dmg | 5 Sep 2006 02:34
Picon
Picon
Favicon

Latest patch... to libpano12


Hi Everybody,

I just committed a patch that Ippei Ukai created to get hugin to
compile using libpano12 under OS X 10.4

I think it will make sense to release a new version (or libpano12),
given that the patch is needed by hugin.

It will also be the first release (I think) since we moved to svn.

--
daniel

--
Daniel M. German                  "The only bad press is an obituary."
                                    Dennis Rodman
http://turingmachine.org/
http://silvernegative.com/
dmg (at) uvic (dot) ca
replace (at) with  <at>  and (dot) with .

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
Bruno Postle | 5 Sep 2006 10:44
X-Face

Re: Latest patch... to libpano12

On Mon 04-Sep-2006 at 17:34 -0700, dmg wrote:
>
> I think it will make sense to release a new version (or libpano12),
> given that the patch is needed by hugin.

Ok, though there are a couple of changes we need to make so that it 
is possible to install both pano12 and pano13 on the same system:

1. panoinfo needs to be renamed pano12info

2. PTmender, PTblender, PTtiff2psd and PTuncrop need to be removed 
from pano12 as there are more up-to-date versions available with 
pano13.

Sorry, I was going to do this, but haven't found time.

After this the pano13 PTOptimizer needs renaming to PToptimum or 
something.

--

-- 
Bruno

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
Daniel M. German | 7 Sep 2006 06:25
Picon
Picon
Favicon

Re: Latest patch... to libpano12


Hi Bruno

 Bruno Postle twisted the bytes to say:

 Bruno> On Mon 04-Sep-2006 at 17:34 -0700, dmg wrote:
 >> 
 >> I think it will make sense to release a new version (or libpano12),
 >> given that the patch is needed by hugin.

 Bruno> Ok, though there are a couple of changes we need to make so that it 
 Bruno> is possible to install both pano12 and pano13 on the same system:

 Bruno> 1. panoinfo needs to be renamed pano12info

 Bruno> 2. PTmender, PTblender, PTtiff2psd and PTuncrop need to be removed 
 Bruno> from pano12 as there are more up-to-date versions available with 
 Bruno> pano13.

I just committed the changes:

----------------------------------------------------------------------
2006-09-06  dmg  <dmg@...>

	* README.PTmender: REplace text with a note pointing to pano13 for
	their latest version.

	* PTmender.c, PTblender.c, PTtiff2psd.c, PTuncrop.c: Removed the files.
	
	* Makefile.am: Removed all executables
(Continue reading)

Carlos Eduardo | 7 Sep 2006 21:11
Picon

Problems compiling hugin 0.6.1 on FreeBSD

Hi people,

I am now having some problems with boost thread lib to compile hugin. I
already have boost lib installed and have compiled hugin 0.5.4 before
(that's the actual version of the freebsd port). Now my compilation
stops on the configure. I bypassed the configure analisys (changing the
configure file) just to try to see if the compilation would go ok and it
didn't. I guess the boost is compiled ok, but don't know if there is
some special option. The boost package seams to be compiled with
threads. This is the default installation and the package gives the
message "You have built the Boost library with thread support. Don't
forget to add -pthread to your linker options when linking your code.".
Well, I guess the -pthread option is beeing used. The config.log shows
"BOOST_CPPFLAGS='-pthread -I/usr/local/include'". The configure result
is as follows. Any help is welcome.

