Available from
http://poppler.freedesktop.org/poppler-0.9.1.tar.gz
WARNING: This is a unstable release, it is actually 0.10 Beta 2 release, it should work like any release from the 0.8 branch, but do not blame us if it makes your chair break.
Differences with 0.9.0:
Core:
* Fix crash on some AESv2 encrypted files (bugs #13972, #16092, #17523)
* Improve parsing of broken files (bug #17568)
glib frontend:
* Minor improvements to the demo application
utils:
* pdftohtml: Generate the outline file in the same place of the other generated files (bug #17504)
Testing, patches and bug reports welcome.
A blog about random things and sometimes about my work translating and developing KDE and anything
Wednesday, September 17, 2008
Poppler 0.9.1 (0.10 Beta 2) released
Wednesday, September 10, 2008
Poppler 0.9.0 (0.10 Beta 1) released
Available from
http://poppler.freedesktop.org/poppler-0.9.0.tar.gz
WARNING: This is a unstable release, it is actually 0.10 Beta 1 release, it should work like any release from the 0.8 branch, but do not blame us if it makes LHC destroy the world.
Differences with 0.8 branch:
Core:
* Initial JavaScript support
* Annotation improvements
* Improvements in the Arthur based renderer
* Improvements in the Cairo based renderer
* Added a JPEG2000 decoder based on OpenJPEG
* Small fixes in ActualText implementation
* Fix jpeg rendering when not using the libjpeg based decoder
* Movie fixes
* Do not get out of memory on documents that specify huge fonts
* Emulate Adobe Reader on documents with duplicate keys in Dictionaries
* Forms improvements
Qt4 frontend:
* Annotation improvements
* Forms improvements
* Add the possibility of extracting embedded fonts
* Initial Movie support
* Documentation improvements
* Small improvements in the PS exporter
glib frontend:
* Annotation improvements
* Attachment fixes
utils:
* updated man pages
* Added -listenc to pdfinfo and pdftotext
Testing, patches and bug reports welcome.
http://poppler.freedesktop.org/poppler-0.9.0.tar.gz
WARNING: This is a unstable release, it is actually 0.10 Beta 1 release, it should work like any release from the 0.8 branch, but do not blame us if it makes LHC destroy the world.
Differences with 0.8 branch:
Core:
* Initial JavaScript support
* Annotation improvements
* Improvements in the Arthur based renderer
* Improvements in the Cairo based renderer
* Added a JPEG2000 decoder based on OpenJPEG
* Small fixes in ActualText implementation
* Fix jpeg rendering when not using the libjpeg based decoder
* Movie fixes
* Do not get out of memory on documents that specify huge fonts
* Emulate Adobe Reader on documents with duplicate keys in Dictionaries
* Forms improvements
Qt4 frontend:
* Annotation improvements
* Forms improvements
* Add the possibility of extracting embedded fonts
* Initial Movie support
* Documentation improvements
* Small improvements in the PS exporter
glib frontend:
* Annotation improvements
* Attachment fixes
utils:
* updated man pages
* Added -listenc to pdfinfo and pdftotext
Testing, patches and bug reports welcome.
Friday, September 05, 2008
Looking for aKademy-es organizers
Somehow i managed to forget to announce here we are looking for organizers for this year KDE Spain users and developers meeting. If you are interested have a look at http://www.kde-espana.es/akademy-es2008/ubicacion.php. We are accepting proposals up to september 15th.
Tuesday, September 02, 2008
Poppler 0.8.7
Just two fixes, but you really want it as fixes a regression in Form handling introduced in Poppler 0.8.6.
Got it while it's hot http://poppler.freedesktop.org/poppler-0.8.7.tar.gz
Got it while it's hot http://poppler.freedesktop.org/poppler-0.8.7.tar.gz
Monday, September 01, 2008
Looking for people
Dear Lazyweb,
At the poppler project we are trying to relicense all our patches as GPLv2 or later so that in case that xpdf is ever relicensed to something newer than gpl2 only we are not stuck because of patches.
The problem is that we have problems getting answers from everyone, so i'm posting this so that in case you know some of the missing people in person or have IM adreses of them, you can bug them for me.
And now onto the names:
- Marco Pesenti Gritti
- Kjartan Maaras
- Timothy Lee
- Carl Worth
- Ed Catmur - specially problematic has his mail server bounces mails
At the poppler project we are trying to relicense all our patches as GPLv2 or later so that in case that xpdf is ever relicensed to something newer than gpl2 only we are not stuck because of patches.
The problem is that we have problems getting answers from everyone, so i'm posting this so that in case you know some of the missing people in person or have IM adreses of them, you can bug them for me.
And now onto the names:
- Marco Pesenti Gritti
- Kjartan Maaras
- Timothy Lee
- Carl Worth
- Ed Catmur - specially problematic has his mail server bounces mails
Etiquetes de comentaris:
gpl,
poppler,
relicensing
Sunday, August 24, 2008
KHTML rocks
'nuff said. Keep it rocking guys!
Sunday, August 17, 2008
Maemo scratchbox on amd64
So it seems the maemo scratchbox refuses to install on amd64, has anyone succeded in doinng so? Can i fool the script, or do i need to run a virtual machine with linux i386 inside and then run the scratchbox inside that?
Sunday, August 03, 2008
KDE 4.1 adoption seems high
KDE 4.1.0 was released this very same week and already 17% of Konqueror users of the last month to my blog are using Konqueror 4.1, seems quite high to me
Catalan Summer of Code
So Catalan government decided to run a project similar to Google's Summer of code program. That's cool, the bad news is that the law that grants the money was approved as of May 30th and i just discovered the site because i was doing some random browsing.
So if you can read catalan go to http://www.lafarga.cat/beques/bases-de-la-convocatoria and start thinking what to do for Free Software.
Actually it seems like it's non-published on purpose so either:
a) It fails and politics can say "See Catalonia has no interest for Free Software" and they can go on paying Microsoft for doing sucky and patchy translations of their products to Catalan (yes they pay twice, license and translation)
b) Someone's son needs to make sure he is the chosen one, so the less publicity the better.
So if you can read catalan go to http://www.lafarga.cat/beques/bases-de-la-convocatoria and start thinking what to do for Free Software.
Actually it seems like it's non-published on purpose so either:
a) It fails and politics can say "See Catalonia has no interest for Free Software" and they can go on paying Microsoft for doing sucky and patchy translations of their products to Catalan (yes they pay twice, license and translation)
b) Someone's son needs to make sure he is the chosen one, so the less publicity the better.
Saturday, August 02, 2008
Bug your distro to package OpenJPEG
The new poppler major version that will come out in a month or so, sports a new JPEG2000 decoder based in OpenJPEG that is better than the current one, for example, it renders Google scanned docs, that is good. Yay! The problem is that actually not much distros package OpenJPEG (i could only find Debian), so go to your distro bug tracker and bug them to include OpenJPEG.
And BTW:
And BTW:
Monday, July 28, 2008
Poppler 0.8.5 released
Available from
http://poppler.freedesktop.org/poppler-0.8.5.tar.gz
Changes since 0.8.4:
Core:
* Fix crash on PDF that define a page thumbnail but it's not a Stream
* Fix crash when Annots object is not of the desired type
* Fix crash when obtaining fonts in PDF where XObjects link themselves in loops
* Fix crash on documents with an IRT object
* Saving should work much better now
* Plug some memory leaks in Annotation handling
Utils:
* pdftohtml: Don't crash on documents that specify an invalid named dest for a link
* pdftohtml: Make html output to keep all the spaces with
* pdftohtml: Improve a bit text layout
* pdftohtml: Make xml output valid xml
Testing, patches and bug reports welcome.
As a side note, one of the four crashes was a CVE while the other three where not, of course the four are as critic as the other.
http://poppler.freedesktop.org/poppler-0.8.5.tar.gz
Changes since 0.8.4:
Core:
* Fix crash on PDF that define a page thumbnail but it's not a Stream
* Fix crash when Annots object is not of the desired type
* Fix crash when obtaining fonts in PDF where XObjects link themselves in loops
* Fix crash on documents with an IRT object
* Saving should work much better now
* Plug some memory leaks in Annotation handling
Utils:
* pdftohtml: Don't crash on documents that specify an invalid named dest for a link
* pdftohtml: Make html output to keep all the spaces with
* pdftohtml: Improve a bit text layout
* pdftohtml: Make xml output valid xml
Testing, patches and bug reports welcome.
As a side note, one of the four crashes was a CVE while the other three where not, of course the four are as critic as the other.
Saturday, July 19, 2008
kpovmodeler looking for help
Andreas Zehender wrote
So if you are interested mail him!
I'm looking for a skilled KDE developer with some knowledge in povray and 3D stuff, who will follow future KDE technology changes and keep kpovmodeler compilable and stable.!
So if you are interested mail him!
Thursday, July 03, 2008
Away from the internet
I'll be without internet connection until the 19th/20th of July so if you mail me, no i'm not ignoring you, i'm just cycling around sardinia :-)
If you need a quick answer for anything i might now, just write the appropiate mailing list for that question, i'm not that special i'm the only one that can asnwer your questions.
If you need a quick answer for anything i might now, just write the appropiate mailing list for that question, i'm not that special i'm the only one that can asnwer your questions.
Sunday, June 29, 2008
On football, national feelings et al
Be warned, this is not a KDE nor Free Software post, if you don't want to read it, don't do it.
Be warned 2, this post is going to be controversial, please don't write saying "You idiot" or similar/worst things, i'm trying to be as respectful as i can when dumping my brain into this post.
So it seems Spain just won Eurocup 2008, lots of people in the streets making noise and using firecrackers/fireworks leftovers from Sant Joan that was just a few days ago.
The ones that know me well know that i don't have a big Spanish feeling, i feel more Catalan and European, but to tell the truth i have mixed feelings about this Spain win.
On the good side, the spanish team is formed by young talented players, that don't act like if they were gods and that do a rather good football. And most of that players feel sympatic and someone you might empathize with, so they winning feels good.
On the other side, looking at the flags spanish followers waved at the stadium you can find three types at very similar rates of use:
* Normal Spanish flag: Nothing to say
* Spanish flag with a bull: Mixed feelings again, the bull is a animal that i probably would say i like if you ask me, but that bull flag is most of the times used either to support bullfighting or to support a rancid view about Spain. And i disagree with both
* Franquist flag: Yes, you might wonder why around one third of people supporting Spain soccer team wave a flag that means you don't like democracy nor any kind of personal rights. I do wonder too.
And then you have people like someone in my neighborhood that play cara al sol, a semi-official franquist song, each time Spain wins.
And that makes me wish Germany did win.
Sadly, i have to agree with what a coworker says: "España es un pais de pandereta" (not sure if makes sense translated but that'd be "Spain is a country of tambourine")
Be warned 2, this post is going to be controversial, please don't write saying "You idiot" or similar/worst things, i'm trying to be as respectful as i can when dumping my brain into this post.
So it seems Spain just won Eurocup 2008, lots of people in the streets making noise and using firecrackers/fireworks leftovers from Sant Joan that was just a few days ago.
The ones that know me well know that i don't have a big Spanish feeling, i feel more Catalan and European, but to tell the truth i have mixed feelings about this Spain win.
On the good side, the spanish team is formed by young talented players, that don't act like if they were gods and that do a rather good football. And most of that players feel sympatic and someone you might empathize with, so they winning feels good.
On the other side, looking at the flags spanish followers waved at the stadium you can find three types at very similar rates of use:
* Normal Spanish flag: Nothing to say
* Spanish flag with a bull: Mixed feelings again, the bull is a animal that i probably would say i like if you ask me, but that bull flag is most of the times used either to support bullfighting or to support a rancid view about Spain. And i disagree with both
* Franquist flag: Yes, you might wonder why around one third of people supporting Spain soccer team wave a flag that means you don't like democracy nor any kind of personal rights. I do wonder too.
And then you have people like someone in my neighborhood that play cara al sol, a semi-official franquist song, each time Spain wins.
And that makes me wish Germany did win.
