Gaim-svn 18019-1 bug on avatar resize
I'm getting a 100% cpu load and a crash when trying to open a conversation window with gaim-svn 18019-1.
I check with the project page, and it's is already documented.
I'll just leave the heads-up: If you have a gaim-svn that works fine, don't upgrade to this version.
my mistake about gaim-svn in aur, someone might have posted the pkgbuild in the forums at some point. Maybe the name at the top will give it away
If you want the latest version, grab this PKGBUILD (or the other posted) and install versionpkg - then run versionpkg in the same dir as this PKGBUILD. That should build you the current version
#Contributor: SaintShakaJin <saintshakajin>
pkgname=gaim-svn
pkgver=18081
pkgrel=1
pkgdesc="A GTK+-based messaging client"
url="http://gaim.sourceforge.net"
depends=('startup-notification' 'libao' 'gnutls>=1.2.0' 'gtk2' 'gtkspell'
'audiofile' 'perl' 'python' 'libxss' 'dbus' 'ncurses' 'nss' 'aspell')
provides=('gaim')
conflicts=('gaim' 'gaim-cvs' 'gaim-devel')
makedepends=('subversion' 'intltool' 'pkgconfig')
source=()
md5sums=()
_svnmod=gaim
_svntrunk=https://svn.sourceforge.net/svnroot/gaim/trunk
build() {
cd $startdir/src
msg "Getting source..."
svn co $_svntrunk $_svnmod -r $pkgver
cd $_svnmod
msg "Starting make..."
./autogen.sh --prefix=/usr
--enable-aspell --enable-gnutls=yes --enable-nss=yes
--enable-dbus --with-x
--disable-mono --disable-gstreamer
--disable-gevolution
--disable-tcl --disable-tk
--disable-doxygen
make || return 1
make prefix=$startdir/pkg/usr install || return 1
rm $startdir/pkg/usr/lib/perl5/*/$CARCH-linux-thread-multi/perllocal.pod
mv $startdir/pkg/usr/lib/perl5/?.?.? $startdir/pkg/usr/lib/perl5/current
find $startdir/pkg -name '*.la' -exec rm {} ;
Similar Messages
-
the recent version of gaim-svn-17980-1 is unstable. as soon as it logs in in yahoo (i have saved the password, so when i start gaim, it signs in automatically), it crashes. This is not the case with gaim-svn-17840-1 though. I had to go back to that. Did anyone else had the same experience or any solution?
my mistake about gaim-svn in aur, someone might have posted the pkgbuild in the forums at some point. Maybe the name at the top will give it away
If you want the latest version, grab this PKGBUILD (or the other posted) and install versionpkg - then run versionpkg in the same dir as this PKGBUILD. That should build you the current version
#Contributor: SaintShakaJin <saintshakajin>
pkgname=gaim-svn
pkgver=18081
pkgrel=1
pkgdesc="A GTK+-based messaging client"
url="http://gaim.sourceforge.net"
depends=('startup-notification' 'libao' 'gnutls>=1.2.0' 'gtk2' 'gtkspell'
'audiofile' 'perl' 'python' 'libxss' 'dbus' 'ncurses' 'nss' 'aspell')
provides=('gaim')
conflicts=('gaim' 'gaim-cvs' 'gaim-devel')
makedepends=('subversion' 'intltool' 'pkgconfig')
source=()
md5sums=()
_svnmod=gaim
_svntrunk=https://svn.sourceforge.net/svnroot/gaim/trunk
build() {
cd $startdir/src
msg "Getting source..."
svn co $_svntrunk $_svnmod -r $pkgver
cd $_svnmod
msg "Starting make..."