(...)
checking if Panotools package is complete... yes
checking for boostlib >= 1.31... yes
boost found
checking whether the Boost::Thread library is available... yes
checking for main in -lboost_thread-gcc-mt-1_32... no
checking for main in -lboost_thread-gcc-mt-s-1_32... no
checking for main in -lboost_thread-gcc-1_32... no
checking for main in -lboost_thread-gcc-s-1_32... no
checking for main in -lboost_thread-mgw-mt-1_32... no
checking for main in -lboost_thread-mgw-mt-s-1_32... no
checking for main in -lboost_thread-mgw-1_32... no
checking for main in -lboost_thread-mgw-s-1_32... no
checking for main in -lboost_thread-mt-1_32... no
(Continue reading)

Pablo d'Angelo | 7 Sep 2006 23:26
Picon

Re: Problems compiling hugin 0.6.1 on FreeBSD

Carlos Eduardo wrote:

> checking if Panotools package is complete... yes
> checking for boostlib >= 1.31... yes
> boost found
> checking whether the Boost::Thread library is available... yes
> checking for main in -lboost_thread-gcc-mt-1_32... no
> checking for main in -lboost_thread-gcc-mt-s-1_32... no
> checking for main in -lboost_thread-gcc-1_32... no
> checking for main in -lboost_thread-gcc-s-1_32... no
> checking for main in -lboost_thread-mgw-mt-1_32... no
> checking for main in -lboost_thread-mgw-mt-s-1_32... no
> checking for main in -lboost_thread-mgw-1_32... no
> checking for main in -lboost_thread-mgw-s-1_32... no
> checking for main in -lboost_thread-mt-1_32... no
> checking for main in -lboost_thread-mt-s-1_32... no
> checking for main in -lboost_thread-1_32... no
> checking for main in -lboost_thread-s-1_32... no
> checking for main in -lboost_thread-gcc-mt... no
> checking for main in -lboost_thread-gcc-mt-s... no
> checking for main in -lboost_thread-gcc... no
> checking for main in -lboost_thread-gcc-s... no
> checking for main in -lboost_thread-mgw-mt... no
> checking for main in -lboost_thread-mgw-mt-s... no
> checking for main in -lboost_thread-mgw... no
> checking for main in -lboost_thread-mgw-s... no
> checking for main in -lboost_thread-mt... no
> checking for main in -lboost_thread-mt-s... no
> checking for main in -lboost_thread... no
> checking for main in -lboost_thread-s... no
(Continue reading)

Bruno Postle | 8 Sep 2006 00:34
X-Face

Re: Latest patch... to libpano12

On Wed 06-Sep-2006 at 21:25 -0700, Daniel M. German wrote:
>
> > On Mon 04-Sep-2006 at 17:34 -0700, dmg wrote:
> > 
> > > I think it will make sense to release a new version (or libpano12),
> > > given that the patch is needed by hugin.

Ok, I've created a pano12 2.8.5 release candidate source archives:

http://panotools.sourceforge.net/snapshots/

There is also an *untested* win32 binary zip archive built with 
mingw.

Since there has been something broken with about half the recent 
releases, it would be good to test this one properly first.  
Particularly it needs testing with PTStitcher, since that is the 
whole point of this release.

If somebody else can build this, I'll ask the panotoolsng list for 
testers.

--

-- 
Bruno

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
(Continue reading)

Carlos Eduardo | 8 Sep 2006 04:01
Picon

Re: Problems compiling hugin 0.6.1 on FreeBSD

Is it that simple? I don't know... it looks like something of the boost
lib have been found:
        checking for boostlib >= 1.31... yes
        boost found
        checking whether the Boost::Thread library is available... yes
And on FreeBSD these libs are in a quite standard place (at least in my
point of view :) ):
        /usr/local/lib/libboost_thread.a
        /usr/local/lib/libboost_thread.so
        /usr/local/lib/libboost_thread.so.2
Do I have to change something? I tried to run "./configure
--with-boost=/usr/local/lib" but the result has been the same.

Can it be some function available on newer versions? I have boost-1.32.0
and have also tested with 1.33.1 in another machine.

Thanks!