Sadly, i have to agree with what a coworker says: "España es un pais de pandereta" (not sure if makes sense translated but that'd be "Spain is a country of tambourine")
Monday, June 23, 2008
Installed OpenSuse 11.0
Just installed OpenSuse 11.0 in the laptop. They've done an good job in overall putting KDE 4 in a distro but i had two important problems, so i'm going to blog about them so that if someone has the same problem can find it.
First, the installer misdetected my monitor resolution, then i told him the good one, but that ended up in a messed xorg.conf that applied zoom onto the desktop, i had the remove the Option "PreferredMode" line from the file.
Second, knetworkmanager won't connect to my wifi network, i had to update (using a wired connection) and then it worked like a charm.
First, the installer misdetected my monitor resolution, then i told him the good one, but that ended up in a messed xorg.conf that applied zoom onto the desktop, i had the remove the Option "PreferredMode" line from the file.
Second, knetworkmanager won't connect to my wifi network, i had to update (using a wired connection) and then it worked like a charm.
Friday, June 13, 2008
KDE, plugins, GPL and closed source applications
Here i was today, thinking about this stuff (yeah you can say i'm weird).
In KDE all of our core libs are LGPL so if someone owns a Qt commercial license he can write closed source applications using kdelibs.
In KDE we also have this nice technology called KParts that let's you load plugins for lots of things on runtime. One of the KParts possibilities is to load a KPart for a specific mimetype.
Let's say the above closed source application uses KParts to preview things, and the user of that closed source application decides to preview a PDF.
In that moment, the KPart system will lookup and load the OkularPart that is able of viewing PDFs.
And there is where my mind broke, OkularPart is GPL and is being used by a closed source application.
"NO! VIOLATION!" Does my mind shout, but how can the closed source developer prevent it? All he did is use a LGPL library, so, maybe this is not a violation? And if it is, what are we doing in the KDE project to help the closed source developer be a good citizen? Maybe we should tell the kpart system what kind of Parts are acceptable?
Can someone highlight me?
In KDE all of our core libs are LGPL so if someone owns a Qt commercial license he can write closed source applications using kdelibs.
In KDE we also have this nice technology called KParts that let's you load plugins for lots of things on runtime. One of the KParts possibilities is to load a KPart for a specific mimetype.
Let's say the above closed source application uses KParts to preview things, and the user of that closed source application decides to preview a PDF.
In that moment, the KPart system will lookup and load the OkularPart that is able of viewing PDFs.
And there is where my mind broke, OkularPart is GPL and is being used by a closed source application.
"NO! VIOLATION!" Does my mind shout, but how can the closed source developer prevent it? All he did is use a LGPL library, so, maybe this is not a violation? And if it is, what are we doing in the KDE project to help the closed source developer be a good citizen? Maybe we should tell the kpart system what kind of Parts are acceptable?
Can someone highlight me?
Etiquetes de comentaris:
closed source,
gpl,
kparts
Sunday, June 08, 2008
Dead monitor (again)
On wednesday, my Benq FP937s+ died again, it does not want to power on, the same thing that happened 15 months ago. Not good to have the most expensive Benq monitor at the time it was bought to die twice in less than 3 years :-(
Luckily i've still 3 months of guarantee so i'll be getting a new^Wrepaired one for free.
Luckily i've still 3 months of guarantee so i'll be getting a new^Wrepaired one for free.
Saturday, June 07, 2008
Looking for PDF with JPX streams
I'm currently working on a patch for poppler that makes use of OpenJPEG to decode JPX streams instead of the xpdf inherited code because the inherited code does not work for some PDF and i think that there's no reason not to use an already existing solution if it works. And that's the question, i need more PDF with JPX streams to be sure my 170 lines of code + OpenJPEG can replace the 3200 of xpdf code without any regression.
So if you could
or as said in the comments
and send me (aacid@kde.org) a link to the pdf or the file pdf itself of all the files that match.
So if you could
cd /
find -name "*pdf" -exec grep -H JPXDecode {} \;
or as said in the comments
locate -0 .pdf | xargs -0 -n 1 grep -H JPXDecode
and send me (aacid@kde.org) a link to the pdf or the file pdf itself of all the files that match.
Tuesday, June 03, 2008
Poppler 0.8.3 released
Available from
http://poppler.freedesktop.org/poppler-0.8.3.tar.gz
Changes since 0.8.2:
Core:
* Fix crash when reading some PDF with annotations
* Fix crash on PDF that reference Optional Content elements that don't exist
* Fix leaks on error conditions
* Do not limit CharCodeToUnicodeString to 8 characters
* Support for surrogates outside the BMP plane
Qt 3 frontend:
* Fix crash when reading PDF with password
* Fix leak when calling scanForFonts()
Qt 4 frontend:
* Fix the text() method
Splash renderer:
* Fix compilation with --enable-fixedpoint
Version 0.8.4 is scheduled for July 3rd
Testing, patches and bug reports welcome.
http://poppler.freedesktop.org/poppler-0.8.3.tar.gz
Changes since 0.8.2:
Core:
* Fix crash when reading some PDF with annotations
* Fix crash on PDF that reference Optional Content elements that don't exist
* Fix leaks on error conditions
* Do not limit CharCodeToUnicodeString to 8 characters
* Support for surrogates outside the BMP plane
Qt 3 frontend:
* Fix crash when reading PDF with password
* Fix leak when calling scanForFonts()
Qt 4 frontend:
* Fix the text() method
Splash renderer:
* Fix compilation with --enable-fixedpoint
Version 0.8.4 is scheduled for July 3rd
Testing, patches and bug reports welcome.
Tuesday, May 20, 2008
GDB hidden gem
Well at least it was hidden to me :D
Sometimes it's difficult to keep in view the surrounding lines of code when you are using GDB. list command helps but it's not that user friendly. So the other day a coworker told me about GDB TUI. You can enable it with Ctrl+X A and voilà you get the full listing of the code in a nice browseable view on the top :-)
And a screenie for your viewing pleasure:
Sometimes it's difficult to keep in view the surrounding lines of code when you are using GDB. list command helps but it's not that user friendly. So the other day a coworker told me about GDB TUI. You can enable it with Ctrl+X A and voilà you get the full listing of the code in a nice browseable view on the top :-)
And a screenie for your viewing pleasure:
Monday, May 12, 2008
KDE and strigi analyzers
If you are compiling kde strigi analyzers and installing them into a different path where strigi is installed on, you should be aware that strigi won't find your kde analyzers unless you instruct it to.
Let's say your strigi is in /usr, so it is looking for plugins in /usr/lib/strigi and your kde is in /home/kdesvn/installed/ so the KDE strigi analyzers are on /home/kdesvn/installed/lib/strigi.
You should define the environment variable STRIGI_PLUGIN_PATH to /home/kdesvn/installed/lib/strigi:/usr/lib/strigi
That in case you use strigi 0.5.9 that supports more than one path to be defined in STRIGI_PLUGIN_PATH, in case you are using 0.5.7 like Ubuntu users, you are out of luck as STRIGI_PLUGIN_PATH can only be one path, so you might want to define STRIGI_PLUGIN_PATH to /home/kdesvn/installed/lib/strigi and then go there and create symbolic links from all the ones in /usr/lib/strigi.
Anyway if you are using Ubuntu strigi your analyzers won't work because what seems to be a packaging bug.
Let's say your strigi is in /usr, so it is looking for plugins in /usr/lib/strigi and your kde is in /home/kdesvn/installed/ so the KDE strigi analyzers are on /home/kdesvn/installed/lib/strigi.
You should define the environment variable STRIGI_PLUGIN_PATH to /home/kdesvn/installed/lib/strigi:/usr/lib/strigi
That in case you use strigi 0.5.9 that supports more than one path to be defined in STRIGI_PLUGIN_PATH, in case you are using 0.5.7 like Ubuntu users, you are out of luck as STRIGI_PLUGIN_PATH can only be one path, so you might want to define STRIGI_PLUGIN_PATH to /home/kdesvn/installed/lib/strigi and then go there and create symbolic links from all the ones in /usr/lib/strigi.
Anyway if you are using Ubuntu strigi your analyzers won't work because what seems to be a packaging bug.
Tuesday, May 06, 2008
KUbuntu KDE 3.5.9 kded lock fixed
So this is the great thing of Free Software, people care about their programs/distributions/wathever and we got a relatively quick fix for the ugly problem about kde locking on Kubuntu. Cheers to Jonathan for fixing it.
Saturday, April 26, 2008
Poppler 0.8.1 released
Available from
http://poppler.freedesktop.org/poppler-0.8.1.tar.gz
Changes since 0.8.0:
core:
* Do not call FT_Done_Face on a live cairo_font_face_t as it might cause crashes
* Do not take into account Colorspace resource subdictionary for image XObjects
* Downsample 16 bit per component images to 8 bit per component so they render
build system:
* Link to pthread when the system needs it
windows:
* Fix comparing against NULL instead against INVALID_HANDLE_VALUE when calling FindFirstFile
0.8.2 version is scheduled for May 26th
Testing, patches and bug reports welcome.
http://poppler.freedesktop.org/poppler-0.8.1.tar.gz
Changes since 0.8.0:
core:
* Do not call FT_Done_Face on a live cairo_font_face_t as it might cause crashes
* Do not take into account Colorspace resource subdictionary for image XObjects
* Downsample 16 bit per component images to 8 bit per component so they render
build system:
* Link to pthread when the system needs it
windows:
* Fix comparing against NULL instead against INVALID_HANDLE_VALUE when calling FindFirstFile
0.8.2 version is scheduled for May 26th
Testing, patches and bug reports welcome.
Thursday, April 24, 2008
Q_FOREACH is your friend
It's real, foreach makes writing loops much more convenient, but you have to know how to use it, see the following code:
There are two problems in the usage of foreach in this code.
The first problem is serious, it is trying to use foreach to modify the values of vector. The code compiles, but if you run it, you'll notice nothing was changed, why oh why?! may you ask. Because Foo f is a local variable to the foreach, so what you are doing is modifying a variable that exists in this very same line and nothing more.
The second problem is that you are constructing far much more Foo objects than needed, if you look at the output the copy constructor is called each time in the foreach, so the correct way for the debug line would be
So remember, always that your foreach iterator is not a basic value like an int, double or pointer, use const &, it will be faster and will protect you against the try to use foreach to modify values problem, because
will not compile.
#include <QList>
#include <QtDebug>
class Foo
{
public:
Foo()
{
qDebug("Empty constructor");
}
Foo(int aa, int bb, int cc) : a(aa), b(bb), c(cc)
{
qDebug("Constructor");
}
Foo(const Foo &f)
{
a = f.a;
b = f.b;
c = f.c;
qDebug("Copy constructor");
}
int a, b, c;
};
int main(int /*argc*/, char ** /*argv*/)
{
QList<Foo> list;
list << Foo(1, 2, 3);
list << Foo(4, 5, 6);
list << Foo(7, 8, 9);
list << Foo(10, 11, 12);
list << Foo(13, 14, 15);
foreach(Foo f, list)
f.a = f.b = f.c = 0;
foreach(Foo f, list)
qDebug() << f.a << f.b << f.c;
}
There are two problems in the usage of foreach in this code.
The first problem is serious, it is trying to use foreach to modify the values of vector. The code compiles, but if you run it, you'll notice nothing was changed, why oh why?! may you ask. Because Foo f is a local variable to the foreach, so what you are doing is modifying a variable that exists in this very same line and nothing more.
The second problem is that you are constructing far much more Foo objects than needed, if you look at the output the copy constructor is called each time in the foreach, so the correct way for the debug line would be
foreach(const Foo &f, list)
qDebug() << f.a << f.b << f.c;
}
So remember, always that your foreach iterator is not a basic value like an int, double or pointer, use const &, it will be faster and will protect you against the try to use foreach to modify values problem, because
foreach(const Foo &f, list)
f.a = f.b = f.c = 0;
will not compile.
KUbuntu 8.04 released with rock solid 3.5.9 goodness...
