Re: [hugin-ptx] Google Summer of Code 2011: Call for Mentors

[Bruno Postle]

On Fri 25-Mar-2011 at 22:11 -0400, Yuval Levy wrote:

On March 23, 2011 02:58:25 am David Haberthür wrote:

http://wiki.panotools.org/GSOC_2011#Mentors If you are willing to be a
mentor this year, please let us know, any help is needed and appreciated.
Please reply to this email and/or add yourself to the panotools wiki.


done.  I understand the Melange application is still not fully functional for
mentors invitation and students have a deadline before mentors.  Who is
approving mentors and students while you are away on holiday, Habi?


I'm registered as backup admin, please let me know if I'm missing 
any important steps.


--
Bruno

--
You received this message because you are subscribed to the Google Groups "Hugin and 
other free panoramic software" group.
A list of frequently asked questions is available at: 
http://wiki.panotools.org/Hugin_FAQ
To post to this group, send email to hugin-ptx@googlegroups.com
To unsubscribe from this group, send email to 
hugin-ptx+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/hugin-ptx

[hugin-ptx] Re: hugin plugin interface - developers please liaise

[kfj]

On 26 Mrz., 16:37, Yuval Levy  wrote:
> On March 26, 2011 06:47:59 am kfj wrote:

> > I can't be bothered to
> > look at Windows now that I've finally made the move to Linux.
>
> Understand.  I recall you sounded quite differently when you first joined this
> group.  Keep in mind that not everybody is as progressive as you are, and even
> for those who are, the choice of platform is a personal one.  There is no
> absolute best platform, there is just the platform that works best for me and
> it may not necessarily be the one that works best for you.

I didn't mean to say Linux is 'the absolute best platform'. But you
know (and convinced me at the time) that stuff like packet management
was very helpful indeed. So made the move. I remember when in the
nineties I first started working on UNIX machines - it was liberating.
Back then it was all still pretty much command line stuff, and all of
the sudden I could actually work from the command line, instead of
having to wrestle with MSDOS batch files. Now the same thing happened
again, sort of - only that now Linux finally has arrived on my Laptop
without me having to do without half it's capabilities.

> > It might have been cleaner to go through with
> > the innovation in a coherent testbed, getting it to a good working
> > status and then porting it to the specific platforms, rather than
> > having to develop for all platforms at once.
>
> Isn't this how it is done?  I mean, when you develop a new feature, it first
> works on your platform, whatever that platform is.  Then you start sharing it
> with others who have slightly different platforms, e.g. different versions of
> Ubuntu; then different versions of Linux; and then different POSIX systems;
> and then completely "exotic" (from your perspective) systems, solving the
> issues that present themselves as you go along?

Hmmm... - suppose you're right. I just wish there had been more
success solving issues on 'exotic' systems...

> > Everyone can run Linux since it's free.
>
> Everyone can go jogging once a day for 30 minutes since it's free and it is
> proven to improve health and prolong life.  It does not mean that everyone
> feels like doing it.

30 minutes a day is just over 2% of your lifetime (and a larger
percentage of your waking time, which is probably worth more than the
sleep you get during the extended lifespan). So it'd better increase
your lifespan by more than 1.6 years ;-)

What I meant is that you could just lump together a suitable Linux
plus all the tools and libraries, make a VM from it and offer it for
download without having to pay anyone any licenses. The other systems
don't allow that. I'm not proposing everyone should migrate, but I
think VirtualBox is free for all systems, so it'd be easy to let
everyone have a VM like that and give them an easy start on coding for
hugin without having to worry about platform-specific issues. Just an
idea.

> My problem with a "reference platform" is that even "Linux" is a very broad
> concept.  As you are finding out, some Linux distros already switched to
> Python 3.0 while others are at 2.7 and some are even still at 2.6.  Which one
> would be your "reference platform"?  And wxWidgets?  There are some new things
> in 2.9.x that solve some of the UI issues we're all pestering about.

Point taken. Reference platform is probably the wrong concept.

> > I found it awkward trying to remotely diagnose difficulties
> > people had trying to get my code to run on systems I have no access
> > to.
>
> Here I think you are being too conscientious.  Sure it is awkward, but it is
> really not your problem.

It sort of is, though, because, really, I'm not so terribly interested
in the interface itself but I want to use it. If it never makes it to
mainstream status, all effort I put into using it is reduced to my
personal pleasure, where I'd much rather share it with the bunch.