./autogen.sh --prefix=/usr
--enable-aspell --enable-gnutls=yes --enable-nss=yes
--enable-dbus --with-x
--disable-mono --disable-gstreamer
--disable-gevolution
--disable-tcl --disable-tk
--disable-doxygen
make || return 1
make prefix=$startdir/pkg/usr install || return 1
rm $startdir/pkg/usr/lib/perl5/*/$CARCH-linux-thread-multi/perllocal.pod
mv $startdir/pkg/usr/lib/perl5/?.?.? $startdir/pkg/usr/lib/perl5/current
find $startdir/pkg -name '*.la' -exec rm {} ; -
I reported a bug today I feel could do some nasty thing to document when you use resize() to re-size a layer. Particularly if you have used suspend history. It looks like resize for some unknown reason backs the document up one history state before doing the re-size. I have tested this in CS5, CS6, CC and CC 2014 all fail the same way. Nasty Nasty...
I can program around that bug if history states are being recorded for the script. However I like to suspend history state recording in scripts so that its easy to back up in history after the script is used if need be.
However that bug can destroy the document and also remove prior history states in so that its impossible to recover the document from history for prior state has been replaced with the recording of the use of the script. How many states are removed would depend on how many .resize(); the script did. Nasty Nasty.
Here are some screen of the bug in action. The fist before and after look the way the script is designed to work. However I did an extra select all before using the script. You see its no longer in history in the after.
The second before and after same as first without the extra select all note a guide line has been removed.
In the third before and after I changer the order of my of steps before the script I first drag out the guides then place in the image. You see that the image has been removed the wrong layer has been resized and the place removed from history. Nasty Nasty.
The script works when history is not suspended for I programmed the script to work around the bug... http://www.mouseprints.net/old/dpr/FitImageToGuides.jsxYes I tested CS5 its there. I never install CS4 and lost my CS3 because of disk drive crashes. Did not test CS2 for CS2 selection bounds does not work and I would need to add my work around for that bug.
I only have two machine on my I7 Windows 7 system I have perpetual CS5 and CS6 installed and on my workstation also Windows 7 I've installed my subscription versions along with CS2. There are differences between CS6 13.0.1.3 and subscription version 13.1.2. Adobe seems to have given up on being backwards compatible starting with CS4. Even Bill Gate's realized Bugs were bad for business has them exterminated Adobe does not seem to accept that bug are bad for business. So now I have CS2, CS5, CS6, CS6 CC and CC 2014. I use file system links to add plug-in and script so all installed versions us the single copy of the add-ons I install. -
[svn] 2599: FXG Bug Fix
Revision: 2599
Author: [email protected]
Date: 2008-07-23 15:17:33 -0700 (Wed, 23 Jul 2008)
Log Message:
FXG Bug Fix
MXMLG-116 only edge pixels expand when resize this bitmapGraphic
In BitmapGraphic we constrain the dimensions of the rectangle that hold the bitmap fill to be the minimum of either the image's dimensions or the drawing dimensions. This occurs only when repeat is false.
Bug: MXMLG-116
QA: Yes
Doc: Yes
Review: Ryan
Ticket Links:
http://bugs.adobe.com/jira/browse/MXMLG-116
http://bugs.adobe.com/jira/browse/MXMLG-116
Modified Paths:
flex/sdk/trunk/frameworks/projects/flex4/src/flex/graphics/BitmapGraphic.as -
[svn:fx-trunk] 11319: Allow for resizing of sub-apps after they initialize
Revision: 11319
Author: [email protected]
Date: 2009-10-30 11:42:01 -0700 (Fri, 30 Oct 2009)
Log Message:
Allow for resizing of sub-apps after they initialize
QE Notes: Two Mustella Tests need updating:
MP_Controls_Tests2:MP_ComboBox_closeOnMouseDown_untrusted
MP_Controls_Tests2:MP_ComboBox_closeOnMouseDownOutside_untrusted
The new sub-app resize closes the dropdown. Probably have to wait a bit longer before opening the dropdown
Doc Notes: New API in ISystemManager
Bugs: SDK-23892
Reviewer: Darrell
API Change: yes
Is noteworthy for integration: No
tests: checkintests mustella/MarshallPlan
Ticket Links:
http://bugs.adobe.com/jira/browse/SDK-23892
Modified Paths:
flex/sdk/trunk/frameworks/projects/airframework/src/mx/managers/WindowedSystemManager.as
flex/sdk/trunk/frameworks/projects/framework/src/mx/core/UIComponent.as
flex/sdk/trunk/frameworks/projects/framework/src/mx/managers/ISystemManager.as
flex/sdk/trunk/frameworks/projects/framework/src/mx/managers/SystemManager.as
flex/sdk/trunk/frameworks/projects/framework/src/mx/managers/systemClasses/MarshallingSup port.asAfter commenting out nearly everything in the above autostart.sh, I also found that using . $GLOBALAUTOSTART will also cause the same issue. Once I commented that out, SLiM logged me into Openbox just fine.