...unless you try to automount USB devices, because that'll give you a deadlocked kded! HORRAY! It's quite interesting to see how a bug caused for a patch to use LUKS is not fixed or the patch reverted when such an used feature has such a big bug like deadlocking kded...
Tuesday, April 22, 2008
Spanish keyboard, dead ~ or not?
If you are using a Spanish keyboard layout, you may have noticed that in some Xorg update, AltGr+4 stopped typing a ~ and then you had to do AltGr+4+space or AltGr+4+AltGr+4 to get a ~.
This is because a non spanish user requested that change and Xorg people changed it.
I asked it to be reverted as there's no use for a dead ~ in Spanish as we don't have ã nor õ (we have ñ but we have a whole key for that) and the xkb maintainer asked me to run some kind of votation so that people decides what they like more.
So please, if you have an opinion on that issue go to http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=477197 and http://bugs.freedesktop.org/show_bug.cgi?id=9763 to get all the facts and write your opinion to 477197@bugs.debian.org
This is because a non spanish user requested that change and Xorg people changed it.
I asked it to be reverted as there's no use for a dead ~ in Spanish as we don't have ã nor õ (we have ñ but we have a whole key for that) and the xkb maintainer asked me to run some kind of votation so that people decides what they like more.
So please, if you have an opinion on that issue go to http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=477197 and http://bugs.freedesktop.org/show_bug.cgi?id=9763 to get all the facts and write your opinion to 477197@bugs.debian.org
Dissappointed with SoC results
So, selected KDE SoC projects are out, and i have to say that personally i'm somewhat disappointed.
Personally i think we've done bad decisions in leaving proven KDE people with sensible projects out, while betting for too many non-proven newbies.
Besides that, the IRC meeting had, in my opinion, too much power over the final outcome, and people like me or David that could not attend it, saw how their mentored projects got dismissed when probably if we had been there that would not have happened so easily.
But well, that huge power of the IRC meeting is also probably due to Google's webpage sucking when you have to rate over 250 applications.
So life's hard, nothing to see here, keep moving.
Personally i think we've done bad decisions in leaving proven KDE people with sensible projects out, while betting for too many non-proven newbies.
Besides that, the IRC meeting had, in my opinion, too much power over the final outcome, and people like me or David that could not attend it, saw how their mentored projects got dismissed when probably if we had been there that would not have happened so easily.
But well, that huge power of the IRC meeting is also probably due to Google's webpage sucking when you have to rate over 250 applications.
So life's hard, nothing to see here, keep moving.
Sunday, April 13, 2008
fglrx blocking Xorg on logout
If you are running Kubuntu Hardy with the flgrx drivers and logging out does not work due to the Xorg process locking up at 100% of CPU usage, you should thank Stefan Carslöv for tracking down the issue to /etc/ati/authatieventsd.sh being unproperly configured, you need to change
to:
XDM_AUTH_MASK=/var/lib/xdm/authdir/authfiles/A$1*
to:
XDM_AUTH_MASK=/var/run/xauth/A$1*
Thursday, April 10, 2008
Throwing Away Promo CDs is Illegal
Or so says Universal. The world has gone mad.
Wednesday, April 02, 2008
iParty X was awesone
This weekend i was with some other spanish KDE dudes (ereslibre, apol, ana`, nefertum and pgquiles) at the Aditel iParty X where we did some talks about KDE. It was a very nice experience being at a Free Software only party.
We celebrated ereslibre's anniversary with a KDE cake and a huge greeting :-)
You can see the photos at http://www.kde.cat/aacid/fotos/ipartyx/ and the Qt4 Dance at http://www.youtube.com/watch?v=EZxmKzbbmEU
We celebrated ereslibre's anniversary with a KDE cake and a huge greeting :-)
You can see the photos at http://www.kde.cat/aacid/fotos/ipartyx/ and the Qt4 Dance at http://www.youtube.com/watch?v=EZxmKzbbmEU
Wednesday, March 26, 2008
Poppler 0.8 Released
So we made it and released poppler 0.8. Congrats to everyone involved.
Changes since 0.7.3 include:
* Fix caching of members in the glib frontend causing issues with rendering
* Change glib public api to have a correct naming
* Some better error handling on corner cases
* Check the document stream is seekable when opening it
* Build fixes with autotools and with cmake
* Fix infinite recursion on some malformed documents when consulting the fonts
* Fix possible crash when asking for Movie contents
I've created a branch for 0.8 releases for bugfixes, i'll try to keep a monthly release schedule like we had with 0.6
If you want to create a directory holding branch 0.8 you can do:
git clone ssh://yourUser@git.freedesktop.org/git/poppler/poppler poppler-0.8
cd poppler-0.8/
git checkout --track -b poppler-0.8 origin/poppler-0.8
if you don't have an account first line should be
git clone git://git.freedesktop.org/git/poppler/poppler poppler-0.8
Testing, patches and bug reports welcome.
Changes since 0.7.3 include:
* Fix caching of members in the glib frontend causing issues with rendering
* Change glib public api to have a correct naming
* Some better error handling on corner cases
* Check the document stream is seekable when opening it
* Build fixes with autotools and with cmake
* Fix infinite recursion on some malformed documents when consulting the fonts
* Fix possible crash when asking for Movie contents
I've created a branch for 0.8 releases for bugfixes, i'll try to keep a monthly release schedule like we had with 0.6
If you want to create a directory holding branch 0.8 you can do:
git clone ssh://yourUser@git.freedesktop.org/git/poppler/poppler poppler-0.8
cd poppler-0.8/
git checkout --track -b poppler-0.8 origin/poppler-0.8
if you don't have an account first line should be
git clone git://git.freedesktop.org/git/poppler/poppler poppler-0.8
Testing, patches and bug reports welcome.
Wednesday, March 19, 2008
Poppler 0.7.3 (0.8 RC2) released
This is the second and last release candidate of what will be Poppler 0.8
Changes include:
* Fix regression in Splash renderer
* Fix off-by-one write in Splash
* Plug some minor leaks in Optional Content code
* Improve error handling when creating a document in the glib frontend
So only one regression and some minor fine tuning. Seems we are on track for
releasing 0.8.0 next week
glib tests are known not to build using cmake, so if you are using cmake
disable glib tests for now.
Testing, patches and bug reports welcome.
Changes include:
* Fix regression in Splash renderer
* Fix off-by-one write in Splash
* Plug some minor leaks in Optional Content code
* Improve error handling when creating a document in the glib frontend
So only one regression and some minor fine tuning. Seems we are on track for
releasing 0.8.0 next week
glib tests are known not to build using cmake, so if you are using cmake
disable glib tests for now.
Testing, patches and bug reports welcome.
Wednesday, March 12, 2008
Poppler 0.7.2 (0.8 RC1) released
So one week after the scheduled date, we have released Poppler 0.7.2.
This is the first release candidate of what will be Poppler 0.8
From now on no API nor ABI changes are allowed unless TOTALLY necessary. Only bugfixes until we release and branch 0.8
We will try to follow the planned schedule with a 1 week delay applied, that is:
* March 19 -> Poppler 0.7.3 (0.8 RC 2)
* March 26 -> Poppler 0.8.0
And now onto the changes
Major Changes:
* Improve font matching not forcing default values onto Fontconfig
* Add preliminary annotations support in the glib frontend
* Initial Movie support in the core
* Make GDK dependency optional in glib bindings
Minor Changes:
* Make the core able to read mime types of embedded files
* Qt4 API for accessing mime types of embedded files
* Handle correctly check state of optional content groups
regarding parents state
* Avoid setting singular CTM matrices on the Cairo backend
* Improved Qt4 API to get character position
* Qt4 api documentation improvements
* Qt4 minor stability fixes
* Proper lib64 Qt detection
* Fix build when compiling without cairo support
Testing, patches and bug reports welcome.
This is the first release candidate of what will be Poppler 0.8
From now on no API nor ABI changes are allowed unless TOTALLY necessary. Only bugfixes until we release and branch 0.8
We will try to follow the planned schedule with a 1 week delay applied, that is:
* March 19 -> Poppler 0.7.3 (0.8 RC 2)
* March 26 -> Poppler 0.8.0
And now onto the changes
Major Changes:
* Improve font matching not forcing default values onto Fontconfig
* Add preliminary annotations support in the glib frontend
* Initial Movie support in the core
* Make GDK dependency optional in glib bindings
Minor Changes:
* Make the core able to read mime types of embedded files
* Qt4 API for accessing mime types of embedded files
* Handle correctly check state of optional content groups
regarding parents state
* Avoid setting singular CTM matrices on the Cairo backend
* Improved Qt4 API to get character position
* Qt4 api documentation improvements
* Qt4 minor stability fixes
* Proper lib64 Qt detection
* Fix build when compiling without cairo support
Testing, patches and bug reports welcome.
Tuesday, March 04, 2008
Konqueror 4.0.2 is more translated than firefox :-)
With the imminent release of KDE 4.0.2, Konqueror will not only have a bigger Acid3 score than Firefox, it will also include 50 languages (49 translations + original English) which equals 50 Firefox 2.0.0.12 translations (44 official + 1 beta + 5 language packs) so if you belong to a language that still is not well translated into KDE join the amazing l10n team of KDE and help us surpass Firefox there too ;-)
Etiquetes de comentaris:
kde,
konqueror,
translation
Sunday, March 02, 2008
Spending my time writing useless software
So here i am, having a TODO list of over 70 items and having to finish my slides for a class i'll do on Saturday about KDE and what do i do? Port and improve my gopher ioslave to KDE4. *sigh*
Wednesday, February 27, 2008
Poppler 0.7.1 (0.8 Beta 2) released
Available from
http://poppler.freedesktop.org/poppler-0.7.1.tar.gz
This is the second release of what will be the 0.8 stable series. We will be having a roughly weekly release schedule until we hit 0.8 stable version. Target is March 19 for 0.8.0
Major Changes:
* Really distribute CMake files as optional build tool
* Initial Optional Content support in core and in the Qt4 frontend
Minor Changes:
* Allow grouped checkboxes to be selected individually
* Qt4 demo program improvements
* Keep cairo and cairo_shape consistent
* Safety checks on Splash renderer so that it does not draw outside the
allocated bitmap
* Do not try to display bitmaps of invalid size
* Fix building with exceptions
* Improvements for building with MSVC and CMake
Testing, patches and bug reports welcome.
http://poppler.freedesktop.org/poppler-0.7.1.tar.gz
This is the second release of what will be the 0.8 stable series. We will be having a roughly weekly release schedule until we hit 0.8 stable version. Target is March 19 for 0.8.0
Major Changes:
* Really distribute CMake files as optional build tool
* Initial Optional Content support in core and in the Qt4 frontend
Minor Changes:
* Allow grouped checkboxes to be selected individually
* Qt4 demo program improvements
* Keep cairo and cairo_shape consistent
* Safety checks on Splash renderer so that it does not draw outside the
allocated bitmap
* Do not try to display bitmaps of invalid size
* Fix building with exceptions
* Improvements for building with MSVC and CMake
Testing, patches and bug reports welcome.
Saturday, February 23, 2008
PostScript is turing complete
Yes, PostScript is turing complete so next time you get a different output when printing with kpdf than the one you get when doing print preview in kpdf, specially if the print preview looks correct, blame whoever decided to use such a complex language for printers and also the drivers of your printer for not being able to understand it, do not blame kpdf.
Etiquetes de comentaris:
kpdf,
postscript,
turing
Tuesday, February 19, 2008
Poppler 0.7.0 (0.8 Beta 1) released
Available from
http://poppler.freedesktop.org/poppler-0.7.0.tar.gz
This is the first release of what will be the 0.8 stable series. We will be having a roughly weekly release schedule until we hit 0.8 stable version.
Besides lots of small improvements/bugfixes, the big things of this release are:
* Saving support
* Partial annotation support
* Forms improvements
* Add support for ActualText entries
* Display characters outside of unicode BMP with TT font
* CJK rendering fixes
* Implement Adobe Glyph Naming convention for fonts
* CMake as optional build tool
* Better font scaling for non embedded fonts
* Preserve PDF page labels when we output as postscript
I've created a tag for 0.7.0 but not a branch. We will keep developing on master for the short 0.7 series, so please, only commit bugfixes, frontend improvements and "preapproved core changes" like saving/forms/optionalContent changes.