> You are being very kind and trying to help them, but
> it is ultimately their problem and you can stop conceptually at the "reference
> platform".

I don't think an attitude of 'it's their problem' is helpful here. If
I've learned one thing during this involvement with the project, it's
that a 'me and them' attitude is wrong and doesn't work.

>We have three code maturity criteria [0].  Don't feel obliged to
> go all the way.  "Works for me" is good enough for a development branch and if
> other people will find it useful (and indeed there is strong indication that
> your python_scripting branch is very useful) they will take it on from there
> and help it build on the major supported platforms.  Your help will be
> appreciated, but nobody will expect you to do diagnose and solve difficulties
> on other platform.  Your contribution, as-is is an extremely welcome gift to
> the Hugin community.

I did look at the code maturity criteria. Of course I made my best
effort to satisfy a and c. And I had lots of help along the way -
thanks to everyone! But then I tried to instigate having b

Re: [hugin-ptx] [PATCH 0 of 1] patch for bug lp:486303 (Every illegal character aren't well detected and forbidden)

[Yuval Levy]

On March 23, 2011 07:12:27 am Thibault Lemaitre wrote:
> As described on the bug report
> https://bugs.launchpad.net/hugin/+bug/486303, I did some modifications to
> take care of the illegal character.

Thomas has commented on this on the tracker report.

 
> I try to send this patch by hg pull but I was forbidden, told me please if
> this mailing-list is the best place to submit patch.

You mean hg push?  This works only for approved contributors.  While we are 
very liberal with welcoming new contributors and granting repository write 
access, there are some basic rules of etiquette and my impression is that you 
have not followed them.

Why are you asking if the mailing list is the best place to submit a patch 
*after* submitting?  Makes the question a rhetoric one.  To answer it:  It is 
not.  The best place to submit a patch is in the tracker [0].

That said, we do accept patch submission anywhere but only the tracker 
guarantees that the patch is persistent enough to be actually reviewed.  A 
patch submitted to the mailing list gets buried by newer ML messages and might 
get lost in time.  We try to review patches when resources are available.  
Submission and review do not guarantee acceptance.  In this case, Thomas 
explained to you why your patch was not accepted.

 
> Last information, I'm not on the ML, don't forget me in your eventual
> answers.

Basic etiquette:  if you want to communicate on a communication channel, make 
sure you join it.  You could have kept the communication in the tracker which 
is good as well.  Simply dropping your stuff onto a mailing list and leaving 
the burden of contacting you to the mailing list participant is not polite.

Yuv

[0] 


signature.asc
Description: This is a digitally signed message part.

Re: [hugin-ptx] [REVIEW] Pushing Hugin changes to VIGRA upstream – Hugin changes

[Yuval Levy]

On March 26, 2011 06:05:56 am Lukáš Jirkovský wrote:
> I'll branch out the repository when (and if) the proposed changes get
> their way to upstream. Till that the branch would be useless.

I beg to differ.  Branching in Hg is extremely cheap.  Branch out, and save 
the patches to that you want to contribute upstream in the branch as well.

The use is that it makes your contribution persistent and easy to reproduce.  
hg pull your branch, apply the patches to the vigra source code that is 
already on the system and run.  As opposed to:  search for the patches to the 
vigra source code in a mailing list archive that will forget them over time; 
search for the patch to the Hugin codebase on a little known server that may 
or may not be available a few months down the road; apply patches to Hugin; 
apply patches to vigra source code that is already on the system and run.

Anyway, I hope your changes get accepted upstream asap.  Then it will take 
time for the updated vigra version to trickle down the distributions and in a 
few years we can clean up another part of the foreign source from the Hugin 
repo.

Yuv


signature.asc
Description: This is a digitally signed message part.

Re: [hugin-ptx] Re: hugin plugin interface - developers please liaise

[Yuval Levy]

On March 26, 2011 06:47:59 am kfj wrote:
> I feel I have helped already

Indeed you did, thank you very much for your contributions.  I respect that it 
is your time and you rule.  Feel free to chime in any time again, and have a 
happy long summer holiday!


> >  Plus, you'll get a Google T-Shirt ;-)
> 
> No, not even then.

sorry, was trying to bribe you ;-)


> I can't be bothered to
> look at Windows now that I've finally made the move to Linux.

Understand.  I recall you sounded quite differently when you first joined this 
group.  Keep in mind that not everybody is as progressive as you are, and even 
for those who are, the choice of platform is a personal one.  There is no 
absolute best platform, there is just the platform that works best for me and 
it may not necessarily be the one that works best for you.


