Sometimes I really want to know how Eclipse implements one feature or another, because Eclipse sometimes feels like magic. Usually it turns out they are also just cooking with water.

Most open source code is easily available by checking out a project from some SCM. For eclipse it is not as easy. But there is a nice tutorial how to access eclipse source code by Lars Vogel. This post is just so I can recover the link next time instead of googling for an hour. Thanks, Lars!

I finally bought myself an eBook reader. To be precise, I got the Sony PRS-600 Reader Touch Edition. Two reasons:

• I bought a lot of technical literature in digital (PDF) format and want to read it without looking at a CRT while on a train.
• The idea of having a lot of reference material with me appeals me.
• Many of my books are read only once (especially Star Trek etc.) and I don’t want to waste paper and the energy for shipping.

First of all, the display is only really great with a powerful light source. The reflection is a bit of a nuisance but I can live with it. What I don’t like is the PDF support of the device. Viewing a normal page, the font faces are so tiny that it is not readable in any way. The zoom feature works around that, but you can’t switch pages from there. Instead you have to unzoom, go to the next page and zoom again. Crappy! There is also the reflow option, which does not mix with source code figures in my books.

I am glad that most of my electronic books are from The Pragmatic Bookshelf, which means that I can download them in Epub as well, all without DRM. Thanks, guys!

For the PDFs I would like to carry, I googled a while and stumbled across this blog post. The example to work with source documents worked fine for The Little Book of Semaphores at least. But even with BRISS I did not (yet) manage to crop books I bought from APress. Most likely I will try doing it via a bit of python scripting and the poppler bindings for python.

I also bought my first DRM content: Star Trek Vanguard: Harbinger. I really enjoyed the book, but the download was a PITA. After buying, you only get a link to a ASCM file, which is not useful by itself. One needs Adobe Digital Editions to download it. So I installed it into a virtual machine which has XP installed. After that, I was able to read it on the PC, which wasn’t the goal of the exercise. To put it on the reader, one needs to authorize it from the same Windows software and upload the file.

All that stuff took almost an hour and of course the digital edition of the book is more expensive than getting it from amazon germany. But at least I did not have to wait for delivery. Having finished the book I wanted to buy the second part, but of course it is only available for canadian residents on ebooks.com. I researched a number of ebook sites which mostly sell it only in the US. Funny enough, the first hits when searching where pirated versions of the book in full text. Why do they make it so hard to get stuff legally? *sigh*

Just in case I forget the link again: 7 weeks ago (says git) I merged upstream 2.8.11.0 into the Debian packaging git. The changes are available at a personal alioth repo if anybody cares.

Edit: I just wanted to file a bug, only to notice that I did this weeks ago. Need more memory!

Note to self: It is actually possible to select a single test module or a single test function in py.test.

But passing the file name as argument to a py.test invocation selects only the doctest from that file (wtf!?). Instead, you need to call it like this:

$ py.test -k test_module    # to run the tests from test_module.py
$ py.test -k test_func      # to run tests having the function name test_func

This is documented in the section advanced test selection and running modes of the py.test documentation, although I fail to see how this is advanced.

I was planning to attend DebConf New York this year, but for a number of reasons I decided not to go. Fortunately, Ferdinand Thommes organized a MiniDebConf in Berlin at LinuxTag and I managed to attend. Thanks, Ferdinand!

There were a number of interesting Talks. I especially liked the talk of our DPL, and those about piuparts and git-buildpackage. In contrast to the other LinuxTag talks, we had a livestream of our talks and recorded (most) of them. The kudos for setting this up goes to Alexander Wirt, who spent quite a few hours to get it up and running.

I have to apologize for being late in bringing my Notebook, which was intended to do the theora encoding of the livestream. This was a misunderstanding on my part, I should have known that this is not going to be setup in the night before show time… So to compensate the extra hours he had to put in for me, I offered to do the post processing of the videos.

Basic approach for post processing

The main goal of post processing the videos was (of course) to compress them to a usable size from the original 143 GB. I also wanted to have a title on each video, and show the sponsors at the end of the video.