Testing, patches and bug reports welcome.
http://poppler.freedesktop.org/poppler-0.7.0.tar.gz
This is the first release of what will be the 0.8 stable series. We will be having a roughly weekly release schedule until we hit 0.8 stable version.
Besides lots of small improvements/bugfixes, the big things of this release are:
* Saving support
* Partial annotation support
* Forms improvements
* Add support for ActualText entries
* Display characters outside of unicode BMP with TT font
* CJK rendering fixes
* Implement Adobe Glyph Naming convention for fonts
* CMake as optional build tool
* Better font scaling for non embedded fonts
* Preserve PDF page labels when we output as postscript
I've created a tag for 0.7.0 but not a branch. We will keep developing on master for the short 0.7 series, so please, only commit bugfixes, frontend improvements and "preapproved core changes" like saving/forms/optionalContent changes.
Testing, patches and bug reports welcome.
Saturday, February 02, 2008
Okular and libspectre
Since KDE 4.0 okular has been using libspectre to render PS files. Due to libspectre being too new we decided to package it inside okular sources in the KDE 4.0.x tree. That has been corrected in trunk because everyone knows having other libraries inside your own source tree sucks, so for now on you need libspectre to get PS support on okular. The downside of that is that i could only find Debian and Pardus libspectre packages, so go to your distro and file a bug and tell them to package libspectre.
Intel publishes graphic cards specs
So now, besides paying people to work on drivers, specs are available to everyone else, cool!
Monday, January 28, 2008
Suggestions for improving a message
Today speaking with fellow KDE devels we realized some of them had the region set to the default C region, that basically means something like USA region but not completely, so not something you want when living not in USA.
Currently the name for that text entry for that is "Default", that may lead to problems as you might think the "Default" region is the correct, but it most probably is not. Some people suggested to switch to "Not Set" which is not completely accurate as it is set to C region, but will probably help "non-purist" users to set the correct Region. Other options go through "Default (Not Set)" or "Not Set (C Region)", etc.
So i'm asking the wider audience here, what would you think it would be the better wording?
To see it live go to systemsettings->Regional & Language, choose "Country/Region & Language" option on the left bar and "Country or Region: YourCurrentRegion (change...)" is what we are talking about.
Currently the name for that text entry for that is "Default", that may lead to problems as you might think the "Default" region is the correct, but it most probably is not. Some people suggested to switch to "Not Set" which is not completely accurate as it is set to C region, but will probably help "non-purist" users to set the correct Region. Other options go through "Default (Not Set)" or "Not Set (C Region)", etc.
So i'm asking the wider audience here, what would you think it would be the better wording?
To see it live go to systemsettings->Regional & Language, choose "Country/Region & Language" option on the left bar and "Country or Region: YourCurrentRegion (change...)" is what we are talking about.
Saturday, January 26, 2008
Poppler 0.6.4 released
Poppler 0.6.4 is available from http://poppler.freedesktop.org/poppler-0.6.4.tar.gz
Changes since 0.6.3:
Qt4 frontend:
* Fix crash on links that point to a non existant page
* Make Document::renderHints return the correct render hints
* Fix infinite loop when parsing LineAnnotation
core:
* Fix crash in the Splash renderer when T3 fonts are badly defined
* Draw underlined Links correctly
utils:
* Fix two use after free bugs in HtmlOutputDev.cc
build system:
* Fix build on mingw32
tests:
* Distribute the glade file of pdf-inspector
Changes since 0.6.3:
Qt4 frontend:
* Fix crash on links that point to a non existant page
* Make Document::renderHints return the correct render hints
* Fix infinite loop when parsing LineAnnotation
core:
* Fix crash in the Splash renderer when T3 fonts are badly defined
* Draw underlined Links correctly
utils:
* Fix two use after free bugs in HtmlOutputDev.cc
build system:
* Fix build on mingw32
tests:
* Distribute the glade file of pdf-inspector
Friday, January 18, 2008
ktank project seeking for developers
KTank project, formed by DrIDK and Ikit, two new french KDE developers that can be often found on #kdegames, aims to create a nice Tank vs Tank game for KDE.
They already have some nice 2D openGL engine
But of course as any other project out there they need more [wo]manpower, specially they are looking for someone that can create nice sounds and songs (rockets have to sound BIG) and also someone with some GGZ knowledge for the multiplayer side, so if you want to be involved in a cool KDE game write to istdasklar free fr or drop by the IRC kdegames channel.
They already have some nice 2D openGL engine
But of course as any other project out there they need more [wo]manpower, specially they are looking for someone that can create nice sounds and songs (rockets have to sound BIG) and also someone with some GGZ knowledge for the multiplayer side, so if you want to be involved in a cool KDE game write to istdasklar free fr or drop by the IRC kdegames channel.
Tuesday, January 08, 2008
KDE 4.0.0 dinner in Barcelona
If you are from Barcelona and want to come to the dinner we organize for Saturday 12 night with KDE programmers, translators and supporters, drop me a mail at aacid@kde.org before Friday night, everyone is welcome!
Thursday, January 03, 2008
libspectre 0.2.0 now available
A new libspectre release is now available from http://libspectre.freedesktop.org/releases/libspectre-0.2.0.tar.gz
That means more work for when KDE 4.1 branch opens, damn!
Changes since libspectre 0.1.0
==================================
API additions:
- Rendering unstructured documents: it's a bit confusing calling
get_page for a document with no pages. The following functions has
been added for rendering unstructured documents:
* spectre_document_render
* spectre_document_render_full
* spectre_document_get_page_size
- Tile rendering: spectre_page_render_slice has been added to allow
rendering rectangles of the page. (Bug #13754)
API changes:
- spectre_render_context_set_page_size determines the size of the
output image, but it doesn't affect the scale factor anymore.
- spectre_render_context_[set|get]_scale now require the scale factor for
both the X and Y dimensions.
That means more work for when KDE 4.1 branch opens, damn!
Changes since libspectre 0.1.0
==================================
API additions:
- Rendering unstructured documents: it's a bit confusing calling
get_page for a document with no pages. The following functions has
been added for rendering unstructured documents:
* spectre_document_render
* spectre_document_render_full
* spectre_document_get_page_size
- Tile rendering: spectre_page_render_slice has been added to allow
rendering rectangles of the page. (Bug #13754)
API changes:
- spectre_render_context_set_page_size determines the size of the
output image, but it doesn't affect the scale factor anymore.
- spectre_render_context_[set|get]_scale now require the scale factor for
both the X and Y dimensions.
Saturday, December 22, 2007
Translate!
Hello, this is Albert speaking from my new KDE 4 i18n/l10n coordinator position.
If you have always wanted to help KDE and you have knowledge of a language different than english you can do it! Join today your local Translation team and you will improve your english but also your native language skills :-)
Translations for KDE4.0 are closing in about 12 days but there will be a 4.0.1 so don't worry of being late!
And congrats to Portuguese team that has a 100% translation!
If you have always wanted to help KDE and you have knowledge of a language different than english you can do it! Join today your local Translation team and you will improve your english but also your native language skills :-)
Translations for KDE4.0 are closing in about 12 days but there will be a 4.0.1 so don't worry of being late!
And congrats to Portuguese team that has a 100% translation!
Etiquetes de comentaris:
i18n,
kde,
l10n,
translation
Sunday, December 16, 2007
libspectre 0.1.0 now available
First libspectre release is now available from:
http://libspectre.freedesktop.org/releases/libspectre-0.1.0.tar.gz
What is libspectre
==================
libspectre is a small library for rendering Postscript documents. It provides a convenient easy to use API for handling and rendering Postscript documents.
Dependencies
============
libspectre depends on libgs which is available at http://www.ghostscript.com
More information about libspectre
=================================
Web site: http://libspectre.freedesktop.org
The programming manual: http://libspectre.freedesktop.org/manual
Bug reports and feedback: https://bugs.freedesktop.org/ Product: libspectre
And yes, in the TRUE cross-desktop spirit this library is programmed by gnome and KDE people :-)
http://libspectre.freedesktop.org/releases/libspectre-0.1.0.tar.gz
What is libspectre
==================
libspectre is a small library for rendering Postscript documents. It provides a convenient easy to use API for handling and rendering Postscript documents.
Dependencies
============
libspectre depends on libgs which is available at http://www.ghostscript.com
More information about libspectre
=================================
Web site: http://libspectre.freedesktop.org
The programming manual: http://libspectre.freedesktop.org/manual
Bug reports and feedback: https://bugs.freedesktop.org/ Product: libspectre
And yes, in the TRUE cross-desktop spirit this library is programmed by gnome and KDE people :-)
Friday, December 14, 2007
Valgrind 3.3.0
The cool valgrind crew has released a new version, from the NEWS file:
*********
3.3.0 is a feature release with many significant improvements and the
usual collection of bug fixes. This release supports X86/Linux,
AMD64/Linux, PPC32/Linux and PPC64/Linux. Support for recent distros
(using gcc 4.3, glibc 2.6 and 2.7) has been added.
The main excitement in 3.3.0 is new and improved tools. Helgrind
works again, Massif has been completely overhauled and much improved,
Cachegrind now does branch-misprediction profiling, and a new category
of experimental tools has been created, containing two new tools:
Omega and DRD.
*********
Omega is cool it actually tells you when a leak just happened instead of having to wait to the end of program execution to know you have a leak :D
And i already blogged on how useful helgrind can be to detect problems in thread programs, but i'll say it again, helgrind rocks too!
*********
3.3.0 is a feature release with many significant improvements and the
usual collection of bug fixes. This release supports X86/Linux,
AMD64/Linux, PPC32/Linux and PPC64/Linux. Support for recent distros
(using gcc 4.3, glibc 2.6 and 2.7) has been added.
The main excitement in 3.3.0 is new and improved tools. Helgrind
works again, Massif has been completely overhauled and much improved,
Cachegrind now does branch-misprediction profiling, and a new category
of experimental tools has been created, containing two new tools:
Omega and DRD.
*********
Omega is cool it actually tells you when a leak just happened instead of having to wait to the end of program execution to know you have a leak :D
And i already blogged on how useful helgrind can be to detect problems in thread programs, but i'll say it again, helgrind rocks too!
Wednesday, December 12, 2007
Poppler 0.6.3 and Poppler-data 0.2.0 released
Poppler 0.6.3 is available from http://poppler.freedesktop.org/poppler-0.6.3.tar.gz
Changes since 0.6.2:
poppler core:
* Fix crash in extra debug code
glib frontend:
* Make sure passwords are passed correctly to poppler core
Qt frontend:
* Fix crash on documents that specify an empty date
build system:
* Disable gtk tests if the user disabled glib frontend
Poppler-data 0.2.0 is available from http://poppler.freedesktop.org/poppler-data-0.2.0.tar.gz and contains lots of new cmaps and nameToUnicode and unicodeMap files so if you are a non strictly latin1 user be sure to grab it.
Changes since 0.6.2:
poppler core:
* Fix crash in extra debug code
glib frontend:
* Make sure passwords are passed correctly to poppler core
Qt frontend:
* Fix crash on documents that specify an empty date
build system:
* Disable gtk tests if the user disabled glib frontend
Poppler-data 0.2.0 is available from http://poppler.freedesktop.org/poppler-data-0.2.0.tar.gz and contains lots of new cmaps and nameToUnicode and unicodeMap files so if you are a non strictly latin1 user be sure to grab it.