> Sometimes it makes me want to extract a streamlined version from the
> current body of code that runs exclusively on Linux and is
> unencumbered by platform issues (I know you'll sympathize with the
> notion).

Actually I am not that sympathetic to the notion, and I'll explain to you why 
in a little bit.


> compromise between new code and the byzantine status quo.

Life is always a compromise.  You can try the "green field approach", take 
only the learning from the current code base and write something completely 
new, from scratch.  Hat off if you can pull that through.  In most cases it is 
simply not economical.  Time is the currency in the Free software world.

> It might have been cleaner to go through with
> the innovation in a coherent testbed, getting it to a good working
> status and then porting it to the specific platforms, rather than
> having to develop for all platforms at once.

Isn't this how it is done?  I mean, when you develop a new feature, it first 
works on your platform, whatever that platform is.  Then you start sharing it 
with others who have slightly different platforms, e.g. different versions of 
Ubuntu; then different versions of Linux; and then different POSIX systems; 
and then completely "exotic" (from your perspective) systems, solving the 
issues that present themselves as you go along?

 
> Everyone can run Linux since it's free.

Everyone can go jogging once a day for 30 minutes since it's free and it is 
proven to improve health and prolong life.  It does not mean that everyone 
feels like doing it.

My problem with a "reference platform" is that even "Linux" is a very broad 
concept.  As you are finding out, some Linux distros already switched to 
Python 3.0 while others are at 2.7 and some are even still at 2.6.  Which one 
would be your "reference platform"?  And wxWidgets?  There are some new things 
in 2.9.x that solve some of the UI issues we're all pestering about.

Going streamlined will necessarily leave users and developers behind and 
fracture the community.  It will cost more than the cost of supporting 
multiple platforms.


> I found it awkward trying to remotely diagnose difficulties
> people had trying to get my code to run on systems I have no access
> to.

Here I think you are being too conscientious.  Sure it is awkward, but it is 
really not your problem.  You are being very kind and trying to help them, but 
it is ultimately their problem and you can stop conceptually at the "reference 
platform".  We have three code maturity criteria [0].  Don't feel obliged to 
go all the way.  "Works for me" is good enough for a development branch and if 
other people will find it useful (and indeed there is strong indication that 
your python_scripting branch is very useful) they will take it on from there 
and help it build on the major supported platforms.  Your help will be 
appreciated, but nobody will expect you to do diagnose and solve difficulties 
on other platform.  Your contribution, as-is is an extremely welcome gift to 
the Hugin community.

Thank you Kay!

Yuv

[0] 



signature.asc
Description: This is a digitally signed message part.

Re: [hugin-ptx] Strange error while making a pano

[Seb Perez-D]

On Sat, Mar 26, 2011 at 14:06, Gerald  wrote:

> enblend: warning: failed to detect any seam
> enblend: mask is entirely black, but white image was not identified as
> redundant
>

See http://hugin.sourceforge.net/docs/manual/Hugin_FAQ.html

Use --finemask

Cheers,

Seb

-- 
You received this message because you are subscribed to the Google Groups 
"Hugin and other free panoramic software" group.
A list of frequently asked questions is available at: 
http://wiki.panotools.org/Hugin_FAQ
To post to this group, send email to hugin-ptx@googlegroups.com
To unsubscribe from this group, send email to 
hugin-ptx+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/hugin-ptx

Re: [hugin-ptx] [GSoC-11] Student Introduction : Neeraj Gupta

[Neeraj Gupta]

Yuv,

Thanks for quick reply.

You are eligible but you (and any student applying that has not previously
> contributed) need to show proficiency.
> 1. set yourself a build environment - instructions for the different
> operating
> systems are linked at [0]
> 2. do something small but meaningful to change the code.  Pick a bug report
> or
> feature request from our tracker [1].  A few of them are tagged gsoc [2] to
> help you choose.
> 3. produce a patch and post it here.
>
>
Wish to inform you that at present I have successfully set up  Hugin build
environment on my system(using Ubuntu 10.10). Currently I am going
documentation(doxygen) and trying to get a brief overview of the code. I
will start working on a bug soon. :)