Carlos E. G. Carvalho           OpenIT Solucoes Tecnologicas
Consultor Unix/Internet         Tel. +55 21 2517-6000
http://www.OpenIT.com.br
http://www.MyFreeBSD.com.br

Em Qui, 2006-09-07 às 22:26 +0100, Pablo d'Angelo escreveu:
> Carlos Eduardo wrote:
> 
> > checking if Panotools package is complete... yes
> > checking for boostlib >= 1.31... yes
> > boost found
> > checking whether the Boost::Thread library is available... yes
(Continue reading)

Daniel M. German | 12 Sep 2006 10:30
Picon
Picon
Favicon

Re: clens


| 9. clens and TIFF's
|     Posted by: "Brian Puccio" brian@... brianmpuccio
|     Date: Mon Sep 11, 2006 7:28 pm (PDT)
| 
| I've compiled clens on OSX, got that working with nona instead of  
| PTstitcher and wrote a script to get it all working nicely so I can  
| just right click an image and defish my fisheye pictures and end up  
| with a rectilinear image. However, clens doesn't seem to take any  
| other image format aside from jpeg. Does anyone know if there is a  
| reason why clens doesn't handle TIFF's? Its been a while since I've  
| programmer in C (read: high school 7 years ago), but I see the file  
| format input check in main.c, just not sure if there's a reason that  
| it can't handle TIFF's.
| 

I just checked the source code. It is a "feature" that it only
supports JPEgs. it actually checks if the input file is a jpeg, but
there is no reason why it would not work for TIFFs. It probably needs
5-10 lines of code. 

BUt I remember correctly clens has been superseded by another tool,
but I just can't remember what its name is. Can anybody help me here?

| Thanks!
| 
| (If there is an open source alternative to clens that can handle TIFF  
| for input, I'd be open to using that, I prefer a CLI program.)
| 
| 
(Continue reading)

Pablo dAngelo | 12 Sep 2006 11:48
Picon

Re: clens

Hi all,

Daniel M. German wrote:

> | 9. clens and TIFF's
> |     Posted by: "Brian Puccio" brian@... brianmpuccio
> |     Date: Mon Sep 11, 2006 7:28 pm (PDT)
> | 
> | I've compiled clens on OSX, got that working with nona instead of  
> | PTstitcher and wrote a script to get it all working nicely so I can  
> | just right click an image and defish my fisheye pictures and end up  
> | with a rectilinear image.
> 
> BUt I remember correctly clens has been superseded by another tool,
> but I just can't remember what its name is. Can anybody help me here?

The fulla tool distributed with hugin (OSX binary available on the hugin homepage) http://prdownloads.sourceforge.net/hugin/hugin-0.6.1_mac_tools.tar.gz?download
does everything that clens does, and can also correct vignetting and TCA at the same
time.

Actually clens has a bug that does introduce some distortion,
if the output image is cropped to avoid black areas at image borders.

It is therefore preferable to use fulla.

ciao
  Pablo
_______________________________________________________________________
Viren-Scan für Ihren PC! Jetzt für jeden. Sofort, online und kostenlos.
Gleich testen! http://www.pc-sicherheit.web.de/freescan/?mc=022222
(Continue reading)

Florian Merz | 15 Sep 2006 10:49
Picon
Picon

compilation error introduced in revision 565 (_isnan())

Hi list,

svn revision 565 intruduced errors during compilation for me. The problem is 
the usage of _isnan() in PTcommons.
The first patch below seems to make it compile for me, but I don't know 
about its correctness.

Additionally I'm using autoconf 2.6, which seems to work fine but is not 
accepted by the bootstrap script.
The second patch below fixes this issue for me.

Greetings,
 Florian

Patch PTcommon.c to use isnan() instead of _isnan():

Index: PTcommon.c
===================================================================
--- PTcommon.c  (Revision 569)
+++ PTcommon.c  (Arbeitskopie)
 <at>  <at>  -741,11 +741,11  <at>  <at> 
                        //a value of "-1.#IND00".  This is not a number, and 
probably indicates
                        //a divide by zero error somewhere in the mapping 
function.  This should
                        //be solved, but, for now, discard this value and 
keep going
-                       if (!_isnan(Dx)) {
+                       if (!isnan(Dx)) {
                                if ((int)Dx < ROIRect->left) ROIRect->left = 
(Continue reading)


Gmane