Monday, December 03, 2007
KDE-EDU Paris meeting summary
So here's my KDE-EDU Paris meeting summary:
* 14 nice people, some people already knew like annma and carsten and new faces like jpwhiting and SaroEngels
* Fixed issues everywhere ktuberling, kalzium, kgeography, blinken, kate, catalan translation, ... Ok, not all are KDE-EDU related but we are KDE developers after all too, right? ;-)
* Fixed an ugly bug together with Jure 'JLP' Repinc about kickoff search not finding items when their GenericName was not defined, it results in ktrader syntax you have to check that properties exist before using them or you get a tree evaluation error, so it is not (GenericName contains Foo) or (Name contains Foo) but ((GenericName exists) and (GenericName contains Foo)) or (Name contains Foo)
* Together with apol we helped a bit Vladimir making Step compilable standalone and as part of playground-edu and BTW if you have not tried Step, do it NOW! It totally ROCKS.
* Helped SaroEngels with an okular problem on CanNotSayTheNamePlatform, it seems KMimeType::findByPath always returns text/plain on the CanNotSayTheNamePlatform so it's actually not a problem in okular.
* 14 nice people, some people already knew like annma and carsten and new faces like jpwhiting and SaroEngels
* Fixed issues everywhere ktuberling, kalzium, kgeography, blinken, kate, catalan translation, ... Ok, not all are KDE-EDU related but we are KDE developers after all too, right? ;-)
* Fixed an ugly bug together with Jure 'JLP' Repinc about kickoff search not finding items when their GenericName was not defined, it results in ktrader syntax you have to check that properties exist before using them or you get a tree evaluation error, so it is not (GenericName contains Foo) or (Name contains Foo) but ((GenericName exists) and (GenericName contains Foo)) or (Name contains Foo)
* Together with apol we helped a bit Vladimir making Step compilable standalone and as part of playground-edu and BTW if you have not tried Step, do it NOW! It totally ROCKS.
* Helped SaroEngels with an okular problem on CanNotSayTheNamePlatform, it seems KMimeType::findByPath always returns text/plain on the CanNotSayTheNamePlatform so it's actually not a problem in okular.
Saturday, December 01, 2007
Thursday, November 29, 2007
okular, libgs and crashes on Debian and derivatives
If you are trying postscript rendering on okular, be advised you need to have a recent version, the more recent the best. (around 8.60 or so should good enough)
Unless you use Debian or any of its derivatives that reuse their patches (like Ubuntu) because they are using a patch (06_libpaper_support.dpatch) that makes libgs crash when you reuse the gs instance.
Fortunately for the patch creator there is not any program i know that reused the gs instance, until now. Okular is reusing it and now we crash because of someone patching library code and not being prudent.
So next time you touch library code be prudent! And if you are not the main developer of it be more than prudent!
Unless you use Debian or any of its derivatives that reuse their patches (like Ubuntu) because they are using a patch (06_libpaper_support.dpatch) that makes libgs crash when you reuse the gs instance.
Fortunately for the patch creator there is not any program i know that reused the gs instance, until now. Okular is reusing it and now we crash because of someone patching library code and not being prudent.
So next time you touch library code be prudent! And if you are not the main developer of it be more than prudent!
Etiquetes de comentaris:
debian,
ghostscript,
libgs,
okular,
ubuntu
Wednesday, November 21, 2007
Translations, Merchandising and a plug
This saturday we hold the KDE 4 Catalan translation marathon, if you want to join KDE Catalan translation team it's the moment! Come and help us making the translation ready!
I have some merchandising left from Akademy-es, 5 baseball caps, 3 t-shirts (1 S, 2 XXL) and "lots" of stickers, if you are from Spain contact me. Of course all money goes to the e.V.
And now the shameless plug!
A friend has started a new Software Company in Barcelona, called AteneaTech, they specialize on Web Accessibility and Usability and with SaaS and any kind of custom development you need. So if you need such a service from anyone, think on them and i'll win a dinner for each new customer that mentions me ;-)
I have some merchandising left from Akademy-es, 5 baseball caps, 3 t-shirts (1 S, 2 XXL) and "lots" of stickers, if you are from Spain contact me. Of course all money goes to the e.V.
And now the shameless plug!
A friend has started a new Software Company in Barcelona, called AteneaTech, they specialize on Web Accessibility and Usability and with SaaS and any kind of custom development you need. So if you need such a service from anyone, think on them and i'll win a dinner for each new customer that mentions me ;-)
Etiquetes de comentaris:
ateneatech,
catalan,
kde,
kde4
Thursday, November 15, 2007
helgrind is back!
Today i was having a hard time debugging some code of mine that used threads and i thought of helgrind, as of valgrind 3.2.3 trying to use it gives you
Helgrind is currently not working, because:
(a) it is not yet ready to handle the Vex IR and the use with 64-bit
platforms introduced in Valgrind 3.0.0
(b) we need to get thread operation tracking working again after
the changes added in Valgrind 2.4.0
If you want to use Helgrind, you'll have to use Valgrind 2.2.0, which is
the most recent Valgrind release that contains a working Helgrind.
Sorry for the inconvenience. Let us know if this is a problem for you.
So i downloaded valgrind 2.2.0, but it complained about me having a too new glibc, so i downloaded valgrind from svn and YES! it's there :-) It made me realize i needed to add QDeepCopy to some QStrings and it is working now :-) Yet again valgrind to the rescue!
Helgrind is currently not working, because:
(a) it is not yet ready to handle the Vex IR and the use with 64-bit
platforms introduced in Valgrind 3.0.0
(b) we need to get thread operation tracking working again after
the changes added in Valgrind 2.4.0
If you want to use Helgrind, you'll have to use Valgrind 2.2.0, which is
the most recent Valgrind release that contains a working Helgrind.
Sorry for the inconvenience. Let us know if this is a problem for you.
So i downloaded valgrind 2.2.0, but it complained about me having a too new glibc, so i downloaded valgrind from svn and YES! it's there :-) It made me realize i needed to add QDeepCopy to some QStrings and it is working now :-) Yet again valgrind to the rescue!
Saturday, November 10, 2007
Poppler 0.6.2 released
Available from http://poppler.freedesktop.org/poppler-0.6.2.tar.gz
Changes since 0.6.1:
poppler core:
* Fix CVE-2007-4352, CVE-2007-5392 and CVE-2007-5393
* Fix a crash on documents with wrong CCITTFaxStream
* Fix a crash in the Cairo renderer with invalid embedded fonts
* Fix a crash with invalid TrueType fonts
* Check if font is inside the clip area before rendering
it to a temporary bitmap in the Splash renderer. Fixes crashes on
incorrect documents
* Do not use exit(1) on DCTStream errors
* Detect form fields at any depth level
* Do not generate appearance stream for radio buttons that are not active
* mingw fixes
build system:
* Require fontconfig >= 2.0
* builddir != srcdir fixes
Qt4 frontend:
* Improved documentation
misc:
* Fix FSF address
Testing, patches and bug reports welcome.
Changes since 0.6.1:
poppler core:
* Fix CVE-2007-4352, CVE-2007-5392 and CVE-2007-5393
* Fix a crash on documents with wrong CCITTFaxStream
* Fix a crash in the Cairo renderer with invalid embedded fonts
* Fix a crash with invalid TrueType fonts
* Check if font is inside the clip area before rendering
it to a temporary bitmap in the Splash renderer. Fixes crashes on
incorrect documents
* Do not use exit(1) on DCTStream errors
* Detect form fields at any depth level
* Do not generate appearance stream for radio buttons that are not active
* mingw fixes
build system:
* Require fontconfig >= 2.0
* builddir != srcdir fixes
Qt4 frontend:
* Improved documentation
misc:
* Fix FSF address
Testing, patches and bug reports welcome.
Thursday, October 25, 2007
Spanish computer shops to sell Free Software
I read from a Mercè Molist article (spanish) that the "Spanish Association of Small and Medium Sized Computer and New Technologies Companys" will inform their clients about free software, computers more compatible with it and how to claim back Windows fee if you bought a Windows preloaded computer. A really nice move, let's see what's its future.
Etiquetes de comentaris:
apemit,
free software,
mercè molist
Saturday, October 20, 2007
Paris here we go
Just bought my tickets for KDE-edu meeting in Paris. In 40 days i'll join the edu KDE crew to ensure KDE keeps being the best educational desktop out there :-)
BTW, from yesterday to today the plane ticket price was like 15€ more expensive, so if you are planning to come, reserve FAST!
BTW, from yesterday to today the plane ticket price was like 15€ more expensive, so if you are planning to come, reserve FAST!
Friday, October 19, 2007
Iberia is weird
So i go to Iberia.es to book some flights. I try to get Barcelona->Paris (just one way) for 30th November. Price is 632 €. OUCH. Then i try Barcelona<->Paris (roundtrip) coming back home the 2th of December and it costs 136€. Errr, what?
Anyone with insight in how flight companies work can explain that nonsense? Or is it just that French hate us and they let us go to do some tourism and come back but not just go? :-D
Update: The french hating us is a JOKE. I've been twice to Paris and it has been great.
Anyone with insight in how flight companies work can explain that nonsense? Or is it just that French hate us and they let us go to do some tourism and come back but not just go? :-D
Update: The french hating us is a JOKE. I've been twice to Paris and it has been great.
Wednesday, October 17, 2007
Zaragoza here we go
Just bought my tickets for aKademy-es 2007. In 30 days i'll join all the Spain KDE crew and do some talks, party and coding, all in the correct proportion. Now i just need to think what to talk about... And BTW you are ALL invited to join us and enjoy the event, it's free!
Etiquetes de comentaris:
akademy-es,
kde,
zaragoza
Friday, October 12, 2007
Poppler 0.6.1 released
Available from
http://poppler.freedesktop.org/poppler-0.6.1.tar.gz
Changes since 0.6.0:
poppler core:
* Fix printing with different x and y scale
* Fix crash when Form Fields array contains references to non existent objects
* Fix crash in CairoOutputDev::drawMaskedImage()
* Fix embedded file description not working on some cases
Qt4 frontend:
* Fix printing issue
* Avoid double free
* Fix memory leak when dealing with embedded files
glib frontend:
* Fix build with --disable-cairo-output
* Do not return unknown field type for signature form fields
build system:
* Support automake-1.10
* More compatible sh code in qt.m4
utils:
* Fix build on Sun Studio compiler
Testing, patches and bug reports welcome.
http://poppler.freedesktop.org/poppler-0.6.1.tar.gz
Changes since 0.6.0:
poppler core:
* Fix printing with different x and y scale
* Fix crash when Form Fields array contains references to non existent objects
* Fix crash in CairoOutputDev::drawMaskedImage()
* Fix embedded file description not working on some cases
Qt4 frontend:
* Fix printing issue
* Avoid double free
* Fix memory leak when dealing with embedded files
glib frontend:
* Fix build with --disable-cairo-output
* Do not return unknown field type for signature form fields
build system:
* Support automake-1.10
* More compatible sh code in qt.m4
utils:
* Fix build on Sun Studio compiler
Testing, patches and bug reports welcome.
Sunday, October 07, 2007
CMake HEAD got faster reinstalls
On Friday Brad King announced on kde-buildsystem@kde.org "Repeat installations are now lightning fast", today i tried the CVS version of CMake (cmake version 2.5-20071007)
Well, it's not lightning fast, but it's a considerable win, here kdebase went from about 240 seconds to 75 seconds, so congrats to the CMake team and thanks for improving CMake in a continous way!
Well, it's not lightning fast, but it's a considerable win, here kdebase went from about 240 seconds to 75 seconds, so congrats to the CMake team and thanks for improving CMake in a continous way!
Thursday, October 04, 2007
QImage::Format_ARGB32_Premultiplied is your friend
If you have looked at QImage in Qt4 you'll have noticed that there is the typical QImage::Format_RGB32 and QImage::Format_ARGB32 plus QImage::Format_ARGB32_Premultiplied.
Maybe you have wondered why is there, well, i'll tell you, it's FAST.
I have done a simple test rendering 100 times gear-flowers.svg wallpaper in a qimage of 1024x768. The results follow
RGB32: 20351 msec
ARGB32: 39273 msec
ARGB32_Premultiplied: 20248 msec
So if you look at the numbers, you'll see ARGB32_Premultiplied is actually faster than RGB32 that is not even producing correct results.