Regards
Neeraj

-- 
You received this message because you are subscribed to the Google Groups 
"Hugin and other free panoramic software" group.
A list of frequently asked questions is available at: 
http://wiki.panotools.org/Hugin_FAQ
To post to this group, send email to hugin-ptx@googlegroups.com
To unsubscribe from this group, send email to 
hugin-ptx+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/hugin-ptx

[hugin-ptx] Strange error while making a pano

[Gerald]

enblend: warning: failed to detect any seam
enblend: mask is entirely black, but white image was not identified as
redundant
enblend: info: remove invalid output image "2011-03-09-
BIBLIOTHEQUE-001_fused.tif"
make: *** [2011-03-09-BIBLIOTHEQUE-001_fused.tif] Erreur 1



Output options

Hugin Version: 2011.0.0.472072ea810d
Project file: /tmp/huginpto_Xd7iip
Output prefix: 2011-03-09-BIBLIOTHEQUE-001
Projection: Equirectangular (2)
Field of view: 360 x 162
Canvas dimensions: 16744 x 7535
Crop area: (0,153) - (16744,6997)
Output exposure value: 14.88
Selected outputs
Exposure fusion
* Fused and blended panorama


System information

Operating system: GNU/Linux
Release: 2.6.35-28-generic
Kernel version: #49-Ubuntu SMP Tue Mar 1 14:39:03 UTC 2011
Machine: x86_64

AMD Phenom  9550 Quad-Core

-- 
You received this message because you are subscribed to the Google Groups 
"Hugin and other free panoramic software" group.
A list of frequently asked questions is available at: 
http://wiki.panotools.org/Hugin_FAQ
To post to this group, send email to hugin-ptx@googlegroups.com
To unsubscribe from this group, send email to 
hugin-ptx+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/hugin-ptx

[hugin-ptx] Re: hugin plugin interface - developers please liaise

[kfj]

On 26 Mrz., 03:07, Yuval Levy  wrote:
> On March 23, 2011 11:46:21 am kfj wrote:
>
> > When I'm online, I'll
> > gladly help, but I won't be available continuously, so I feel I
> > wouldn't make a good mentor.
>
> OK, so not a primary mentor.  Still, I suggest you join the mentors when Habi
> sends out the invitation, so you get to give your two cents on the selection
> of the students and on the evaluation of their work.

No. I'm very conscientious. I do not want to commit to anything at the
moment, not even some sort of 'secondary' mentorship, because I'd
still feel responsible and that's precisely what I don't want over the
summer. I'm taking a holiday.

I've been working a fair part of every day for the last three months
for hugin. Part of this was coding - a great part of it coding script
and plugin code, another great part trying to get a discussion going
about how to proceed, which has been very time-consuming, and a third
part where I have made an effort to make the code transparent, comment
it well and write documentation to go with it, so that anyone can pick
it up and figure it out. I want to leave it at that for now and have a
break. My offer is to help if help is needed when I'm about, and
that's it. I feel I have helped already, it's all there in the code,
the comments and the documentation.

>  Plus, you'll get a Google T-Shirt ;-)

No, not even then.

> > I'm using Quantum GIS (QGIS)
>
> AFAIK it uses PyQt, the sole GUI framework known to me that already has
> support for Python 3.  PySide is in the making and is likely to come in
> earlier than wxWidgets.  But then this would bring about the old debate of Qt
> vs. wxWidgets...

What I found more interesting than the GUI side is the use of plugins,
which are used for a lot of stuff. I was musing whether it wouldn't be
possible to use some of their code for hugin. But that's future stuff.
First hsi/hpi has to be established and the interface question solved,
then we can start actually picking interesting bits from the vast
Python module universe to add functionality to hugin which is not
encumbered by DLL hell and doesn't have to be linked to it either.

> back in 2006/7 it was DLL-hell.  Static linking was not only convenient:  IIRC
> only magicians with mystical connection were able to get a dynamically linked
> Hugin work, and it would be frail on the users system (read: disappearing DLL
> and other funny things).

> I don't know if with the current crop of Windows O/S things have improved with
> regard to the protection of DLL and other installed files.  Definitely worth a
> try, if somebody out there wants to try.

I think things have changed quite a bit since then, though I can't
think right now what the relevant article was that made me think that.
I have the suspicion that getting cmake to compile a minGW version is
difficult because the distinction between 'compiling for/on a Windows
platform' and 'using MSVC' is blurred. Anyway, I can't be bothered to
look at Windows now that I've finally made the move to Linux.