Thanks alexandrite. -
[svn] 3149: Fix bugs introduced by the display object sharing optimization.
Revision: 3149
Author: [email protected]
Date: 2008-09-08 16:58:35 -0700 (Mon, 08 Sep 2008)
Log Message:
Fix bugs introduced by the display object sharing optimization. This checkin should fix ordering problems when dynamically adding/removing graphic elements, and ordering problems when changing graphic element properties that require a display object (rotation, alpha, filters, etc).
Bugs:
MXMLG-219: BitmapGraphic component content always rendered from 0, 0 origin.
MXMLG-220: setting visible property of Rect and Ellipse does nothing.
MXMLG-222: setting the alpha of a rectangle changes z-order when there are 3 or more rectangles
SDK-16754: Z-order incorrect when a Group is present
QA: Yes - the test files from these bugs (or something equivalent) should be added to our standard test suite
Reviewer: Deepa
Ticket Links:
http://bugs.adobe.com/jira/browse/MXMLG-219
http://bugs.adobe.com/jira/browse/MXMLG-220
http://bugs.adobe.com/jira/browse/MXMLG-222
http://bugs.adobe.com/jira/browse/SDK-16754
Modified Paths:
flex/sdk/trunk/frameworks/projects/flex4/src/flex/core/Group.as
flex/sdk/trunk/frameworks/projects/flex4/src/flex/graphics/BitmapGraphic.as
flex/sdk/trunk/frameworks/projects/flex4/src/flex/graphics/graphicsClasses/GraphicElement .asThis is the great thread ;-)
My favorite threads are below.
Munky posted great reply :8}
SQL Query Help
order in a connect by -
[svn] 2810: FXG Bug Fix
Revision: 2810
Author: [email protected]
Date: 2008-08-11 17:48:51 -0700 (Mon, 11 Aug 2008)
Log Message:
FXG Bug Fix
MXMLG-210 - [Gumbo_Priority] Unable to create second item after masked item
Masks were broken for GraphicElements after the lifecycle consolidation work. Added hacks to Group to get the masks working again. The graphic element mask code and the commitProperties logic will eventually change, removing the need for this hack.
Bug: MXMLG-210
QA: Yes
Doc: No
Review: Evtim
Ticket Links:
http://bugs.adobe.com/jira/browse/MXMLG-210
http://bugs.adobe.com/jira/browse/MXMLG-210
Modified Paths:
flex/sdk/trunk/frameworks/projects/flex4/src/flex/core/Group.as -
[svn] 2844: FXG Bug Fix
Revision: 2844
Author: [email protected]
Date: 2008-08-14 11:16:44 -0700 (Thu, 14 Aug 2008)
Log Message:
FXG Bug Fix
SDK-16467 NPE from BitmapGraphic.updateDisplayList if repeat=false but source is not yet defined
Added null check for source in updateDisplayList
Bug: SDK-16467
QA: Yes
Doc: No
Review: Deepa
Ticket Links:
http://bugs.adobe.com/jira/browse/SDK-16467
http://bugs.adobe.com/jira/browse/SDK-16467
Modified Paths:
flex/sdk/trunk/frameworks/projects/flex4/src/flex/graphics/BitmapGraphic.as -
[svn] 1571: Fix bugs ASC-3310, ASC-3337
Revision: 1571
Author: [email protected]
Date: 2008-05-06 10:58:32 -0700 (Tue, 06 May 2008)
Log Message:
Fix bugs ASC-3310, ASC-3337
ASC-3337:
ASC was emitting multinames for [] access with the wrong namespace set. It was using the def_namespaces, which is a list of open namespaces that conflicting definitions could be found in (public, private, protected, internal for a class). This set does not include any other open namespaces. This changes asc to use the set of all open namespaces, which includes namespaces open by default, and explicitly by the user.