So QImage::Format_ARGB32_Premultiplied should be probably renamed to QImage::Format_ARGB32_UseMeBecauseImFast :D
Maybe you have wondered why is there, well, i'll tell you, it's FAST.
I have done a simple test rendering 100 times gear-flowers.svg wallpaper in a qimage of 1024x768. The results follow
RGB32: 20351 msec
ARGB32: 39273 msec
ARGB32_Premultiplied: 20248 msec
So if you look at the numbers, you'll see ARGB32_Premultiplied is actually faster than RGB32 that is not even producing correct results.
So QImage::Format_ARGB32_Premultiplied should be probably renamed to QImage::Format_ARGB32_UseMeBecauseImFast :D
Wednesday, October 03, 2007
Performance of radeon free drivers
The other day, i updated to KUbuntu gutsy and saw it ships a free graphics driver for my Ati X300 card (r300 based). I switched to it from the non-free driver but i have went back.
Why? Well, actually it does work, but it is slow, for example, hiding and showing again a normally sized ktuberling, takes around 180ms on the fglrx driver while it takes 550ms on the radeon driver.
But well, after all, as far as i remember they reverse engineered the driver, so i should congratulate the people doing it, it's a bit slow but works, so people that NEED a free driver or where the non-free one does not work can actually use the card now, thanks :-)
This is with driver 6.7.194 which i would say is latest released version.
UPDATE: Setting the driver to use EXA makes it go as fast or faster than fglrx driver in this scenario of swapping ktuberling windows, now i ask me why this is not the default :-S
UPDATE 2: Now blogger.com newsticker makes CPU go to 100% on both konqueror and firefox. I think i'll wait for the next iteration of the driver.
Why? Well, actually it does work, but it is slow, for example, hiding and showing again a normally sized ktuberling, takes around 180ms on the fglrx driver while it takes 550ms on the radeon driver.
But well, after all, as far as i remember they reverse engineered the driver, so i should congratulate the people doing it, it's a bit slow but works, so people that NEED a free driver or where the non-free one does not work can actually use the card now, thanks :-)
This is with driver 6.7.194 which i would say is latest released version.
UPDATE: Setting the driver to use EXA makes it go as fast or faster than fglrx driver in this scenario of swapping ktuberling windows, now i ask me why this is not the default :-S
UPDATE 2: Now blogger.com newsticker makes CPU go to 100% on both konqueror and firefox. I think i'll wait for the next iteration of the driver.
Friday, September 28, 2007
Why small KDE projects are useful for the big KDE picture?
On Wednesday, Burkhard Lück began sending mails to the kdegames mailing list about the shortcomings we had regarding translations not being loaded, not being possible, etc.
One of the most reported issues was that translations of theme names were not used even though they had been translated. That shocked us as we use the typical KDE way of using .desktop files, so we investigated a bit, and came up with with a very small test case that showed that the kdelibs class for accessing .desktop files was not working correctly.
In less than two hours we had the bug fixed. So that's why small KDE projects like kdegames are useful for KDE, they are as good testers as anyone else.
One of the most reported issues was that translations of theme names were not used even though they had been translated. That shocked us as we use the typical KDE way of using .desktop files, so we investigated a bit, and came up with with a very small test case that showed that the kdelibs class for accessing .desktop files was not working correctly.
In less than two hours we had the bug fixed. So that's why small KDE projects like kdegames are useful for KDE, they are as good testers as anyone else.
Monday, September 24, 2007
KUbuntu Gutsy issues
Yesterday i updated to what will be KUbuntu 7.10 to 7.04. Here are the results:
* They still poppler patch kpdf even if the 3.5.7 packages for KUbuntu 7.04 did not, so expect that i still treat KUbuntu users as second class users when reporting kpdf bugs.
* There's a ¿kernel? bug that hogs the CPU https://bugs.launchpad.net/ubuntu/+source/udev/+bug/119315/
* My network configuration was lost, resulting in not being able to access the internet anymore, that is because now my second network interface is supported (that's good) and so what before was my eth0 is now eth1 and nothing worked (that's bad)
* They fixed a bug i reported about exit dialog accelerators not working correctly
* kdesu hangs randomly https://bugs.launchpad.net/ubuntu/+source/kdebase/+bug/144397
* XF86Mail key is hardcoded to a new mail message and you can't reassign it to something else like for example opening KMail https://bugs.launchpad.net/ubuntu/+source/kdebase/+bug/144398
* They added a ugly user list to kdm that when disabling it results in an even uglier kdm https://bugs.launchpad.net/ubuntu/+source/kdebase/+bug/129309
All in all, my opinion is quite negative at the moment.
* They still poppler patch kpdf even if the 3.5.7 packages for KUbuntu 7.04 did not, so expect that i still treat KUbuntu users as second class users when reporting kpdf bugs.
* There's a ¿kernel? bug that hogs the CPU https://bugs.launchpad.net/ubuntu/+source/udev/+bug/119315/
* My network configuration was lost, resulting in not being able to access the internet anymore, that is because now my second network interface is supported (that's good) and so what before was my eth0 is now eth1 and nothing worked (that's bad)
* They fixed a bug i reported about exit dialog accelerators not working correctly
* kdesu hangs randomly https://bugs.launchpad.net/ubuntu/+source/kdebase/+bug/144397
* XF86Mail key is hardcoded to a new mail message and you can't reassign it to something else like for example opening KMail https://bugs.launchpad.net/ubuntu/+source/kdebase/+bug/144398
* They added a ugly user list to kdm that when disabling it results in an even uglier kdm https://bugs.launchpad.net/ubuntu/+source/kdebase/+bug/129309
All in all, my opinion is quite negative at the moment.
Saturday, September 22, 2007
New KTuberling playground and looking for a description
Thanks to Jocelyn Etienne KTuberling in KDE 4.0 will feature another nice playground

Besides, the other day, Robert Knight, our konsole overlord warned me that normal menu users, current name of the game is "Potato Guy" and the description is "Game for Children". First is actually wrong because of this new playground and second is really really precise, eh ;-)
So i'm open to any suggestion for both name and description.
Besides, the other day, Robert Knight, our konsole overlord warned me that normal menu users, current name of the game is "Potato Guy" and the description is "Game for Children". First is actually wrong because of this new playground and second is really really precise, eh ;-)
So i'm open to any suggestion for both name and description.
Kate welcomes Find Selected
Thanks to quick work by Sebastian Pipping (sping) and some small contribution from my side, kate will have in KDE 4 the "Find Selected" feature.
Basically it's the find feature, but starts find automatically on the selected text and does not show the find dialog. At start it may seem not useful enough, but those used to Eclipse (ctrl+k) and nedit (ctrl+h) will surely welcome this new addition to kate
Basically it's the find feature, but starts find automatically on the selected text and does not show the find dialog. At start it may seem not useful enough, but those used to Eclipse (ctrl+k) and nedit (ctrl+h) will surely welcome this new addition to kate
Monday, September 03, 2007
Poppler 0.6 released
Poppler 0.6 is available from http://poppler.freedesktop.org/poppler-0.6.tar.gz
Changes since RC2:
- CairoOutputDev fixes
- Allow pdftoppm to read/write from stdin/stdout
- API work on Qt4 frontend
- Fix pdfimages produces inverted image for black & white image
- Fix error on the NFKC text matching routine
- Add support for word and line selections
- Do not enforce %%EOF at the end of file
- Pad zeroes instead of aborting when rendering 1-bit images
and the stream is too short
- Update glib bindings documentation
Testing, patches and bug reports welcome.
Changes since RC2:
- CairoOutputDev fixes
- Allow pdftoppm to read/write from stdin/stdout
- API work on Qt4 frontend
- Fix pdfimages produces inverted image for black & white image
- Fix error on the NFKC text matching routine
- Add support for word and line selections
- Do not enforce %%EOF at the end of file
- Pad zeroes instead of aborting when rendering 1-bit images
and the stream is too short
- Update glib bindings documentation
Testing, patches and bug reports welcome.
Monday, August 27, 2007
Junior Jobs for KBackgammon
KBackgammon is KDE backgammon client. SHOCK!!!
It did not get as much love as other kde games for the KDE 4.0 release but we decided to keep for reasons that are not important now.
What is important is that i created 3 Junior Jobs for people that want to start with KDE 4 development and don't know where to.
If you want to have a look at them see:
Bug 149275: Enable antialias in KBackgammon
Bug 149276: Convert config dialogs to .ui files
Bug 149277: Remove engineSet usage
All three should be "fairly" straightforward even for a newcomer.
So get your brain ready and start coding.
It did not get as much love as other kde games for the KDE 4.0 release but we decided to keep for reasons that are not important now.
What is important is that i created 3 Junior Jobs for people that want to start with KDE 4 development and don't know where to.
If you want to have a look at them see:
Bug 149275: Enable antialias in KBackgammon
Bug 149276: Convert config dialogs to .ui files
Bug 149277: Remove engineSet usage
All three should be "fairly" straightforward even for a newcomer.
So get your brain ready and start coding.
Saturday, August 25, 2007
Modern art is ...
... curious. Have a look at this photo from me at the Pompidou Center of Paris.

You have to have guts to tell people that's art.
You have to have guts to tell people that's art.
Etiquetes de comentaris:
modern art,
paris,
pompidou
Wednesday, August 15, 2007
poppler 0.6 Release Candidate 2 released
Today i released poppler 0.5.91 aka 0.6 Release Candidate 2
Changes include:
- Various memory leaks fixed
- Compile with --enable-fixedpoint. Bug #11110
- Header cleanup
- Remove dependency on debugxml. Bug #11187
- Allow access to more document metadata in glib and qt4 frontends
- Several glib API frontend improvements
- Fix crash on accessing embedded files
- Build on Sun Force compiler
- Render '*' instead of the actual content in password form fields
- Fix pdftohtml complex output. Bug #9746 and #11610
- Windows build fixes
- Improve Japanese font support. Bug #11413
- Do not exit the program on files that confuse libjpeg
- Update required cairo version to 1.4
- Fix CVE-2007-3387
Testing, patches and bug reports welcome.
Changes include:
- Various memory leaks fixed
- Compile with --enable-fixedpoint. Bug #11110
- Header cleanup
- Remove dependency on debugxml. Bug #11187
- Allow access to more document metadata in glib and qt4 frontends
- Several glib API frontend improvements
- Fix crash on accessing embedded files
- Build on Sun Force compiler
- Render '*' instead of the actual content in password form fields
- Fix pdftohtml complex output. Bug #9746 and #11610
- Windows build fixes
- Improve Japanese font support. Bug #11413
- Do not exit the program on files that confuse libjpeg
- Update required cairo version to 1.4
- Fix CVE-2007-3387
Testing, patches and bug reports welcome.
Tuesday, August 14, 2007
Weird PS files needed
As you may know, i'm currently working on adding PS support to okular. It's currently more or less working, the problem is most files i have around here are result of a few programs "printing" to PS, so most of them are basically "the same in their internals" and thus are quite unuseful for testing. What i need are PS files with different page sizes, different page rotations, etc. the weirder the better. If you have or know where to get some of this kind of files, please mail me at aacid@kde.org with the details.
Thursday, August 02, 2007
Avoid Ya.com
<BEGIN RANT>
Let me begin with something everyone agrees with "All spanish ISP suck".
That said i have to suggest any spanish reader out there to avoid ya.com.
I've been a Ya.com ADSL user for two years now, and besides minor hiccups and their broken DNS servers i was fairly satisfied, but not anymore.
On July 16th, my internet connection stopped working, it is still not working, and their technical service as marked my incidence as WONTFIX. They have told me all is fine in their side and what i have to do is call the company that is providing them the service (they are mere resellers) and tell them to fix my problem.
WTF?¿? Why do i have to call that company i'm not paying anything for my connection. I'm paying ya.com for my connection and any normal people would think it's them who have to fix my problems. But well, i did call the other company, four times, and obviously they told me "hi Mr Astals you are not our client and we can't help you, sorry"
So i'm giving up from Ya.com and just going to some other company to see if i have more luck.