> This was all with MSVC.  minGW is another set of problems.  There have been a
> few people who have tried over the year, this list archives is full of
> reports, partial successes and frustrations.  It is not known to me that
> anybody has succeeded building Hugin with minGW in the past three years.  Yes,
> it would have great benefit, but somebody must be really determined to get it
> done.

Sometimes it makes me want to extract a streamlined version from the
current body of code that runs exclusively on Linux and is
unencumbered by platform issues (I know you'll sympathize with the
notion). As a playground for innovation, without all the read tape
along the lines of 'you can't do that because on <...>, lib<...> is
only available up to 17.33b. I had the feeling a lot of (not only my)
time went into trying to find a compromise between new code and the
byzantine status quo. It might have been cleaner to go through with
the innovation in a coherent testbed, getting it to a good working
status and then porting it to the specific platforms, rather than
having to develop for all platforms at once.

Everyone can run Linux since it's free. It'd even be quite feasible to
have a virtual machine ready with all the needed development tools,
libraries and documentation so all it'd take would be to install
VirtualBox and have a standardized hugin development environment right
away. I found it awkward trying to remotely diagnose difficulties
people had trying to get my code to run on systems I have no access
to. What am I to do if I'm told: 'it does not work. It says 'unknown
error 2237'. (or such). I feel the multi-platform complexity is a real
problem. Trickle down from a reference platform might be more
effective.

Kay

-- 
You received this message because you are subscribed to the Google Groups 
"Hugin and other free panoramic software" group.
A list of frequently asked questions is available at: 
http://wiki.panotools.org/Hugin_

Re: [hugin-ptx] [REVIEW] Pushing Hugin changes to VIGRA upstream – Hugin changes

[Lukáš Jirkovský]

On 26 March 2011 02:46, Yuval Levy  wrote:
> On March 25, 2011 04:51:14 pm Lukáš Jirkovský wrote:
>> Because the patch is bigger than the maximal allowed size of an
>> attachment I had to store it somewhere else:
>
> How about branching out the repository?
>
> Thanks for the contribution!
> Yuv
>

Hello Yuv,
I'll branch out the repository when (and if) the proposed changes get
their way to upstream. Till that the branch would be useless.

Lukas

-- 
You received this message because you are subscribed to the Google Groups 
"Hugin and other free panoramic software" group.
A list of frequently asked questions is available at: 
http://wiki.panotools.org/Hugin_FAQ
To post to this group, send email to hugin-ptx@googlegroups.com
To unsubscribe from this group, send email to 
hugin-ptx+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/hugin-ptx

Re: [hugin-ptx] Re: [REVIEW] Pushing Hugin changes to VIGRA upstream

[Lukáš Jirkovský]

On 26 March 2011 08:03, Pablo d'Angelo  wrote:
> Hi Lukas,
>
>> Please review them and comment on them, because I'd like to send
>> them to the upstream soon. I'm not sure about copyrights in some cases
>> (eg. canvasSize patch) but I think most of it is Pablo's work. I'll look
>> it up in the subversion history.
>>
>
> great, looks good, feel free to submit those to vigra!
>
> ciao
>  Pablo
>

Hello Pablo,
thank you for giving it the thumbs up, especially given the fact it's
mostly your work.

I'll wait some time if there was some response from others, meanwhile
I'd like to update the copyright headers of the affected files.

Have a nice day,
Lukas

-- 
You received this message because you are subscribed to the Google Groups 
"Hugin and other free panoramic software" group.
A list of frequently asked questions is available at: 
http://wiki.panotools.org/Hugin_FAQ
To post to this group, send email to hugin-ptx@googlegroups.com
To unsubscribe from this group, send email to 
hugin-ptx+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/hugin-ptx

[hugin-ptx] Re: [REVIEW] Pushing Hugin changes to VIGRA upstream

[Pablo d'Angelo]

Hi Lukas,

> Please review them and comment on them, because I'd like to send
> them to the upstream soon. I'm not sure about copyrights in some cases
> (eg. canvasSize patch) but I think most of it is Pablo's work. I'll look
> it up in the subversion history.
>

great, looks good, feel free to submit those to vigra!

ciao
 Pablo

--
You received this message because you are subscribed to the Google Groups "Hugin and 
other free panoramic software" group.
A list of frequently asked questions is available at: 
http://wiki.panotools.org/Hugin_FAQ
To post to this group, send email to hugin-ptx@googlegroups.com
To unsubscribe from this group, send email to 
hugin-ptx+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/hugin-ptx