost can sol
On Tue, Jul 31, 2007 at 10:15:58PM +0200, Joost Verburg wrote:
> Bo Peng wrote:
> >>Bo Peng wrote:
> >>>Unfortunately yes. I just checked that 1.5.0/win can not even open a
> >>>properly compressed .lyx file created under linux.
> >>So the format of compressed files has also been changed on Linux?
Andre Poenitz <[EMAIL PROTECTED]> writes:
> On Tue, Jul 31, 2007 at 02:25:59PM -0500, Bo Peng wrote:
>> > Maybe it could be better to write own iostream with zlib compression? Or
>> > you
>> > could find one in the net?
>> >
>> > The Boost problem might be some deep into cacheing/buffering the out
"Bo Peng" <[EMAIL PROTECTED]> writes:
>> > So this is the second report on corrupted compressed .lyx files within
>> > a week. Maybe it's time to revert the boost zipping stuff.
>
> I can confirm this problem under windows, but not under linux. It
> might be a zlib windows packaging problem or a bo
On Tue, Jul 31, 2007 at 02:25:59PM -0500, Bo Peng wrote:
> > Maybe it could be better to write own iostream with zlib compression? Or you
> > could find one in the net?
> >
> > The Boost problem might be some deep into cacheing/buffering the output.
> > Even more - implementation of STL might cause
On Tue, Jul 31, 2007 at 02:24:17PM -0500, Bo Peng wrote:
> On 7/31/07, Joost Verburg <[EMAIL PROTECTED]> wrote:
> > Bo Peng wrote:
> > > But I am not sure if the zlib way is compatible with the iostreams
> > > compressed files, which can be properly used under linux.
> >
> > LyX 1.4.5 still uses zl
On Tue, Jul 31, 2007 at 01:09:55PM -0500, Bo Peng wrote:
> > > So this is the second report on corrupted compressed .lyx files within
> > > a week. Maybe it's time to revert the boost zipping stuff.
>
> I can confirm this problem under windows, but not under linux. It
> might be a zlib windows pac
Joost,
Attached please find a patch, against 1.5.0tag, that use gzstream (the
one 1.4.x uses) instead of boost::iostreams. The only thing I can say
is that it compiles, and can open a compressed file (by 1.5.0) under
linux. I do not have a good windows environment to test so I will
leave this patc
Bo Peng wrote:
Bo Peng wrote:
Unfortunately yes. I just checked that 1.5.0/win can not even open a
properly compressed .lyx file created under linux.
So the format of compressed files has also been changed on Linux?
I do not think so. The problem is with windows. I do not really want
to rever
> Bo Peng wrote:
> > Unfortunately yes. I just checked that 1.5.0/win can not even open a
> > properly compressed .lyx file created under linux.
>
> So the format of compressed files has also been changed on Linux?
I do not think so. The problem is with windows. I do not really want
to revert to g
Bo Peng wrote:
Unfortunately yes. I just checked that 1.5.0/win can not even open a
properly compressed .lyx file created under linux.
So the format of compressed files has also been changed on Linux? Then I
think 1.5.1 should be released quickly and use zlib for compression
again. Otherwise
> Maybe it could be better to write own iostream with zlib compression? Or you
> could find one in the net?
>
> The Boost problem might be some deep into cacheing/buffering the output.
> Even more - implementation of STL might cause problems under Windows :-(
1.4.x uses that approach, and we switc
On 7/31/07, Joost Verburg <[EMAIL PROTECTED]> wrote:
> Bo Peng wrote:
> > But I am not sure if the zlib way is compatible with the iostreams
> > compressed files, which can be properly used under linux.
>
> LyX 1.4.5 still uses zlib, right? Does this mean that 1.5.0 is not able
> to open compressed
Bo Peng wrote:
But I am not sure if the zlib way is compatible with the iostreams
compressed files, which can be properly used under linux.
LyX 1.4.5 still uses zlib, right? Does this mean that 1.5.0 is not able
to open compressed files created by 1.4.5?
Joost
> My suggestion is to remove this menu item for now (and only under
> windows), using the following patch. How to properly fix this issue is
> something for Jose/Jurgen to decide.
By the way, if you would also like to prevent the case of "open
compressed file generated under linux, save a corrupte
> Please provide me with a patch to make LyX 1.5.0 use zlib again. The
> Windows installer still includes zlib, so no other changes are needed. I
> will then release 1.5.0-2.
But I am not sure if the zlib way is compatible with the iostreams
compressed files, which can be properly used under linux
> In the meantime I came to believe so as well. So why did we rip off
> zlib initially?
Isn't this boost::iostreams thing beautiful? At least to me, the
elegency of the following code was the biggest reason to replace
gzstream with boost::iostreams.
if (compressed)
gz_.push(io::gzip_decompres
> > So this is the second report on corrupted compressed .lyx files within
> > a week. Maybe it's time to revert the boost zipping stuff.
I can confirm this problem under windows, but not under linux. It
might be a zlib windows packaging problem or a boost::iostreams bug,
who knows.
But this is a
48 matches
Mail list logo