On the Free Software side this means you won't have non-gui-blocking text search in Okular in less than two weeks (while i have done it for weeks) and that someone else will have to release poppler 0.5.9 RC2 (or wait until i get internet again)
</END RANT>
Let me begin with something everyone agrees with "All spanish ISP suck".
That said i have to suggest any spanish reader out there to avoid ya.com.
I've been a Ya.com ADSL user for two years now, and besides minor hiccups and their broken DNS servers i was fairly satisfied, but not anymore.
On July 16th, my internet connection stopped working, it is still not working, and their technical service as marked my incidence as WONTFIX. They have told me all is fine in their side and what i have to do is call the company that is providing them the service (they are mere resellers) and tell them to fix my problem.
WTF?¿? Why do i have to call that company i'm not paying anything for my connection. I'm paying ya.com for my connection and any normal people would think it's them who have to fix my problems. But well, i did call the other company, four times, and obviously they told me "hi Mr Astals you are not our client and we can't help you, sorry"
So i'm giving up from Ya.com and just going to some other company to see if i have more luck.
On the Free Software side this means you won't have non-gui-blocking text search in Okular in less than two weeks (while i have done it for weeks) and that someone else will have to release poppler 0.5.9 RC2 (or wait until i get internet again)
</END RANT>
Wednesday, July 18, 2007
No internet
Since Sunday night i'm without internet at home, so all you that are expecting replies and internet actions from me (Chusslove, Yukiko, Miquel...) please have some more patience, the ADSL company told me "We are working on it".
Wednesday, July 11, 2007
Canon Ixus 850 (2)
So finally i got the Canon Ixus 850, it costed a bit less than 350€ and came with lots of extras (1GB memory, case, tripod and second battery). As far as i've been testing it i'm quite happy with it, quite fast when taking photos, quite good image quality and no problems using it with digikam. So that's it, i do not regret the buy, maybe it helps for others in the same situation.
Eclipse Europa can't use lineedits history
Yesterday i just installed the new Eclipse version at work and suddenly i realized i could not use up/down arrows to use lineedit's history. I filed a bug against their bugzilla that got closed as duplicate of a WONTFIX/NOTOURBUG bug :-(
Actually it seems the bug is on gtk but it quite astonishes me the fact that i'm using the same gtk i was and it worked on previous version of Eclipse.
So that forces us to ask the question: As a software developer, it's better to release a product that does not work and wait for the library we use to fix itself or should we workaround the bug as much as we can?
I'm not sure which one i'd pick with my developer hat, but as a user i tell you i want things to work :D
Actually it seems the bug is on gtk but it quite astonishes me the fact that i'm using the same gtk i was and it worked on previous version of Eclipse.
So that forces us to ask the question: As a software developer, it's better to release a product that does not work and wait for the library we use to fix itself or should we workaround the bug as much as we can?
I'm not sure which one i'd pick with my developer hat, but as a user i tell you i want things to work :D
aKademy 2007 photos
aKademy 2007 was great! I did not blog as much as i wanted (actually i blogged 0 because blogger.com was unaccessible from the aKademy network) so here come the photos i made as small substitution. They suck a bit, but i've never been a good photographer so please bare with me! http://kde.cat/aacid/fotos/akademy2007/
Monday, June 25, 2007
Canon Ixus 850
I'm planning buying a Canon Ixus 850 if i can get one for a reasonable price (between 300 for a barebone one and 350€ with extras) before going to akademy 2007. Anyone has first experience with it and want to give some comment? Please abstain people that have a Olympus/Sony/wathever :-)
See you on Saturday on Glasgow :-)
See you on Saturday on Glasgow :-)
Thursday, June 21, 2007
SpreadKDE rebirth
Thanks to Pier-Olivier Thibault & Luke Parry we got a new and shiny SpreadKDE webpage. Paraphrasing Pier-Olivier
*****
We'd like to invite you to our new website http://spread.kde.org where you'll find all the information you need. As of today, the main feature we implemented is a managing system for people who are willing to organize events on the behalf of KDE. If you have any question about spread.kde.org, feel free to ask them here on kde-promo@kde.org or on #spreadkde on Freenode.
We strongly suggest you that if you're willing to organize an event, to make it so through the spread.kde.org website as it will help us make the website more usable for you all.
*****
As always help is welcome so if you want to help KDE building spread.kde.org have a look at Building SpreadKDE
*****
We'd like to invite you to our new website http://spread.kde.org where you'll find all the information you need. As of today, the main feature we implemented is a managing system for people who are willing to organize events on the behalf of KDE. If you have any question about spread.kde.org, feel free to ask them here on kde-promo@kde.org or on #spreadkde on Freenode.
We strongly suggest you that if you're willing to organize an event, to make it so through the spread.kde.org website as it will help us make the website more usable for you all.
*****
As always help is welcome so if you want to help KDE building spread.kde.org have a look at Building SpreadKDE
Saturday, June 16, 2007
KTuberling themers wanted
If you can draw and always wanted to help KDE but did not know how, this is your opportunity!
As you may know KTuberling got a nice SVG engine a month ago. The problem is that i am not able to draw, and even if the KDE world has some nice artists, they are overloaded with work, so more people on board never hurt.
Contributing themes for KTuberling is VERY easy. All you have to do is draw a nice SVG image.
As example, here are the two themes that we have currently
potato-game.svg and default_theme.svg, the themes.HOWTO file and the two .theme files for the SVG files potato-game.theme and default_theme.theme
If you have any code question do not hesitate to contact me at <aacid _AT_ kde _DOT_ org>
For graphical guidance contact kde games graphics dude at <johann _DOT_ ollivierlapeyre _AT_ gmail _DOT_ com>
KDE fame is waiting for you!
As you may know KTuberling got a nice SVG engine a month ago. The problem is that i am not able to draw, and even if the KDE world has some nice artists, they are overloaded with work, so more people on board never hurt.
Contributing themes for KTuberling is VERY easy. All you have to do is draw a nice SVG image.
As example, here are the two themes that we have currently
potato-game.svg and default_theme.svg, the themes.HOWTO file and the two .theme files for the SVG files potato-game.theme and default_theme.theme
If you have any code question do not hesitate to contact me at <aacid _AT_ kde _DOT_ org>
For graphical guidance contact kde games graphics dude at <johann _DOT_ ollivierlapeyre _AT_ gmail _DOT_ com>
KDE fame is waiting for you!
Sunday, June 10, 2007
New Strigi blog
Due to problems with kdedevelopers, Jos has opened a new blog at http://strigi.sourceforge.net/?q=blog/2. While clee adds him to the planet, see his latest post about strigi, Xesam and akademy 2007 at http://strigi.sourceforge.net/?q=node/32.
More news tomorrow at the same time!
More news tomorrow at the same time!
Wednesday, June 06, 2007
TaxiPilot looking for developers
Thomas Friedrichsmeier, TaxiPilot's author is looking for development helpers to port his small but nice game to KDE 4. So if you are looking for a nice introduction to KDE 4 development do not hesitate to contact him!
Monday, June 04, 2007
okular and PS files
I'm having some problems deciding what to do with PS files on okular, seems strange, eh, since PS is so old and unchanging one would think it's a solved problem. At the moment we are using libgs instead of calling the gs binary to render the files but there are some things i am not sure, here is the list of pros and cons of using libgs.
Pro:
* Using a library seems the right thing to my programmer mind
* We don't have to interact with gs using non-portable X properties like kghostview does
* We don't have to write the images to disk and then load them like ligature does
Con:
* Everyone (kghostview/evince/ligature) uses the gs binary
* libgs is not much widespread among distro packages, although that is improving and okular using it would probably "fix it"
* if libgs crashes it takes down the whole process
* libgs API is not what one would like it to be
So anyone out there that has the magic solution?
Of course, testers and constructive comments are more than welcome for the ghostscript okular backend, actually for all okular parts :-)
Pro:
* Using a library seems the right thing to my programmer mind
* We don't have to interact with gs using non-portable X properties like kghostview does
* We don't have to write the images to disk and then load them like ligature does
Con:
* Everyone (kghostview/evince/ligature) uses the gs binary
* libgs is not much widespread among distro packages, although that is improving and okular using it would probably "fix it"
* if libgs crashes it takes down the whole process
* libgs API is not what one would like it to be
So anyone out there that has the magic solution?
Of course, testers and constructive comments are more than welcome for the ghostscript okular backend, actually for all okular parts :-)
Etiquetes de comentaris:
ghostscript,
gs,
libgs,
okular,
ps
Wednesday, May 30, 2007
poppler 0.6 Release Candidate 1 released
Today i released poppler 0.5.9 aka 0.6 Release Candidate 1, this is the first time i do a release on the poppler project, so hope i did it right :-D
Changes include:
- Merge xpdf 3.02 changes
- Qt4 frontend is not marked anymore as unstable
- Support for Sound objects
- Support for Opening/Closing page actions
- Support for page duration
- Improve PS Tokenizer performance thanks to Scott Turner
- Various speed ups by Krzysztof Kowalczyk
- Beginning of Interactive Form support by Julien Rebetez
- xpdfrc is no longer used for anything
- Add AbiWord output device and pdftoabw program by Jauco Noordzij
- Fix security issue MOAB-06-01-2007
- Lots of bugs fixed
Testing, patches and bug reports welcome.
We are targetting 27th June for 0.6 release, but that may slip depending on feedback from this release.
Changes include:
- Merge xpdf 3.02 changes
- Qt4 frontend is not marked anymore as unstable
- Support for Sound objects
- Support for Opening/Closing page actions
- Support for page duration
- Improve PS Tokenizer performance thanks to Scott Turner
- Various speed ups by Krzysztof Kowalczyk
- Beginning of Interactive Form support by Julien Rebetez
- xpdfrc is no longer used for anything
- Add AbiWord output device and pdftoabw program by Jauco Noordzij
- Fix security issue MOAB-06-01-2007
- Lots of bugs fixed
Testing, patches and bug reports welcome.
We are targetting 27th June for 0.6 release, but that may slip depending on feedback from this release.
sjfonts 2.0.2 released
Today i released sjfonts 2.0.2, only change is the addition of ĞğŞşİı characters to SteveHand font. Thanks goes to Serdar Soytetir!
Sunday, May 27, 2007
Where's my ballot paper?
I'm upset, VERY upset. I just came back from the electoral college for Spanish 2007 municipal elections and i had to spend at least 5 minutes in front of all the stacks of ballot papers looking for the one i wanted.
And the problem was not that there were much possibilities to vote (that would have been a GOOD THING). The problem is that someone had put other party ballot papers on the top of the stack of the party i wanted to vote, so i had to actually realize there was two stacks with the same ballot papers and then start removing ballot papers from the top of both until i actually found the one i wanted to get.
I feel shame of living in the same neighborhood than someone so dull and simple that wants to fuck that small party i vote and doesn't have a better idea than putting an even small party ballots on top of it.
And then i went to complain to municipal representative and he said something that can be summarized as "Shit happens", Very helpful, eh? :-/ He could have said "I'll double check next time", even if it had been false it would have made me feel better.
And the problem was not that there were much possibilities to vote (that would have been a GOOD THING). The problem is that someone had put other party ballot papers on the top of the stack of the party i wanted to vote, so i had to actually realize there was two stacks with the same ballot papers and then start removing ballot papers from the top of both until i actually found the one i wanted to get.
I feel shame of living in the same neighborhood than someone so dull and simple that wants to fuck that small party i vote and doesn't have a better idea than putting an even small party ballots on top of it.
And then i went to complain to municipal representative and he said something that can be summarized as "Shit happens", Very helpful, eh? :-/ He could have said "I'll double check next time", even if it had been false it would have made me feel better.
Etiquetes de comentaris:
ballot paper,
municipal elections
Monday, May 21, 2007
aKademy 2007 GPG Signing Party
Anne Wilson has kindly offered to coordinate the key signing party for aKademy 2007. If you are going to be on Glasgow go to http://akademy2007.kde.org/conference/keysigning.php, follow the instructions and you will get your key signed by fellow KDE people :-)
Sunday, May 20, 2007
Car's working again
Yesterday the car was not working, but it was due to the battery was empty, a dude from the RACC came and started the car with his battery. So my father and I went around to do some kilometers and now everything is working again :-)
Saturday, May 19, 2007
Bring on good old mechanical things!