ASC-3310:
LintEvaluator was returning ObjectValue.undefinedValue sometimes from a method that was supposed to be returning a TypeValue. This caused a ClaassCastException later as the return value was cast to a TypeValue. Change to return noType (*) instead of the undefinedValue.
smokes: asc tests, tamarin tests, flex checkintests
Ticket Links:
http://bugs.adobe.com/jira/browse/ASC-3310
http://bugs.adobe.com/jira/browse/ASC-3337
http://bugs.adobe.com/jira/browse/ASC-3337
http://bugs.adobe.com/jira/browse/ASC-3310
Modified Paths:
flex/sdk/trunk/modules/asc/src/java/macromedia/asc/embedding/LintEvaluator.java
flex/sdk/trunk/modules/asc/src/java/macromedia/asc/semantics/CodeGenerator.java -
[svn] 3791: Fix bug when resetting stylesheet on FxApplication
Revision: 3791
Author: [email protected]
Date: 2008-10-21 13:23:09 -0700 (Tue, 21 Oct 2008)
Log Message:
Fix bug when resetting stylesheet on FxApplication
SDK-17691 RTE loading styles
When the FxApplication replaces its style sheet, it unloads the previous skin and loads in a new one. With the new skin, it creates a new instance of the contentGroup. However, when the content is added to the contentGroup, it has already been parented by the previous version of the contentGroup. This causes the RTE because the content is not in the correct state.
The solution is to set the old contentGroup's content property to null when we remove the skin. We also force the contentGroup to validate immediately because the new contentGroup was processing its content array before the oldGroup had removed the content as children.
QE Notes: Needs testing
Doc Notes: none
Bugs: SDK-17691
Reviewer: Glenn
Ticket Links:
http://bugs.adobe.com/jira/browse/SDK-17691
http://bugs.adobe.com/jira/browse/SDK-17691
Modified Paths:
flex/sdk/trunk/frameworks/projects/flex4/src/mx/components/FxContainer.asThe password history check is not honored in a password reset. It always been like that and it is documented:
"Password History Length Constraint: If all of the following conditions are true, the following constraints MUST be satisfied:
The requesting protocol message is a password change (as compared to a password set)."http://msdn.microsoft.com/en-us/library/cc245669.aspx
So changing it is not a correcting a bug but asking to change the design. From a pure support perspective, you switched from bug request to design change request. Which might still be arguable if you have a solid business case.
Password set is given through an extended right, so by default only administrators or operators have this right. So it comes down to trust those individual. If you cannot, then you don't give them the right and you implement other identity
management solution (Password Self Reset for example).
Other fancy and not useful information:
http://msdn.microsoft.com/en-us/library/cc245692.aspx
Note: Posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose. -
Revision: 4793
Author: [email protected]
Date: 2009-02-02 11:20:06 -0800 (Mon, 02 Feb 2009)
Log Message:
Fix bug SDK-17734 Path with width and height set different than path data has incorrect bounds
Fix: When calculating the bounds position we should take into account the implicit scaling factor actualSize/naturalSize. Also did some refactoring, getting rid of the protected method calculateTopLeft.
QE Notes: None
Doc Notes: None
Bugs: SDK-17734
Reviewer: Ryan
tests: mustella gumbo/layout/GraphicElement
Ticket Links:
http://bugs.adobe.com/jira/browse/SDK-17734
http://bugs.adobe.com/jira/browse/SDK-17734
Modified Paths:
flex/sdk/trunk/frameworks/projects/flex4/src/mx/graphics/Ellipse.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/graphics/Path.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/graphics/graphicsClasses/GraphicElement.a sThank you so much for replying.