My basic idea to implement that consisted of the following steps:

1. Create a title animation template.
2. Generate title animations from template for all talks.
3. Use a video editor to create a playlist of the parts title – talk – epilogue.
4. Run the video editor in batch mode to generate the combined video.
5. Encode the resulting video as ogg theora.

As always with technology, it turned out that the original plan needed a few modifications.

Title animations

Originally I wanted to use Blender for the title animation, but I knew it is quite a complicated bit of software. So I looked for something simpler, and stumbled across an article that pointed me towards Synfig Studio for 2D animation. This is also in Debian, so I gave it a try.

I was delighted that Synfig Studio has a command line renderer which is just called synfig and that the file format is XML, which would make it simple to batch-create the title animations. My title template can be found in this git repository.

Batch creation of title animations

I used a combination of make and a simple python script to replace the author name and the title of the talk into the synfig XML file. The data for all talks is another XML file talks.xml. Basically, I used a simple XPath expression to find the relevant text node and change the data using the ElementTree API of lxml python module.

The same could be done using XSLT of course (for a constant replacement, see this file) but I found it easier to combine two XML files in python.

Note that I create PNG files with synfig and use ffmpeg to generate a DV file from those. Originally, I had synfig create DV files directly but those turned out quite gray for some reason. I am now unable to reproduce this problem.

Combining the title animation with the talk

For joining the title animation with the talk, I originally went with OpenShot, which somebody of the video team had running at the conference. My idea was to mix a single video manually and just replace the underlying data files for each talk. I expected that this would be easy using the openshot-render command, which renders the output video from the input clips and the OpenShot project file. However, OpenShot stores the video lengths in the project file and will take those literally, so this did not work for talks of different play times…

I considered working with Kino or Kdenlive but they did not look more appropriate for this use case. I noticed that OpenShot and Kdenlive both use the Media Lovin’ Toolkit under the hood, and OpenShot actually serializes the MLT configuration to $HOME/.openshot/sequence.xml when rendering. I first tried to read that XML file from python (using the mlt python bindings from the python-mlt2 package) but did not find an API function to do that. So I just hard coded the video sequence in python.

I ran into a few gotchas on the way:

• it’s easy to segfault MLT from Python (something is wrong with the refcounting)
• generating ogg theora using MLT from Python (which uses libavformat) resulted in a bad audio/video synchronisation of the output (more than 1 second for 1 hour of video)
• generating dv from MLT and running ffmpeg2theora on the result is disk-bound on my quad core system – so use a pipe to ffmpeg2theora

Things to improve

While the results look quite okay for me now, there is a lot of room for improvement.

• What bothers me the most is the unequal audio volume between talks. We should apply something like replay gain for normalization. I did not find an easy way to add this to the processing pipeline, otherwise it would have been included already.
• Original recordings have different aspect ratios, so after combining them, there is a letterbox added. This is the case for Goswin’s talk on Multiarch for example. I’d rather have the title distorted or letter boxed, if I get around to it, I will fix it in the script.
• Most of the recordings concentrate on showing the slides. As we did not keep the raw camera data, but recorded the DVswitch output, we missed the opportunity to show the speaker more often in full screen. Looking at the same slide becomes quite boring after half a minute, so it would be better
  to show the speaker more often. That would require quite a bit of man power for post processing though, I have to admit that I did not watch all the hours of video.
• We could cut a bit off the talk at start and end, when the speaker discusses where he is available for questions. It is not of interest anymore after the conference is over.

Availability

• git repo (browsing): http://www.landschoff.net/git/?p=mdc-video.git;a=summary
• git repo (clone): git://landschoff.net/mdc-video
• Videos: http://meetings-archive.debian.net/pub/debian-meetings/2010/minidebconf-berlin/ (note: most of these are in german, apart from the DPL and the LXDE talk)

It has been some time since I blogged about something. I wanted to write this post for a while, but never got around.

Since beginning of february 2010, I have a new job and got a new work notebook: A shiny Dell Latitude E6400. Interestingly, the company usually relies on getting Lenovo Notebooks due to good Linux support, but we were unable to find a notebook without integrated camera and decent specs. We also researched some HP Notebooks. We ended up with a Dell since the website makes it really easy to find a system that fits your requirements.