That may seem a strange sentence coming from a computer science person like me, but at this moment i'm hating electronics.
Today i lost a lunch with some friends from one of the local LUGs because the Renault Megane we bought second hand two weeks ago decided to act weird. This car is equipped with a proximity sensor (maybe RFID?) that comes in a card, so to open it you just have to go near the car and open it. And to start it you just have to press the brake pedal and push the "Start" button.
My father did this at 6 am and came back home at 6:30 am to get the keys of the old car because he could not get it started.
This morning one our before the lunch time i did the same, went into the car, pressed the brake pedal and pushed the Start button, and a "Electronics Problem" message came up on the status panel. I still had time so i could have went lunching with the metro, so i got off the car and pressed the close button on the card to close the car.
No reaction.
So i pull off a key that is inserted on the card and use it to close driver's door, but doing that still leaves co-pilot door and back door open.
ARG!
So i call try to call previous owner of the car, my uncle, and he says he never had this problem.
Then i go home, pick up the car's manual and see it says there's a mechanical lock on doors you can use to lock them, good! Back onto the car i discover there's no suck lock on my car :-@
I go home again, look into the internet a Renault number, phone them and i explain them the problem. The dude says that i should get onto the car, press the "lock doors button" and then go off it and close driver's door with the key. Back onto the car i try that, doesn't work either, opening driver's door unlocks all other doors.
At this point my mother and I were already pretty desperate. At that moment i remembered a friend whose sister had a Megane too. I phoned her and luckily she was with her sister. And bam! she had the solution, it seems the car thought it was powered on, so all i had to do was put the card onto the slot that there's on the car, push the "Start" button twice and then a "Please remove the card" message appeared, so the car thought it was off, we got off the car and could even close all the doors using the card buttons.
We still did not try to get the engine started again, that will have to wait for tomorrow, but at least we could close the doors! Unfortunately more than hour had passed with all that and i could not make to the lunch :-(
So next time you design something cool with electronics, please think of the failure scenario and that a mechanical way to fix it is always welcome!
Today i lost a lunch with some friends from one of the local LUGs because the Renault Megane we bought second hand two weeks ago decided to act weird. This car is equipped with a proximity sensor (maybe RFID?) that comes in a card, so to open it you just have to go near the car and open it. And to start it you just have to press the brake pedal and push the "Start" button.
My father did this at 6 am and came back home at 6:30 am to get the keys of the old car because he could not get it started.
This morning one our before the lunch time i did the same, went into the car, pressed the brake pedal and pushed the Start button, and a "Electronics Problem" message came up on the status panel. I still had time so i could have went lunching with the metro, so i got off the car and pressed the close button on the card to close the car.
No reaction.
So i pull off a key that is inserted on the card and use it to close driver's door, but doing that still leaves co-pilot door and back door open.
ARG!
So i call try to call previous owner of the car, my uncle, and he says he never had this problem.
Then i go home, pick up the car's manual and see it says there's a mechanical lock on doors you can use to lock them, good! Back onto the car i discover there's no suck lock on my car :-@
I go home again, look into the internet a Renault number, phone them and i explain them the problem. The dude says that i should get onto the car, press the "lock doors button" and then go off it and close driver's door with the key. Back onto the car i try that, doesn't work either, opening driver's door unlocks all other doors.
At this point my mother and I were already pretty desperate. At that moment i remembered a friend whose sister had a Megane too. I phoned her and luckily she was with her sister. And bam! she had the solution, it seems the car thought it was powered on, so all i had to do was put the card onto the slot that there's on the car, push the "Start" button twice and then a "Please remove the card" message appeared, so the car thought it was off, we got off the car and could even close all the doors using the card buttons.
We still did not try to get the engine started again, that will have to wait for tomorrow, but at least we could close the doors! Unfortunately more than hour had passed with all that and i could not make to the lunch :-(
So next time you design something cool with electronics, please think of the failure scenario and that a mechanical way to fix it is always welcome!
Etiquetes de comentaris:
electronics,
mechanics,
renault
Tuesday, May 08, 2007
KTuberling SVG
You may have read the list of games not making the cut for KDE 4.0 at Johann's blog. KTuberling was in the position of "We all love (or have relatives that love it) it but nobody is working on it".
So i got up my mind and started coding the SVG port, in two days i already had it working, with exception of text, everyone that has tried to draw text in Qt, knows is not so easy as it might seem, but i hope to get it working in less than a week.
Mandatory screenshot:

I have not made any optimization attempt and it works quite snappy on my machine, but it is a fast machine so if anyone with a slower machine could test it and give me his opinion i'd be happy.
Of course any comment is welcome (except "You should be working on something more important")
So i got up my mind and started coding the SVG port, in two days i already had it working, with exception of text, everyone that has tried to draw text in Qt, knows is not so easy as it might seem, but i hope to get it working in less than a week.
Mandatory screenshot:
I have not made any optimization attempt and it works quite snappy on my machine, but it is a fast machine so if anyone with a slower machine could test it and give me his opinion i'd be happy.
Of course any comment is welcome (except "You should be working on something more important")
Wednesday, May 02, 2007
Started aKademy Arrival and Departure pages at wiki
So i started Arrival at aKademy 2007 and Departure from aKademy 2007 when you can voluntarily write yourself to try to find partners to make the trip from and to the hostel more KDE oriented :-)
See you in Glasgow :-)
Update: Seems that was already started on techbase, wonder why someone decided TECH base is a good place for that, but oh well, please go to http://techbase.kde.org/Events/Akademy2007/Attendees if you want to share your information.
See you in Glasgow :-)
Update: Seems that was already started on techbase, wonder why someone decided TECH base is a good place for that, but oh well, please go to http://techbase.kde.org/Events/Akademy2007/Attendees if you want to share your information.
Sunday, April 29, 2007
kde-i18n-ca: Extend and embrace
So for KDE 4 we switched from using KDE's own mimetype database to shared-mime-info from freedesktop (thanks David!) this has lots of advantages i don't need to explain, the only obvious problem i could find was the lack of translation to Catalan, but that can easily fixed when you are a translator yourself and you know how to trick others to finish your work, so thanks to Josep Ma. Ferrer and me, shared-mime-info has now a translation to Catalan :-) And the good thing of sharing things is that gnomies, xfceis, roxies and all others using shared-mime-info will benefit from our translation :-)
Thursday, April 26, 2007
Cannot find konq-kubuntu.rc problem and fix
On one of the five machines i've upgraded from Edgy to Feisty, i got a weird error with Konqueror. When starting it, i got a very weird Konqueror without Address line for example and two symptomatic errors
konqueror: WARNING: KXMLGUIClient::setXMLFile: cannot find .rc file
konq-kubuntu.rc
konqueror: WARNING: KXMLGUIClient::setXMLFile: cannot find .rc file
konq-kubuntu.rc
For us that know some KDE programming that means Konqueror was trying to find the konq-kubuntu.rc
file to create it's gui but failed to find it.
Quick solution was looking for some file that referenced konq-kubuntu.rc I found it at .kde/share/apps/konqueror/profiles/webbrowsing
As i had another working machines it was easy to see that XMLUIFile=konq-kubuntu.rc should be XMLUIFile=konqueror.rc and now it is working like a charm :-)
konqueror: WARNING: KXMLGUIClient::setXMLFile: cannot find .rc file
konq-kubuntu.rc
konqueror: WARNING: KXMLGUIClient::setXMLFile: cannot find .rc file
konq-kubuntu.rc
For us that know some KDE programming that means Konqueror was trying to find the konq-kubuntu.rc
file to create it's gui but failed to find it.
Quick solution was looking for some file that referenced konq-kubuntu.rc I found it at .kde/share/apps/konqueror/profiles/webbrowsing
As i had another working machines it was easy to see that XMLUIFile=konq-kubuntu.rc should be XMLUIFile=konqueror.rc and now it is working like a charm :-)
Etiquetes de comentaris:
konq-kubuntu.rc,
konqueror,
kubuntu
Wednesday, April 25, 2007
aKademy registration deadline #1
Remember, if you want the aKademy organization team to handle accommodation booking for you, you should register and PAY before April 30th.
Of course if you can handle booking accommodation for yourself you are allowed to register later :D
Of course if you can handle booking accommodation for yourself you are allowed to register later :D
Sunday, April 22, 2007
Back from Albacete
I'm back from the nice conference organized by LinuxAlbacete people. I reused the talk Aaron gave at Guademy about KDE 4, and people seemed to like it quite a bit, i'm not as good as Aaron giving speeches but i know how to do jokes in spanish and Aaron not ;-)
It has been a pleasure beign there and eating some mancheguitos with you :-) Good luck for next year and keep up the good work!
Saturday, April 14, 2007
kdemail.net is alive again
I noticed it because i started getting spam on the address again *sight*
Short note to those who were subscribed to mailing lists using kdemail service: Check you are still subscribed, but as administrator of the kde-edu mailing list i remember seeing some messages saying "foobar@kdemail.net has been automatically unsubscribed due too high number of bounces"
Update: Seems only smtp and mail aliases work for now. As imap and pop3 connections are refused. :-/
Short note to those who were subscribed to mailing lists using kdemail service: Check you are still subscribed, but as administrator of the kde-edu mailing list i remember seeing some messages saying "foobar@kdemail.net has been automatically unsubscribed due too high number of bounces"
Update: Seems only smtp and mail aliases work for now. As imap and pop3 connections are refused. :-/
Thursday, April 12, 2007
KDE SoC Projects
Bigger annoucement later at The Dot, but for the anxious http://code.google.com/soc/kde/about.html
Monday, April 09, 2007
Printing hack in KPDF 0.5.7
KPDF is usually good enough at printing documents, but sometimes it fails completely.
And there's nothing worse that having a document you need to print and not beign able to do it. To fix that situation i just added a small hack to KPDF version that will be released with KDE 3.5.7.
In that version there will be a PDF Options tab with a Force rasterization checkmark that will let you force KPDF to render each PDF page to an image and print the image instead convert PDF commands into PS commands. This is bad as you lose the vector side of having a PS, but in cases KPDF did show the document correctly, but printing failed, i think it will be a welcome addition.
And there's nothing worse that having a document you need to print and not beign able to do it. To fix that situation i just added a small hack to KPDF version that will be released with KDE 3.5.7.
In that version there will be a PDF Options tab with a Force rasterization checkmark that will let you force KPDF to render each PDF page to an image and print the image instead convert PDF commands into PS commands. This is bad as you lose the vector side of having a PS, but in cases KPDF did show the document correctly, but printing failed, i think it will be a welcome addition.
Tuesday, April 03, 2007
HPLIP team rocks
So i had a problem printing a PS file. First i did was contact our resident KDE printing guru, Kurt Pfeifle.
He ensured me the file seemed correct as three different PostScript printers printed the file correctly so it might be a problem in the HPLIP driver of my HP PSC 1610
Today i contacted the hplip-help mailing list:
16:37:23 - my mail reaches the list
17:39:34 - an HP engineer contacts me so so i send him the file in private (as it contains semi-sensitive data)
18:17:04 - i send the file
19:12:43 - confirmation that they can reproduce the problem (yay! for bug reproductivity)
23:30:38 - i get a patch for latest HPLIP release making my PS file print correctly
Now try to do this on windows...
He ensured me the file seemed correct as three different PostScript printers printed the file correctly so it might be a problem in the HPLIP driver of my HP PSC 1610
Today i contacted the hplip-help mailing list:
16:37:23 - my mail reaches the list
17:39:34 - an HP engineer contacts me so so i send him the file in private (as it contains semi-sensitive data)
18:17:04 - i send the file
19:12:43 - confirmation that they can reproduce the problem (yay! for bug reproductivity)
23:30:38 - i get a patch for latest HPLIP release making my PS file print correctly
Now try to do this on windows...
Subscribe to:
Posts (Atom)