Yes I have removed and reinstalled WMP.
I had good results with the PD6 application installed on the default path onto the C: drive with the one exception that if the application was launched by accident and the user data path was not available, the PD6 application would blow away my custom user path registry settings. Now that I know what they are I have made a .reg file to repair my registry to my desired user data paths.
Installing the application on the removable drive appeared to help prevent me from launching the application by accident and overwriting my registry with default user paths.
So which is the less of the two evils?
If the application directory is not available, windows media player still tries to launch the .msi for installing PD6.
If I install the application to the C: drive but the user data to the removable drive, launching the PD6 application without the user data drive will still corrupt my registry settings for a user data path.
Both these issues seem like a logical (if not easy) fix that should be done in the PD6 application and installation package. I mean really, cannot anyone tell me why windows media player is checking the PD6 application directory? Why in PD4 did we have an option control for setting the user data path from the PD4 application? Why is this option not in the PD6 application, just the installer?
I am given a choice during installation to move the user data to another non default location. Why else would this be provided if not to accommodate my kind of request to store the user data into an alternate location other than “My Document”. Certainly Palm is not trying to force the users on how to protect and store their personal data?
Post relates to: Centro (Verizon) -
Revision: 2841
Author: [email protected]
Date: 2008-08-14 04:45:09 -0700 (Thu, 14 Aug 2008)
Log Message:
Fix bug loading untrusted applications that have not been sized.
QE:YES
Doc:
Checkintests: YES
Reviewer: Alex
Bugs: SDK-16470
SWFLoader:
Fixed a security violation trying to access untrusted content. The problem happened in doScaleContent() when the loaded content had not been sized yet. The fall back is to check the class type of the content, but the SWFLoader did not have access to the content. The security violation was caught and the FlexLoader size is set when it shouldn't be. The problem is solved by checking access rights before accessing the content and not generating an exception.
Ticket Links:
http://bugs.adobe.com/jira/browse/SDK-16470
Modified Paths:
flex/sdk/branches/3.0.x/frameworks/projects/framework/src/mx/controls/SWFLoader.asIgnoring what is causing the errors for a moment, what would help is a way to get the application loader to restart from the point of error rather than going back to the beginning of the "Loading applications" step each time.
I imagine this to be something of a big ask. Any other suggestions in a similar vein?
Nick. -
[svn] 4803: Fix bug SDK-18434 Bounds for rotated rounded rect is incorrect
Revision: 4803
Author: [email protected]
Date: 2009-02-02 16:06:34 -0800 (Mon, 02 Feb 2009)
Log Message:
Fix bug SDK-18434 Bounds for rotated rounded rect is incorrect
Fix: Calculate the bounds of the transformed corner ellipses in the case of matrix transformation. Refactor the Ellipse's helper method and move it to MatrixUtil.as so that it's shared between classes.
QE Notes: None
Doc Notes: None
Bugs: SDK-18434
Reviewer: Ryan
tests: checkintests, mustella gumbo/layout/graphicelement
Ticket Links:
http://bugs.adobe.com/jira/browse/SDK-18434
http://bugs.adobe.com/jira/browse/SDK-18434
Modified Paths:
flex/sdk/trunk/frameworks/projects/flex4/src/mx/graphics/Ellipse.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/graphics/Rect.as
flex/sdk/trunk/frameworks/projects/framework/src/mx/utils/MatrixUtil.as -
*-svn AUR mega bug.
SourceForge ([url]http://sourceforge.net/docs/A04[/url]) wrote:
Project Subversion (SVN) Service: Online [1] Last updated: 2007-01-05 Pacific
Recent Updates »
( 2007-07-09 10:43:54 - Project Subversion (SVN) Service ) As announced, support for the deprecated subversion access method (svn.sourceforge.net) was removed. Please use the PROJECT.svn.sourceforge.net access method that is described in our docs.
This is affecting A LOT of *-svn packages from AUR. They are fetching their SVN sources from https://svn.sourceforge.net and THIS ISN'T WORKING ANYMORE, because they must now specify the project name BEFORE "svn" (https://[projectname].svn.sourceforge.net).