There is still room for improvement though – you still have to select between Vostro, Latitude, Precision Mobile and XPS before you are allowed to configure the system. There is a filter on the left side of the shop, but filtering for a Core 2 Duo and 4 GB RAM won’t show you the notebook I got. Instead there is some overpriced Latitude (~ €3000) and an XPS system. I just tried the XPS, but was unable to find a Core 2 Duo processor in the options.

Anyway, somehow I made it through and got the following configuration:

I installed Ubuntu karmic (amd64) the day it arrived and I am impressed. I have yet to find a serious problem with hardware support. Everything just worked out of the box, installation was done after half an hour (I needed 10 minutes to find a network cable since the installer did not support the wireless). I wanted to install Debian on it but so far I did not get around to actually do it. But I expect sid to perform just as well with lenny probably lacking some of the needed drivers.

So why am I writing this? The main point is that you can buy a notebook today and have it fully supported wrt. Linux. Open Source has come a long way!

Second: If I replace my personal notebook (a Sony Vaio), I will probably buy Dell as well. Definitely nothing from Sony, because I did not find any way to change the display brightness for the 3 years old system, which makes it usable only next to a power outlet.

For quite a while we are receiving TV via Internet. The original reason is that I got at most 2 MBit/s via ADSL due to line noise, but VDSL supports 25 MBit/s on the same line. Of course, you couldn’t order the DSL line without also getting a phone flat rate and IPTV.

Most of the channels are encrypted and can only be received with the original “Media Receiver”, which is a proprietary (and Microsoft) solution. But at least, I can watch the public service broadcasts using these playlists for VLC or mplayer.

Recently, Eclipse started ignoring clicks on dialog buttons for me. This seems to be due to some changes in gtk 2.18. It does not use native windows for all widgets anymore, and SWT seems to rely on it.

Thanks to this blog post, I have this fix in my bashrc:

alias eclipse="GDK_NATIVE_WINDOWS=true eclipse"

The Debian bug tracker also knows about this problem, which is partly fixed for the eclipse packages. Bad luck that I am using a download from eclipse.org.

Working on another laptop, I was missing my quiltrc settings. I think, I got them from a this email by Marco d’Itri. Perhaps it is of help to somebody (for example, when I go searching for it again).

BTW, I changed it locally to check only for debian folder in parent directories and create debian/patches automatically if needed:

for d in . .. ../.. ../../.. ../../../.. ../../../../..; do
    if [ -d $d/debian ]; then
        if ! [ -d $d/debian/patches ]; then
            mkdir $d/debian/patches
        fi
        export QUILT_PATCHES=debian/patches
    fi
done

I’d rather end up with a patches dir inside the debian folder instead of having it at a random place in the package sources when adding the first quilt patch.

I wonder if we should ship this with the quilt package!?

Update: After discussion with James Vega on IRC, the quiltrc evolved into this version:

d=.
while [ ! -d "$d/debian" -a `readlink -e $d` != / ]; do
    d="$d/.."
done

if [ -d "$d/debian" ]; then
    export QUILT_PATCHES=debian/patches
    if ! [ -d $d/debian/patches ]; then
        mkdir $d/debian/patches
    fi
fi

This works with any directory depth while the original would only support 5 levels from the package base directory.

Ouch! This is the second time I am running into this so I’ll write it up this time.

I used libtool via automake for building a python module. Quite unhelpfully, libtool happens to build only the static library. It does that even though I passed the -module option, which should make it clear that a static library does not help at all. There is also no mention why it skips the shared library.

After some research, my memory came back: libtool requires to pass the -rpath option to actually build shared libraries. I did not want to pass that options, since it is considered harmful where multiple variants of a library are available. It is even less useful in my specific case, as the module is only used inside the source tree for unit testing.

Solution: I changed the Makefile.am to use

_helper_la_LDFLAGS = -module -rpath /freaking/libtool/requires/at/least/a/dummy/path

Interestingly, the same scheme (apart from the wording) is used in libtool’s own source.