Can any TU fix all of those packages?
Thanks in advance.
Last edited by Alejandro Nova (2007-07-11 04:26:58)you could try alerting the maintainers/devs of such scripts as yaourt/pacbuild/etc.
alternatively
function sfs() {
pattern="://svn.\(sourceforge\|sf\).net/"
if [[ "${_svntrunk}x" != "x" && `echo $_svntrunk | grep "$pattern"` ]];
then _svntrunk=`echo $_svntrunk | sed "s@$pattern@://${_svnmod}.sourceforge.net/@g"`
fi
alias build="sfs;build;"
in /etc/makepkg.conf should be a sufficient workaround while the pkgbuild maintainers catch-up .
unfortnately i am nowhere near an AL system so i can't confirm that it works.. but simple test with
_svntrunk=https://sf.net/svn/project/trunk
_svnmod=project
function build() {
echo $_svntrunk
#::how i tested it::
#file1: mkpkg > above function + . pb \n build
#file2: pkgbuild >above code
shows promising results
sorry my bash programming isn't as good as I'd like feel free to rip it apart if you like
hope it works;)
Last edited by raeven (2007-07-12 19:38:00) -
[svn] 1941: Fix bug with writing out a MorphShape2 line style when it has a fill
Revision: 1941
Author: [email protected]
Date: 2008-06-04 12:13:17 -0700 (Wed, 04 Jun 2008)
Log Message:
Fix bug with writing out a MorphShape2 line style when it has a fill
connected to it. Had been calling encodeMorphFillstyles() directly
with an array of one element, but that adds a count byte (of value 1)
which should not be there. So I decomposed new method
encodeMorphFillstyle() out of encodeMorphFillstyles(). So not a big
change, but lots of indentation changes.
Other than that all changes were replacing the use of constant numbers
with static constants from FillStyle and also adding an equals()
implementation for FocalGradient.
reviewed by Pete Farland, passed flex checkin tests
Modified Paths:
flex/sdk/trunk/modules/swfutils/src/java/flash/swf/TagDecoder.java
flex/sdk/trunk/modules/swfutils/src/java/flash/swf/TagEncoder.java
flex/sdk/trunk/modules/swfutils/src/java/flash/swf/types/FocalGradient.javaHello,
We also have this problem ...
We are concluded, (??) When there is a note on the first or second line that starts in the block, justification and rebuilt and the problem happens! even made to change color of text on the blank page occurs ... our solution to an extra line at the beginning of the block without note or make it past the note on the previous page ...
when in a true solution?
Translation from French
Bonjour,
Nous avons aussi ce problème …
Nous somme arrivé à la conclusion, (???) quand il y a une note sur la première ou la deuxième ligne qui commence dans le bloc, justification et refaite et le problème arrive !!! même le faite de changé de couleur sur du texte les page blanche survient … notre solution mettre une ligne supplémentaire en début du bloc sans note ou faire passé le note sur la page précédente …
à quand une vrai solution?
Philippe
Maybe you are looking for
-
Import from digital camera no longer works
iPhoto 6.0.6 I suddenly can't import photos from my Casio digital camera. It worked fine two months ago, with the same computer and the same camera. When I connect the camera to the computer, and turn the camera on, the camera appears on the Desktop,
-
HT201317 Manually adding photos to Photo Stream
I've manually added photos to the upload photo from my windows desktop but they are not appearing on my iPhone's Photo Stream. I think I have everything set up and turned on correctly...
-
I use Photoshop CS6 on Macbook retina, icons and texts on the toolbar are serrated. Can you help me?
-
How do I print on the back of 6x4 photo I have a photo smart 5520
I have 50 6x4 photos for wedding invite.and I need to print times ext on revers.
-
Remove unicode control character
hey, so i've asked this around a few places and recieved no answers, but the arch community seems like a bunch of smart people and i've recently switched over from ubuntu. is there a way to remove input methods and unicode control characters from the